lij:

m ij

II Pf

l!'

{!

■ M\

Mi

i

u

iit) il

1

II

'il ' I I 1 i

!

l'lll

i ' \\

11

11

lift'

Pil

1111

:

I II 11 il

ïlliHilli ! II

lil!'

lift

lil'- llll il' 1111$

il

THE

CORALLINACEAE OF THE SIBOGA-EXPEDITION

Siboga-Expeditie

LXI

THE CORALLINACEAE OF THE SIBOGA-EXPEDITION

BY

A. WEBER-van bosse and M. FOSLIE

Eerbeek Tiondhjem

With XVI plates and 34 textfïgures

-=■1=4 <§>(«!-=-

late E. J. BE1LL

PUBLISHERS AND PRINTERS

LEVDEN — 1904

THE CORALLINACEAE OF THE SIBOGA EXPEDITION

BY

A. WEB ER-VAN BOSSE and M. FOSLIE

Eerbeek. Tvondhjem.

With XVI plates and 34 textfigures.

INTRODUCTION

BY

A. WEB ER— van bosse.

No algae excite such general interest as calcareous algae. They are not only a delight
and at the same time a constant puzzle to botanists, but also to geologists who are compelled
to study them, for they meet them in various formations in layers of enormous thickness, tor
instance in the Scandinavian Baltic-silurian district; near Syracuse, where the renowned Latomiae
are quarries in old Lithothamnion-banks; in the "Leithakalke" near Vienna or in the Dolomites.
Zooloeists ag-ain findins: them in a living state, will fain ask what these brilliant coloured
organisms are, that are often found in such enormous masses in tropical, temperate and arctic
regions ; either branched and forming knolls perhaps as big as or bigger than a man's hst, and
o-iving shelter to innumerable animals; or crustlike, and covering corals, stones and other algae
with thin or thick crusts.

The study of calcareous algae is, like all algological study, one of the younger branches
of botanical science. In 1S16 Lamarck l) in his renowned book "Histoire des Animaux sans

1) LAMARCK. Hist. nat. d. Animaux sans vertèbres. Paris 1S16, t. II, p. 203.
SIBOGA-EXPEDITIE LXl.

O-

Lted the name of Nullipores for the "millépores avec pores polypifères peu ou

! apparens", in opposition to "mill avec pores polypifères toujours apparens". This

name Nulli as outlived its time and cannot be maintained. Philippi l) first and afterwards

have shown in a (Kar and convincing manner that a great number of so called

ent and fossil are Lithothamnia. Gümbel 3) in 1871 divided the Nullipores into

Nulliporcn aus dem Pflanzenreiche : Lithothamnium" anti "die Nulliporen des

Thierreichs : Dactyloporidae".

Xhe Da "idae which Gümbel still considered as belonging to the anima) kingdom

hav, d by Munier-Chalmas 4) as calcareous algae belonging to the group

of 1; ladiaceae. < >ld fossil forms of this family are known according to Stolley ') from the

a Baltic-Silurian, and recent representatives live in the wanner temperate and tropical

ls under the generic names of Cymopolia, Neomeris, Acetaöularia, etc.

If nou these two groups the Lithothamnia in the broad sense and the Dactyloporidae
( Dasycladiaceaej are taken out of the old group of the Nullipores^ one feels inclined to ask,
do there still remain any Nullipores^.

( >n the reefs in the tropical seas calcareous algae are found in great quantities: Hali/ueda,
Galaxaura, Liagora, Peyssonelia - ■ this last one remarkable because it sometimes imitates
incrusting Lithothamnia - and many others, but none of these will be considered a Nullipore
by botanists. A. C. Seward 6) speaking of Nullipore* says: "the genus Nullipore has been
used in an extremely wide sense embracing organisms which are for the most part members of
the vegetable kingdom but including others that must still be retained as animals". This shows
rly that the genus of the Nullipores has in tact been broken up into various families ot
which none tan prei, nd to the name of Nullipore.

Still the old name persists ; I was puzzled several times, while reading the beautiful
illustrated account of Prof. Agassiz7) on the coral reefs, to understand what the author meant
by Nullipores. ( )n page 73 for instance we read "'Nullipores and other calcareous algae" on
page 229 "large blocks ot algae and Nullipores". Atter all that I have seen of reefs, I
believe that all these bright-coloured Nullipores are algae, belonging to very different families
and genera. The old name brings only confusion to a botanical mind ; it can only be used in
a very broad sense to indicate all the marine vegetable organisms, which contain chalk in
their membranes, thus including all calcareous algae.

The Siboga Expedition, sent out for the exploration of the deep-sea basins and the
marine fauna and flora not only of these but also of less deep and of shallow waters in the
Malay Archipelago, visited a great number of islands during her cruise of a year. Whenever

1 Phili ' o Pfl ind. Wiegmann's Archiv 1837, p. 387.

. Zur naheren K< I ithal il] .. Denkschr. d. K. Akad. d. Wissensch. Wien 1858.

I ithamnium und Dactylopora). Abh. d. K. Bay. Akad. d. Wissensch. 1S74, Bd. XI.
ur les Alg. calc. appartenant au groupe des Siphonées verticilt. (Dasycladées Haiv.). Comptes
rend.

ind ihre Verbreit. im Skand. balt.-sil. Gebiet. Schrift d. Natuvw. Ver. fut Schleswig-

! n - [894, vol. II. p. iS.
f the lïnp. Pacific. Cambridge l . s. A. 1903.

there was an opportunity her staff would go ashore or land on the numerous reefs that
surround the islands to collect animals and plants. Great was my astonishment to find Litho-
thamnia on almost every reef; in not less than 55 stations did we gather these organisms in
greater or smaller quantities, indeed almost on every occasion, where marine algae could be
gathered. This impressed me mnch, as up to the time of the Siboga Expedition little was known
about Lithothamnia from the Malay Archipelago. Vox Martens ') in his standard-work "Die
Tange von Ost-Asien", attributed our slight knowledge of them to the neglect by botanists
to collect them. I was well acquainted with the fact that Darwix 5) had mentioned three
Nullipores of the Keeling- Atoll, that Heydrich 8) had lately described four new species from
New-Guinea, and that Martin ') had described Lithothamnium Rosenbergi from the Tertiary
of Timor and New-Guinea, still I was not prepared to meet with such masses of these algae
nor above all to find them almost everywhere.

I am sorry to say that two geological papers had escaped my notice, for as early as
1891 Walther 3) had described subfossil banks of Lithothamnia from the Palk Straits and in
1898 — the year we left Europe — W. Stanley Gardiner6) published two papers, one on
the coral reefs of Funafuti, the other on the building of Atolls. In these papers he mentions
several times the presence of great numbers of Nullipores recent and fossil on the coral reefs
of Funafuti, and their presence everywhere in great heaped up masses impressed him so
strongly that he writes in the latter paper p. 121: "The chief organisms, so far as I have
seen, engaged in building up coral reefs are massive and incrusting Xulliporcs" .

The results of the Siboga confirming, as they do, all those of previous and of
simultaneous investigators 7) deserve special notice as throwing some light on the conditions
under which Lithothamnia grow. Lithothamnia, as everybody knows, belong to the family of
the Corallinaceae. Areschoug 8) united them with Hapalidinm, Mclobesia and Mastophora in
the group of the Melobesieae in opposition to the Coral lincac (verae) to which he referred
Amphiroa, Cheilosporwn, Arthrocardia, Jania and Coralliua. The inarticulated fronds of the
Melobesieae and the articulated fronds of the Corallincae are still a good character, whereby
to distinguish the two groups, although that of the Melobesieae has undergone many changes
since Areschoug's Monograph appeared, as will be seen in the following pages by Foslie.

The Siboga Expedition found Lithothamnia, as already stated, at 55 stations; their
favourite habitats as a rule being where a strong current was running. If a strong tidal or
other current swept along the Siboga, while she lay at anchor, this was enough to induce us

1) G. v. in' Martens. Die Tange von Ost-Asien. 1866. p. 107.

2) Ch. Dakwin. A naturalist's voyage round the world. 1860, p. 498.

3) F. Heydrich. Beitr. zur Kenntn. der Algentlora von Kaisev Wilhelm's Land (Deutsch Xeu-Guinea). Ber. d. D. Bot. Gesellsch.
1892, p. 483. — Beitr. z. Kenntn. d. Algenflora von Ost-Asien. Hedwigia 1894, p. 299. — Neue Ralkalgen v. D. Xeu-Guinea. Biblioth.
botanica 1897.

4) K. Martin. Samml. d. Geol. Mus. in Leiden 1SS1. Sedimente Timor» p. 1. Tertiar vod Xeu-Guinea p. 65.

5) P. L Walther. Die Adamsbrücke u. d. Korallenriffe d. Palkstrasse. Erganz. Heft Pet. Mittlr. 1891. — Lithogenesi» d.
Gegenwart. Jena 1893.

6) W. Stanley Gardiner. The coral reefs of Funafuti, Roturaa and Fiji. Proc. Cambridge Phil. Soc. IX, 189S. The building
of Atolls Proc. Intern. Congress Zoology 1898.

7) Agassiz found "Nullipores'" or huge masses of calcareous algae on almost every reef in the Pacific Ocean which he visited
during the winter 1S99 — 1900.

8) L E. ARESCHOUG. Corallineae in Agardh's Genera et Species Floridearum 1S51. p. 506.

te» throw out a dredge and very often success would crown our endeavours and the dredge
would bring up Lithothamnia. We kept a small, strong dredge for this work, for the chances
of losing our net and even the whole dredge were often great. The Lithothamnia appeared to
thrive best in from 10 — ^o m. but we as often fouiul them on the reefs near to or below
low-water mark, though seldom in great quantities; and again several little bits, too small for
determination or harren, came up at various times with the lead from a depth of 75, 83,
04 and 120 m.

Near the coast of Haingsisi, an island near the S.W". point of Timor, the Siboga anchored

<%>M

Li ihothaniQ ion-bank uf Haingsisi,

±ji>ï

twice (Stat. 60 & 303); the second time good-luck favoured us, it was springtide, the water
sank very low and we could observe that the whole reef, which stretched from the shore out
into the sea, consisted chiefly of Lithothamnion erubescens f. Haingsisiana. It was remarkable
that the branching knolls remained quite dry during several hours of the day, exposed to the
: tropical sun and that this seemed not to injure them. This shows clearly that at
some species of this genus do not fear the rays of the sun, as others appear to do in
the Gulf of Naples. Along the coast of arctic Norway species growing in shallow water, always

- according to a communication in a letter from Foslie —

but this is owing to the permanent Hght, day and night, that reigns in these regions during
the summer months. Even outside the polar circle as far south as Trondhjem the influence of
the northern summer is visible on the Lithothamnia. On the south coast of Norway, where
Foslie also collected Lithothamnia, they are as a rule much darker.

The reef of Haingsisi was literally covered vvith knolls of L. eruèescens, as the two
text-figures show. The knolls were of very different size; we found little ones only i — 2 cm.
in diameter and other big ones that measured 9 cm. Often two or more knolls had grown
together and formed one big knoll; the living ones were red all round, the dead and the

Lithothamnion-bank of Haingsisi.

dying ones were totally or partially white. Other species and genera as Litliothamnion australe,
Lithophylhun Okamurai and Goniolitkon Reinboldi were also to be found, but they were
numerically of 110 importance compared with the ubiquitous L. eruöescens.

This Lithothamuioii-hank. struck me because it is such a unique sight to see the ground,
as far as the eye can reach, covered by the pretty beautifully pink-coloured knolls, which are
heaped up so . close together that, while walking, one crushes them continuallv making a
peculiar noise as of broken china. We encountered however other and perhaps more instructive
Lithotha/////iou-ba.nks during our voyage.

War Lumu-Lumu Island, Stat. 78, a coral island that rises to the surface of the sea
from the Borneo-bank in the Makassar Straits, the Siboga anchored for a couplé of days.
Durin» our stay on the Borneo bank our little dredge was often thrown out behind the ship
and hauled up at ber forecastle and every time it came up filled with various species of
Lithothamni miolithon Reinboldi coming up oftenest. All along the east-side of Borneo

and in the Sulu Archipelago we met Lithothamnia in such quantities, that they formed a
conspicuous character of the reets. It was also here that I found for the first time Archaeo-
lithothamni ; its thick branches and beautiful soft pink colour were fascinating to the

ivhile the knowledge, that representatives oi this genus were described by Rothpletz from
the furassic and Pliocene periods, made it doubly interesting to the mimi.

in shallow channels near Tual, Stat. 258, of Low-Key Island the dredge was filled with
Lithothamnion australe f. tualensis. We were at work for hours with the steam-launch and
constantly the dredge brought np Lithothamnia even to the vexation of our zoologists. For
though Lithothamnia will give shelter as a rule to a great number of quite different animals,
which adapt themselves outwardly in form and colour in a striking manner to the plants on
which they live, this is not always the case. The Lithothamnia of Tual were not favourable
to the maintenance of animal life, being small with widely spreading branches and thus affording
but a had shelter to snails, worms, star-fish and other animals.

The most remarkable bank was one south of the Island Saleyer, Stat. 66, mentioned
on the charts as a "coral bank" and extendine; between the islands of Bahuluwangr and
Tambolungan. The Siboga had vainly tried to find a good anchorage on the east-side of
Saleyer: the night advanced rapidly and therefore Commander Tvdeman resolved to anchor
for the night on the above named coral bank, where he was sure to find from 8 — 10 m.
water. How great was our astonishment the next morning when coming on deck, we saw a
distinct red colour at the bottom of the sea regularly interrupted by narrow white bands. It
was nu coral bank on which we were lying but an enormous bank oi Lithothamnia. that g'ave
a red colour to the bottom of the sea, and the white bands we remarked consisted of heaped
up dead joints of Halimeda.

Weber x) examined the whole bank and his view of Lithothamnion-ba.iiks, the conditions
under which they develop themselves and expecially of this particular bank may well be
recapitulated here.

For the development oi Lithothamnia the bottom upon which the roundish knolls grow
must be more or less flat. A current must be running over it, which turns the knolls liphtly
over, for the knolls want light and sea-water on every side or they would soon lose their
colour, which means death to them. We observed this on the bank of Haingsisi where
half-buried knolls were quite white on their underside. But the current must not be too strong.
When- Lithothamnia throve best many tiny Florideae grow on them and this would be impossible
it the current turned the knolls roughly over. The structure of this peculiar bank near Saleyer
\\ 1 bi r explains as follows :

et ] 'Expi lition du Siboga 1902, p. 38.

"Between the islands of Bahuluvvang and Tambolungan our bank extends in a direction
from North to South covered by a layer of water averaging from 6 to 9 metres deep ; but
towards the West and East of the bank the layer of water deepens fairly rapidly, attaining a
depth of 10, 14 to 18 metres, becoming at last very deep. On this bank Lithothamnia live in
great abundance; living corals are entirely wanting. The net, in fact, only brought up scanty
and dead fragments, which the current had clearly carried there after detaching them from
neighbouring coastal reefs. The force of the current is great. During the ebb it sets west, while
the dood tide sets east. If the bank is bare of corals - - which is seldom the case in the
Archipelago - - this is due to the current which, carrying along sand and débris of Halimeda,
covers up the corals that are trying to establish themselves there and smothers them out of
existence. This is not the fate of the Lithothamnia. As we have said, the current rolls them
sometimes in one way sometimes in another accordino- as it is running- East or West, fiood or
ebb. In places where the play of forces is suitably balanced between the strength of the current,
the vveight of the nodules of Lithothamnia and the weight and mass of the detritus, the nodules
remain free and uncovered and are able to form a bank. But in places where local accumulations
of detritus are formed, in the same direction as the current, those bands of which we have
spoken appear; these are less favourable to the development of Lithothamnia and become less
and less so in proportion as they grow more accentuated".

In Arctic seas Foslie ') tells us, that Lithothamnion-banks often consist of only one
species. Near Nowaja-Zemlya and Spitsbergen Lithothamnion glaciale will cover the bottom
for several miles and on the coast of Iceland and northern coast of Norvvay Lithothamnion
i ngeri will occur in masses. So also Lithothamnion investiens, a species nearly related to the
former, covers in certain places on the north-coast of Norway the bottom for several miles.
In the Malay Archipelago it appeared to us, that one or other species usually was predominant
but other species would not be entirely lacking, and incrusting forms can almost always be
found, when attentivelv searched for. For the Lithothamnia, besides contributine greatlv to the
bulk of a reef or bank by their branching species, will also play an active part in the building
up of a reef by their incrusting species, which grow over shells, corals or other Lithothamnia,
cementing all these together into one solid mass. The value of this cementing cannot be over-
estimated on a reef, where the waves of the sea are continually washing off the looselying masses.
A most dangerous enemy to Lithothamnia is another family of algae : the boring algae.
These algae were first made known by Bornet and Flahault -). They belong as a rule to
filamentous blue, green or even red algae. The study of these plants has as vet only been
begun and their universal distribution and their power of disintegrating even rocks, will become
more and more recognized.

■ Here and there in literature I have met with the expression "green Lithothamnia" . I do
nut believe that there exists one single green species of Lithothamnion. If knolls or crusts of
these plants are green, this is due to boring algae, which are found according to my experience

1) M. Foslie. The- Lithothamnia of the Laccadive and Maldive Islands. Fauna and Geogr. of the Mald. and Lace. Archipelagoes.
vol. II, part. 4.

2) Bornet et Flahault. Sur quelques plantes vivant dans Ie test calcaire de» Mollusqües. Buil. Soc. bot. de France. t. XXXVI.

in oreat quantities in all Lithothamnia of the tropical seas. And not only Lithothamnia but also
corals have to suffer from these algae. Duerden *) writing about algae as coral disintegrating
agents says p. 326: "the surface of many corals, even when covered with the polypal tissue is
strongly green in colour .... (due) to the presence of green filamentous algae".

Wonderful is Hfe <>n a coral reet and the various agents now building up and then
destroying again whal scarcely was developed, wil! ever afford a tempting object of study
t<. Naturalists.

In the above quoted paper on the Lithothamnia of the Laccadive and Maldive Islands
Fosi ie observes that "the Lithothamnia of the Tropics have been rather disregarded". In latter
years, thanks to several collections that were sent for investigation either t<> Foslie or to
Heydrich our knowledge of these plants bas much increased and the collections of the Siboga,
to which were added some specimens brought home by the Dutch New-Guinea Expeclition,
will help to <^ïve us a fairly good idea of the distribution of Lithothamnia in the Indian Seas.
[E has described in the following pages 32 species of Lithothamnion and other genera,
belon^iiiL; to the group of the Melobesieae Aresch. •. amongst these seven are new to science,
and three, which I "had studied while still on board H. M. Siboga, were published shortly after
my return, in our joint names, and here I should like to say, how happy I am to have secured
for the collections of the Siboga the valuable collaboration of Mr. Foslie. His extensive
knowledge of all existing Lithothamnia gives us hope, that he will bring order into this most
confusing group.

It was first understood between us, that Mr. Foslie should work out the incrustine and I
the branchingf Lithothamnia. but I soon discovered that I should be most grrateful if he would
undertake the namins: of them all. The Corallineae verae. of which the Siboga material contained
several genera and species were to be worked out by me alone and the paper in which both
groups Melobesieae Aresch. and Corellineae verae were treated, appear under our joint names.

The Corallineae verae: Amphiroa^ Cheilosporum, Jania and Corallina, slender plants
as thev may be, are very active reef-builders, indeed, taking their smallness into account, it is
astonishing to observe how much sand may be kept together by their tiny branches. They
illustrate often in a wonderful manner how true the words of Walther are: "Das Wesen der
Riffbildung beruht im Sandfansren".

Corallineae verae were gathered by the Siboga to a clepth of 36 m. On the Borneo
bank beautiful, long-creeping specimens of Cheilosporum spectabile Harv. were found. Large
squarrose Amphiroae were abundant along the south-east coast of Timor, indeed almost every
reef yielded one or another species of this genus. Still I could not detect more than a single,
new species among my collection. The French investigators who explored the eastern part ot
the Archipelago during the beginning of the nineteenth century had already gathered them all,
and they wen; described by Lamouroux.

I would like to express here my sincere thanks to Prof. Lignier of Caen for lending
me so liberally the Amphiroae out of Lamouroux's Herbarium.

EN. Al 1 . a\ disintegrating Agents. Buil. Amer. Mus. of Nat. Hist. 1902. XVI.

But if the study of the Amphiroae showed clearly that only one new species was to bc
looked for amongst my material, at least this study led me to a careful investigation of the
whole group, which had to be worked out afresh. How and why this was done will be
explained in that part of the paper that treats of the Corallineae verae.

I should like also to express my thanks to Prof. Rothpletz of München and to Prof.
Martin of Leiden, who both gave a large collection of fossil slides containing Corallindceae
to Mr. Foslie and me for investigation. In the collection of Prof. Martin I detected a fossil
Amphiroa from the Tertiary of Celebes and a fossil Jania from a more recent stratum at
Ambon, and in a slide from the Tertiary of New-Guinea that I owed to the kindness of Prof.
Wichmann of Utrecht, I found another species of Amphiroa.

To all those, who gave me good advice and helped me on, though I don't name them
here, I feel heartily indebted and especially so to my kind friends Mr. and Mrs. Gepi- who
looked over this paper and improved my English.

SIBOGA-EXPEDITIE LXI.

I.

LITHOTHAMNIONEAE, MELOBESIEAE, MASTOPHOREAE

i;v

M. FOSLIE.

With plate I— XIII.

I. LITHOTHAMNIONEAE Fosl. T).
Lithothamnion Phil. (Fosl. emend.).
a. C r u s t ac e a 2).
i . Lithothamnion siamense Fosl.

Corall. in SCHMIDT, Fl. Koh Chang II, p. 19; excl. f. simulans.
f. typica. Fig. 3; PI. I, fig. 1 — 2.

Lithothamnion siamense f. minuta Fosl. 1. c.

f. pseudoramosa Fosl. mscr. PI. I, fig. 3 — 9.

The crust forming short branch-like excrescences, mainly by covering up undissolved
conceptacles, other species or extraneous objects, and often less clinging to the substratum than
in the typical form.

Stat. 49'. Sapeh Strait, Sumbawa. 69 m. Coral and shells.

Stat. 64. Kambaragi Bay, Tanah Djampeah. Up to 32 m. Coral, coral sand.

St.it. 109. Pulu-Tongkil, Sulu Archipelago. 13 m. Lithothamnion bottom.

Stat. 125. Sawan, Siau Island. 27 m. Stone and some Lithothamnion.

Stat. 234. Nalahia Bay, Nusa-Laut Island. 46 m. Stony.

Stat. 258. Tual, Kei Islands. 5 — 20 m. Lithothamnion, sand and coral.

ij '' 1 ) \ : 1 1 Vidensk. Selsk. Skrifter 1902, Aarsberetning (Trondhjem 1903) I have divided the Corallinaceae in ihe

following sectioi ionally without particular argument: Chaetolithoneae : Chaetolithon\ Lithothamnioneae : Phymatolithon,

Clalkt Archaeolithothamnion\ Schmitzielleae : SchmitzUlla\ Choreonemeae : Choreonema\ Melobesieae:

Dermatolithon^ I.ithophvlium: Mastophoreae : Mastophora\ Corallineaic : Amjikiroa, Chtilosporum^ Corallina.

2) In Nonv. Lithoth. and Rev. Syst. Surv. Melob. I divided Lithothamnion into Innata and Evanida. Instead <>f this I

' iiistacea. Subramosa and Ramosa.

1 1

As remarked 1. c, this is one of the most delicate species of the genus Lithothamnion
attached to hard objects. It sticks to different other calcareous algae, especially dead or stunted
ones, calcareous sediment, corals, dead shells of molluscs, or occasionally stones. When the
plant occurs on a level, or fairly level, substratum, the crust is smooth and generally has a
thickness of 30 — 60 y.. In this case it clings rather closely and often firmly to the substratum
(PI. I, fig. 1 — 2). Frequently, however, new crusts will be formed, one above the othei/, or
the primary crust will develop secondary incrustments. Such a complex of crusts may attain a
thickness of up to 300 [j.. The species is apt to being overgrown by other calcareous algae,
and is often itself attached to other species. A section of the plant with its substratum will,
therefore, now and then exhibit layers of two or several species growing one above the other
or even alternating with lower animals, e. g. Bryozoa.

If the substratum is rough, or if foreign bodies have attached themselves to the primary
crust, the surface of the plant fmally becomes more or less uneven. The formation of new
crusts on fertile specimens, expanding over undissolved conceptacles of sporangia or cystocarps,
or the occurrence of the plant on other crustlike species which are sparingly branched, e. g.
stunted Lithothamnion fruticulosum f. clavulata, gives rise to the form which I have named
f. pseudoramosa. The form is certainly but slightly differentiated and ought perhaps not to be
kept distinct. However, as it is externally often rather diverging from f. typica, I classify it as
a separate form (PI. I, fig. 3 — 9). Besides it does not appear to adhere so intimately to the
substratum, but shows a tendency to partly detaching itself from the latter.

As regards the structure of the species, vertical sections show that the hypothallium ')

frequently consists of a few layers of cells whose lower anticlines are

often but slightly converging to the substratum, sometimes not

converging at all, and apparently never marked coaxial. The cells are

elongated and may be up to about 25 f/., long, though they rarely exceed s

12 — 18 u. The perithallium, issuing from the hypothallium and as =/^^Z)ODOOCDd

. . JODddqdOD

a rule taintly developed, is formed by cells which are partly roundish, ~]{~]( '1/' ]j ?r 1/ — 1/ — \r~)\~

partly square, 6 — 10 p.. in diameter, often, however, a little elongated, ~)(~]r~|( }r~)r~}f~)ni — ir

with the longer diameter now in horizontal, now (and more frequently) "VTAOf-) f~) OiOO DC

in vertical direction, in the latter case being exceptionally up to 18 a. ^rinr\nnO^

long and 10 a. broad ; fig. 3"). Flg' 3'

Lithothamnion siamense Fosl. f. typica.

The reproductive organs hitherto known are sporangia and A. Vertical section of the crust; x 72.

, • , -t-rr v • 1 1 t^i r /;- Part of the perithallic laver in ver-

cystocarps which occur on dirterent individuals. 1 he tormer are ,,

J 1 tical section; X 390.

subprominent or almost superficial, 300 — 400 u. in diameter, when

seen from above, at first convex, afterwards more or less flattened, with a thin roof, perforated

1) The two kinds of tissue which form the thallus in most of the algae in question were indicated by SOLMS-LAUBACH as
Markstrang and Hinde. Areschoug used for the former the denomination of hypothallium^ and for the latter Rothpletz pvoposed the
name of perithallium. 1 have adopted the latter name in the same sense. As to the former, I also use the term of hypothallium for the
basal or rhizoidal layer, composed of rows more or less bent or coaxial in a number of crustlike fornis. If this tissue consists of only one
basal cell-row, as the case is in some other forms, I simply name it the basal cells or one-layered hypothallium. In attached, branching
forms there is frequently some difference in size and sometimes even in shape between the said hypothallium in the crustlike part of the
plant and the inner layers of radiating tissue in the branches. In describing the structure I therefore name the latter medullary hypothallium.

2) All the figures showing the structure of the species are from undecalcified sections which have been made thin by grinding.

I 2

1,\ about 40 delicate muciferous canals. The root' finally falls off, and dissolved conceptacles only
leave smal! elevated edges, or a scar more or les distinguishable. The four-parted sporangia are
60 — 70 ■>.. long, and 25—30 ;j.. broad. The conceptacles of cystocarps are almost superficial,
conical, sometimes however rather low, abruptly passing into a very short and thin tip, which
soon falls away, 550 450 ■>.. in diameter, when seen from above. Antheridia are unknown.

The typical form of the species in sterile stage is easily confounded with young Squa-
mariaceae, and is sometimes hardly distinguishable from young specimens of Lithothamnion
simulans. The form pseudoramosa is in habit approaching young and stunted specimens ot
Lithothamnioti fruticulosum f. clavulata, for which it is easily mistaken, sometimes being hardly
distinguishable from it also as to structure. It is of no infrequent occurrence that it covers
specimens apparently belonging to the form mentioned.

Specimens of the present species having been collected in April, May, July and December
are sparingly provided with cystocarps and sporangia. In the Gulf of Siam the species bears
these organs in the month of February.

Occurrence: On the whole this species occurs rather scarcely. The comparatively greatest
number has been found at the stations 49a, 64, 258. The species mostly has been found associated
with Lithothamnion fruticulosum and Lithothamnion australe^ but almost always it shares the
substratum also with stunted specimens of other calcareous algae and a series of lower animals.

Area: Xorth Pacific: The Gulf of Siam.

2. Litliothamnion bandanum Fosl. mscr. Fig. 4; PI. I, fig. 10.

Thallus crustlike, up to 0,5 mm. thick, closely clinging to the substratum. Conceptacles
of sporangia subprominent, 250 — 450 u.. in diameter.

Stat. 240. Banda. Lithothamnion bank in 18 — 36 m.

Of this species only a single specimen is known. It is 1 cm. in diameter, and forms
a crust, up to about 0,5 cm. thick round the substratum, whose form it assumes. Some
wartlike excrescences seem only to be owing to the form of the substratum, and it is
questionable whether such ones really are developed from the crust at all (PI. I, fig. 10).

A vertical section exhibits a faintly developed hypo-
thallium, formed by rows of cells whose lower anticlines are
but very sligthly converging to the substratum. The cells
attain a length of 12 — 20 a., and a breadth of up to 10 a. —
The cells of the perithallium are rather irrecnilar and have

rig. 4. Lithothamnton bajiaanum frosl. 1 ö

Part of tiie perithallium in vertical section, with often thick walls. The\' are partly roundish, partlv (and

transversaly cul feeble hypothallic layer , . ,, , ,

;j mostly) vertically elongated, 7—15 or sometimes up to i8p.

long, and 6 — 8 or seldom up to ioix. broad. There is no
distinct stratification to be seen in a section, but new hypothallic layers here and there are
formed over conceptacles or small extraneous objects (Fig. 4).

The conceptacles of sporangia are at first convex, partly very slightly rising above the
surface of the thallus, partly subprominent, finally fairly flattened, rather crowded in any part

of the crust, 250 — 450 u.. in diameter, when seen from above. Two or three conceptacles
sometimes become confluent. The roof is intersected with about 30 muciferous canals. The
conceptacles become overgrown and appear in great number in a section of the crust. A few
ones examined were partly emptied, partly attacked by animals who had destroyed the sporangia.
The specimen was picked up towards the end of November, therefore the plant is likely to
bear mature sporangia by this time.

In habit this species on one hand approaches a little certain forms of Lithothamnion
Engelhartii, and on the other hand L. fumigatum. It is however most closely allied to L.
maldivicum both in habit and structure, but the conceptacles are much smaller, and as regards
structure it mostly shows smaller and less thick-walled cells than the said species.

Occurrence: The only specimen met with was found among a number of stunted
calcareous algae from the quoted station apparently belonging to Lithothamnion fruticulosum ,

6. Subramosa.

3. Lithothamnion fragilissimum Fosl. mscr. Fig. 5; PI. I, fig. 11 — 16.

Thallus crustlike or more or less leaf-like, 150 — 500 fjt. thick, horizontally extended,
partly attached, scantily and irregularly prolificating. Conceptacles of sporangia convex, but
very little prominent, at length somewhat flattened, crowded, 350 — 450 u. in diameter.

Stat. yS. Lumu-Lumu-shoal, Borneo-bank.
Stat. 81. Pulu Sebangkatan, Borneo-bank.
Stat. 234. Nalahia Bay, Nusa-Laut Island. Reef.

The plant occurs partly, and particularly when young, clinging rather closely to the
substratum, partly, and most frequently, but loosely covering the latter, here and there attached
to it. At lenght the greater part of the thallus gets free. It is 150 — 500 u.. thick, decreasing
in thickness towards the margin, somewhat shining, rather
brittle, but on the other hand of a rather hard consistencv.
Judging from a solitary almost entire specimen it is some-
what lobed (PI. I, fig. 1 1). The species is scantily prolifica-
ting. The prolifications partly more or less cling to the
primary crust, partly they are semi-circular and somewhat
rising, or irregularly plicate, mainly when covering up
extraneous objects. Besides it is often provided with small
and irreeular excrescences, which also are due to its
covering up different objects. The plant is concentric zonate,
distinctly so in the part turning downwards, in the part
turning upwards indistinctly.

As to the Structure, a radial section of the crust * ?2' ?• ^rt of the perithaUic fcyer in vertical section;

X 390. C. Radial section near the margin; X 72.

shows a marked and rather vigorous coaxial hypothallium,

which in the central parts of the crust frequently sends forth upwards rather long perithallic rows.

On the contrary, in the peripherical portions the perithallium is very faint (Fig. 5 C). The

EoaaaGL
DaOODDOC

DaaaoDoc

Doaaooac

DCDOOOOcoc
DO CO 30g<2C

-ir— ii — 11 — innnr

Fig. 5. Lithoth fragilissimum Fosl.

./. Vertical section towards the central part of the crust;

1-1

hypothallic cells are 18 28^. long and 7 — 14 ,u- broad, partly with fairly thick walls, partly
thin-walled. In the perithallic layer the cells are partly square or somewhat rounded, partly a
little vertically or horizontally elongated, frequently 5—8 <i. in diameter, with rather thick
walls, here and there however commonly elongated in vertical direction and up to about i2u.
long, very seldom more excepl between crowded conceptacles of sporangia, where the cells
may be somewhat longer. I Fig. 5 1.

The reproductive organs hitherto known are sporangia. The conceptacles are mostly
densely crowded on different parts of the crust, sometimes partly confluent, and are in old
imens finally overgrown. They are not much prominent, with slightly convexed roof, at
length somewhat flattened, circular or oblong in circumference, but often not sharply defined,
-450 ij., in diameter, when seen from above. The roof is thin and intersected with 30 — 40
delicate muciferous canals. The sporangia are four-parted, about 80 u. long und 30 u. broad,
with enduring partition walls between the former. The plant bears these organs in the months
of June and November, but rather scantily.

The species comes nearest to Lithothamnion Engelhartii from the south coast of Australia.
It is distinguished from the latter in habit particularly by a thallus more spread and more freely
growing. Besides it is brittier. The conceptacles are a little larger and less sharply defined, and
there is also some difference as to structure. Young or sterile specimens are, on the other
hand, hardly distinguishable from Lithothamnion simulans.

Occurrence: This species seems to have been best developed at the station 8 1 , from
which, however, only a few specimens, more or less fragmentary, are in hand. From the station 7 8
only a small fragmentary crust has been brought home, together with L. simulans, and from
the station 234 there is only one young specimen, accrete with two or three other species,
among which a Squamariacea and perhaps a young L. simulans.

By the kindness of Professor K. Martin I have had the opportunity of examining some
sections of fossil sediments from the Geological and Mineralogical Institut' at Leyden, rnainly
from Xew Guinea. These sections include a great number of calcareous algae, among which
two of the types of Lithothamnion Rosenbergi K. Mart. !). This species I classified in Rev. Syst.
Surv. ") as Archaeolithothamnion Rosenbergi according to the description given by Rothpletz
in Foss. Kalkalg. 3). Having afterwards been acquainted with the original description and the
pictures of the species, I find it questionable, whether it includes any Archaeolithothamnion at
all. At any rate it includes at least two species, one of which may be identical with Lithotham-
nion fragilissimum. big. 6 represents a reproduction of the pictures mentioned. Judging from

1 K. Mv. I Versteinerungsführenden Sedimente Timors. — Sammlungen des geolog. Reichsmuseums in Leiden. I, Heft i

(1881), p. 12. t. III. lig. 6—7.

K. Mai iis. Kine ïertiaerformation von Neu-Guinea uad benachbarten Insein. 1. c. Heft 2 fiSSi1. p. 70.

Systematical Survey of the Melobesieae. — Det Kgl. Norske Vidensk. Selsk. Skrifter. 1900, n° 5.
Rotiii'Lei Kalkalgen aus den Familien d:r Codiaceen und Corallinaceen. — Zeitschr, d. Oeutschen geolog.

1 111. Heft 2 (1S91), p. 295.

15

n-0 ruffl

A

Fig. 6. Lithothamnion Rosmbergi K. Mart.
JX 50; B X 23; CX 32. (After Martin).

the description 1. c, I think the fig. 6 A (the fig. 6 of the original) is to be lookecl upon as

the proper type of the species. But of this one I have only seen the picture. To judge from

the picture, the supposed sporangia are of

another form than usually found with Archaeo-

lithothamnion, but this may be owing to an

accidental obliquity of the section. They are,

however, according to the picture surrounded

with a structureless m a s s , which seems

to show that this part of the figure cloesn't

represent any calcareous alga, whereas the

parts situated above and below are likely to

be such ones. According to Martin 1. c. the

cells here are 10 u. long and broad. In

sections of fossil sediments with calcareous

algae, as well as of recent forms of the latter,

we will frequently observe that layers of

crustlike calcareous algae occur alternating, occasionally even quite regularly, with certain

species of lower animals. It is not improbable that fig. 6 A represents such a state of things.

As a matter of course I don't venture to express any decided opinion, judging from the picture

alone, and I therefore still subsume the species under Archaeolithotliamuioii.

The form reproduced in fig. 6 B is found in section "123 Soëk v. Rosenberg (8)". It
is "von dem Eilande Soëk im Geelvink Busen, an der Nordküste von Neu-Guinea". —
"Unter den Organismen herrscht Lithothamnium hier so sehr vor, dass es geradezu gestein-
bildend auftritt und man in Dünnschliffen ausser dieser Pflanze oft nur noch sparliche, die
Zwischenraume ausfüllende Kalkspathkörnchen wahrnimmt" J). This form is no doubt to be referred
to the group of calcareous algae which form thin crusts or foliaceous thallus like LitJiotlicuunion
lichenoides, L. fragilissimum and other species. The thickness of the crust in the section is
330 — 450 tj.. The central cells of the coaxial hypothallium are 22 — 32 p. long and 8 — 11,
seldom up to 14 u.. broad, with rather thin walls. The cells of the perithallium are commonly
squarish, 6 — 10 p. in diameter, with the greater length partly in horizontal, partly in vertical
clirection. In the former case they may be only 5 p. by 9 p., in the latter case very seldom
up to 14 p. long. This form being compared with L. fragilissimum, the perithallic layers of
both species will show almost complete conformity. The structure of the hypothallium is also
much the same as in L. fragilissimum, but the cells are here mostly longer. I have previously
pointed out that the hypothallium can be much varying with several calcareous algae and show
a comparatively greater variety than the perithallium, a fact which is — partly at all events -
owing to the shape and nature of the substratum. In the case in hand the variety is even less
than in several recent species. The material, however, is scarce, and the fossil form is sterile.
Therefore, I don't venture to identify it at present, but at any rate I must consider it as a

1) K. Martin. Lithothamnium in cretaceUchen und jüngeren Ablagerungen tropischer Insein. — Centialbl. für Mineralogie
etc. 1901, p. 163.

i6

species closelj connected with /. fragilissimtim. In the sections "121. Soëk. v. Rosenberg (6)"
a few sin. ill fragments occur, which seem to belong to the same species. It hére appears
ther with numerous fragments of Amphiroa and other calcareous algae, among which
Mastophora melobesioides classified below.

In the same section as fig. 6 />' also the calcareous alga represented in fig. 6 C is to
be found. It probably represents a species different from the above mentioned one. The cells
are here usually 11 — 20 or sometimes up to 25 u.. long and mostly 7 — 10 u.. broad. It is,
however, hardly to be ascertained what group it belongs to.

According to Martin 1. c. the species, in the sense taken by him, also occurs at Timor,
where the fossil sediments "sind stellenweise so vorherrschend aus Kalkalgen gebildet, dass
man sie nur als fossile Lithothamnien-Banke bezeichnen kann. Das gilt besonders für das
Vorkommen von der kleinen Insel Samauw, gegenüber Kupang". - It is unknown to me
which of the forms mentioned above is here represented.

4. Lithothamnion simulans Fosl. mscr.

Lithothamnion siamense f. simulans Fosl. Corall in SCHMIDT, Fl. Koh Chang II, p. 19.
f. typica. Fig. 71 PI. I, fig. 24 — 25.

I 1 iSLIE 1. c.

f. crispescens Fosl. mscr. PI. I, fig. 21 — 23.

Thallus more or less lamellate and crispate, or plicate-leaf-like.

Stat. 78. Lumu-Lumu-shoal, Borneo-bank. 34 m. Coral and Coralsand.

Stat. 81. Pulu Sebangkatan, Borneo-bank. 34 m. Coral bottom and Lithothamnion.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion bottom.
Sand and coral.

Stat. 125. Sawan, Siau Island. 27 m. Stone and some Lithothamnion.

Stat. 213. Saleyer and Surroundings. Reef.

Stat. 240. Banda-anchorage. Lithothamnion bank in 18 — 36 m.

Stat. 258. Tual, Kei Islands. 3 — 12 m. Lithothamnion, sand and coral.

Stat. 261. Elat, West coast of Great-Kei Island. Reef.

Stat. 2S2. Between Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and Litho-
thamnion.

Stat. 299. Buka or Cyrus Bay, South-coast of Rotti Island. 34 m. Mud, coral and Lithothamnion.

Stat. 315. East of Sailus Besar, Paternoster Islands. Up to 36 m. Coral and Lithothamnion.

Stat. 322. 1 ' ,, mile south of Tandjong Lajar, South coast of Bawean Island. 32 m. Coral.

In 11. Koh Chang 1. c. I described the above plant as a form of Lithothamnion siamense
and remarked: "I have been in doubt whether f. simulans ought not perhaps to be considered
as a separate species, but as I have seen only a solitary well developed specimen of this form,
I do not venture1) to keep it distinct, especially since the material also of f. minuta is scarce
and some of the specimens no doubt are somewhat stunted". The material in hand from the

t «ill appear from the connexion 1. c. p. 30, 1. 19 from below "venture" is to be read instead of the word "hesi:

17

Siboga expedition shows, however, that the two forms mentioned represent different species,
and that the tvpical specimen from the Gulf of Siam has been young. The plant assumes
thicker and more extensive crusts, vvhich however are rather varying, and the conceptacles ot
sporangia generally prove to be larger than in the specimen mentioned. I therefore place it
here as an independent species.

In a young stage the plant clings closely and rather firmly to the substratum. Even
when advancecl, it may be fairly closely adherent, though it is in part gradually detaching
itself. On a large and even substratum the crust becomes rather smooth (PI. I, hg. 24). If the
substratum is small and branched, it certainly conforms fairly to it in shape, but is more
loosely attached to it or partly growing quite freely. PI. I, fig. 25 shows a form growing
on corals, partly loosely encompassing the branches of the coral, partly even distended between
the latter. This form comes also nearest to the typical one. If the plant is charged with
extraneous objects, new crusts are formed above, which give rise to irregular or branchlike
formations. Cp. PI. I, fig. 21, representing a form intermediate between f. typica and f. crispescens,
however most closely allied to the latter This form is distinguished by its thallus being more
or less lamellate and crispate, or plicate-leaf-like (PI. I, fig. 22 — 23). It corresponds partly to
Lithoihamnion lJ/ii/if>pü f. crispata, partly to L. Engelhar tii f. imbricata and f. pseudocrispata*),
with which forms it has partially much in co'mmon, though distinctly separated particularly as
to the conceptacles of sporangia. The form seems peculiarly to rise, when the substratum is
irreeular or of less firm consistency.

The species is attached to different objects, partly to shells of molluscs, partly and
particularly to corals, calcareous sediment or other calcareous algae. Thus the specimen pictured
in pi. I, fig. 24 has coalesced with Archaeolilhotkamnion crxtliracum. which it partly covers.
A small portion of the last mentioned alga is to be seen at the bottom of the picture. The
same specimen is also quite covering a young Goniolithon Reinboldi. It sometimes anastomoses
with Lithoihamnion fragilissimum, and it even occurs on rather fine-
branched species, as Lithophyllum moluccense and L. Bamleri, but
is here less typical. So also when it sticks to Archaeolithothamnion
timorense. The plant is thus less dependent on the substratum than
several other species. ^^^^mp.ju^^.

Sections of the crust show that the hypothallium is frequently ^i l±;iiHH^(f;ééf(i£^$0iiiL
marked coaxial. The cells of this layer are 12 — iS ij., or now and ' v

then up to 22 p.. long by a breadth of 6 — 1 2 <j.. The perithallium is Fig. 7. LitkothamnhnsimuiansSoA.
slightlv developed, especially in such parts of the crust which are A- Vertical section of a crust vather

ö J c i. j 1 adherent to the substratum; X 72- B.

growing more or less freely. Sometimes it may, however, be rather Radial section of a part of a crust
vigorous and even stratified. The cells are square, often with rounded e

corners, 5 — 7 u. in diameter, or slightly vertically elongated, in the latter case here and there
up to 1 2 ij., long; tig. 7.

1) This form from Tasmania, referred by Hevdrich to Lithothamn, -iim Hauck, I have classified in Melob. Arb. Heydr.

p. 27 as a denominated form of L. Engelliartii. It is, however. possible that it only represents a more irregular L. Engelhartii f. imbricata.
I have only seen a sterile specimen.

SIBOGA-EXPEDIT1E LXt. 3

[8

The reproductive organs hitherto known are sporangia and cystocarps which occur on
different specimens. The tonner are subhemispheric, however now and then at length somewhat
flattened in Üic central parts, 500(400) — 700 ;/. in diameter, when seen from above. Tliey
are sometimes to be found overgrown by a new-formed layer of tissue extending itself over
the roof betere the Iatter gets dissolved. The sporangia are four-parted, about 140 — 160 by
60 — 80 •).. The conceptacles of cystocarps are conical, frequently low, but rather acute,
600(500) — 800 ■/.. in diameter when seen from the surface. The species is furnished with the
said organs, especially sporangia, in the months of January, February, June, November and
December, mostly however rather scantily.

As to the relation to other species is to be observed that young specimens are rather
approaching to L. siamense\ but at an older stage the species is much differing from the
Iatter. It approaches nearly to L. fragilissimuiii, from which it sometimes is hardly distinguish-
able in a sterile stage, whereas on the other hand fertile specimens are easily recognized. In
habit it also comes near to certain forms of L. Engelhartii. On the other hand it is much
like L. funafutiense f. purpurascens^ particularly as regards the conceptacles of sporangia.
Although the material of the Iatter form is very scanty, I now suppose it ought to be considered
as a distinct species.

Occur rence: There are only a couple of specimens from each of the stations above
quoted, from the station 125 even only a sterile and uncertain specimen. This fact seems to be
indicative that the species occurs but sparingly over a great part of the East-Indian Archipelago.

A.rea: North Pacific: The Gulf of Siam.

5. Lithothamnion prolifer Fosl. mscr. Fig. 8; PI. I, fig. 17 — 20.

Thallus crustlike, irregularly extended, 0,5 — 1 mm. thick, scantily prolificating, with
irregular excrescences. Conceptacles of sporangia subprominent, at length rather flattened,
800 — 11 00 ij., in diameter.

Stat. 78. Lumu-Lumu-shoal, Borneo-bank. 34 m. Coral and coralsand.

Stat. 81. Pulu Sebangkatan, Borneo-bank. 34 m. Coral bottom and Lithothamnion.

The plant forms irregular crusts on other calcareous algae, or different hard objects. The
crust is 0,5 — 1 mm. thick and of a hard consistency. In a young stage the plant clings to the
substratum and fairly conforms to it in shape, but afterwards it detaches itself little by little.
In the downwards turned detached part of the crust it is distinctly concentric zonate, partly
even forming rather deep striae, or furrowed. It is scantily prolificating, and besides it develops
irregular or short branchlike excrescences. The specimens in hand show several such excrescences,
which, however, are only owing to the form of the substratum or to foreign bodies having
attached themselves to the plant or its substratum, and then having been overgrown with
new-formed tissue. The excrescences seem to occur rather scarcely in typically developed
PI. I, tig. 19 — 20). The specimen pictured in pi. I, fig. 17 shows branchlike
formations which are apparently large and vigorous, still mainly owing their rise to the form

19

of the substratum. This specimen, however, is on the whole somewhat differing, and only bears
a solitary conceptacle of sporangia which is but slightly developed.

On a radial section of the crust the hypothallium appears vigorously developed, but the
cells are smaller, than usual vvith species whose crusts develop themselves
partially freely. The central cells are mostly but slightly elongated, 8 — 15
or up to iSu.. long and 6 — 1 2 ;;.. broad and sends forth upwards rather
long bent dichotomous rows of cells, but downwards short ones. It is,
however, to be noticed that for the time I only possess somewhat oblique
sections which are not properly representative of the structure of the
hypothallium. The perithallium is also vigorously developed and either
not stratified or indistinctly so to judge from the material examined. The Lithothamnion proiif er YoA.

,. r , . . i-i 11 i-ii Vcrtical section of the

cells of this layer are square, mostly with rounded corners, slightly perithaiiic layer- x 72.
elongated horizontally or vertically, 4 — 7 u.. in diameter, or more definitely

vertically elongated, 7 — 10 or sometimes up to 12 u.. long. The walls of the cells are generally
fairly thick (Fig. 8).

The reproductive organs known are sporangia. The conceptacles are subprominent or
in part but very slightly raised above the surface of the plant, finally rather flattened, 800 — 1 100 u.
in diameter, when seen from above. They are partly not sharply defined. The roof is intersected
with about 100 muciferous canals. The sporangia are four-parted, 150 — 200 by 80 — 100 p..,
with enduring partition walls. The plant is sparingly furnished with these organs in the
month of June.

This species is found associated with Lithothamnion fragilissimum and L. simulans, and
partly shares substratum with the former. Young specimens resemble L. fragilissimum a little
in habit. As to reproductive organs hitherto known it comes very near to L. Philippi. On
the other hand the plant is next akin to L. funafutionse, from which species, however, it is
distinguished by mostly larger conceptacles and by detaching itself from the substratum. Besides
there is some divergency as to structure.

Occurrence: The species seems to be of no frequent occurrence at the stations
mentioned above. It was found best developed at the station 81.

c. R a m o sa.

6. Lithothamnion. fruticulositm (Kütz.) Fosl. Fig. 9, ex parte.

List of Lith. p. 6 (non Norw. Lith.); Spongites fruticulosa Kütz. Polyp. Calcif. p. 33 et Tab.
Phyc. 19, t. 99. Syn. L. fasciculatum /3 fruticulosum Hauck, Meeresalg. p. 274.

f. pteridoides Fosl. mscr. PI. II, fig. 1 — 4.

Resembling f. crassiuscula except that the conceptacles of cystocarps are less distinctly
defined, lower and with a more distinct acute tip.
f. clavulata Fosl. PI. II, fig. 5 — 9.

Corall. in SCHMIDT, Fl. Koh Chang, p. 17.

20

Stat. (.4. Kambaragi Bay, Tanah Djampeah. Up to 32 m. Coral, coralsand.
Stat. 66. Bank between islands of Bahuluwang and Tambolungan, south of Saleyer. 8 to iom.
I u ad coral, I talimeda, Lithothamnion.

Stat. 78. Lumu-Lumu-shoal, Bi 10 bank. 34 m. Coral and coralsand.

Stat. ui. Muaras-reef, inner side: East coast of Borneo.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 1 2 m. Lithothamnion-bottom.

Sand and coral.

Stat. io6. Off Kapul Island, Sulu Archipelago. 13 m. Coral-bottom.

Stat. 109. Off Pulu Tongkil, Sulu Archipelago. 13 m. Lithothamnion-bottom.

Siat. [23. North Bay. Biaru Island. 36 — 27 m. Stone and Lithothamnion-bottom.

Stat. 125. Off Sawan, Siau Island. 27 m. Stone and some Lithothamnion.

Stat. 133. Lirung. Salibabu Island. Reef.

Stat. 136. Ternate-anchorage. 23 m. Mud and stone.

Stat. 155. Piapis Bay (Telok Sapira) Northwest coast of Waigeu Island. Reef.

Stat. 240. Banda-anchorage. From 9 — 45 m. Lithothamnion bank in 18 — 36 m.

Stat. 250. Kilsuin, West-coast of Kur Island. 20 — 45 m. Coral and Lithothamnion.

Stat. 258. Tual-anchorage, Kei Islands. 22 m. Lithothamnion, sand and coral.

Stat. 277. Kulewatti (Sollot) Bay, Dammer Island. Reef.

St.it. 282. Between Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and
Lithothamnion.

Stat. 2yo,. Buka or Cyrus Bay, South-coast of Rotti Island. 34 m. Mud, coral and Litho-
thamnion.

Stat. 322. 1 ' '., mile south of Tandjong Lajar, South-coast of Bawean Island. 32 m. Coral.

In Fl. Koh Chang 1. c. I referred some small and not well developed specimens from
the Gulf of Siam to the above species, the only specimens of this species then known from the
Pacific. They resemble delicate specimens of a form from the Mediterranean and the Adriatic
which I proposed to name f. clavulata.

I have afterwards referred a few other and similar specimens from the Maldives to the
same form of the species '). As I then remarked, the latter also agree in habit ancl structure
witli somewhat stunted growths from the Mediterranean and the Adriatic. The only difterence
appears to be that the conceptacles of cystocarps are more conical or more acute than generally
seen in specimens from the last named regions. As remarked, there seems to be no reason to
consider this aio-a as a form so differing- from the Mediterranean and Adriatic f. clavulata as
to be specially denominatecl. Besides, in delicate branching specimens from the said regions the
cystocarpic conceptacles sometimes may be even as conical as in those from the Maldives,
especially when they are developed in the tip of the branches, and then somewhat differing
from those in a coarse f. crassiuscula.

The mentioned collection from the Maldives also contained some other and coarser
specimens which partly coincide with L. fruticulosum f. crassiuscula. (1. c. pi. XXIV, fig. 2),
partly appear to be somewhat differing (1. c. pi. XXIV, fig. 3). I have however been unable
to distinguish even the latter specimen from the form crassiuscula, ancl therefore I consider it
only to represent an extreme variation of the said form. Cp. 1. c. p. 465.

1) M. Fosme. The Lithothamnia of the Maldives and Laccadives. — The Fauna and Geography of the Maldive and Laccadive
1 1 [902 Pag. 464, PI. XXIV — XXV.

2 I

In another paper ') I have pictured three specimens from the south coast of Australia
together with specimens from the Mediterranean and the Adriatic. Also the Australian form
seems to be identic with f. crassiuscula. Cp. 1. c. pi. I, fig. 4 — 9. It appears to be abundant
in certain localities in Port Philip Bay, Victoria. It is richly provided with conceptacles of
sporangia. The other reproductive organs are however quite wanting in a rather large number
of specimens examined, while in Adriatic specimens cystocarps are rather more common than
sporangia, but on the other hand antheridia are very scarce.

I have shown in the paper mentioned that the Mediterranean as well as especially
the • Adriatic forms of the present species are much varying in habit and partly even in
structure too. So also in regard to the forms from the Maldives. As remarked 1. c, this appears
partly to depend on local conditions, partly on the fact that L. fruticulosum is as a rule
more burdened with extraneous objects than most other Lithothamnia, and therefore more
irregularly developed.

In the collection in hand from the Siboga Expeclition occur a number of specimens
which, as far as ascertainable, are identic with or
very nearly related to L. fruticulosum. However,
many of them have been dead or water worn when
collected, and germinating plants of one or other
species have often attached themselves to such
specimens and are gradually covering them more
or less. But above all, the specimens are much
burdened with extraneous objects, and, therefore,
they are more or less stunted or irregularly deve-
loped. The nature of the bottom seems in many
cases to have been unfit for the development of
the species, i. e. it has been rather loose, and there
has been too little of hard objects for fastening
the plant, which is often burdened also with a
mixture of clay and sand. It is frequently attached
to fragments of corals or dead calcareous algae,
anastomosed with or attached to other living species,
but on the other hand also often serving as a
substratum to other and especially crustlike Litho-
thamnia, as well as numerous Squamariaceae.
I have met with up to seven or eight species
anastomosed in one piece not more than 2 — 3 cm. in diameter, or one species fully or in
part covering the other. Fig. 9 shows samples of such anastomosed species. In addition there
are specimens which are partly covered with young crusts of other species, closely and firmly

Fig. 9. Anastomosed or partly overgrown specimens of:
.1. (Stat. 282).

Lithothamnion fruticulosum (Kütz.) Fosl.
Lithothamnion australe Fosl. forma.
Lithophyllum Okamurai Fosl.?
Lithophyllum sp.
Squamariacea.
(Corals, bryozoa, worms).
B. (Stat. 109).

Lithothamnion fruticulosum (Kütz.) Fosl.

Lithothamnion australe Fosl. forma.

IJthothamnion sp.

Lithophyllum Okamurai Fosl.?

Lithophyllum sp.

Mastophora vielobesioides Fosl.

Pcyssonnelia sp.

Another Squamariacea.

(Corals, bryozoa, worms) \ nat. size.

1) M. FOSLIE. Die Lithothamnien des Adriatischen Meeres nnd Marokkos, — Wissensch. Meeresuntersuchungen, Abt. Helgo-
land. (In press.).

22

adhering and provided with reproductive organs, which on a cursory examination might be
considered as a new thickening layer of the thallus of the species in question. The specimens
in this collection are even more burdened with lower animals than those from the other area
mentioned, especially small corals, Bryozoa and numerous worms as wel! as boring algae,
penetrating and disturbing the tissue. Thus, an intense struggle for existence is to be seen
partly between the algae mutually, partly and especially between plant and animal. There is
scarcely a single specimen not being burdened with such objects of one kind or other. Numerous
new hypothallic layers are to be found in sections not only of the crustlike part of a specimen,
but very often in branching forms too, and as a rule owing to extraneous objects. In other
sections is often to be seen a conglomerate of divers calcareous algae (especially Lithothamnia
as well as Squamariaceae) and animals, one more or less covering the other, sometimes even
forming almost regularly alternating layers.

It is extremely difncult to clefine such stunted specimens with certainty, in several cases
quite impossible. Besides, the very greatest number of specimens apparently belonging to the
species in question are sterile, in some cases provided with sporangia, seldom are to be seen
a few conceptacles of cystocarps, and not a single one with antheridia. Such specimens scantily
fertile have been collected in the months of May, June, July and December. Therefore, it must
be the subject of future investigations to show to what extent the forms here mentioned really
belong to the species to which the)- are now referred. It may also be noticed that several
specimens come near to the following species, L. australe, ancl a sure delimitation can not be
founded on the materials present, although, on the other hand, there is no doubt that the
said species in fact are specifically distinct.

The form named f. pteridoides is partly resembling f. crassiuscula in habit, partly a
little differing from it (PI. II, fig'. i — 4). The specimens of this form are, however, mostly young
and not well developed. The crust is sometimes thicker than in typical specimens of f. crassiuscula,
and on the other hand the branches are shorter, but this is owing to extraneous objects being
covered up or other species attaching themselves to the crust and being overgrown. The
consequence of this is a later and less vigorous development of the branches, such as to be
seen in numerous specimens of L. fruticulosum from the Adriatic. As to structure the form
in the main agrees with the other forms of the species. So also as regards the conceptacles
of sporangia, which, however, are sometimes more confluent than in the other forms. The
chief difference which is now to be ascertained is in the conceptacles of cystocarps. 1 have
certainly seen but a few ones in young specimens, and — as formerly remarked — even these
organs are somewhat varying in this species. The said conceptacles in f. pteridoides are of
aboul the same diameter as in f. crassiuscula. Seen from above they are, however, less distinctly
defined, lower, with a more acute ancl distinct tip than in the latter form, the conceptacles of
which are frequently low conical, though more arched, or approaching a subhemispheric-conical
shape, and the tip indistinct and apparently sooner falling away. It may however be observed
as mentioned in Lithoth. Adriat. Meer., the conceptacles of cystocarps in f. crassiuscula
may sometimes be rather high and almost superficial, sometimes low, but little prominent, and
that when emptied they often become overgrown by new-formed tissue.

The specimens which I have referred to f. clavulata partly resemble those from the
Gulf of Siam and the Maldives mentioned above, e. g. the branches are more delicate and
often less developed than in typical specimens of this form (PI. II, fig. 9). Other specimens
resemble typical f. clavulata in habit and strncture, or they are in part somewhat coarser
(PI. II, fig. 5 — 8). As remarked in Lithoth. Adriat. Meer. this form runs into f. crassiuscula
without any proper limit. Cp. 1. c. pi. I, fig. 4 — 6 and 10 — 15. It apparently stands in alrnost
the same relation to f. pteridoides in the present collection, but cannot be elucidated till better
material is known. The conceptacles of sporangia resemble those in Adriatic specimens, and
as to the conceptacles of cystocarps, which are very scarce, I refer to my remarks above
under the Maldive form.

In Lithoth. Adriat. Meer. I classified a delicate and sparingly branching form under the
name of f. soluta. Cp. 1. c. pi. I, fig. 18 — $j- I have 1. c. in detail compared it with stunted
forms of other species arising from external causes or certain local conditions. As observed 1. c,
this form perhaps represents a species being in evolution, and as yet too little differentiated
to be looked upon as independent. Judging from the material in hand, however, this doesn't
seem to be the case ; it seems to be a dwarf-form which has risen from local conditions. There
is no certain limit to be drawn between it and f. clavulata. As also observed 1. c, the form
certainly diverges a great deal from the typical L. fruticulosum. Still the divergency is but
little greater than, for instance, that of L. calcareum f. subsimplex from f. valida of the same
species, viewed apart from quite monstrous forms.

In the collection in hand occur sorae specimens which stand in nearly the same relation
to f. clavulata as the Adriatic f. soluta mentioned. The)- have probably risen in the same way
as the latter, e. g. selfloosening or broken branches of delicate specimens continuing their
growth lying loose on the bottom. In habit these specimens, however, are less diverging from
a small or stunted f. clavulata than the majority of the Adriatic specimens of f. soluta. On
the other hand they are sometimes so near approaching to L. australe f. miuutula that the
line can hardly be drawn with certainty, the less so, because these forms in the collection in
hand are represented, without any exception, by sterile specimens.

Occurrence: In the sense taken above the species appears to be commonly distributed
over a great part of the investigated area. However, trom most ot the stations occur only a
few specimens, and from some stations only one has been brought home ; the greatest number
havs been brought home from stat. 109 and 240. A considerable bank of Lithothamnia at
stat. 66, between the islands of Bahuluwang and Tambolungan, south of Saleyer, is mentioned
by M. Weber x) and Madame A. YVeber 2). This bank is not unlikely, at least in part, built
up with the species in question. There occurs, however, but a solitary small specimen from this
place. The specimen is in the lower part composed of a conglomerate of tubes of worms,
Bryozoa and other animals together with dead and living calcareous algae. Among the latter
is Mastophora melobesioides \ covered partly with lower animals, partly Peyssonnelia sp. and the

1) Siboga-Expeditie I. Max Weber. [ntroduction et Description de L'Expédition. p. 38.

1) A. Weber-van Bosse. Études sur les Algues de I. 'Archipel Malaisien. — Annales du Jardin Botanique de Buitenzen-.
Ser. 2, Vul. II, Leide igoi, p. 131.

24

species in question. The upper layers of this irregularly crustlike piece with wart-like or short
1 ranch-like ex< i s are formed by a very smal] crust oi Lithothamnion sp. as well as

Lithophyllum sp., but in the main b) a plant not well developed which seems to belong to
f. pteridoides of the species in question.

Ar ca: North Atlantic: The southern part of the Atlantic coast of Europa (f. crassius-
ai/a, rare), Mediterranean (f. ramulosa, i. crassiuscula, f. clavulata), Adriatic (the same forms
as well as f. solutd)\ Iftdic: The Red Sea (not quite certain), the Maldives (f. crassiuscula^
f. clavulata), and the south coast of Anstralia (f. crassiusculd)\ North Pacific: The Gulf of
Siam (f. clavulatd).

-. Lithothamnion australe Fosl. Fig. 10.

L. coralloides f. australis Fosl. Norw. Lith. p. 62, ex parte; New or crit. Lith. p. 8, fig. 6 — 7;
L. australe Fosl. Rev. Syst. Surv. Melob. p. 13.

t'. tualensis Fosl. mscr. Fig. 11; PI. II, fig. 10 — 17.

Thallus irregularly branching, shrub-like, 1 — 2 cm. in diameter; branches short and more
or less spreading, often curved and knotty, about 1,5 mm. thick. Conceptacles of sporangia
slightlv prominent, 400 — 550 u.. in diameter.

f. ubiana Fosl. mscr. PI. II, fig. 18 — 24.

Thallus forming roundish nodules about 2 cm. in diameter, with rather straight and
subfastigiate branches 1 — 2 mm. thick, partly rather crowded, partly somewhat spreading.

f. brachiata Fosl. mscr. PI. II, fig. 25 — 38.

Thallus irregularly, rather sparingly branched, frequently with spreading and diverging
branches 1 — 2 mm. in diameter. Conceptacles of sporangia 400 — 600 ti. in diameter.

f. minutula Fosl. mscr. PI. II, fig. 39 — 62.

Thallus more or less branching, roundish or irregular, frequently 0,5 — 1 cm. in diameter;
branches partly rather crowded, partly more or less spreading, frequently about 1 mm. thick.

Stat. 79 . Pulu Kabala-dua, Borneo-bank. 22 m. Coralsand.

Stat. 86. Dongala, Palos Bay, Celebes. Reef.

Stat. 91. Muaras-reef, inner side : East coast of Borneo.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion-

bottom. Saml and coral.
Stat. 99. Off North Ubian. 16 — 23 m. Lithothamnion-bottom.
Stat. 106. Off Kapul Island, Sulu Archipelago. 13 m. Coral-bottom.
Stat. ioy. Off Pulu Tongkil, Sulu Archipelago. 13 m. Lithothamnion-bottom.
Stat. 149. Fau-anchorage, West coast of Gebé Island. 31 m. Coral.
Stat. 240. Banda-anchorage. Lithothamnion bank in 18 — 36 m.
Stat. 258. Tual-anchorage, Kei Island. 22 m. Lithothamnion, sand and coral.
St.it. 277. Kulewatti (Sollot) Bay, Dammer Island.
St.it. 282. Betvveen Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and

I ,ithothamnion.
Stat. 299. Buka- or Cyrus Bay, South coast of Rotti Island. 34 m. Mud, coral and Lithothamnion.
Stat. 303. Haingsisi, Samau Island. Reef.
Stat. 315. Easl of Sailus Besar, Paternoster Islands. Up to 36 m. Sand, coral and mud.

25

In Norw. Lithoth. ]) I placed among Lithothamnion coralloides Crn., then taken in a
rather wide sense, a form which I named f. australis, and which has afterwards been partly
ranked with Lithothamnion calcareum (Pall.) Aresch. The knowledge of L. coralloides, as well
as of L. calcareum, was then insufficiënt, and decided fertile specimens were not in hand. As
to the mutual relation of the tvvo species mentioned, I refer to Lithoth. Adriat. Meer, where
I have presumed that L. coralloides includes partly L. calcareum, partly L. fruticulosum,
in the sense taken by Crouan ), however, in the main only L. calcareum. Simultaneously with
Norw. Lithoth. I published a paper where I classified f. australis from the Gulf of California8).
This form constitutes the basis of Lithothamnion australe, entered into Rev. Syst. Surv.
Melob. '). The said form from the Gulf of California is here named f. amcricana.

The question whether the forms from the East-Indian Archipelago, named above, actually
belong to the same species as the American form is not to be settled at present, the material
partly in the main consisting of sterile specimens, partly being too small. There are several
specimens in the collection in hand which are in almost full conformity with the said American
form. It is, however, to be observed that specimens of rather widely different species may
often resemble each other in habit as well as in structure, and accordingly cannot be defined
with certainty, when sterile. Besides it cannot — as yet — be ascertained to what extent
there is a connection — as to this group of the algal flora — between the East-Indian
Archipelago and the Pacific coast of America. I have, below, in treating of Lithothamnion
ertiöesceus, mentioned some instances that seem to be inclicative that the distribution of some
species has a very wide range. At any rate, however, I do not find reason for the present to
classify the forms in question as independent species (one or more), but I provisionally place
them with L. australe.

The form tualcnsis mostly forms small shrub-like masses which appear to have developed
themselves freely at the bottom, and generally are i — 2 or up to 3 cm. in diameter. It is
subdichotomously or — oftener — repeatedly irregularly branched. The branches are partly
somewhat crowded, partly — and more commonly — rather spreading, more or less curved,
with short axes, frequently about 1,5 mm. thick, either slightly tapering or now and then
slightly enlarged towards the tip, with rounded ends (PI. II, fig. 10 — 17). This form comes
near to L. fruticulosum, and particularly old specimens approach to certain forms of this
species. Future investigations will ascertain the relations between them. I here place the form
under L. australe, to which it — judging from the material in hand — is most closely allied.
Among the specimens brought home a comparatively great number prove to have been wholly
or partially dead, when collected. On these were often developed crusts of other calcareous
algae, or new crusts of the same form, together with animals and other extraneous objects,
had caused a very irregular development. Such specimens frequently show an appearance rather

1) M. Foslie. The Norwegian Fovms of Lithothamnion. — Det Kgl. Norske Videnskabers Selskabs Skrifter 1S94. Trondhjem
1S95, p. 62.

2) P. L. et H. M. Crouan. Florule de Finistère. Paris, 1867, p. 151.

3) M. Foslie. New or critical Lithothamnia. — Det Kgl. Norske Vidensk. Selsk. Skrifter 1895, Trondhjem 1895, p. 8.

4) M. Foslie. Revised Systematical Survey of the Melobesieae. — Det Kgl. Norske Vidensk. Selsk. Skrifter, 1900, N". 5,
Trondhjem 1900, p. 13.

SIBOGA-EXPEDITIE LXI. 4

20

differing from specimens typically developed. Selfloosening or — I presume - - more usually
broken branches often seem to continue growing.

form ubiaua constitutes roundish nodules which also seem to have been freely
developed ai the bottom. It is generally 2 cm. in diameter and is distinguished by rather
rivet, subfastigiate and straight branches, which, however, are partly rather crowded and knotty,
partly - and more frequently somewhat spreading. The branches are, as a rule, longer

than in the form tualensis and 1 — 2 mm. thick, partly tapering, partly — but more rarely —
vvith somewhat thickened ends (PI. II, tïg\ iS — 24). Otherwise this form runs into the following
one, and there is no distinct line between them.

The form brachiata is distinguished by spreading branches, more or less curved, sometimes
rather elongated, of about the same diameter as in the form ubia?ia. There are partly delicate
and sparingly branched specimens of this form. At any rate some of these specimens no doubt
have risen from broken branches, either of young specimens, or of the last order of older
specimens, continuing their growth. Thus there are such delicate specimens where one or more
surfaces of fracture are covered with new-formed tissue.

The form of the species apparently most divergent is f. minutula. It is 0,5 — 1 cm. in
diameter, partly very sparingly branched with spreading and frequently short branches, partly
and oftener the branches are rather crowded, and sometimes it assumes an almost subfastieiate

o

shape. The branches are either terete and slightly attenuating, or cylindrical, or now and then
slightlv thickened towards the rounded ends, about 1 mm. thick. There is a rather ereat
number of specimens of this form from two stations, and they are of obvious conformity.
Cp. pi. II, hg. 39 — 50 from stat. 109, and hg. 51 — 62 from stat. 2S2. As I mentioned in
treating of L. fruticulosum, the form was sparingly mixed with specimens which seem to belong
to f. clavulata of the latter species or to be closely connected with f. soluta, and which are
on the whole hardly distinguishable. The other specimens, however, come nearest to L. australe
in the sense here taken, though they are also mixed up with fragments or small specimens
which may behing to another form not to be distinguished with certainty. There are several
specimens showing a surface of fracture more or less distinct and covered with new-formed
tissue. I take, however, f. minutula in the sense indicated by the specimens pictured. The
form approaches partly to L. australe f. tualensis, partly to f. brachiata, and it might be
supposed to have risen from both these forms and be considered as stunted specimens now of
one, now of the other. But this is not probable. It is perhaps more likely representing an
independent species, though all the specimens being sterile, it cannot for the time be settled.
A sure definition of this form (as well as of the others) will only be realised by a number of
fertile specimens or by investigations on the spot whether, for instance, the form constitutes
small and in the main independent banks, as the northern species Lithothamnion norvegicum
almost always sterile, which it somehow recalls, or if it only occurs together with other forms
of the species.

What I remarked in treating of L. fruticulosum as to foreign bodies causing an irregular
development, also refers to the forms ranked with the species in question, though certainly to
a less extent. There are numerous specimens, particularly of the three forms first mentioned,

in which small corals are found having attached themselves and then having been overgrown,

o ooi

as a series of Bryozoa, attaching themselves to the plant and often overgrown by new layers
of accretions, bring about an abnormal thickening of the branches of the plant. If the plant
comes into contact with a foreign body, there may even be developed a rather extensive crust,
as in several other species in which such a development is not normal. As is the case with
L. fruticulosum, L. calcareum and others, it will equally be found with L. australe that
germinating plants attach themselves partly to living specimens, partly to dead ones of the
same species. If simultaneously or later on Bryozoa or other organisms attach themselves to
such specimens, a more irregular development will be the result. In cases of recurrence, coarse
forms will come into existence, greatly differing from typical specimens in appearance. Such a
peculiar monstrous form of L. calcareum I have described and pictured in Lithoth. Adriat. Meer.
Cp. 1. c. pi. 3, fig. i — 7. In the collection in question there are some specimens which are most
closely allied to f. ubiana, but coarser than typical specimens of the. form, and approaching
very nearly to certain forms of L. fruticulostim. These specimens at all events seem to have
partly risen in the said way. Others in which two or more specimens seem to have been melted
together, finally assume a shape little typical. If several species are growing gregarious in the
same locality, and germs of such ones attach themselves to a different species or to a common
substratum, complexes of species will rise like those under L. fritticulosum mentioned; fig. 9.

As to the structure of this species, it is to be observed that also in this respect the
line distinguishing it from other species near of kin is hardly to
be drawn with certainty. Indeed it is in part impossible. This is
the case in the first place with branched forms of L. fruticu-
losum. On the other hand it is all but impossible to draw the
line in the said respect between L. australe and L. pulchrum.
Besides the forms of L. australe show mutually some variety,
as specimens which are quite like each other in habit, are
sometimes rather divergent in structure, and even in one and
the same specimen the size of the cells is often rather varying.
This variation in structure often also corresponds with a varying
consistency, which in some specimens is looser than in other
ones. A section of L. australe f. amerieaua, parallel with the Fis- I0-

Lithotkamnion australe Fosl. f. amcricana.

longitudinal axis of a branch, exhibits distinct cup-shaped layers part of amedian section of abranch 5x72.
of tissue. The cells of the medullary hypothallium are 14 — 22 a.

long and 7 — 14 u.. broad, with somewhat rounded corners. The perithallic cells are 8 — 18 u..
long and 7 — 12 u. broad, or here and there roundish and only ó — S u.. in diameter; fiV. 10.
A similar section of f. tualensis shows a structure in the main coinciding with that in the latter
form. The medullary hypothallic cells are 12 — 22 y.., frequently 14 — 18, seldom up to 25 >j..
long by a breadth of 7 — 12 p.., often with rather rounded corners. The perithallic cells are
rectangular, 9 — 18 u.. long by 8 — 10 u.., now and then however roundish; fig. 11. The cells
are in general a little shorter, proportionately to the breadth, in this form than in other forms,
and the walls of the cells are often thicker. The structure of f. ubiana is rather varying.

2 8

II.

I osl. f. tualemis.

Part of a n I a branch; X 72.

I'hc cells are partly of the same of the last mentioned form, partly throughout

maller, partly - and as a genera) thing — ■ longer,
the hypothallic ones sometimes being up to 30 <j.. long. Such
is also the case with f. brac h ia la . the cells of which are
also partly a little longer throughout than those off. tualensis,
partly, however, quite like the latter in structure. On the
other hand the structure of f. brachiata is even approaching
to that of Lithothamnion erubescens with more regular cells
and thinner walls, than usual in any form of L. australe.
Also the structure of f. minutula is rather varying and shows
conformity partly with that of f. tualensis, partly with that
of f. brachiata.

As to structure, L. a?tstrale thus presents about the
same variety as L. fruticulosum, though perhaps in a less
degree, as in the latter a considerable difference is prevailing
between specimens crustily spread with short branches and branched specimens which have lain

free on the bottom. The cells of the latter are always longer than
those of the former, and in the Adriatic f. ramulosa always lying
free on the bottom the hypothallic cells are generally even longer
than in a much branchine f. crassiuscula. and can attain a leneth of
up to about 30 u. L. australe is, therefore, most closely connected
with branched forms of the said species.

The varying habit and irregular development above mentioned
due to extraneous objects, of course appear also in sections of the
plant. But, on the other hand, irregular developments in the structure
are often to be seen without any apparent external alteration, being
however also in such cases caused by damage in some way or
other, or by delicate extraneous objects attached to the plant (Fig. 1 2).
The latter represents a small part of a median section of a branch
of a rather coarse specimen which comes nearest to f. ubiana. Here
a part of the branch has been damaged or burdened with delicate
foreign bodies. A new and even marked coaxial hypothallium is
then formeel, resembling the said layer of tissue in several crustlike
species. This hypothallium issues from the surrounding undamaged
parts of the perithallic layer towards the centre of the damaged
part. When this part has been covered with the new hypothallium,
the latter sends forth perithallic rows in connection with new thickening

Fosl. f. 1 ■ , r 1 1 1 1-11 -1 1

Qch layers m the parts 01 the branch which have remained unnurt.

As mentioned above, I have met with very few reproductive
organs in L. australe. Thus only a couple of specimens off. tualensis
i with a few conceptacles of sporangia. These specimens were taken in the month

men, with a new-
ithallium covi

■ 72.

of December. The conceptacles are but little prominent and not sharply defined, when seen from
above, 400 — 550 [j.. in diameter. They appear, however, not to be fully developed. The sporangia
are unknown. I have seen a specimen of f. brachiata from stat. 303 (pi. II, fig. 25) furnished
with a couple of conceptacles, these too not sharply defined and about 600 — 700 u.. in diameter.
Another specimen from stat. 99 (pi. II, fig. 31) collected in June, is the only one of this species
that shows a fairly rich development of the said organs. Here too most of them are not fully
developed and partly quite young. The conceptacles are a little more prominent than tnose
of f. tualensis and very like those of the specimen last mentioned, partly however not sharply
defined. They are partly circular in circumference, partly approaching an oblong shape, when
seen from above, rather scattered here and there in the branches, although two or three some-
times may be almost confluent, 400 — óoo u. in diameter. I have not seen the sporangia. In
f. ubiana and f. minutula I have found 110 conceptacles of sporangia. And cystocarps and
antheridia are wanting in all the four forms.

I have also seen specimens from Samoa (Bot. Mus. Hamburg), which are partly connected
with L. australe in the sense here taken, partly approaching to certain forms of L. fruticulosum.
But also these specimens were sterile.

It seems unquestionable to me that f. ubiana and f. brachiata belong to one and the
same species, even though there are specimens which apparently belong to the former, but —
as stated above — on the other hand are nearly approaching to certain forms of L. fruticu-
losum. There are also some specimens, particularly from stat. 2 8 2, which are partly closely
allied to L. australe f. ubiana or f. brachiata, partly approaching to L. fruticulosum f. ramulosa
occurring in the Mediterranean and the Adriatic, or to forms of L. calcareuui occurrine in the

O

said area as well as in the Atlantic Ocean. Among these is found a flabellate and sterile
specimen, about 3. cm. in diameter, which in habit bears a striking resemblance to L. calcarcuin
f. palmatifida. On the other hand it is closely approaching to some of the doubtful, likewise
sterile, specimens above mentioned which are perhaps representative of a species or form as
yet undescribed. It cannot, however, be elucidated till fertile specimens are known. In treating of
f. tualensis and f. minutula I have already suggesteel that these forms are possibly representing
independent species.

In addition to the relations stated to the species already mentioned it is to be observed that
L. australe sometimes obviously approaches to young specimens of L. pulckrum, in habit even
in a higher clegree to young Archacolithothamuion timoreuse. Sometimes it may also be hardly
distinguishable with certainty from young and sterile specimens of L. crubeseens f. haingsisiana.

Occurrence: Forma tualensis is known from stat. 7913, from which, however, only a
single specimen is in hand, which presumably belongs to this form, and from stat. 258, where
the plant seems to form a rather large bank; (cp. the introduction). The form minutula too
is only known from two stations, viz. stat. 109 and 258. The two other forms are taken at
almost all the stations quoted, partly alone, partly and mostly associated with each other and
with other species.

Area: South Pacific: Samoa (not certain); Xorth Pacific: The Gulf of California (f.
au/ericana).

8. Lithothamnion Dickiei Fosl. Fig. 13—14-

New

Lithothamnion polymorphum Dickie, Linn. Soc. Journ. Bot. Vol. XV, Pag. 452.

Lithothamnion imbricatum Dickie, herb.; nee Zan.

Island. 20 111. Coral and sand.

Stat. Ni. 1'ulu Si Borneo-bank. 34 m. Coral bottom and Lithothamnion.

With some doubt 1 ascribe a couple of specimens to this species. The plant is up to
the present unsuffkiently known, and its delimitation is vet hardly to be ascertained. Fig. 13

exhibits a fairly entire specimen and
three fragments which are the types
of the species kept in the British
Museum. The former is about 7.5
by 5,5 cm. in diameter and about
2 cm. thick. It has perhaps been at
first attached to some hard object,
but has afterwards detachecl itself and
continuecl its growth freely at the
bottom. There are also a few frag-
ments kept in Kew Gardens herbarium.
In Suppl. Alg. Tahiti ') the plant is
mentioned as follows: "A calcareous
alga in rounded masses forming the
bottom in 1 o fathoms off Great Island
(Santa Cruz Major), Zambonga. The
dredge came up filled with these
masses". The thallus has a short mainstem which is either flattened and upwards broadening,
or very short and almost terete, and from which issue branch-systems repeatedly subflabellate
or irregular, spreading almost in one plane, one above the other. The branches, more or less
compressed, are about 1,5 or up to 2 mm. thick, and frequently much confluent with each
other. The tip of the branches are partly rounded, partly almost truncate. The specimens or
fragments examined are very scantily furnished with conceptacles of sporangia. The latter are
subprominent, flattened in the central parts, or occasionally almost clisc-shaped, though sometimes
not sharply defined, 350 — 400 u. in diameter when seen from above. The roof is intersected
with about 30 coarse muciferous canals. Sporangia are not known.

From the scanty material hitherto known it is impossible to settle the clirections of

variation vvithin the species. It seems, however, to come nearest to Lithothamnion eruèescens,

next described, though less in liabit than in structure. When I first described the

plant, I had reason to suppose that it was distinct from any other species then known. Nor did

Fig. 13. Lithothamnion Dickiei Fosl.
The type specimens from Tahiti; .-/, seen a little from the side: nat. size.

collected by II. X. MOSELEY of II.M.S. "Challenger", chierly obtained in Torre- Straits. Coasts
1 1. Bot. Vol. W. p. 43-'" 8. Algae collected "ii the Keef- of Tahiti. 9. Supplement to

its relationship now presumecl to the saicl species suggest itself, the type specimens being very
different Cp. fig. 13 and 15. However, in L. Dickiei the more or less plane branch-systems
one above the other differ from those in any form of L. erubescens. Besides, in the type specimen
the cells are more irregular than in the latter, partly smaller, partly however of about the same
size, although as a rule a little broader proportionally to the length.

The specimen in the collection in hand from stat. 61 is attached to a piece of a coral.
In habit it agrees rather well with
L. Dickiei, though it is younger and
the branch-systems are less plane
and the branches in part a little
thinner than in the type specimen
above mentioned (Fig. 14 A). It is
unfortunately sterile, and in regard
to structure it almost coincides with
L. erubescens, but, on the other hand,
it is not likely to be ascribed to
the latter. I must at any rate at
present consider it to be a iorm of
the species in question. The other
specimen, which was brought home from stat. 81, is very stunted and bnrdened with several
extraneous objects (Fig. 14-5"). As to structure, it coincides even more with L. erubescens than
the specimen from stat. 61, but on account of other characters it seems also to have to be
referreel to the species in question. It bears a few conceptacles of sporangia not well developed.

O ecu r re nee: Only one specimen known from each of the stations above quoted.

Area: South Pacific: Tahiti.

Fig. 14 Lithothamnion Dickiei l-'o-l.
A, stat. 61; B, stat. 81; nat. size.

9. Lithothamnion erubescens Fosl. Fig. 15 — 16; PI. III, fig. 20.

New or crit. calc. Alg. p. 9.

Syn. Lithothamnion mamillare Dickie, ex parte; nee Harv.

f. haingsisiana A. Web. et Fosl. Fig. 17; PI. III, fig. 1 — 19, 21 — 22.

Three new Lithoth. p. 4.

f. subflabellata Fosl. mscr. PI. III, fig. 23 — 25.

The branch-systems more or less fiabellate, and the branches compressed or subcompressed.

Stat. 60 and 303. Haingsisi, Samau Island, Timor. Lithothamnion in 3 m. and less. Reef.
Stat. 91. Muaras-reef, inner side: East coast of Borneo. Karang Lintang and Pulu Palabangan

Island. Lithothamnion near the islands.
Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion-bottom.

Sand and coral.
Stat. 149. Fau-anchorage. West coast of Gebé Island. 31 m. Coral.

ila B< si. Reef.
Stal Lithothamnion bank in 18 — 36 ni.

Sta; .-ii Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and

Lithothamni

As mentii (1. c), I have not been able to draw any definite line between

the type of tliis sp from the coast of Brazil and the nearly allied forms occurring in the

Indic and I ans. The former I have named f. americana. Cp. fig. 15, representing

the three specimens of this form hitherto
known, on which was based the original
description of the species. Fig. 1 5 B shows
one of the specimens seen from the side.
A reproduction of the same specimen is to
be found on pi. III, fig. 20, here seen from
above. Also from Bermuda a couple of
specimens are known, which seem to have
to be ascribed to the same form of the
species, though the cells are here partly a
little smaller than in the type specimens.
From the same spot I have classified a form
which I have named f. prostrata l).

Before entering upon the forms occur-
ring in the East-Indian Archipelago which
seem to have to be ascribed to the same
species, I want to make some remarks as
to the geographical distribution of this group
of algae.

Viewed apart from some few species

of the genus JMcloècsia, which are forming

thin crusts on other algae and are widely

distributed in the Atlantic as well as the

Indian Ocean and parts of the Pacific,

the geographical distribution of the other

unarticulated calcareous algae of this tribe is still rather uncertain and unsufficiently explained.

I have carlier pointed out the fact that they are often very local. As efficients of essential

importance are also to be considered the facts that they partly, or rather usually, lay other

claims to the physical quality of the bottom than the majority of other algae, and that their

rrence is very dependent on sets of currents and tide. Owing to these circumstances the

are often much varying, and occur here and there in forms more stunted than perhaps

group of other Florideae, and in forms which may be almost unrecognizable and generally

only occur as sterile or sparingly feftile. This is, however, also to a great extent due to the

Fig. 15. Lithothamnion crubescens f. americana Fosl.

The type specimens from Fernando do Noronha;

B also represented pi. III, hg. 20. seen from above; nat. size.

e. — Det Kgl. Norske Vidensk. Selsk. Skiïfter. 1900. n" 0. Trondhjem 1901, p. 3.

fact that these algae are much harassed by lower animals or on the whole much encumbered
with different extraneous objects, and, therefore, have to carry on a hard struggle for existence.
Consequently these algae must be supposed - - at any rate to some degree — ■ to be more
checked than others in their distribution over larger areas. It is possiblé that also other efficients
make their play contributing to impede the distribution of some species.

However, it must be considered as proved that some of the forms in question are
distributed over great parts of the Pacific and the Indic Oceans, probably also of the Atlantic.
But the limits of this distribution are rather uncertain, and above all it is as yet dimcult to
realize, whether any species occurring in the Malay or Pacific Archipelagoes concerns the
Pacific or Atlantic coasts of America. This seems, however, to be the case with e. g. Lithophyllum
oncodes, which probably also occurs both on the southern coast of California and in the West
Indies. On the other hand, Goniolithon Reinboldi, apparently widely distributed in the Indian
Ocean and common in the Malay Archipelago, has up to the present not been met with
farther to the east, than at the Sandwich Islands in the North Pacific. In addition, a great
deal may be said in favour of the notion that among the species of calcareous algae hitherto
described from different areas, there are some identical ones, or one is likely to be considered
only as a somewhat varying form of the other. Thus Lithophyllum pallescens Fosl. ') (including
L. californiense Heydr.) from the Gulf of California is so closely allied to Lithophyllum
Okamurai, classified in this paper, from the Pacific coast of Japan and several places in the
Malay Archipelago, that the limit is hardry to be drawn. This is also the case with Lithophyllum
decipiens, from the coast of South America 2) and apparently also of the West Indies, in relation
to Lithophyllum Ycurfoi, mentioned below, from Japan and some other places in the Pacific.

For the present, I therefore, find no reason to classify the forms of Lithotha-mnion
eruöescens from the East-Indian Archipelago as an independent species, the less so, as it has
been impossible to me to draw any certain line. On the other hand it is certainly to be
observed that the direction of variation within a species is hardly to be stated only from some
few specimens, as is the case with f. americana of the said species. In fact there are independent
species both of the genus Lithotha-mnion and Lithophyllum which are so closely allied to each
other that their mutual relations only can be fully characterized on the ground of a fairly
considerable material from several places. On the other hand there are instances where, judging
from a scanty material, plants were considered as fairly well defined species which afterwards
a larger collection have shown to be but varying forms, parly even but local forms of one
and the same species. It is further to be observed that the only reproductive organs hitherto
known of L. erttöescens are sporangia. The other reproductions organs are unknown, and it
will probably be the discovery of these organs that will decide the delimitation of the species.

By comparing the specimens of L. eruöescens f haingsisiana pictured in pi. III, fig. i — 19
and 21 — 22, with the specimens of f. americana reproduced in fig. 15 and pi. III, fig. 20,
no difference in habit worth mentioning will urge itself between older and well developed

1) M. Fosi.ie. New ov critical Lithothamnia. — Det Kgl. Korske Vidensk. Selsk. Skrifter 1S95. Trondhjem 1895, p. 4.

2) M. Foslie. Calcareous Algae from Fuegia. — Wissensch. Ergebnisse der Schwedischen Expedition nach dtn Magellans-
landern 1895 — 97. Stockholm 1900, p. 71.

SIBOGA-EXPEDITIE LXI. 5

34

specimens of both forms. 1 li< ■ form haingsisiana is partly attached to other hard objects, as
the specimens of f. americana hitherto known, and then gradually forms a thin crust over the
substratum, partly it develops itself freely, or detached parts of the plant continue their growth
freely, on the bottom. In the latter cases it may - in a young stage - show an appearance
somewhal diverging from older and densely branched specimens (PI. III, fig. i — n). This
form is - for the rest somewhat varying both in habit and in structure. Among the rather

great numb cimens from stat. 303 there are some less typical ones, showing a more

irregular branching, and less fastigiate (PI. III, fig. 12 and 151. Others an- a little coarser
and show transition to i. subjlabellala. As stated 1. e., the branches of I. americana are mostly
1.25—1.75 mm. thick. The branches of f. haingsisiana are generally of the same thickness.
But particularly in specimens that are more or less aj ; 1 >ach:.":g to f. subfiabellata, the branches
may be about 2 mm. thick. The majority of specimens from stat. 240 are also less typical,
partly in habit, partly also in structure, particulars of which are found below. They are also
somewhat stunted and much burdened with extraneous objects. (PI. III, fig-. 21 — 22). The
mens in hand trom stat. 193 partly resemble typical f. haingsisiana, partly they are a
little coarser. A single specimen from each of the stations 93 and 149 is in habit most closel]
allied to the same from stat. 240. In f. americana as well as in f. haingsisiana the upper part
of the plant may be somewhat worn by the action of a very strong current. In such specimens
the branches are more or less truncate. Live specimens of f. haingsisiana are pink-coloured,
judging from a coloured specimen in the collection. In dried state it is mostly whitish or slightly
purplish, frequently, however, with a greenish shade. It attains a diameter of about 9 cm.

The form subfiabellata is distinguished by its compressecl and less crowcled branches or
more or less flabellate branch-systems. It is a little coarser than f. haingsisiana. the branches
usually being 2 — 2,25 mm. thick (PI. III, fig. 23 — 25). Of the specimens here picturecl fig. 23

represents an intermediate form between the forms mentioned.
Both seem to have occurred associated with each other. The
form subfiabellata is coarser and also in other respects some-
what diverging from the form prostrata known from Bermuda.
As regards structure, I have mentioned before (1. c.)
that the cells of f. haingsisiana are often slightly broader, with
slightly thicker walls, than those of the type of the species.
From an examination of a number of specimens off. haingsi-
siana. the structure proves to be so far varying that in this
respect too no line is to be drawn. In f. americana the cells of
the medullary hypothallium are in general about twice as long
as broad, partly however longer, partly and often shorter, or
12 — 22, here and there but 10, frequently 14 — 1S <j_. long and
7 — 10, seldom up to 12 u.. broad, with rather thin walls. The
perithallic cells are partly square, 4 — 6 u.. in diameter, partly
and most frequently vertically elongated and 7 — 12 or seldom
up to ! mg (Fig. 16). In f. haingsisiana the medullary hypothallic cells resemble those

Vvv\t t i Vu

i ;\ branch: X 72.

33

of f. americana, but they are on the whole less regular. There is similarity also as reoards

the perithallic cells, though those of f. haingsisiana sometimes show a little thicker walls than

those of f. americana, of which on the other hand only a single

specimen has been examined (Fig. 1 7). In the above mentioned speci-

mens from stat. 240 the structure partly fully resembles typical f.

haingsisiana, partly the medullary hypothallium shows mostly a little

shorter cells with corners partially rounded and thicker walls. Besides |§

the perithallic layer is more vigorously developed, frequently with a 1|

little smaller and more thick-walled cells than in typical specimens.

In these specimens a new hypothallium often stretches itself partially

over the branches and covers roughnesses or extraneous objects. Such

is also the case with the above mentioned specimens from stat. 93

and 149. These specimens of the plant somewhat differing and more

or less stunted come ver}- near to f. madagascarensis ]), especially in

structure. I have however seen but a single specimen of the latter

from Madagascar, and besides a few quite young ones from the Pacific

coast of Japan, which seem to belong to the same form. It may be

that f. madagascarensis represents an independent species, but this

cannot at present be decided.

The conceptacles of sporangia in i. americana are somewhat
prominent and rather crowded in the upper part of the branches,
300 — 400 ij., in diameter, when seen from above. The roof is often '• ' ' A- Web- et FosL

Part of a median section of a

rather nattened in the central parts and here traversed by about 20 branch; x 72.

muciferous canals. In f. haingsisiana they are of about the same size

and more or less prominent, exceptionally however up to 450 u.. in diameter. The conceptacles
at length become overgrown by new-formed tissue, but such ones are not commonly seen in
the specimens examined, and they are in most cases peripherical. The sporangia are four-parted,
about 90 p.. long and 50 u.. broad.

The very greatest number of the specimens collected are sterile. It seems to be the
same case with this species as with several northern or arctic ones that reproductive organs
are as a rule no more developed in old specimens, but mostly to be found only in somewhat
younger specimens. - - On the other hand, the said organs are to be found even in quite young
specimens of certain species, especially crustlike ones. The form haingsisiana is scantily furnished
with sporangia in the months of February, June, August and November, and f. subflabellata
in the month of November.

The species in question seems to be most closely allied to Lithothamnion Dickiei, the
species last mentioned. It resembles a little certain forms of Lithothamnion falsellum in habit,
but is quite different in structure. It also recalls the northern species Lithothamnion dimorphum.
On the other hand it comes very near to Goniolithon frutescens f. congesia in habit. Cp. pi. III,

Fig. 17.
Lithothcnnnion crubescens Fosl.

ij M. Foslie. New Forms of Lithothamnia. — Det Kgl. Norske Vidensk. Selsk. Skiifter, 1901, X". 3. Tïondhjem 1901, p. 3.

36

pi. X, fier. [o— ii, and Lithoth. Mald. and Laccad., fig. 5 — 6. Even in structure it somewhat
approaches the latter spreid especially as regards the shape and size of the cells of the medullary
hypothallium, luit the perithallic layer is differing. Thus heterocysts are missing which are
characteristic of Goniolithon. Besides in G. frutescens there is hut indistinct, if any, stratification
in the medullary hypothallium, vvhereas L. erubescens exhibits distinct hut often rather irregular
cup-shaped layers of tissue, and the consistency is looser in the former than in the latter.
Therefore, even stcrile specimens of both species are easily separated when closely examined.
Young specimens of L. erubescens may he confounded with young ones of other species, for inst.
Archaeolithothamnion timorense, especially however Lithothamnion australe.

Occurrence: At stat. 303 the plant forms a large and extended bank in the lowest
part of the litoral and upper part of the sublitoral region, sparingly mingled with specimens of
Lithothamnion australe f. brachiata, Goniolithon Reinboldi and Lithophyllum Okamurai. perhaps
also other species (Fig. 1 — 2). lts occurrence here is recorded by Madame A. Weber-van Bosse 1).
At stat. 240 it seems also to have occurred in some number. On the other hand there are but
a few specimens from stat. 193, and a solitary and nut well developed specimen from each of
the stations 93 and 149.

Area: North Pacific: Japan (Hinga), not certain ; Indic: Madagascar (f. madagascarensis);
South Atlantic: Chaloup Bay, Fernando do Xoronha, Brazil (f. americand)\ North Atlantic:
Bermuda (f. americana, t. prosiratd).

10. Lithothamnion pulchrum A. Web. et Fosl. Fig. 18; PI. IV.

Three new Lithoth. p. 3.

Stat. 96. South-east side of Pearl-bank, Sulu Archipelago. 15 m. Lithothamnion bottom.
Stat. 315. Kast of Sailus Besar, Paternoster Islands. Up to 36 m. Coral and Lithothamnion.

The plant forms roundish shrubdike masses up to about 10 cm. in diameter. The branches,
issuing from a slightly thickened centre, are subdichotomously or irregularly divided, partly rather
spreading and somewhat curved, partly more crowded. often with rather long axes and side-
branches issuing at an almost right angle, terete or subcompressed, either cylindrical, or slightly
ing, or nöw and then slightly thickenins.;' towards the tip, with the ends rounded, 2 — 3,
frequently about 2,5 mm. in diameter.

As remarked 1. c, the species seems mostly to develop itself freely at the bottom.
Among the gathered specimens I have found none that had developed itself attached to any
hard object. But when extraneous objects attach themselves to the plant, it develops a thin
crust and tends to cover the object (PI. IV, fig. 6). In the specimen here pictured a shell of
a mollusc is enclosed between the outermost branches. The shell is in part covered with a thin
crust issuiiiL; trom the branches that have been in contact with it. So also in the specimen
ted pi. IV, fig. 9. Here several branches have been covered with some lower animals,

1. Études sur les Algues de 1. 'Archipel Malaisien. — Annales du Jardin Botanique de Buitenzorg.
1901, p. 132. — Cp. also the introduction of the present paper.

37

ssgsg

apparently Bryozoa, even filling the intervals between some crowded and in part dead branche
These Bryozoa are again for a great part covered with a crust of the plant. The incrustment
rises from the nearest branches alive, and by and by it develops new branches too. Still in
other specimens new crust-like formations are developed over dead branches or extraneous objects,
sometimes issuing from branches being alive, sometimes, as it seems, formed from germinating
plants which have attached themselves to the said branches. In this case too new branches are
developed in a rather early stage of the said formation (PI. I\T, fig. 5 and 10 — 11). In the
specimen last quoted a small and young specimen of the same species has been brought into
contact with two of the outermost branches. Then a crustlike formation some mm. broad has
been developed, issuing from both the said branches in a semicircle around one of the branches
of the said young specimen, thus fastening the latter, although but partially anastomosed with it.
These cases seem to indicate that the species perhaps sometimes attaches itself to small objects,
at first forming a crust over the latter, and afterwards loosening itself. In addition broken or
loosened branches probably continue their growth freely at the bottom.

A section parallel to the longitudinal axis of a branch exhibits superposed, generally
clistinctly separate and rather regular cup-shaped layers of tissue, the cells of which are arranged
in rows radiating in the shape of a fan, such as in most of the branching species of this genus.
The cells of the medullary hypothallium are rectangular, often however with somewhat rounded
corners, the length being frequently twice the
breadth or larger by a half. The length varies
between 12 u. (occasionally less) and 30 a., fre-
quently 16 — 22, the breadth between 7 and 14 p..
The lower cells of each of the said cup-shaped
layers appear longer, rectangular, the uppermost
ones shorter or often almost square, as in most
of the Lithothamnia forming similar layers. The
cells of the perithallium are squarish or rectangular,
frequently with thicker walls than the former,
6 — 14 u.. long and 6 — 9 ij., broad; fig. 18.

The organs of reproduction are unknown
in the species in question. However, as remarked
1. c. the plant must be considered to be a true
Lithothamnion , judging from a solitary cavity
resembling an emptied conceptacle of sporangia,
which appears just below the surface in a section of a branch. This supposed conceptacle is
about 550 u.. in diameter by a height of about 220 y.., and is overgrown with a few layers
of cells. It is however to be observed that the walls of the conceptacle have partly been torn
away during the grinding of the section; fig. 18. Therefore, it cannot be realized whether it
actually represents the said organ, although there is every probability it does. In another, but
fragmentary, section of a rather young specimen, in habit and structure fully resembling the
species in question, are to be found a few other overgrown conceptacles, which also seem to

Fig. 18. Lithothamnii um A. Web. et Fosl.

Part of a median section of a braDch, with a supposed

conceptacle of sporangia overgrown; x 72.

38

indicate that the plant is a Lithothamnion. They are only 200—300 by 75-110-;.. in diameter,
but the sections are apparently not median ones.

[•],,. species on one hand resembles certain forms of the arctic Lithothamnion tophiforme
in habit and on the other hand it ma) sometimes be difficull to distinguish it from Archaeo-
lithothamnion timorense. Especially young specimens are sometimes almost quite like certain
forms of this in habit and are in this respect hardly distinguishable. Cp. pi. IV and

pi YI11, fio\ 1 —14. The surface is however a little shining and smoother than in the said

ies. Even in structure they uiten much approach one another. Thus the cells of the medullary
hypothallium are of about the same size in both, in L. pulchrum however less regular and
frequently a little broader, and the corners are more rounded. Besides, the cup-shaped layers
issue are more distinct in the species in question, whereas on the other hand in A. timorense
they partly appear indistinct, partly and most frequently are not to be seen in a section. The
perithallic layer is more distinctly marked towards the hypothallic one, and the cells of the
tormer are generally smaller than in A. timorense. The most certain distinguishing feature is,
however, the presence of small oblique intermediate cells in A. timorense, which are quite
wanting in L. pulchrum. This character allows of the distinction even of sterile specimens. But
the said cells are often not easily descried when the section is not quite thin. In addition, young

mens are hardly distinguishable from certain forms of Lithothamnion australe.

Occurrence: There are several specimens from stat. 315, where the species has
probably been rather abundant, but from stat. 96 there is only a solitary, though well developed
one. A few and likewise sterile specimens from stat. 282 belong perhaps to this species. They
have been brought home together with some specimens which partly belong to Archaeolitho-
thamnion timorense, partly are somewhat differing and uncertain.

Archaeolithothamnion (Rothpl.) Fosl.

1. Archaeolithothamnion erythraeum (Rothpl.) Fosl.

Rev. Syst. Surv. Melob. p. 8.

A. ROTHPLETZ. Über einc neue Pflanze (Lithothamnium erythraeum n. sp.) des Rothen
Meeres. Bot. Centralbl. 1893, N" 14.

f. dura (Heydr.) Fosl. PI. V.

L. c. p. 8. Sporolithon ptychoides f. dura Heydr. in Bei', der Deutsch. Bot. Gesellsch. 1897, p. 6j.

f. mol/is (Heydr.) Fosl. PI. VI.

L. c. p. 8. Sporolithon ptychoides f. mollis Heydr. 1. c. p. Gj.

Syn. Sporolithon ptychoides Heydr. 1. c. p. 415.

Syn. Sp. molle Heydr. 1. c. p. 416.

Syn. Sp. crassum Heydr. 1. c. p. 417.

St.it. 47. Bima. Rei f.

Stat. 71. Makassar and surroundings. (Tuin Barang, Samolana, Ley-Ley). Up to 32 m.

Stat. 81. Pulu Sebangkatan, Borneo-bank. 34 m. Coral bottom and Lithothamnion.

Stat. 86. Dongala. Palos Bay, Celebes. Reef.

39

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion
bottom.

Stat. 96. South-East side of Pearl-bank, Sulu Archipelago. 15 m. Lithothamnion bottom.

Stat. 99. Off North-Ubian. 16 — 23 m. Lithothamnion bottom.

Stat. 109. Off Pulu Tongkil, Sulu Archipelago. 13 m. Lithothamnion bottom.

Stat. 123. North Bay, Biaru Island. 36 — 27 m. Stonc and Lithothamnion bottom.

Stat. 125. Off Sawan, Siau Island. 27 m. Stone and some Lithothamnion.

Stat. 133. Lirung, Salibabu Island. Up to 36 m. Mud and hard sand.

Stat. 169. Atjatuning, West coast of New-Guinea. Keef.

Stat. 172. Gisser, anchorage between this island and Ceram-Laut. 18 m. Coral- and Litho-
thamnion bottom and Reef.

Stat. 193. Sanana Bay, East coast of Sula Besi. Reef.

Stat. 213. Saleyer and surroundings. Reef.

Stat. 220. Pasir Pandjang, West coast of Binongka. Reef.

Stat. 234. Nalahia Bay, Nusa-Laut Island. Reef.

Stat. 240. Banda-anchorage. Lithothamnion-bank in 18 — 36 m.

Stat. 250. Kilsuin, West-coast van Kur Island. 20 — 45 m. Coral and Lithothamnion.

Stat. 261. Elat, West-coast of Great-Kei Island. Reef.

Stat. 277. Kulewatti (Sollot) Bay, Dammer Island. Reef.

Stat. 296. Noimini, South coast of Timor. Reef.

Stat. 299. Buka- or Cyrus Bay, South coast of Rotti Island. 34 m. Mud, coral and Lithothamnion.

Stat. 315. East of Sailus Besar, Paternoster Islands. Up to 36 m. Coral and Lithothamnion.

A part of the type specimen of this species, from El Tor in the Red Sea, is represented
pi. V, hg. 1, and two other fragments from Obock in the Gulf of Aden pi. V, hg. 3 and 4.

As formerly stated by myself, Sporolithon ptyekoides Heydr. is identic with the above
species. Sp. ptyekoides was at first divicled by Heydrich 1. c. into two forms, f. dttra and f.
mollis. Afterwards he considered f. dura as an independent species, the type of Sp. ptyekoides
in a more restricted sense than formerly. At the same time he divided f. mollis into two
separate species, which he named Sp. inollc and Sp. crassum, and the latter in its turn was
divided into two forms, f. typica and f. variabilis. From a comparison of the descriptions 1. c.
as well as the illustrations 1. c. pi. III, fig. 15 — 23, it will be realized that there is no real
distinction between the three species last named and A. erytkraeum, neither as regards habit
nor as to structure. The latter is a rather varying species, the habit of which often depends
on local conclitions and the shape of the substratum to which the plant is attached. Therefore,
I have classified only two forms, mainly in the same sense as they were first described by
Heydrich 1. c, viz. f. dura and f. mollis of which iragmentary but authentic specimens from
El Tor, clistributed by Heydrich himself, are represented pi. V, fig. 2 and pi. VI, fig. 1.

The form dura is generally distinguisheel by thick and densely crowded branches (pi. V).
In f. mollis the branches frequently are thinner and less crowded than in the former (pi. VI).
Numerous transitions are, however, to be found, so that a certain limit can not be drawn.
Both forms partly grow gregariously, partly in different localities.

As far as hitherto known, the species in question always attaches itself to some hard
object, corals, stones or other calcareous algae. Detached parts of the plant, however, seem to
continue their growth freely on the bottom, although apparently seldom. The plant sometimes
fully surrounds the object, especially when the latter is a small one. It partly develops excres-
cences or branches in a rather early stage, partly in a later stage and particularly when the

4o

substratum is large. In the latter case the crust frequently grows thicker than in the former.
In this respect it reseml I other species, lor inst. Lithothamnion fruticulosum. It is

as a rule much burdened with extraneous objects ol one kind or other. Thus the inner or
lower parts of a specimen ma) sometimes be almost fully destroyed by boring-muscles, or

r specimens are traversed with numerous, wider or finer, passages made by worms. In
addition, different animals attach themselves to the surface of the plant and become more or
overerown by new formed tissue. This contributes to an irregular development of the
plant. New crusts are often developed over the primary one, sometimes even more crusts, one
above the other, and such new formations often expand over the primary excrescences or short
branches. In such cases the branches not seldom become coarser than otherwise (PI. V,
So also when germinating plants attach themselves to more or less knotty corals,
dead specimens of the same species, or to other calcareous algae with wartdike excrescences
or short branches. < >n pi. V, fig. i i is represented a specimen with a coral nucleus. The latter
has been covered with Goniolithon Rcinboldi, and this species has in its turn been almost fullv

red with the species in question. Here and there in the said specimen appear two crusts
of . I. erytkraeum, one above the other, the first crust covered with a thin layer of other
organisms, apparently lower animals, and then the other one developed over the said layer,
both, In 'wever, covering G. Reinboldi. Besides wartdike excrescences or short branches are also
in this specimen developed apart from those which have risen from covering up the densely
crowded branches of G. Reinboldi. Some uncovered branches of the latter species provided
with conceptacles are to be seen in the uppermost part as well as to the right on the quoted
picture. The branches of the species in question are on the whole much varying, and in one
and the same specimen are often to be found some ones several times thicker than others
(PI. V, fig. 2, 4, 5, 10, 12). Sometimes the plant becomes worn and rubbed in the upwards
turning part by the action of a strong current, in the same way as often to be seen in
Lithothamnioji dimorphum and other species. The edges of such worn branches now and then
get covered by new formations of tissue issuing from the next undamaged parts of the plant
(PI. V, hg. 7).

In regard to the structure of the forms above mentioned I refer to my former remarks
on this species '). I have there shown that the measures given by Heydrich 1. c. are not
correct. The hypothallic layer of the crust is composed of more or less bent rows, the cells
of which are up to about 40 u. long, frequently however shorter. The cells of the medullary
hypothallium are 10 — 22, frequently 12 — 18 u. long and about 9 — 10 u.. broad. The perithallic
cells are partly about 8 u.. square, partly of nearly the same length as those of the medullary
hypothallium. However, now and then appear rather irregularly alternating long and short
In addition, the minute square and oblique intermediate cells sometimes appear almost
everywhere in a section, sometimes apparentlv but here and there.

Die reproductive organs known are cystecarps and sporangia. The former seem, however,

I die HEYDRiCH'schen Melobesien-Arbeiten eine sichere Grundlage: — Uet. Kgl. Norske Vidensk. Selsk.

rondhjem 1901. N". 2. p. (6.

4'

rather seldom to occur. There are a rather great number of specimens in the Siboga collection,
but only a couple are provided with the said organs. The conceptacles are convex or subhemi-
spheric-conical, about 300 — 600 u. in diameter, when seen from the surface, but not sharply
defined. Externally they are rather like the corresponding organs in certain forms of Lithophyllu m,
but otherwise they come near to those in Goniolithon, as the carpospores arise from any part
of the conjugation cell. Sporangia are on the other hand common, and the sori are partly to
be found overgrown in great numbers by new formed tissue, partly, however, they are not to
be seen in a section. The sporangia are frequently 75 — 90 u.. long by a breadth of 35 — 55 f/..1).
Fertile specimens have been collected nearly all the year.

Occurrence: Only one specimen has been brought home from each of the stations
4;, 109, 125, 133, 240, 296 and 311, a fact which seems to indicate that the species has been
scarce in these places. From the other stations are 2 — 8 specimens. It is uncertain whether the
plant forms banks at any of the quoted stations, alone or in company with other species, but
most probably it does not, as it is nearly always attached to other hard objects, but apparently
occurring both in the litoral and sublitoral regions. The species has also been met with at the
following places on the north coast of New-Guinea, here collected by Prof. A. Wichmann:
Pulu Adi; Pulu Kelamala, east of Pulu Adi; Anggar-Mios Island, Geelvink Bay; and Sari
(Sjari), west coast of Geelvink Bay 3).

Area: Indic : The Red Sea ; the Gulf of Aden; the Maldives.

2. Archaeolitliothamnion Siöogae A. Web. et Fosl. PI. VII.

Three new Lithoth. p. 3.

Stat. 86. üongala, Palos Bay, Celebes. Reef.

Stat. 96. South-east side of Pearl-bank, Sulu Archipelago. 15 m. Lithothamnion bottom.

Stat. 99. Off North-Ubian. 16 — 23 m. Lithothamnion bottom.

This species is shrub-like or roundish, up to about 7 cm. in diameter. The thallus is
subdichotomously or irregularly branched, with axes of up to three orders. In typically developed
specimens the branch-systems issue in all directions from the centre of the thallus. The branches
are more or less spreading and curved, partly short, partly elongated and up to about 2 cm.
long, terete or slightly compressed, frequently knotty, with the ends rounded, 3 — 6, generally
4 — 5 mm. in diameter. I have not seen any attached specimens, but the plant is perhaps
attached at first to some hard object, although afterwards it detaches itself and lies free on
the bottom. However, a partly dead specimen in the collection is anastomosed with a roundish
extraneous object and apparently also fragments of another species, and besides it is burdened
with extraneous objects. New layers of tissue are here and there developed over the dead parts
of the plant, or partly stretched over the said objects as crustlike formations, from which short
branches are developed, such as in Lithothamnion pulcJirum and other species generally lying

1) The measures of the sporangia in the species of this genus refer to the size of overgrown cavities due to the said organs.
The sporangia themselves are therefore in fact a little smaller.

2) The Dutch New Guinea Expedition 1902 — 1903.

SIBOGA-EXPED1T1E I.XI. 6

I-

■ iii the bottom (PI. VII, fig. [6 Anotl I stunted .specimen is anastomosed with two

others, sterile and undeterminable calcareous algae, corals, sponges and Bryozoa, which seem

nclose a common nucleus of some hard object. The species in question has also in this

developed crustlike formations, from which issue short branches. The plant is in general

rather burdened with extraneous objects, especially Bryozoa. Detached branches often appear

to continue their orowth, and the surface of fractures is soon effaced by new-formed tissue.

In .ui parallel to the longitudinal axis of a branch the medullary hypothallium is

composed of cells which are frequently rectangular, and the length of which is i', — 2 times
the breadth. The size varies bei ween to and 22. seldom up to 28 ■>.. in length, generally
15 — 20 -/. 1 ii ■)- in breadth. They gradually decrease towards the perithallium, the cells

of which are usually smaller than those of the hypothallium, varying from 5 — 8 u.. squarish and
up to about 12 il. in length. The limit between the said layers is however very indistinct. Here
and there in the hypothallium as well as the perithallium are to be seen alternating short and
loiiver cells. rather more commonlv in this species than in the other ones of this g-enus. Minute
square and oblique intermediate cells are numerous; they are 2 — 4 «,. in diameter.

In the specimens examined only a small number of sporangia occur. They are frequently
65 — 85 •!.. long and 30 — 45 u.. broad. The other reproductive organs are unknown.

The species in question on one side approaches to coarse forms of A. timorense. Voung
as well as stunted specimens of both species are sometimes not easily distinguished. However,
the plant is frequently much coarser than the said species, and the cells are as a rule a little
smaller. On the other hand it occasionally recalls A. erythraeum.

Occurrence: The plant seems to have been fairly common at the stations 96 and
90. while only two rather young specimens have been brought home from stat. 86.

3. Archaeolithothamnion timorense Fosl. PI. VIII, fig. 1 — 14.

Thallus freely developed at the bottom, shrub-like, 3 — 5 cm. in diameter, subdichoto-
mously or irregularly branched; branches more or less spreading and curved, frequently knottv,
terete or subcompressed, most often tapering, 1,5 — 2,5 mm. thick. Sporangia 70 — 90 u.. long
by 35 — 50 ij., broad.

Stat. 99. Off North-Ubian. 16 — 23 m. Lithothamnion bottom.

Stat. 144. North of Salomakie (Damar) Island. 45 m. Coral bottom and Lithothamnion.
Stat. 277. Kulewatti (Sollot) Bay, Dammer Island. Reef.

Stat. 2'6z. Between Xnsa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and Litho-
thamnion.
Stat. 315. East of Sailus Besar, Paternoster Islands. Up to 36 m. Coral and Lithothamnion.

The plant is shrub-like or roundish, up to about 5 cm. in diameter. Judging from the

gathered specimens, it develops itself freely at the bottom. Nor have I seen any specimen of

attached to other objects, but sometimes anastomosed with small ones. In that

the plant little by little develops a thin crust over the object, such as in some of the species

menl lor inst. /.. tenue, L. pulchrum, A. Sibogae. So also it tends to cover Bryozoa

or other animals which fasten themselves on or between the branches, or new formations

43

are developed over old and dead parts of the plant. This contributes to a more irregular
development, with branches more crowded, now and then anastomosing and thicker than
otherwise. The thallus is subdichotomously or irregularly branched, with axes of up to about
three orders. In specimens typically developed the branches issue in all directions from the
centre of the thallus. The branches are frequently rather spreading, partly fairly straight, partly
and more generallv curved, with side-branches which often issue at a rigdit or almost ribht
angle, and rather knotty especially in old specimens. They are terete or a little compressed,
either cylindrical or more frequently tapering, selclom slightly enlarged towards the tip, with
rounded ends, 1,5 — 2,5 mm. in diameter.

A median section of a branch shows partly rather regular cup-shapecl layers of tissue,
such as in most branching species of the genus Lithothamnion, partly the said layers are not
to be seen, or very indistinct, such as in the other species of Archaeolithothamnion. The length
of the cells of the medullary hypothallium is frequently i'.\, — 2', times the breadth. They are
14 — 30 u.. long by a breadth of 7 — 14 u., most commonly however 16 — 22 u. long. The
perithallic cells are partly 7 — 12 u.. square, partly elongated and then as a rule somewhat
shorter than the hypothallic ones. Here and there ,appear irregularly alternating long and short
cells. Intermediate oblique cells 2 — 4 u.. in diameter, also in this species are present in a rather
large number particularly in the perithallic layer.

Some of the specimens collected are provicled with sporangia, mostly, however, in small
number. The said organs are 70 — 90 p.. long and 35 — 50 u.. broad. Cystocarps and antheridia
are unknown.

Among the species of this genus hitherto known, A. timorense is intermediate between
A. Sibogae and A. Schmidtii. On the other hand it is suggestive of certain forms of the
northern Lithothamnion tophiforme. Besides, sterile specimens are sometimes almost undis-
tinguishable from Lithothamnion pulchrum. Thus there are some from stat. 282, which partly
appear to belong to A. timorense (pi. VIII, fig. 2), partly come very near to L. pulchrum,
and are probably referable to the latter, although they are more densely branched, with shorter
axes than usually to be seen in this species. A few of them, on the other hand, partly approach
to A. Schmidtii, partly to delicate forms of A. Sibogae in habit. They cannot at present be
fixed with certainty. The most typically developed specimens of the species are from stat. 315,
but they are here in part rather burdened with extraneous objects (PI. VIII, fig. 1, 3, 4 — 6,
9 — 12, 14). Some of those from stat. 277 are a little coarser, and the branches are less
curved and less knotty (PI. VIII, fig. 7 — 8, 13).

Occurrence: The plant seems to have been rather abundant at stat. 315, but
apparently scarce at stat. 277 as well as at stat. 282. From stat. 99 have been brought home
only a couple of specimens not quite typical, and from stat. 144 a single one.

4. Are haeo lithothamnion Schmidtii Fosl. PI. VIII, fig. 15 — 17.

Corall. in SCHMIDT, Fl. Koh. Chang, p. 16.

Stat. 96. South-east side of Pearl-bank, Sulu Archipelago. 15 m. Lithothamnion bottom.
Stat. 169. Off Adjatuning, West-coast of New-Guinea. Reef.

1 1

The species attaches itsef to corals <>r other hard objects, developing a feeble crust
which attains a thickness of up to aboul 2 mm., frequently however less. From this crust
more or less crowded branches are developed. The Iatter are either simple or subdichotomously
divided, with short axes, partly increasing in thickness upwards, partly and rather seldom not,
or with somewhat spherically thickened ends. The branches are 1,5 — 2 or occasionally up to
3 mm. thick, and thej are often knotty.

A vertical section of the crustlike part of the plant exhibits a rather slightly developed
hypothallic layer, composed oi elongated cells which are up to about 20 <j.. long, frequently
however shorter. In a longitudinal section of a branch the medullary hypothallium is formeel
of cells most commonly 1'., — 2 times longer than broad, or 10 — 18 ij.., occasionally up to 22,
generally however 12 — 16 p.. long and 7 — 11 <j.. broad. The perithallic cells are 6 — 10 u..
square or up to about 14 u.. long. Irregularly alternating, long and very short, cells are seldom
seen in the medullary hypothallium, as well as in the perithallic layer. Besides appear in both
the said layers, particularly however in the perithallic one, square and oblique intermediate
cells 2 — 4 ij., in diameter.

The only reproductive organs known are sporangia. They are 50 — 65, now and then
up to So y.. long by a breadth of 30 — 40 or up to about 45 u..

1 considered 1. c. the species in question to be a well marked one. It has, however,
been stated that the species of this genus at present known are much varying and often not
easily recognizable in their most extreme forms. PI. VIII, fig. 15 represents a part of the type
specimen of this species. It is burdened with several extraneous objects, which are partly covered
with crustlike formations of tissue. The plant is as vet but little known, as till now only a few
specimens have been collected. These are even from rather widely distant areas. However, it is
mnst probably to be maintained as an independent species. It partly approaches to A. erythraeum,
partly it is not easily distinguished from A. timorense, the species last described. It is separated
from the former in the main by its thinner branches and frequently shorter cells as well es
smaller sporangia. On the other hand it seems to be more closely allied to certain forms of
A. timorcn$L\ particularly somewhat densely branched ones, but the Iatter species apparently
always develops itself freely at the bottom and generally exhibits larger cells and sporangia.

Of the two specimens in the Siboga collection, especially that from New Guinea agrees
well with A. Schmidtii (PI. VIII, fig. 17). This is also the case with the other specimen but
lor its exhibiting cells which are often a little larger than in the former one (PI. VIII, fig. 16).

Occurrence: Only one specimen brought home from each of the stations above quoted.

Area: North Pacific: The Gulf of Siam : Indic : The Maldives.

45

II. Melobesieae Aresch. (Fosl. emend.).

Goniolithon Fosl.

a. Crustacea [Herpolithon Fosl.).

i. Goniolithon myriocarpon Fosl. PI. IX, fig. 6 — 7.

Lithoth. Adriat. Meer.

Lithothamnion myriocarpon Fosl. On some Lithoth. p. 19.

Stat. 86. Dongala, Palos Bay, Celebes. Reef.

Stat. 169. Atjatuning, West-coast of New-Guinea. Reef.

Stat. 225. South of Lucipara-group. Reef.

Stat. 277. Kulewatti (Sollot) Bay, Dammer Island. Reef.

In 1897 I described the above species represented by a solitary specimen, picked up
at Massauah in the Red Sea. I have not afterwards met with it until last year, when I got a
specimen for determination, collected at El Tor in the Red Sea by Dr. Plate, which proved
to belong to the same species '). In Lithoth. Adriat. Meer. I have shown that the plant is a
Goniolithon ; but I ani still uncertain whether this species and G. Fosliei, recorded below, really
are specifically distinct. At any rate, they are closely allied to one another. There is no limit
to be drawn in regard to habit, except that the crust as a rule will be thinner in G. myrio-
carpon than in G. Fosliei ; and the surface is frequently rather smooth, now and then even a
little shining in the latter, whereas it is rather dull in the former. The plant attaches itself to
similar hard objects as G. Fosliei, viz. corals, shells or other calcareous algae. Sometimes it
even forms very thin crusts on Archaeolithothamnion erythraeum, in a sterile stage scarcely
distinguishable from the host plant except in a section, as it in habit gives the impression of
a new thickening layer of the latter. PI. IX, fig. 6, represents a part of the type specimen
attached to a bucky, forming incrustations 100 — 200 u. in thickness, closely and firmly adherent
to the substratum. But it also becomes thicker, and it seems at length to attain nearly the
same thickness as G. Fosliei (PI. IX, fig. 7).

As to structure the cells of the perithallic layer are frequently smaller than in G. Fosliei,
and often those of the hypothallium too are smaller. The latter layer of tissue sometimes is
not marked coaxial, which, as far as hitherto seen, always is the case in the said species.
Besides, the heterocysts appear in much smaller number than in G. Fosliei, often not even to
be seen in a section.

There is certainlv, however, no distinct limit to be clrawn in the mentioned respects, the
structure being rather varving in these species as in several others. The distinguishing feature
apparently most significant is clue to the reproductive organs. In the species in question only
sporangia '3) are known. The conceptacles are subprominent or often almost superficial, 450—
650 u.. in diameter, when seen from the surface. Thus they show only about half the size of

1) Th. REINBOLD. Meeresalgen von Tor (Sinai-Halbinsel, Rotes Meer). — Hedwigia. Bd. LXII, 1903, p. 232.

2 ) In the description of the species they are recorded as cystocarps, which is, however, wrong (antheridia uncertain).

the cystocarpic conceptacles of G. Fosliei, which are 800— iioof/.. in diameter, when seen
from above, or sometimes even more. in typical specimens of the latter species sporangia have
not yet heen found. However, in Lithoth. Mald. and Laccad. I mentioned a specimen of G.
Fosliei furnished with the said organs, which are dl about the same size as in G. myriocarpon.

Put the specimen is not quite typical, as also here the eells are a little smaller than in a
tvpical ene from the same place hearing cystocarps, and the heterocysts scantily appearing.
] should be most inclined to consider G. myriocarpon only to be the sporangiadjearing form
of a common species including G. Fosliei as the cystocarpic form, the former developing a
very small number of heterocysts. I have not yet had the opportunity of comparing more
particularly other species of the genus Goniolithon, so as to be able to ascertain a similar state
of things. However, in G. Notarisii sporangia-specimens partly exhibit a rather large number
of h< >ts, partly they are scarce. In G. elatocarpon, a species closely allied, I have not

seen heterocysts. Of this species, however, only a single specimen is known. The latter species
stands in almost the same relation to G. Notarisii as G. myriocarpon to G. Fosliei, e. g. the
cells frequently being smaller. It would be of considerable interest to realize the said relations
particularly in regard to sexual and asexual forms of one and the same species within the
genus Goniolithon, as well as the nearly allied genus Melobesia. Thus in the latter genus too
the mutual relation between e. g. M. farinosa and M. Lejolisii is not yet fully elucidated.
These species are in fact hardly distinguishable except for the heterocysts appearing in the
former, and perhaps a small difference in the size of the cells.

I do not venture, however, at present to unite G. myriocarpon and G. Fosliei, partly
owing to the mentioned difterences in structure, partly and especially because the supposed
relations in regard to the reproductive organs and the different appearance of the heterocysts
are not yet settled.

Occurence: The plant seems to be rather widely dispersed, but apparently not
occurring in great number at any of the quoted stations.

Area: Indic: The Red Sea.

2. Goniolithon Fosliei (Heydr.) Fosl. Fig. 19; PI. IX, fig. 1 — 5.

Lithoth. Adriat. Meer.

Lithothamnion Fosliei Heydr. Corall. in Ber. der Deutsch. Bot. Gesellsch. 1897, p. 58; ex

parte sec. spec.
Lithophyllum Fosliei Heydr. 1. c. p. 410.

Stat. 71. Pulu Barang, near Makassar. Reef.
Stat. 78. Lumu-Lumu-shoal, Borneo-bank. Reef.
Stat. 91. Muaras-reef, inner side: East coast of Borneo.

Stat. 303. Haingsisi, Samau Island. Reef.

As for the apprehension of this species I refer to my remarks in Melob. Arb. Heydr.
]). S and Lithoth. Mald. and Laccad. p. 470. An audiëntie specimen, distributed by Hevdrich
himself, is represented pi. IV, fig. 1, and a vertical section of the same specimen fig. 19 A.

47

In specimens typically developed the plant forms a rather even crust on divers hard
objects, sometimes one crust formed above the other (PI. IX, fig. 2, 3, 5). The crust attains
a thickness of up to about 1 cm., frequently however less. It seldom develops excrescences,
as far as I have seen. Such ones apparently often occur, but they are, in the specimens
examined by me, at any rate as a rule, caused
partly by the shape of the substratum, partly
by covering up extraneous objects. Besides,
they are often due to numerous, finer or wider,
passages made by worms, around the orifices
of which new formations of tissue are by and
by heaped up, or the destruction of the inner
layers, especially those just below the surface
of the plant, induce local formations of tissue
in the shape of more or less wart-like excres-
cences (PI. IX, fig. 4).

A vertical section of the crust exhibits
a rather varying structure, with a vigorous
and marked coaxial hypothallium. The peri-
thallic cells are partly squarish, often however
with rounded corners, partly vertically elonga-
ted and much varying in size, frequently being
largest in specimens which are much attackecl
by animals and attached to an uneven sub-
stratum (Fig. 1 9 B). The cells are most fre-
quently thick-walled. New hypothallic formations
are often to be seen, particularly developed over |
damaged parts of the plant. Heterocysts appear
in great number, and the}- form frequently at
length grown-in rows. These rows are inter-
preted by Heydrich as "Zeilen der Leitungs-
schicht". Cp. 1. c. p. 59, fig. 1. Similar but
much shorter grown-in rows of heterocysts
also appear in Goniolithon frutescens, although
there are frequently to be seen but solitary
cells, which in certain specimens are numerous
almost in any part of the perithallic layer. In
G. Notarisii grown-in heterocysts seldom appear. Cp. Solms-Laubach. Corall. Monogr. t. 1, fig. 2.

The cystocarpic conceptacles in this species are scattered almost all over the crust. They
appear frequently in a rather great number, but they are, on the other hand, never densely
crowded. The conceptacles are subconical, rather low and 800 — 11 00 u.. in diameter, when
seen from the surface, sometimes, however, even up to about 1,5 mm., but often not sharply

Fig. 19. Goniolithon Fosliei (Ileydr.) Fosl.
A. Vertical section of the specimen pictured pi. IX, fig. i ;
/>'. Vertical section of the specimen pictured pi. IX, fig. 2,
with a new hypothallium formed over a conceptacle; X 72.

[S

defined. Antheridia are not yet known with certainty, and as regards the sporangia I refer to
remarks under G. myriocarpon.

< Iccurrence: Also this species appears to be rather widely dispersed, but it apparently
never occurs in any great number at the stations above quoted.

Area: Indic : The Maldives and Laccadives; the Red Sea; Zanzibar.

3. Goniolithon megalocystum Fosl. mscr. Fig. 20 ; PI. IX, fig. 8 — 9.

Thallus crustlike, convex-concave, 0,3 — 0,5 mm. thick. Conceptacles of sporangia sub-
lical, Iow, [,5 — 2 mm. in diameter.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion bottom.
Stat. 129. Kawio and Kamboling Islands. Karkaralong-group. Reef.

There is known but a solitary typical specimen of this species. It is fragmentary, towards
4 cm. long by a breadth of about 2 cm. There is 110 evidence that is has been attached, but
it forms only a part of a larger specimen which probably has been fastened to some hard
object. The crust is -convex-concave and approaches in shape one half of the shell of a muscle
(PI. IX. fig. 8). It is about 0,5 mm. thick in the thickest part, plainly decreasing towards the
margin which is 0,2 — 0.3 mm. thick and somewhat rounded off. In the upwards turning part
the crust is provided with rather faintly distinctive, concentric ridges, and in the lower part it
is concentric striate or shallowly furrowed.

A radial section of the crust exhibits a vigorous hypothallium. The central cells of this

layer are 25 — 35, here and there up to about 40 u.. long,
and 10 — iS u. broad, with rather thick walls. It sends forth
perithallic rows upwards and downwards, which however are
very feeble in the thinner parts of the crust (Fig. 20 A).
The upper layer increases in thickness towards the thickest
part, where it shows an extent of about 200 u.. The cells
of this layer are partly squarish, partly vertically elongated,
about 10 'j.. in diameter, or up to about 18 u. long. A
few heterocysts occur.

The plant bears a few conceptacles of sporangia in

the month of July. They are conical, but low, rather abruptly

passing into a short and thick tip, not distinctly defined,

r,5 — 2 mm. in diameter, when seen from the surface. The

sporangia are thinly scattered over the almost plain bottom,

100 — 140 u. long and about 60 u.. broad.

With some reservation I refer a couple of somewhat differing specimens to the same

species. They not unlikely represent a form corresponding with similar forms of Lithothamnion

fragilissimum and L. simulans. The crust is about 0,3 mm. thick, loosely covering Bryozoa,

Squamariaceae, a young Goniolithon frutescens and other calcareous algae attached to a coral

alocystum Fosl.
.1. Radial section of the type specimen:
B. Vertical section of the form pictured
pi. IX. fig. 9: 72.

49

(PI. IX, fig. 9). It bears a few conceptacles of cystocarps which are conical and about 1 mm.
in diameter, when seen from above. In structure it coincides in the main with the type of the
species (Fig. 20 B). What makes it uncertain, however, whether this form in fact belongs to
the species in question, is that the cystocarpic conceptacles are considerably smaller than those
of sporangia in the type. But the specimens appear to be stunted, and it may be that the
said organs are larger in better developed specimens and show the same variation in size/as
e. g. Goniolitkon Fosliei and other species. Nor can be jndged from a solitary specimen of the
type only hearing a few conceptacles of sporangia, whether the latter are as a rule of the
quoted size, or perhaps partly smaller. Therefore, there seems at present to be no reason to
classify this form as a denominated one, far less as an independent species.

The above species is plainly distinguished from any other species known of this genus.
It seems to stand nearest to G. Fosliei, but distinguishes itself by essential characteristics. In habit,
however, it rather approaches certain forms of Lithothamnion fragilissimum and L. simulans,
and sterile specimens may be easily confounded without closer examination.

Occurrence: As mentioned above, only one specimen is known of the type of the
species. It was collected at stat. 129. Of the said differing form two specimens have been
brought home from stat. 93.

ó. Ramosa [CladoIitJion Fosl.).

4. GoniolitJion Reinboldi A. Web. et Fosl. Fig. 2 1 ; PI. X, fig. 1 — 6.

Lithophyllum Reinboldi A. Web. et Fosl. Three new Lithoth. p. 5.

Lithophyllum Reinboldi Fosl. Lithoth. Mald. et Laccad. p. 466, pi. XXIV, fig. 4.

Syn. Lithophyllum cerebelloides Heydr. in Ber. der Deutsch. Bot. Gesellsch. 1901, p. 405.

Stat. 19. Bay of Labuan Tring, West coast of Lombok. 18 — 27 m.

Stat. 71. Pulu Barang near Makassar. Reef.

Stat. 78. Lumu-Lumu-shoal, Borneo-bank. 34 m. Coral and Coialsand.

Stat. 79'. Pulu Kabala-dua, Borneo-bank. 22 m. Coralsand.

Stat. 91. Muaras-reef, inner side : East coast of Borneo. Lithothamnion-bank near the Island

in shallow water.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion bottom.

Stat. 99. Off North-Ubian. 16 — 23 m. Lithothamnion bottom.

Stat. 125. Off Sawan, Siau Island. 27 m. Stone and some Lithothamnion.

Stat. 193. Sanana Bay, East coast of Sula Besi. Reef.

Stat. 213. Saleyer Island. Reef.

Stat. 234. Nalahia Bay, Nusa-Laut Island. Reef.

Stat. 240. Banda-anchorage. Lithothamnion bank in 18 — 36 m.

Stat. 248. Rumah-Lusi, North-point of Tiur Island. Reef.

Stat. 252. West side of Taam Island. Reef.

Stat. 258. Tual, Kei Islands. 8 — 12 m. Lithothamnion, sand and coral.

Stat. 261. Elat, West coast of Great-Kei Island.

Stat. 282. Anchorage between Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral

and Lithothamnion.

Stat. 299. Buka or Cyrus Bay, South-coast of Rotti Island. 34 m. Mud, coral, and Lithothamnion.

Stat. 303. Haingsisi, Samau Island. Reef.

Stat. 322. i'/i mile south of Tandjong Lajar, South coast of Bawean Island. 32 m. Coral.

SIFOGA-EXPEDIV1E LXI. 7

;o

In classifying this species 1. c. it seemed strange to me that the conceptacles of sporangia,
in a section made thin by grinding, wen- rather differing in shape from those in typically
developed specimens of Lepidomorphzim, one of the two subgenera of Lithophyllum. However,
I classed it wit h the latter, as at the time I had not the opportunity for examining it more
carefullv. nor did I possess good sections. Besides, the conceptacles were smaller than usual in
iolithon, and externally rather resembling those of the genus mentioned above. Afterwards
l have had the opportunity of examining a larger material of the plant, and now l consider
it to be a Goniolithon, although only sporangia are known. The most certain distinguishing
feature between the said two genera is owing to the cystocarps. But also the sporangia are
differing, in as much as in Goniolithon they arise from any part of the disc almost plain or
sometimes even faintly cup-shaped, whereas in Lithophyllum they arise around the peripherical
parts of the disc more or less overarched. Sometimes, however, the snbgenus Lepidomorphum
and the genus Goniolithon are not clearly distinguished only on account of the sporangia, as
these organs, particularly in the said subgenus, often detach themselves, before or towards
maturity, from the bottom of the conceptacle and are then to be found around the walls of the
latter. This is especially the case when the conceptacle is small in proportion to the number
of sporangia developed. In Lithophyllum they often, or rather as a rule, lean against the walls
of the conceptacle even before detached from the disc, as now and then seems to be the case
also in Goniolithon. In addition, heterocysts occur in the genus in question, but in sporangia-
bearing specimens of certain species they seem to be very scarce. In Lithophyllum, on the
other hand, heterocysts do not appear, as far as I have hitherto seen.

In a young stage the plant forms thin crusts on corals or other hard objects (PI. X,
fig. i). It soon develops excrescences or short branches, the latter, however, only once divided
in some of the specimens which I have examined. They are densely crovvded and by and by
anastomosing, so that only the uppermost part is free (PI. X, fig. 2 — 5). In this part they are
frequently 4 — 5 mm. in diameter, sometimes, however, thinner and less confiuent (PI. X, fig. 6).
The shape ot the plant depends on that of the substratum, in general, however, showing a
tendency at length to form roundish balls, which attain a diameter of up to about 8 cm. It
often shares substratum with other species, particularly Archaeolithothamnion erythraeum and
Lithophyllum Okamurai, and it now and then grows over, but, on the other hand, it some-
times also becomes overgrown by such ones. Cp. pi. Y, hg. 11, where this species is almost
covered with A. erythraeum. PI. XI, fig. 16 represents a specimen the greater part of which
belongs to L. Okamurai, being however confiuent with the species in question, which is visible
lowest to the left on the picture, or even to the right between the branches of the former.
Some of the branches of G. Reinboldi are marked with a cross on the picture. PI. IX, fig. 13
also represents a specimen with a few branches ot the species in question, here sharing sub-
stratum partly with /.. Okamurai, partly and especially Goniolithon laccadivicum.

\ vertical section of the crust shows that the hypothallium is composed of a solitary
layer of cells. The latter are vertically elongateel, frequently more or less oblique and 15 — 25 •]..
long. The perithallium is thoroughly formeel by very irregular cells. They are partly rounded,
partly angular, in general 12 — 25 u. in diameter, occasionally, however, somewhat vertically

elongated, with thick walls; fig". 21. Here and there appear short hypothallic rows, which have

been formed over damaged parts of the plant or foreign

bodies. A rather indistinct stratification is to be found, now

and then formed by rows of minute cells. Heterocysts are

very scarce in the sporangia-bearing specimens in hand ; they

are seldom overgrown and then appearing in the inner

layers of tissue.

As mentioned above, sporangia are the only repro-
ductive organs hitherto known in this species. The concep-
tacles appear in great number almost all over the plant, and
they are often so densely crowded that the roofs become
angular. They are at first subconical or subhemispherical, Fig. 21. Goniolithon ReMoidi a. Web. et Fosi.

•,i 1 1 1 • 1 r 11 Part of a vertical section; X 72.

with a coarse, but very snort tip which soon falls away, at

length very low, frequently 350 — 500, or occasionally up to 700 ij., in diameter, when seen
from the surface. The conceptacles by and by become overgrown by new-formed tissue and are
to be found in great number in a section. The plant is furnished with these organs almost all
the year, but particularly old specimens are often sterile.

This species is easily recognized. In habit it approaches to coarse forms of Lithophyllum
Okamurai and L. racemus, sometimes in this respect not easily distinguishable when sterile.
Cp. pi. X, fig. 6 and pi. XI, fig. 14 — 16. It cannot be confounded with any other species
occurring in the Pacific or Indic. But on the other hand it comes near to Goniolithon Börgesenii
from the West Indies. Of the latter, however, only a few specimens are known. They are rather
differing from old specimens of G. Reinboldt in habit, but approach to young ones. These two
species are hardly distinguishable in structure, but in regard to the conceptacles there is some
difference, as the said organs seem frequently to be somewhat larger in the species in question.

Occurrence; Xext to Archaeolithothamnion erythraeum this species seems to be the
most widely distributed one in the Pacific, and apparently in the Indic as well. There have
been brought home but a single or a couple ot specimens from most of the stations cpioted,
many of which partly young, partly rather stunted. At some places, however, it occurs in great
number and vigorously developed, e. g. stat. 91, but it is uncertain whether the plant forms real
banks. It is also known from Pulu Kelamala, east of Pulu Adi, on the north coast of Xew Guinea,
here anastomosed with Archaeolithothamnion erythraeum, collected by Prof. A. Wichmann *).

Area: South Pacific: Samoa; Xorth Pacific: The Carolines, Sandwich Islands ; Indic:
The Maldives and Laccadives, Zanzibar, the Mauritius.

5. Goniolithon laccadivicum Fosl. PI. IX, fig. 10 — 13.

Goniolithon Brassica-florida f. laccadivica Fosl. Lithoth. Mald. and Laccad. p. 469, pi. XXV, fig. 7.

Stat. 91. Muaras-reef, inner side : East coast of Borneo.

Stat. 258. Tual, Kei Islands. 8 — 12 m. Lithothamnion, sand and coral.

1) The Dutch New Guinea expedition 1902 — 1903.

Stat. 501. 1 ast of Rotti [sland. 22 m. Mud. coral and Lithothamnion.

St.u. 311. Sapeh Bay, East coasl oi Sumbawa. Reef.

I classified 1. c. the plant in question as a denominated form with the following
remarks: 1 have heen uncertain whether this plant ought t<> be considered a form of the above
s|>ecies (G. Brassica-florida), or should be regarded as a separate one. Only two fragmentary
pieces are known, and these appear to be rather young. The plant is, however, in several
respects closely allied to G Brassica-fiorida, and, therefore, I place it as a form of the latter
until older specimens may be known. It: somewhat diflers from typical specimens of the species
both in habit and structure, luit on the other hand it must be borne in mind that G. Brassica-
ftorida is a rather varying plant both in habit and even in structure, in the latter respect
owing to the fact that it is frequently attacked by lower animals, especially worms, which nearly
always show a disturbing influence on the normal development of the layers of tissue.

With some reservation I now place the plant as an independent species. It is certainly
not well distinguished from G. Brassica-jlorida, but, according to the material in hand, it is in
some respects so far differing that it seems to be entitled to be kept distinct. Thus it is less
inclincd to form thickened crusts, the consistency is looser, with more regular hypothallic layers
in the branches and frequently rather numerous grown-in rows of heterocysts, ancl the conceptacles
of sporangia are as a rule smaller than in the former. However, as I have already remarked
in Lithoth. Adriat. Meer., it must also be observed that the mutual relation between G. Brassica-
florida and G. mamillare, a species very nearly allied to the former, has not yet been settled.
The species in question seems even to be rather more closely connected with the last named
one. It is, therefore, possible that it will prove to be indentic with, or constitutes a form of,
G. mamillare, or perhaps both will have to be considered as forms of G. Brassica-jlorida.
Typically and well developed specimens of the latter species are till now only known from the
Adriatic, the Mecliterranean and the south coast of Africa. Of G. mamillare only a couple of
certain specimens are known from Bahia (Brazil), i. e. the types in Harvey's herbarium, but
even these are rather young '). Therefore it seems necessary to possess old and well developed
specimens from the American coast, in order to be able to settle the delimitation of these
three species.

In addition to the description of the species is to be observed that the branches in old
specimens are frequently more densely crowded than in the type specimen, the latter and young
one pictured pi. IX, fig. 10. The branches more or less anastomose in the lower part, however
in less degree than in G. Brassica-jlorida, and the crust does not grow so thick as in the
latter, although a nevv crust sometimes may be developed upon the primary one (PI. IX,
fig. 11 — 13). The last quoted specimen is confluent with Litltophxllitm Okamurai ancl Gonio-
lithon Reinboldi. In regard to structure, the medullary hypothallium is more marked than in
the said species, the cells frequently being 15 — 25 ij., long and 10 — 15 0.. broad, i. e. broader
in proportion to the length than in the hypothallium of the crust. Those of the perithallic
layer in the branches are often a little smaller than in the perithallium of the crustlike part

V. II. Harvey. Nereis Australis, or Algae "f the Southern Ocean. Lomion 1S47. p. 109. — As to the occurrence of
Harvey 1. c, at Ugoa Bay on the south coast of Africa, cp. nn remarks in Lithoth. Adriat. Meer.

53

of the plant, though being here more irregular in size. Heterocysts prove to be rather numerous,
and often form grown-in rows. The conceptacles of sporangia, which are the only reproductive
organs known, are to be found somewhat scattered in the branches, and often developed in
the tips of the latter. They are conical, frequently a little contracted in the middle or upper
part, 0,8 — 1 mm. in diameter, when seen from the surface, and they are of about the same
height. Their upper part, more or less elongated, gets dissolved at maturity, and then the
conceptacles often become very low, looking rather differing from those in development. The
sporangia are four-parted and vary between 100 — 160 y.. in length by a breadth of 45 — 60^..
The plant is, in the area in question, furnished with these organs in the months of January,
February, June and December, and at the Laccadives it bears ripe sporangia towards the
end of July.

Besides the relation to the species above mentioned, G. laccadivicurn rather approaches
to G. frutcsccus f. congesta, and young or somewhat stunted specimens are not easily
clistinguished. The branches are however coarser and the conceptacles frequently larger, and
as regards structure, the cells, particularly those of the medullary hypothallium, are frequently
shorter and more thick-walled than in the latter species.

Occurrence: Only one specimen has been brought home from the stations 258, 301,
311, and from stat. 91 a few ones.

Area: Indic: The Laccadives.

6. Goniolithon frutescens Fosl.

Calc. Alg. Funafuti, p. 9.
f. typica Fig. 22; PI. X, hg. 7 — 9.

Foslie 1. c. ; Lithoth. Mald. and Laccad. p. 468, pi. XXV, fig. 4.

f. congesta Fosl. PI. X, fig. 10 — 11.

Lithoth. Mald. and Laccad. p. 468, pi. XXV, fig. 5 — 6.

f. subtilis Fosl. mscr. PI. X, fig. 12 — 13.

Branches rather densely crowded, subfastigiate, terete, about 1 mm. thick.

Stat. 91. Muaras-reef, inner side: East coast of Borneo. Reef.

Stat. 93. Lulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion-bottom.

Sand and coral.

Stat. 220. Pasir Pandjang, West coast of Binongka. Reef.

Stat. 225. South of Lucipara-group. Reef.

Stat. 240. Banda-anchorage. Lithothamnion bank in 18 — 36 m.

Stat. 261. Elat, West coast of Great Kei Island. Reef.

Stat. 277. Kulewatti (Sollot) Bay, Dammer Island. Reef.

The specimens in the Siboga collection of the typical form of this species are partly
attached to other hard objects, partly they have detached themselves and continued their

54

growth freelj at the bottom. They are however more delicate than thosc known from Funafuti
and the Maldives. The branches too are more elongated and less attenuating upwards, but
otherwisi i ing with the said form. Cp. pi. X, fig. 7 — 9 and Lithoth. Mald. and Laccad.

pi. XXV, fig. |. As to structure it fullj coincides with the type, fig. 22.

The form congesta in this collection is not quite
typically developed, being more- irregular, with the branches
less fastigiate than in the type specimen from the Maldives.
The branches are in this form too a little longer than in
the type. Cp. pi. X, fig. 10 — 11, and Lithoth. Mald. and
Laccad. pi. XXV, fig. 5. As mentioned vinder G. laccadi-
vicum, the form in (juestion is sometimes not easily dis
tinguished from the last named species. In habit it also
recalls certain forms of Lithothamnion erubescens.

The most differing form of the species is however
f. subtilis. There is a consiclerable difference in habit between
this form and a coarse f. typica, and still more it cliffers
from a coarse f. flabelliformis \ but the latter form does not
occur in the Siboga collection. The form subtilis is only
met with attachecl to other objects, partly intermingled with
Fig. 22. Gonioiithon frutescens Fosl. filiform algae. It distinguislies itself by rather densely crowded

1 ,,„ section of a bnmch of the branches, which are partly subdichotomously, partly irregularly

type specimen; X 72. l J °

divided and very delicate. The latter are terete, with short
axes, partly a little attenuating upwards, partly not, and about i mm. thick. They often bend
themselves downwards, and continuing their growth they come into contact with the substratum
or extraneous objects. From the downwards turning tip of a branch is then sometimes developed
a small disc-shaped ernst on the substratum, or crustlike formations are formed over the said
extraneous objects. Besides, the branches now and then anastomose with each other. In regard
to structure the form fnlly coincides with the typical one, but the conceptacles are on the
other hand a little smaller. This form much resembles Lithophyllum byssoides Lamour in
habit, and sterile specimens are hardly distinguishable except as regards structure, in which
respect they are, however, quite different.

Occurrence: Only a single and young specimen has been brought home from each
of the stations 9] and 93. At stat. 220 have been collected a number of specimens, which
partly belong to f. congesta, partly form transitions to f. typica, or are referable to the latter,
but mostly not well developed. A couple ot specimens from stat. 240 belong to f. congesta, but
approach in habit to G. laccadivicum. The best developed specimens are from the stations 261
and 277. The form subtilis is only known from stat. 225. Thus the distribution of the species
has a rather wide range within the investigated area, but is more feebly developed and
apparently occurring in less number than at Funafuti and the Maldives.

Area: South Pacific : Samoa, Ellice Islands (Funafuti); Indic : Keeling Islands, the
Maldives and Laccadives.

55

Melobesia Lamour. (Fosl. emend.).

Subgen. Eumelobesia Fosl.

i. Melobesia subtilissima Fosl. mscr.

Thallus forming delicate, monostromatic patches of indefinite shape. Conceptacles /of
sporangia subhemispherical, about 60 u.. in diameter.

Stat. 169. Off Atjatuning, West-coast of New Guinea. Reef.

This species forms delicate patches on Corallina pilifera. The patches are partly very
small, partly somewhat expanding over the branches of the host, but rarely covering parts of
the joints of the latter. The plant is dimcult to detect, as it shows the same light colour as
the host in a dried stage. The peripherical portions of young specimens are sometimes not
confluent, but short subdichotomous rows are formeel, which here and there are anastomosing
as in M. farinosa Lamour. f. callithamnioides (Fkbg.) Fosl. mscr. and M. zonalis (Crn.) Fosl.
The crust is monostromatic except in the neighbourhood of the conceptacles, and it is destitute
of cortical cells. I have at any rate been unable to detect such ones in the material in hand.
In a surface view the cells are partly squarish, 4 — 6 u.. in diameter, partly and especially
towards the peripheral portions rectangular and 7 — 9 u.. long. In a vertical section they are
also squarish or slightly vertically elongated.

The conceptacles of sporangia appear scattered, and they are on the whole rather scarce.
They are subhemispherical and about 60 u.. in diameter, when seen from the surface, with a
single orifice. The sporangia are four-parted, about 25 y.. long and 12 u.. broad. The plant
bears ripe ones in the month of August.

The species in question comes nearest to Melobesia rugulosa Setch. et Fosl. from the
coast of California. It is, however, distinguished by essential characteristics, particularly smaller
cells and smaller conceptacles.

Occurrence: The plant is only known from the station above quoted, where it seems
to occur rather sparingly on the host plant mentioned.

2. Melobesia farinosa Lamour.

Hist. Polyp. p. 315; Rosan. Melob. p. 69.

Stat. 40. Pulu Kawassang, Paternoster Islands. On eelgrass ').

Stat. 43. Pulu Sarassa, Postillon Islands. On eelgrass.

Stat. 45. 70 24' S., n8°i5'.2E. On Turbinaria decurrens. Drifting.

Stat. 53. Bay of Nangamessi, Sumba. On eelgrass.

Stat. 66. Saleyer. On eelgrass.

Stat. 99. North-Ubian. On eelgrass.

Stat. 131. Beo, Karakelang Islands. On Valonia ventricosa.

1) By eelgrass I mean phanerogamous plants with long, linear leaves. These leaves may belong either to Cymodocea, Poscidonia
or Enalis^ three plauts of frequent occurrence in the investigated area. Those specimens covered with Melobesia were sterile, and they
are therefore hardly to be alleged.

56

Stat. [55. Piapis Bay (Telok Sapira) Northwest coast ofWaigeu [sland. On Turbinaria conoides.

Stat. i et, West-entrance Selee (Galewo) Strait, New-Guinea. On Gelidium.

Stat. [65. North-east sul.- of Daram [sland (False Pisangs), East-coast of Misool. On eelgrass.

Stat. 172. ( iissei . l iss.

Stat. 1S4. Kampong Kelang, South-coast of Manipa-island. On eelgrass.

Stat. 213. S.mth [sland near Saleyer. On Laurencia obtusa.

Stat. 220. Pasir Pandjang, West-coast of Binongka. On eelgrass.

Island Leyden, Batavia. On Halimeda macroloba, leg. Dr. BOERLAGE.

Maumeri, Flores. On eelgrass.

Endeh, Flores. On Laurencia concinna1).

Makassar. On Sargassum.

Den Bril, near Makassar. On Laurencia obtusa and Halophila ovalis, leg. SNACKEY.

1 elebes. On Sargassum and eelgrass.
Birakeke, ('elebes. On Sargassum.
Balang-Nipa, Celebes. On eelgrass.

This species attaches itself to a number of different other algae as above stated. It bore
reproductive organs the time when collected, from the month of April to the end of the year,
frequently, however, sporangia and seldom antheridia. The latter organs are developed in
conceptacles which are but 60 — 75 u. in diameter, when seen from the surface.

( Iccurrence; The plant seems to occur in rather great number at most of the
investigated places.

A r e a : Common almost everywhere except in the arctic and antarctic regions.

Subgen. Heteroder ma Fosl.

3. Rlelobesia coronata Rosan.

Melob. p. 64, pi. IV, fig. 9.
f. zonata Fosl.

Xew Spec. or Forms of Melob. p. 9.

Stat. 81. Pulu Sebangkatan, Borneo-bank. On Padina.

Stat. 165. North-east side of Daram Island (False Pisangs), East-coast of Misool. On eelgrass.

Stat. 213. South Island near Saleyer. On Laurencia obtusa.

Endeh, Flores. On Laurencia concinna.

Den Bril near Makassar. On Laurencia obtusa, leg. SNACKEY.

The form of the species considered as typical which underlies Rosanoff's description
1. c, has been found on Pollexfenia pedicellata from Australia, according to specimens in
Lenormand's herbarium. An authentic specimen, which I have examined, forms small orbicular,
reniform or often crenulate crusts. The latter at length grow more or less confluent. The
nproductive organs described by Rosanoff 1. c. as cystocarps are not such ones, but sporangia
which an :loped in more nr less conical conceptacles, with cells as a rule much elongated

1) Where number of station is not quoted, or collector not stated, the specimens have been collected by Madame A. WEBEK.-
VAN BOSSE in iSSS — iSSy. This is also the case with a number of specimens belonging to species of other genera classiried in this paper.
As they are. however. from the same place, though perhaps not the same localities, as the specimens of the same species collected during
the Siboga Expedition, they ave not specially mentioned.

57

around the orifice, forming a corona. These conceptacles are 250 — 350 u.. in diameter, when
seen from above, but they are often not distinctly defined. The sporangia are four-parted,
about 75 — 85 u. long and 35 — 40 u. broad. In other crusts of the same specimen are to be
found conceptacles, which are lower or less conical than the above mentioned ones, and the
cells of the orifice are not or but a little elongated. They are of about the same diameter as
the conceptacles of sporangia, or sometimes a little larger. These are cystocarpic conceptacles,
which seem as a rule not to occur in the same crust as those of sporangia ; but when several
crusts are found on the same substratum, both organs may be met with in the neighbour-
hood of one another. In the same specimen appear also a few other conceptacles, which are
of about the same shape as the cystocarpic ones, but they are only 100 — 150 u. in diameter,
when seen from the surface. These I suppose to be the conceptacles of antheridia.

With some reservation I refer to the species in question the plant stated above from
Borneo bank, forming small, sterile and fragmentary crusts on Padina. The crusts are rather
brittle and have for the greater part fallen away, so that the original shape is no more to be
seen. The plant seems however to belong to M. coronata and to be most closely allied to
the form zo na fa.

In Laurencia obtusa and Laurencia concinna from the places above quoted appear some
very small crusts in company with Melobesia farinosa. They are provided with cystocarpic
conceptacles of about the same size and shape as those in typical M. coronata, with in part
mature carpospores, and the structure coincides in the main with that of the latter. So also in
regard to a few crusts, about 1 mm. large, which together with iSIelobesia farinosa are attached
to eelgrass. They bear conceptacles which probably are young cystocarpic ones.

The specimens here mentioned are partly too fragmentary, partly too feebly developed
to be determined with certainty, especially since the characteristic conceptacles of sporangia
are wanting.

According to Rosanoff 1. c, Dcrmatolithon pustulatum (Melobesia pustulata auct.)
frequently accompanies Melobesia farinosa in the northern temperate zone. It seems, as if the
former is, in the area in question, replaced by Melobesia coronata, which here apparently now
and then accompanies the said species.

Occurrence: The plant seems to occur very sparingly in the places above quoted,
and has hitherto not been met with in a well developed stage.

Area: "Australia" (herb. Lenormand), the type of the species; Indic: South Australia
(f. zonata).

Lithophyllum Phil. (Fosl. emend.).

Subgen. Euli thophyllum Fosl.

a. Crustacea.

1. Lithophyllum oncodes Heydr. PI. XI, fig. 5 — 10.

Lithothamnion oncodes Heydr. Neue Kalkalg. p. 6, t. I, fig. 1 1 ; ex parte.
Lithophyllum oncodes Heydr. in Ber. der Deutsch. Bot. Gesellsch. 1897, p. 410.

SIBOGA-EXPEDITIE LXI. 8

58

Stat. 91- Muaras-reef, inner side: East coast of Borneo. Reef.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. \z m. Lithothamnion-bottom.

Stat. [81. Ambon. Keef.

St.it. 215. North point of Kabia Island. Reef.

St.it. 225. South of Lucipara-group. Reef.

it. 24S. Rumah Lusi, North-point of Tiur Island. Reef.

I do not include in this species the form which, according to Hkydricii 1. c, is furnished
with cystocarpic conceptacles thus described by him: "Die Cystocarp-Conceptakel sind ausserst
selten und bilden auf besonderen Individuen grosse flache und einzelne Erhabenheiten, welche
21 , mm. im Durchmcsser sind; die Offnung ist 125 — 140 u.. gross". This plant cannot be
elassed with the species in question, but probably represents a species of the genus Goniolithou,
likelv with emptied conceptacles judging from the large orifice mentioned in the quoted description.

When the plant grows on an even or fairly even substratum, it forms an even and
smooth crust of a hard consistency. PI. XI, fig. 5 represents such a specimen from Funafuti
almost coinciding in habit with the type specimen 1. c. fig. 11. Here the plant becomes
vigorously developed, partly forming rather thin crusts on branchecl corals, partly, and more
frequently, a thick coat on lumpy corals or other hard objects. It also sticks to consolidated
rock, forming platform or hurricane beach, and attaining a thickness of up to about 1,5 cm.,
frequently, however, less. The figured specimen shows young and circular crusts developed
upon the primary one, and in the upper corner it is confluent with a piece of a coral.

In the Siboga collection occur no specimens so regularly developed as the above
mentioned one, but there are a few rather vigorous ones here too (PI. XI, fig. 9 — 10). The
shape of the crust depends, however, on that of the substratum, as in a number of other and
particularly crustlike species, and sometimes the plant even conveys an impression of developing
excrescences. Cp. pi. XI, fig. 7, which represents a specimen covering a bucky. In this collection
the species in most cases forms more or less irregular and but little expanded crusts in company
with other calcareous algae, Bryozoa, small corals and other animals, attached to hard objects
of different shape and size, mostly, however, dead corals. Sometimes it forms small cap-shaped
crusts on excrescences of corals, together with other algae and sponges, occasionally even
recalling a young Lithophyllum Okamurai (PI. XI, fig. 8). The specimen here figured shows
a looser consistency than usual in this species, probably owing to the substratum.

As regards structure, the perithallic layer of the plant is thus described by Hevdricii 1. c. :
"Die übrige grosse Thallussubstanz wird von keinen bemerkenswerten Schichtungen durchzogen,
weshalb man von irgend einer bestimmten Aufeinanderfolge derselben nicht reden kann ; ab und
zu erscheinen grössere oder kleinere zerstreute Zeilen, die aber dem Inneren ein noch unregel-
massigeres, aber sehr characteristisches Geprage verleihen. Die grosseren Zeilen sind etwa 8 y..,
die kleineren 5 y.. im Durchmesser". — From a specimen distributed by Hevdrich himself and
trom vertical sections of other typical specimens I have found the perithallium to be composed
of cells which are partly squarish or somewhat rounded, 6 — 9 u. in diameter, partly slightly
horizontally or, more frequently, vertically elongated and in the latter case up to about 18 a.
long. The cells are now rather thick-walled, now thin-walled. Besides, in this layer of tissue
are to be found short horizontal rows of larger cells, partly occurring in great number, partly

59

rather scarcely, and differing from the other perithallic cells. The number of cells in each row
generally varies between 6 and 12. They are in general iS — 25 u.. in length by 12 — 15 u.. in
breadth. I cannot with certainty explain the nature of these cells-rows. However, they seem
to be foundations of sporangia, which from some reason or other have not become further
developed. In a horizontal section of a crust these cells are isodiametric and form clusters about
100 or tip to about 150 ;j.. in diameter. They occur in greatest number in specimens which
are richly furnished with overgrown conceptacles of sporangia, but in some parts of a crust they
are quite wanting. Similar formations are to be found in Lithophyllum craspediuvt^ a species
which in structure comes rather near to the species in question. Here too they partly appear
in great number, partly they are very scarce.

Some of the specimens in this collection exhibit cells which in part are more regular, more
commonly vertically elongated and larger than generally seen in this species. They approach to
a form of the species which I have mentioned in Lithoth. Mald. and Laccad. p. 468, but I
did not find the difterence to be so important as to allow of classifying it as a separate
species or form.

The plant is in most cases sparingly furnished with sporangia. It has been collected in
the months of June and September — December.

Occur re nee: There have been brought home but a couple of specimens or even
solitary specimens from stations 91, 93, 181, 215 and 248. This fact seems to indicate that
the plant has been scarce in these places. On the other hand it has apparently occurred in
great number at stat. 225.

A r e a : South Pacific : Tami Island on the north-eastern coast of Xew-Guinea ; Xorth
Pacific: Ellice Islands (Funafuti), California (San Diego) ] ) ; Indic: The Maldives and Laccadives,
the Mauritius; North Atlantic: West Indies (St. Croix?) ').

b. Ram o sa.

2. Lithophyllum Okamurai Fosl.

Five new calc. Alg. p. 4.
f. japonica Fosl. PI. XI, fig. 13 — 19.

Det Kgl. Norske Vidensk. Selsk. Skrifter 1900, Aarsberetning.

f. angularis Fosl. PI. XI, fig. 11 — 12.

Det Kgl. Norske Vidensk. Selsk. Skrifter 1900, Aarsberetning.

Syn. Lithophyllum cephaloides Heydr. in Ber. der Deutsch. Bot. Gesellsch. 1901, p. 271.

Stat. 19. Bay of Labuan Tring, West coast of Lombok. 18 — 27 m. River-mud, coral, coralsand.

Stat. 91. Muaras-reef, inner side: East coast of Borneo. Reef.

Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion bottom.

Stat. 99. North Ubian. 16 — 23 m. Lithothamnion bottom.

1) I am for the present unahle to draw any limit between the type of this species and a specimen in hand from each
of these places.

6o

Stat. [09. 1'uhi ! 1!. Sulu Archipelago. 13 m. Lithothamnion bottom.

Stat. 25S. Tual, Kei Islands. 8 — 12 m. Lithothamnion, sand and coral.
Stat. 303. Haingsisi, Samau [sland. Reef.

When I classified this species 1. c. I considered it to be closely allied to Lithophyllum
fasciculatum, a species which, however, comes very near to L. racemus (PI. XI, fig. tij. I then
possessed specimens only of the form of the species which I have afterwards named f. augularis
(PI. XI. fig. 11 — 12). Simultaneously with the latter I described the form Japouica, which is
strongly resembling L. racemus in habit. (PI. XI, fig. 13). The species includes some specimens
which I have formerly considered referable partly to L. fasciculafum \ partly to L. racemus.

I will not here enter more fully upon the question whether this species is in fact specifically
distinguished from L. racemus, nor on its relationship to other nearly allied species, as L.
Kaiserii Heydr., L. pallescens Fosl. (including L. californiense Heydr.), L. affinc Fosl., L,
fasciculatum (Lam.) Fosl., and L. Andrussowii Fosl. — I have already remarked in Lithoth.
.Malei, and Laccad. that I keep them distinct until they have been more fully examined in a
larger amount of material from different localities. This particularly concerns the first named
three species. I regard L. Okamurai as an independent species owing to the fact that the cells
are as a rule longer in proportion to the breadth than in L. racemus. In this respect, however,
the species in question can hardly be distinguished from L. fasciculatum, the latter hitherto with
certainty known only from the west-coast of Ireland :). Besides, the conceptacles of sporangia
are a little more peripherical than usual in L. racemus, although 110 certain limit is to be drawn
in this respect. In addition, among a rather great number of specimens examined from the coast
of Japan I met with a single one, in which a part of the lower layers of tissue exibits alter-
nating long and short cells. Such ones have never been seen in L. racemus. In this respect
the plant approaches to L. pallescens from the Pacific coast of America. In the latter species
the cells are, however, more frequently alternating long and short, though in such cases often
rather irregularly. This is also the case with L. a/'pue from the Red Sea. The last named
species is partly closely allied to L. racemus, partly to L. pallescens in structure, whereas in
habit it comes near to delicate forms of L. Kaiserii, to which it in part also approaches in
structure. Therefore, it is necessary to have access to a number of well developed specimens
particularly from the Red Sea and from the Pacific coast of America in order to be able to
settle the delimitation of the said species or forms 2).

I have remarked before in this paper that L. Okamurai sometimes also recalls other
species than the above mentioned ones in habit, species which however in other respects are
quite differing. Thus it occasionally comes very near to certain forms of Goniolithon Reinboldi
as to habit. Cp. pi. X, hg. 6. Xow and then it also gets confluent with the latter, as to be
seen on pi. XI, fig. 16, where some branches belonging to G. Reinboldi are marked with a
cross. Besides, pi. IX, fig. 13 represents a specimen which in the main belongs to Goniolithon
laccadivicum , sharing however substratum with the species in question as well as Goniolithon

1) M. Foslie. New or critical calcareous Algae. — Det Kgl. Norske Vidensk. Selsk. Skrifter 1899. n° 3. Trondhjem 1900. p. 30.

2) I intend later on to publish an account of the unarticulated Corallinaceae, with pictures of all the species and forms
known to me.

6i

Reinboldi. The plant also approaches in habit to Archaeolithothamnion erythraeum, though
more seldom, and cannot even in this respect be confounded with the latter except when the
specimens are dead, water worn or much bleached. Such a specimen, anastomosed with the
said species, is represented pi. XI, fig. 17,

In the Siboga collection the species in question is not well represented except from
station 91, where a number of large and well developed specimens have been collected (PI. XI,
hg. 14 — 16). These specimens belong to f. japonica. On the other hand, there occurs no old
specimen of f. angularis, but only a couple of young ones showing transition to this form.
At stat. 303 the species is represented by a delicate form, which partly resembles Lytlwphyllum
Marlothii in habit. It is, however, differing from the latter in structure, as the cells are in
part even a little longer than commonly seen in L. Okamurai. Therefore, I consider it as a
somewhat stunted form of this species (PI. XI, fig. 18 — 19). The plant in question bears
sporangia in the months of February and June, but scantily.

Occurrence: This species has apparently been abundant at stat. 9 1 , the only station
in the investigated area where it occurs typically and well developed. From the other stations
above quoted have been brought home but a single or a few young, frequently stunted and
not always quite certain specimens.

Area: South Pacific : Tami Island on the north-eastern coast of New Guinea ; North
Pacific : The Pacific coast of middle Japan l).

Subgen . Lepidomorph u m Fosl .

a. Crustacea.

3. Lithophyllum Yendoi Fosl. PI. XI, fig. 1 — 4.

New or crit. Lithoth. p. 25.

Stat. 51. Madura Bay and other localities in the southern part of AIolo Strait. Shore.

Stat. 61. Lamakera, Solor Island. Reef.

Stat. 86. Dongala, Palos Bay, Celebes. Reef.

Stat. 155. Piapis Bay (Telok Sapira), Northwest coast of Waigeu Island. Reef.

Stat. 181. Ambon. Reef.

Stat. 193. Sanana Bay, East coast of Sula Besi. Reef.

Stat. 240. Banda-anchorage. Reef.

The type specimen of this species is picturecl pi. XI, fig. 1. In the area investigated
the plant fully coincides with the type from the coast of Japan (PI. XI, fig. 2 — 3). As regards
structure, however, the cells sometimes prove to be a little larger than the measures given 1. c.
But there is no great variation in this respect.

When growing on an even substratum, the plant is almost quite even, the crust closely

1) In the Bvitish Museum a couple of specimens ave kept under the name of Lithothamnion polymorphum from the Rodri^uez
Island in the Indic. They were some years ago placed by myself under Lithophyllum racemus. The specimens also come very near to
L. Okamurai in habit, but are likely to be referred to a densely branched form of L. Kaiserii^ if as remarked above, the latter is on
the whole to be maintained as an independent species. — Cp. George Murrav. An Introduction to the Study of Seaweeds. London
1895. Fig. 83 (p. 242).

and firmly adherent to the substratum. But on an irregular substratum it forms more or less
uni md irregular crusts, which little by little become continent, and sometimes at length

look as it' developing small excrescences. There are two specimens from stat. 240, one of
which i-- attached to an irregular substratum, apparently some volcanic mass. The crusts of

this specimen an- verv irregular, partly with small excrescences which are, however, owing to
the uneven substratum or to the covering up extraneous objects or dead crusts of the same
species. The other specimen is quite different in habit from a typically developed specimen,
formin» numerous and crowded wartlike excrescences or short branches about 1 mm. in
diameter. It has been attached to some rather loose and irregular object. Upon the primary
and at length anastomosed crusts, extraneous objects have attached themselves, and new crusts
have heen developed upon the latter. These crusts have been at least partly water worn, and
new crusts have again been developed together with another appearance of extraneous objects
in a rather irregular way, the crusts of the alga often converging and a little rising, and at
length more or less anastomosing, or covering up the said extraneous objects or roughnesses
of the subjacent parts. In this way the mentioned excrescences are formed. A vertical section
exhibits hvpothallic cells which are partly more elongated than frequently to be seen, sometimes
even forming a rather vigorous layer of tissue, though the hypothallium is in general slightly
developed in this species. The first mentioned specimen belongs without doubt to the species
in question, and also the other one is most probably a form of the same species, only peculiarly
developed owing to the substratum.

In Lithoth. Adriat. Meer. I have mentioned that particularly Lithothamnion Philippii
develops partly a vigorous, partly a rather feeble hypothallic layer, depending on the shape
and nature of the substratum. The consistency is rather varying too, sometimes rather loose,
sometimes firmer than usually met with. A similar condition also appears in Lithothamnion
frutictilosum, GoniolitJion Foslici, Lithophyllum expansum and especially Lithopkyllum incrustans.
Under Lithophyllum oncodes I mentioned a specimen with comparatively loose consistency, —
firm consistency being the general character of the species. The species in question offers a
striking instance of a similar state of things. Thus pi. XI, fig. 2 represents a specimen covering
a piece of rather hard woocl. It is closely adherent to the substratum, although less firmly than
if the plant is attached to stones, as e. g. the specimens pictured pi. XI, fig. 1 and 3. However,
it does not difter in any essential degree from the latter ones, neither in habit, nor in consistency
and structure. On the other hand, the specimen pictured pi. XI, fig. 4 is rather differing in
both the latter respects, particularly, however, as to consistency, and in habit it shows a
more regular outline than usual. It is attached to a piece of very loose wood, i. e. mangrove
( 'Rhizophora) \ forming almost circular and thin crusts which are somewhat bleached. The con-
sistency is looser, if not much looser, than specimens developed on a hard substratum. Besides,
in a vertical section of a crusts, the hypothallic cells are in part larger than usual in this species,
but not el( mgated or less elongated than e. g. ihose in the specimens above mentioned from
stat. 240. They often form almost rhizoid-like sinuosities or clustered crenules towards the
substratum. The perithallic cells are. however, of about the same size and shape as in the
typical form, although sometimes a little larger and oftener faintlv elongated in vertical direction,

63

up to about 10 u.. long. The crusts are rather young, but fertile, and similar young crusts are
to be found in the form attached to the other piece of wood mentioned. Thus the difference
lies principally in the consistency and partly the structure too.

This species is easily recognizable, but, as I have remarked under Lithothamnion
erubescens, it may be questionable whether the plant is in fact distinguishable from Lithophylhixa.
decipicns from the south coast of South America. There are up to the present known but a
few specimens of the latter. It seems to be a little differing particularly as to structure. L.
Yendoi also comes very near to L. subtenellum from the southern Atlantic coast of Europe
and the western coast of Africa. It is furnished with reproductive organs in the months of
April to June, August to September, and in November.

Occurrence: Only one or up to three specimens are known from each station. The
plant has been best and most typically developed at the stations 51, 155 and 193.

Area: North Pacific: The Carolines, the Pacific coast of middle Japan and the Gulf ofSiam.

b. Subramosa.

4. Lithophyllum papillosum (Zan.) Fosl.

Rev. Syst. Surv. Melob. p. 20.

Lithothamnion papillosum Zan. Saggio p. 43 (sine descr.); Hauck, Meeresalg. p. 272, t. II, fig. 4.

f. Cystosirae (Hauck) Fosl.

Lithoth. Adriat. Meer.

Melobesia Cystosirae, Hauck I. c. p. 266, t. III, fig. 1, 2, 6.

Melobesia (?) Cystosirae Fosl. Rev. Syst. Surv. Melob. p. 21.

Stat. 282. Between Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and
Lithothamnion.

As mentioned 1. c. I have been unable to draw any true limit between Melobesia Cystosirae
and Lithophyllum papillosum. Therefore, I placed the former as a form of the latter. In the
Adriatic the plant sticks to stones or other hard objects, e. g. shells of molluscs or a conglo-
merate of Serpulides and calcareous sediments. Besides it is fastened to Cyslosira, exceptionally
to filiform algae heaped together, and it has also been
met with on the branches of Lithothamnion J ' rutiailosum.

In the Sibog^a collection occurs a small and frag-
mentary, almost quite even crust about 3 mm. in diameter
and 0,2 — 0,4 mm. thick. It sticks to a fragment of a
branch of Archaeolithothamnion sp. With some doubt
I refer this crust to f. Cystosirae, to which it agrees in
structure. It must, however, be borne in mind that the
species in question is much varying in this respect, and
the cells are often longer than in the said crust, but often also of exactly the same shape and
size. The hypothallium in this species is formed partly by a solitary row of oblique cells, partly

Fig. 23. Lithophyllum papillosum (Zan.) Fosl. forma.
Vevtical section of the crust; X 72.

°4

several rows of more elongated cells. The. perithallic cells are extremely varying too, from
about i2 up to about 60 u. in length. They are nearly always rather narrow when much
elongated, and often comparatively broader when less elongated. In addition, the cells are
most irregular when the plant grows on an uneven substratum, and new crusts are formed
one above the other, or with numerous new hypothallic formations sometimes being even
marked coaxial. Very long cells and short ones are frequently to be seen in one and the same
section. In this respect L. papillosum to some degree corresponds with Lithophyllum incrustans,
which exhibits sometimes a very slightly developed hypothallium with rather short cells,
sometimes a vigorous hypothallic layer and more elongated cells in the perithallium. This
depends on local conditions and the shape as well as partly also the nature of the substratum.

The mentioned crust in this collection bears some emptied conceptacles (of sporangia?).
They are of about the same shape and size as in L. papillosum, partly, however, a little more
prominent, but the value of this character cannot be elucidated from such a scanty material
of a much varying form. At any rate, however, there seems to be no reason at present to
keep it distinct from the above species.

Occurrence: Only the mentioned small crust is known from the station above quoted.

Area: The Mediterranean and the Adriatic.

c. Ramosa.

5. Lithophyllum Bamlcri Heydr. PI. XII, fig. 1.

Corall. in Bei', der Deutsch. Bot. Gesellsch. 1897, p. 412.

Lithothamnion Bamleri Heydr. Neue Kalkalg. p. 4, t. I, fig. 1 — 3.

Syn. Perispermum hermaphroditum Heydr. in Ber. der Deutsch. Bot. Gesellsch. 1901, p. 409 '|.

Stat. 79l'. Pulu Kabala-dua, Borneo-bank. Reef.

Stat. 282. Between Nusa Besi and the N.E.-point of Timor. Shallow water.

I do not hesitate to class with this species a couple of specimens in the Siboga collection.
One of thein is very small and young. It is attached to a coral, but it shares substratum and
is partially covered with a somewhat stunted Lithothamnion simulans and another calcareous
alga which probably represents a stunted form of Lithophyllum oncodes. The other specimen
seems also to have been attached at first to some hard object, but has afterwards detached
itself. (PI. XII, fig. 1). In habit both specimens fully coincide with the type. Cp. 1. c. t. I,
fig. 1 — 3. But as to structure there is a great difference to judge from the description given
by Heydrich"). A longitudinal section of a branch is described by him as follows: ''Die centralen
Zillen bilden nicht, wie bei der vorigen Species, eine lange und eine kurze Reihe, sondern meist
zwei lange Reihen von 1 p.. Breite und sehr verschiedenen Langen, meist 4 — 6 ju.., und eine
Reihe quadratischer Zeilen von nur 1 u.. Grosse. Die Schichtungen kommen also hier durch die
Aufeinanderfolge von 2 langeren und einer kurzen Zellreihe zu Stande. Die peripherischen Reihen
bestehen aus rundlichen 1 — iV2 u. im Durchmesser fassenden Zeilen mit dichtem Zellinhalt".

I>ie Entwickelungsgeschichte des Corallineen-genus Perispermum Heydrich.
2) F. Heydrich. Neue Kalkalgen von Deutsch Neu-Guinea. (Kaiser Wilhelm's Land)-Bibliotheca botanica H. 41, Stuttgart 1897.

65

However, an authentic fragment which I have examined is in this respect quite different. Thus
in a median longitudinal section of a branch the longer cells of the medullary hypothallium are
27 — 50 u.., sometimes even up to about 60 u.. in Iength and 7 — 12, rather seldom up to 15,
but frequently about 10 u.. in breadth. The shorter cells of the said layer are 12 — 24, or up to
30, mostly 16 — 20 a. long. The perithallic layer exhibits cells which are squarish, 6 — 10 a. in
diameter, or vertically elongated and up to 18, occasionally even up to 24 [i. long. Besid^s,
as to the reproductive organs Heydrich states 1. c. : "Conceptakel 600 — 700 a., oberflachlich,
wenig erhabene Warzchen nur auf der Rückseite der Verzweigungen bildend". On the following
page, however, he states: "Conceptakel mit Tetrasporangien habe ich nicht gefunden. Die
Cystocarp-Conceptakel bilden ausserlich flache meist 1 mm. im Durchmesser fassende Erhaben-
heiten mit wenig erhabener Entleerungsöffnung. Die Höhle ist meist 300 ij., lang, 50 p.. hoch.
Carposporen oval". On the other hand, according to my own measurements they are but
400 — 500 u. in diameter when seen from the surface, subprominent and often not well defined.
In a median vertical section they are however of about the size last quoted, though now and
then somewhat larger. I will not at present express my opinion whether these organs are
cystocarpic, or hermaphroditic, or if the plant in fact is a Lithophyllum, which, however, seems
to be probable. I shall corae back to these questions on some future occasion, and for the
present leave them in abeyance, particularly because of lack of opportunity of examining a
sufficiënt large material. A few conceptacles examined were empty.

In comparing the description of the structure of e. g. Lithophyllum Tamicnsc Heydr.
1. c. with that of L. moluccense Fosl. '), both afterwards identified by Heydrich himself, it will
be seen that his description here is almost as far from being in accordance with the fact as
his description as regards the species in question. This is in the main also the case as to
Archaeolithothamnion erythraetim (Sp. ptychoidcs Heydr.). Cp. Foslie. Melob. Arb. Heydr. p. 16.
The instances of incorrect description might be continued, but I suppose these instances to be
sufficiënt to support my identification of L. Bamleri on a small fragment rather than on the
description of the species 2).

I have above subsumed Pcrispernuim hermaphroditum Heydr. under the species in
question. Of this plant too I have had the opportunity of examining an authentic fragment 3).
It fully resembles L. Bamleri in habit, with in part slightly thicker branches. As to the
structure of the former, Heydrich states 1. c. p. 411: „lm Langsschnitt tritt aber besonders
auffallend die facherförmige Stellung der Zellschichten hervor, die in regelmassig scharf abwech-
selnden Reihen den Thallus durchziehen. Die einzelnen Zeilen eines central orelegenen Fadens

1) M. Foslie. Calcareous Algae from Funafuti — Det Kgl. Norske Vidensk. Selsk. Skrifter 1900, n° r, p. 10.

2) "Nun enthalten leider die Diagnosen vieler Arten nur Angaben über den Habitus; in Folge dessen kommt es recht haufig
vor, dass eine sichere Bestimmung erschwert wird, oder dass man sich vor die Möglichkeit gestellt sieht, eine Pflanze mit demselben
Recht zu zwei oder drei Species zahlen zu könuen. Wollte man dann die fragliche Pflanze so ohne Weiteres einer dieser Species zutheilen,
so möchte dies der Botanik, als einer exacten Wissenschaft, wenig nutzen, und der Autor könnte nachher in die unangeDehme Lage
kommen, dass die nicht beschriebenen Merkmale solcher Species mit denen der von ihm hinzugerechneten Pflanze nicht im Einklang
stehen. Es ist daher jetzt unbedingt nothwendig, dass ein Autor, wenn er Prioritatsansprüche erheben will, die Diagnose nicht nur auf
den Habitus beschrankt, sondern sie auch auf Zeilen und Früchte ausdehnt". Heydrich in Ber. der Deutsch. Bot. Gesellsch. I90i,p. 418. —
I want to add that it is worse when an author cannot be relied on as to the description, or when he quotes incorrectly.

3) I express my best thanks to Dr. Jean Chalon, who has shown me the favour to spare me a part of a specimen of this
species as well as a couple of other ones distributed by Heydrich himself.

SIH0GA-EXPED1TIE LXI. 9

66

wechseln mit kurzen von 14 a. Breite und 20 p. Lange und langen, von glejchfalls 14 ;;..
Breite, aber 40 a. Lange regelmassig ab. Nach der Peripherie zu bleiben sic sich darm in der
('.ros-,,' gleich. Die oberen Zeilen enthalten 1 — 3 langliche, kaum 2 7.. grosse Chromatophoren ;
:gen schliessen die langen mittleren Zeilen bis 80 kleine, kaum 's <i. (!) grosse Chromato-
phoren ein". It would seem from what is quoted that the cell-rows might be alternating one

long and one short, as in Lithophyllum moluccense, but in fact they are
alternating exactly as in L. Bamleri, i. e. usually two long and one
short row (Fig. 24). This alternation is however not always regular, as
in one and the same section sometimes can also be found now a few
long rows, now one long and one short. As to the size of the cells the
plant also fully coincides with L. Bamleri, i. e. my measurements stated
above. A few conceptacles of sporangia occur in the said fragment. They
are convex, but little prominent, 400 — 500 (u. in diameter, when seen
from the surface. The sporangia are four-parted, 55 — 70 u.. long and
about 30 u. broad. These organs mainly coincide in shape and develop-
ment with the genus Litlioplixllum, and they are similar to those in
L. Bamleri. I have certainly not seen the sexual organs in Perispermum.

Fig. 24. Perispermum , , . , .

naphroditum Hevdr. I here cannot, however, be any doubt that those mentioned by Hevdkicii

of a median section j c are ic]entic with the cystocarps described by him under L. Bamleri.

of a branch: X 72.

In addition, it is a significant fact that he 1. c. compares P. hermaphro-
diium with eight other species, several of which are very distantly related, but not with L.
Bamleri, which he received even from the same spot as P. hermaphroditum.

I place the two specimens in this collection under the species-name first given to the
plant, but as above remarkecl, I will not here enter more nearly upon the question whether
the species perhaps represents a separate genus. The organs considered as cystocarpic concep-
tacles certainly seem to be somewhat differing from those in typically developed species of
Lithophyllum, but it is very uncertain whether the value of this difference is of such importance
as to justify the classification of a new genus. In some respect they come near to those in
Goniolithon too. As for Perispermum, however, Heydrich bases this genus on hermaphroditic
sexual organs, which is no generic character. Besides, he states the formation of the procarp
as a co-ordinate character, which he also applies to other genera. I have formerly mentioned
the value of the latter as very uncertain, and in this respect I here refer to Falkenberg's
important work on the Rhodomelaceae '). His remarks are in the main also applicable to the

1) P. FALKENBERG. Die Rhodomelaceen des Golfes von Xeapel und der angvenzenden Meeres-Abschnitte. — Fauna und Flora
olfes vun Neapel. 25. Berlin 1901. — As for the mentioned relation he states as follows p. 103:

"So characteristisch für die ganze Familie der Rhodomelaceen der Bau des Procarps und dessen Entwicklung zum Cystocarp

Het, die in ihren grossen Umrissen constant festgehalten werden, so mannigfache Varianten können daneben in Einzelheiten

auftreten: in der Bildung accessorischer Astchen durch die fertile Pericentralzelle neben dein Carpogonast, in dem Abort dieser Astchen

oder ihrer Entwicklung zu sterilen Fadenbüscheln, in dem Grade der Fusionirung der Cystocarpzellen. Es liegt nahe, diese Charactere

i weiteren systematischen Gliederung der grossen Familie zu verwenden. Dass es zum Theil schwer constatirbare Charaktere sind,

Irde ihrer praktischen Verwendbarkeit zwar Schwierigkeïten in den Weg legen, ihr wissenschaftlicher Werth würde aber dadurch ja

nicht alterirt. Von dem Versuch, auf sie l'nterfamilien zu begründen, habe ich aber fruh Abstand genommen, weil durch ihre Verwendung

hster Verwandtschaft in unnatürlicher Weise hatten aus einander gerissen werden mussen, und diese Schwan-

kungen in Einzelheiten mehr den Charakter unwesentliclier Zufalligkeiten tragen.

6;

Corallinaceae l). Hevdrich further states that the development of the cystocarpic conceptacles
is quite different from that of Sphaeranthera, a genus based by him upon the type of Litho-
thamnion Philippii Fosl. [Lithophyllum decussatum Solms, nee Phil.). I have however shown
in Lithoth. Adriat. Meer 2) that this genus with its single species (Sph. decussata) comprehends
at least two different algae 3) and that the antheridia described by him have nothing to do
with the type of the species 4). For these reasons I keep the species in question provisionally
at least under the genus Lithophyllum.

Occurrence: The two specimens above mentioned are from stat. 282. Afterwards,
however, I met with a small but rather well developed specimen from stat. 79b.

Area: South Pacific: The Tami Island on the north-eastern coast of New Guinea.

6. Lithophyllum moluccense Fosl.

On sorae Lithoth. p. 12 ; Calc. Alg. Funafuti, p. 10 ; New Melob. p. 12.

f. typica. Fig. 25 — 26; PI. XII, fig. 2—4, 6, 8.
FOSLIE. Melob. Arb. Heydr. p. 24.

f. flabelliformis Fosl. PI. XII, fig. 5, 9 — 10.
Melob. Arb. Heydr. p. 24.

f. pygmaca (Heydr.) Fosl. PI. XII, fig. 7, 12 — 13.

Rev. Syst. Surv. Melob. p. 16; Melob. Arb. Heydr. p. 24.
Lithothamnion pygmaeum Heydr. Neue Kalkalg. p. 3, t. I, fig. 8 — 10.
Syn. Lithothamnion Tamiense Heydr. 1. c. p. 1, t. I, fig. 4 — 7.

Stat. 47. Bay of Bima ; near southern fort. Shallow water.

Stat. yS. Lumu-Lumu-shoal, Borneo-bank. 34 m. Coral and coralsand.

Stat. 91. Muaras-reef, inner side: East coast of Borneo. Lithothamnion-bank near the islands,

in shallow water.
Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion-bottom.

Sand and coral.
Stat. 129. Kawio and Kamboling Islands, Karkaralong-group. Reef.
Stat. 131. Beo, Karakelang Islands. Reef.

Stat. 149. Fau Island, lagune. West coast of Gebé Island. Shallow water.
Stat. 193. Sanana Bay, East coast of Sula Besi. Reef.

Diesen Resultaten entsprechend hat auch Schmitz darauf verzichtet, fiir die Abgrenzung der Unterfamilien der Rhodomelaceen
irgerid wie auf den Bau der Procarpien oder Cystocarpien zurückzugreifen. Wenn in den Gattungsdiagnosen von ihm regelmassig der
Bau der Cystocarpwand und der Gonimoblasten berücksichtigt wird, so gescliah das mehr der Uniformiteit wegen, zu der die Darstellungs-
weUe der übrigen Familien drangte, als dass er selbst besonderen Werth auf den Bau der Cystocarpien gelegt hatte".

1) As regards the cystocarps as a genene character, I have in Rev. Syst. Surv. Melob. applied these organs in about the same
sense as indicated by SoLMS-LAUBACH, Corall. Monogr. p. 53 and 64. This character seems to me to be well marked and as a rule
easily ascertainable.

2) Wissensch. Meeresuntersuchungen, Abt. Helgoland. — The paper was sent to be printed in the autumn 1902, but owino-
to unforeseen circumstances it has not yet been printed (March 1904).

3) Lithothamnion Philippii and probably young, but fertile specimens of Goniolithon Brassica-florida,

4) Cp. M. MÖB1US in Ji st's Botanischer Jahresbericht XXIX (1901). Berlin 1903, p. 305: "Die „Antheridien", ebenso unvoll-
kommen beschrieben, wie abgebildet, dürften wohl Parasiten sein".

68

Stat. 213. Saleyer [sland. K

--t.n Pasir Pandjang, West coast of Binongka. Keef.

Stat. 240. Banda. Lithothamnion-bank in iS — 36 m.

St.a. 248. Rumah Lusi, North-point of Tiur Island. Reef.

Stat. 2S2. Between Nusa Besi and the N.E.-point of Timor. Shallow water.

St.it. 311. Sapeh Bay, East coast of Sumbawa. Keef.

Stat. 513. East of Dangar Besar, Saleh Bay. Reef.

. species was, among some other ones, classified by myself without mentioning the
structure, which was however afterwards described 1. c. For this reason, Hevdrich remarks in Ber.
der Deutsch. Bot. Gesellsch. igoi, p. 419, stating the fact that L. moluccense and L. Tamiense
are identic: "Da diese Pflanze nun erst nachtraglich vom Autor sic her gekennzeichnet

ist, so gehort nicht FoSLlE, sondern
mir die Prioritat, denn ich hatte zwei
Jahre früher Lithophylluin Tamiense
30 ausgiebig beschrieben, dass ein Ver-
wechseln mit einer anderen Species
nicht stattfinden konnte; dies war aber
bei der FosLiE'schen ersten Beschrei-
bung recht wohl möglich". Cp. 1. c.
p. 418. However, if a provisional
description of a plant without des-
cription of the structure should induce
the consequence indicated by Heydrich,
several species could be disregarded, among these even species described by Heydrich himself.
Besides, in Melob. Arb. Heydr. p. 17 I have proved that the description of L. Tamiense is
quite incorrect as to structure. This seems to me to be a matter of deeper regret than a not
complete description of a species. Cp. above under Lithophyllum Bamlcri and L. oncodes.
The type specimen from the Moluccas is represented fig. 25.

In regard to the forms of the species I here omit f. unispinosa Heydr. Cp. 1. c. p. 1,
t. I, fig. 4. This form 110 doubt only represents fragments of old specimens particularly of the
typical form. In addition, as to the form described by Heydrich as the typical one I remarked
1. c. : "Ich nehme L. moluccense f. typica in einer anderen Bedeutung als Heydrich 1. c. (L.
Tamiense), indem ich zu dieser Exemplare mit runden, subdichotomen oder unregelmassigen
Zweigen rechne, wahrend ich für die clichotom-facherförmige Form, welche Hevdrich 1. c. f.
txpica benennt, den Namen jlabelliformis vorschlage, entsprechend ahnlichen Formen bei einer
Reihe anderer Arten". Consequently I also here class these forms in the sense quoted. Of the
1 fiygmaea, described by Heydrich 1. c. as an indepentent species, I have certainlv not seen
any authentic specimen '). ( )n the other hand, I have had the opportunity of examining a
specimen from the Mauritius, which Heydrich afterwards also classed with his L. pygmaeum.
But, as I have shown in Melob. Arb. Heydr. p. 23, this specimen cannot be identic with

Fig. 25. Lithophyllum moluccense Fosl.

The type specimen (Herb. M. MÖBIUS);
A seen from above; B seen from below; nat. size.

.10 1. c. only represent- a youüg specimen of f. typica or f. flabellifo

óg

f. pygmaca in the sense first taken, if his description is correct, being on the contrary a typical
L. torquescens Fosl. Referring to my remarks 1. c. I do not think to be wrong as to the
apprehension of the said form. As regards L. torquescens I stated 1. c. p. 24: "Es ist möglich,
dass auch L. torquescens sich blos als eine Form von L. molluccense herausstellen wird, aber
nach dem vorliegenden Material kan dies zur Zeit nicht mit Bestimmtheit entschieden werden.
Unter allen Umstanden ist es nicht identisch mit f. pygviaea". Having now seen a great number
of specimens of the species in question, I consider L. torquescens to be only a form of this
species. It is not represented in the Siboga collection, but there are specimens which form
transition to it. Cp. pi. XII, fig. 11. This form is characterized by very short branches which
are thick at the base and rather attenuating upwards. It somewhat recalls coarse forms of
L. byssoides in habit. As to structure it also proves to run into the species in question.

The forms typica (pi. XII, fig. 2 — 4, 6 — 8) and f. Jlabelliformis (pi. XII, fig. 5, 9, 10)
are but very little differentiated, one often showing transition to the other. Cp. pi. XII, fig. 6,
representing an old specimen of f. typica, partly however shovving transition to f. Jlcibclliforinis.
The form pygmaea is distinguished by its branches rather uniform in thickness and thinner
than in the other forms. (PI. XII, fig. 7, 12, 13). It partly runs into f. typica, partly shows
transition to f. torquescens. This form on the other hand corresponds with Goniolithon frutescens
f. subtilis in habit. Cp. pi. X, fig. 12 — 13. It can even be confounded with the latter if it is
not more closely examined. The consistency is however firmer ancl the structure quite different
viewed apart from the reproductive organs.

As to the structure of this species I refer to my descriptions 1. c. I will only add that
the cells prove to be more varying in size than formerly known, sometimes being thoroughly
large (fig. 26^), sometimes rather
smal! (fig. 26 B). This variation in
structure appears however in part
to depend on local conditions.

The plant is apparently always
at first attached to other hard objects,
particularly corals, but afterwards it
sometimes detaches itself from the
substratum. If the object is small, it
becomes often at length fully encom-
passed by the plant. However, broken
or loosened branch-systems especially
of rather young specimens seem as
a rule to continue their grovvth freely,
conveying the impression of having
been freely developed at the bottom
(PI. XII, fig. 2). The plant is often

rather burdened with foreign bodies which however seem to be of less importance to a normal
development than in several other species. It is furnished with reproductive organs in the

Fig. 26. LilhophyUiim moluccense Fosl.

A. Part of a median section of a brandt of the type specimen;

B. Part of a median section of a branch exhibiting smaller cells; X 72.

months of January and February, and from June to the end oi the year, but in the specimens
collected they most frequently occur in smal! number.

Occurrence: The species has apparently occurred abundantly at the stations gi, 220
and 240, perhaps also at most of the other stations, from which, however, partly only one,
partly a few specimens have been brought home. Nearly all the specimens from stat. 220
belong to f. pygmaea, whereas al the other stations only f. typica has been represented, nou-
and then with scattered specimens of f. flabelliformis.

Area: Pacific: The Moluccas, the Carolines, Tami Island on the north-eastern coast
of New Guinea, and at Queensland; Indic: the Mauritius (f. torquescens).

III. Mastophoreae Fosl.
Mastophora (Decsn.) Harv.

Subgen. E u m a s t o phora Fosl.

1. Mastophora macrocarpa Mont. PI. XIII.

Voy. au Pol. sud. p. 149; Aresch. in J. Ag. Spec. Alg. II, p. 528; Kütz. Tab. Phyc. VIII, t. 100.
Syn. Mastophora macrocarpa Heydr. in Hedvvigia. Bd. 33 (1894), p. 300?

Stat. 58. Seba, Savu. Reef.

Stat. 115. Kast side of Pajunga Island, Kwandang Bay. Reef.

Stat. 125. Sawan, Siau Island. Reef.
Stat. 129. Kawio and Kamboling Islands, Karkaralong-group. Reef.

Stat. 149. Fau Island, lagune. West coast of Gebé Island. Reef.

Stat. 193. Sanana Bay, East coast of Sula Besi. Reef.

Stat. 231. Ambon. Reef.

Stat. 279. Rumah-Kuda Bay, Roma Island. Reef.

Stat. 303. Haingsisi, Samau Island. Reef.
Birakeke, Celebes ').
Den Bril near Macassar. leg. Snac.KEY.

The Kützing figure of this species in Tab. Phyc. 1. c. is a fairly exact representation

ot the plant occurring under that name in the Kützing herbarium, now in the possession of

Madame Weber-van Bosse. The specimen is authentic, having been distributed by Montagne

himself. In addition to the description of the plant is to be stated that the conceptacles of

sporangia in the said specimen are conical and about 750 p.. in diameter, when seen from above.

In the Siboga collection occur a number of specimens which fully agree with the said

ecimen. The plant partly covers and rather closely clings to different other algae (pi. XIII,

-5i 7. 9), partly at length the greater part of it gets free (pi. XIII, fig. 6, 8, 10 — 12),

or it may be loosely creeping over the bottom, to judge from the sand or other extraneous

1) Cp. the footnote 2 under Mclobesia farino

7*

objects with which man)- specimens are burdened in the lower part. It is attached to the sub-
stratum with a larger or smaller number of rhizoids, which also is the case with the lowest
part of e. g. M. Lamoui'oiixii. It often comes near to certain forms of this species in habit.
The thallus, however, is thicker, more uniform in breaclth and less broadening upwards, and the
conceptacles are much smaller. Those of sporangia often occur in rather great number, varying
between 700 and 800 a. in diameter. The sporangia are four-parted, 250 (200) — 350 fx. Wng
and 150 (120) — 200 u.. broad. In a couple of specimens occur a few conceptacles of about the
same diameter as the last mentioned ones.
They are however very low and provided
with a rather elongated tip. I am not sure
whether they represent young conceptacles
of sporangia or perhaps are cystocarpic
ones. Similar, but larger, conceptacles are
met with in very small number in M. Lamou-
rotixii too. The plant bears sporangia,
partly however scantily, in the months of
February, April, and July — November.

A peculiar form of the species from
stat. 193 is represented fig. 27. It forms a
globular-ovoid massdensely heaped together
and somewhat curled in the upwards turning
part. It is furnished with a few conceptacles
of sporangia agreeing with those in the
typical form. This specimen corresponds with certain forms of Corallina ofpcinalis which some-
times occurs in almost similar masses, when growing in much exposed localities and not sheltered
by other algae. The form in question seems to have risen in the same way. Madame A. Weber-
van Bosse kindly informs me that many algae at that station, e. g. Gracilaria, assumed a
similar shape, she supposes owing to the heavy swell that broke on the reef.

Occurrence: The species seems to be commonly dispersed over the greatest part of
the Malay Archipelago.

Area: North Pacific: The Marianne Islands, Formosa l).

Fig. 27. Mastopkora macrocarpa Mont., forma.
Specimen forming a globular-ovoid ma* ; stat. 193; nat. size.

2. Mastophora a films Fosl. mscr. Fig. 28 — 29.

Thallus coriaceous-membranceous, crustily covering Polyopcs constrictus. Conceptacles of
sporangia conical, 1 — 1,2 mm. in diameter.

Sikka, Island of Flores.

This plant crustily covers or encompasses the branches of the named host, partly but

I) With some doubt I class with this species the plant from Formosa mentioned by Heydrich 1. c. He states that "die
Conceptakel sind ausserordentlich gross über die ganze Oberllache verbreitet".

loosely covering, partly rather closely clinging to the latter. It is attached with numerous rhizoids.
New crusts are now and then formed upon the primary one, or covering foreign bodies which

have attached themselves to the plant and
tli en partly also extended over the primary
crust, sometimes at length rather irregularly
developed and a little curled (Fig. 28).
The crust here and there gfrows free, but
it does not form divided branches. It is
very brittle in a dried stage. A vertical
section of the plant exhibits cells which are
rectangular, 50 — 100 fx. long and 15 — 35 y .
broad (Fig. 29). The conceptacles of spo-
rangia are conical, 1 — 1,2 mm. in diameter,
when seen from above. A few sporangia
examined were young, indistinctly four-
parted and 160 — 200 y.. in length by 80 —
100 a. in breadth. The plant is furnished
with these organs in the month of December.
Although there are some species of
this genus which are unsumciently known
and hardly recognizable from the descrip-
tions, I venture to place the plant in
question as an independent species, having been unable to identify it with any formerly described

one. In habit it closely approaches to
young specimens of M. macrocarpa,
but, as far as hitherto seen, it does
not form branches, though it shows a
tendency to divide itself. M. macrocarpa,
on the other hand, forms branches in a
stage as young as the specimens known
(if M. a f/i 'nis. In a section the cells of the latter species are larger than those of the former
one. However, the most certain distinguishing feature lies in the conceptacles of sporangia. I
have measured numerous conceptacles in M. macrocarpa from different places, but I have not
met with larger ones than about 800 u.. in diameter. In the species in question, on the other
hand, they are of the size stated above and about as large as in M. Lamourouxii. Therefore,
the plant seems to be well defïned from M. macrocarpa and cannot, at present at least, be
considered as a form of this species. In several species, particularly of other genera, the
conceptacles are certainly much varying in size which, however, does not seem to be the case
in the here mentioned ones. M. Lamourouxii in a young stage apparently never extends
ily as M. af/iuis, and the thallus is thinner. The conceptacles of sporangia in
M. Lamourouxii vary between 1 , 1 and 1 , 3, seldom up to 1,4 mm. in diameter, and they

Fig. 28. Mastüfhora affinis Fosl.

Cruslily expanding over the branches of Polyopes constrictus\

nat. size.

Fig. 29. Mastophora affinis Fosl.
Vertical section of the crust; X 72.

73

are perceptible in the lower part of the thallus too, i. e. the bottom of the said organs is here
very slightly convex and marked by a faintly impressed annular border. This is never seen
in M. af/niis, nor in M. macrocarpa. Besides, older specimens of M. Lamourouxii are quite
different. The species in question also comes near to M. pacifica, which however is more
delicate, with smaller conceptacles ]).

Occurrence: Only the three specimens figured are known from the station quoted aboVe.

Subgen. Li t hop o rel la Fosl.

3. Mastophora melobesioides Fosl. Fig. 30 — 32.

Det Kgl. Norske Vidensk. Selsk. Skrifter 1902, Aarsberetning (Trondhjem 1903).

Stat. 66. Bank between islands of Bahuluwang and Tambolungan, south of Saleyer. 8 to 10 m.

Dead coral; Halimeda ; Lithothamnion. — The species in question overgrown by

Lithothamnion fruticulosum.
Stat. 81. Pulu Sebangkatan, Borneo-bank. 34 m. Coral bottom and Lithothamnion. — Partly

uncovered, partly overgrown by Lithothamnion prolifer and Archaeolithothamnion

erythraeum.
Stat. 93. Pulu Sanguisiapo, Tawi-Tawi Islands, Sulu Archipelago. 12 m. Lithothamnion

bottom. Sand and coral. — Overgrown by Archaeolithothamnion erythraeum.
Stat. 106. Kapul Island, Sulu Archipelago. 13 m. Coral bottom. — On Lithothamnion fruticulosum.
Stat. 109. Pulu Tongkil, Sulu Archipelago. 13 m. Lithothamnion bottom. — Partly uncovered,

partly overgrown by Lithothamnion fruticulosum.
Stat. 123. North Bay, Biaru Island. 36 — 2701. Stone and Lithothamnion bottom. — Overgrown

by Lithothamnion fruticulosum.
Stat. 133. Lirung, Salibabu Island. Shallow water. — The species in question uncovered.
Stat. 240. Banda-anchorage. Lithothamnion bank in 18 — 36 m. — Overgrown by Lithotham-
nion fruticulosum.
Stat. 282. Between Nusa Besi and the N.E.-point of Timor. 27 — 54 m. Sand, coral and

Lithothamnion. — Overgrown by Lithothamnion fruticulosum.
Stat. 303. Haingsisi, Samau Island. Reef. — Uncovered.
Stat. 315. East of Sailus Besar, Paternoster Islands. Up to 36 m. Coral and Lithothamnion. —

Overgrown by Archaeolithothamnion timorense.
Stat. 322. i'/j mile south of Tandjong Lajar, South-coast of Bawean Island. 32 m. Coral. —

The species in question overgrown by Lithothamnion si mn lans.

I classed this species 1. c. as a subgenus, Lithoporclla, distinguished from Eumastophora
by its irregularly crustlike, undivided and not flexible thallus which is frequently stratified, i. e.
one layer of cells with marginal growth on the other.

The plant forms scaly patches or smaller or larger crusts of indefinite shape which
frequently get confluent and at length form partly small, partly even extended incrustations on
other calcareous algae, Bryozoa, corals and other hard objects, although it has not vet been

1) In Melob. Arb. Heydr. p. 19 I remarked in regard to Melobesia pacifica Ileydr. : "Es ist klar, dass diese Alge keine Mdobesia
ist, dagegen wahrscheinlich ein neues Genus der Squamariaceen reprasentiert; aber dieselbe als solches aufzustellen scheint zur Zeit doch
nicht berechtigt, da ihre Reproduktionsorgane zu wenig bekannt sind". However, in Vidensk. Selsk. Skrifter 1902, Aarsberetning
(Trondhjem 1903) I have shown that the plant is a Mastophora and placed it as JA pacifica (Heydr.) Fosl. — Besides, I remarked that
"altere Teile vom Thallus aus mehreren Zellreihen bestehen". This is however so to be undeistood that new formations sometimes are
developed, one row upon the other, although the cells in certain parts of the thallus can be divided even in vertical direction.
SIBOGA-EXPED1TIE I.XI. 10

74

met wit h en stones (Fig. 30). The ernst clings closely and in general rather firmly to the
substratum, here and there however but loosely covering die Iatter and then developing few

and short rhizoids. Smal]
crusts are fonned partly by
a solitary layer of cells.
partly, and most frequently,
by several ones, which is
the case particularly when
the plant forms extended
crusts. Such crusts attain
a thickness of up to about
2 mm., composed of numer-
ous smaller or larger layers
of cells, one upon the other,
partly anastomosed with the
subjacent, partly not and,
therefore, showing a more
or less scaly or irregular
surface. The plant often
alternates with other cal-
careous algae, or animals,
or it is rather mingled with
divers foreign bodies which
have been overgrown here
and there in the layers
(Fig. 31). Therefore, the shape of the crusts depends on that ot the substratum, and the
layers are in most cases very irregularly fonned on each other.

In a vertical section ot the ernst the cells are shown to be somewhat varying in size,
partly large, partly rather small even in one and the same section. Thus fig. 31 A. exhibits
mostly large cells, about the largest ones which ocenr in this species, but in other parts of the
same section occur cells which are in part even a little smaller than those represented in
fig. 3 1 B. They are always vertically elongated, seldom nearly squarish, with thick walls and
usually a little rounded corners, varying between 30 and 60, or up to about 75 \x. in length
and 15(12) — 25, or up to 30 u.. in breadth.

1 he conceptacles of sporangia, which are the only reproductive organs known, occur
scattered. They are superficial, conical and generally 0,8 — 1 mm. in diameter, when seen from
the surface. The sporangia are four-parted, 1 So — 250 u.. long and 100 — 150 u,. broad. The
conceptacles frequently at length become dissolved, leaving a more or less distinct scar which,
however, soon gets effaced. But sometimes new crusts partially extend themselves over the
conceptacles even before the Iatter are fully developed, or occasionally quite cover such ones
which are emptied and partly undissolved, partly being in a state of dissolution. Therefore,

Fig. 30. Mastophora melobesioides Fosl.

A. The type specimen from the Maldives; />', C, D. Specimens almost fully

covering dead corals and other calcareous algae: Stat. resp. 303, 133, Si: nat. size.

75

in sections can now and then be found conceptacles or parts of such ones which are over-
grown. The plant bears sporangia in
the months of February, June and July.

This species is apt to be over-
grown by other calcareous algae and,
therefore, it is more frequently found
covered than uncovered. It is, however,
easily recognizable even in a sterile
stage. In habit it partly recalls certain
forms of Meloèesia, partly crustlike
Lithothamnia, sometimes rather appro-
aching to e. g. young specimens of L.
simulans, though this species is forming
more even crusts. Besicles, fertile spe-
cimens recall crustlike forms of Gonio-
litkon, to which genus the plant is
nearly related. Otherwise it comes
nearest to Mastophora pacifica, but the
latter species shows mostly longer cells
and also differs in other respects.

Occurrence: The species is
commonly distributed over the greater
part of the investigated area, but is
on the whole scarce. It seems to occur
not only in shallow water, as is probably
the case with most of the other species
of this genus, but it descends to a
rather great depth. In most cases,

however, it has only been found overgrown particularly by other calcareous algae. Cp. the list
of stations with the particulars aclded.

Area: Indic: The Maldives and Laccadives M, the Red Sea 1).

Fig. 31. Mastophora mclobesioides Fosl.

A. Part of a vertical section of the type specimen;

B. Similar section of the specimen fig. 30 C; X 72.

I have above under Lithothamnion fragilissimum mentioned some calcareous algae
occurring in fossil sediments from Xew Guinea, among them also Mastophora mclobesioides.

It has generally been assumed that recent Lithothamnia also are represented in fossil
sediments, i. e. the genera Lithothamnion and Lithophyllum at least. But a comparison between
fossil and recent species has seldom been made, particularly because ol the unsufficient know-
ledo-e of these algae. An exception forms however Lithophxllmn racemus, a species which

1) At the Laccadives it has been found overgrown by Goniolithon laccadivicum, and at El Tor in the Red Sea overgrown by
Goniolithon Fosliei.

76
rather long ago has been considered to occur both fossil and recent. Thus Solms-Lai bai b

6

remarks in Corall. Monogr. p. [8: "Dass sich Lithothamnion Racemus und ebenso das nachher
zu besprechende /.. ramulosum1) seit der Tertiarzeit in unveranderter Weise fortgepfianzt hat,
ist mir unzweifelhaft. In den pliocanen Kalken von Syrakus (Latomia dei Cappucini) und von
Girerend (rupe Atenea) linden sich Banke, die ausschliesslich aus Nulliporen gebildet sind. Ich
habe an ( >rt und Stelle zahlreiche Individuen gesammelt, die sofort als zu den beiden erwahnten
Arten orehörig angesprochen wurden und die auch, zu Hause genauer untersucht, auf Dünn-
schliflfen vollkommen dvn gleichen Ban wie die recenten, überall an Siciliens Kusten vom Meere
ausgeworfenen Pflanzen erkennen Hessen". Also Rothpletz in Foss. Kalkalg. p. 320 mentions
this species living as well as fossil. As regards the connection mentioned, Gümbel 2) aptly
Lrks: "Sie reichen hier (in den Pliocangebilden des Monte Mario), inclem sie in den diluvialen
Muschelbanken unzweifelhaft sich fortsetzen, den Gebilden der Jetztzeit die Hand, welche
von den Wellen Iosgelöst und an den Strand gespült sich mit Muschelschalen und sonstigem
Meeresauswurfe stellenweis zu festem Gestein vereinigen". The species described by him
I. c. p. 39 under the name of Lithothamnion pliocaenum from Monte Mario most probably
also represents a fossil form of Lithophyllum racemus3), bi addition, the arctic Lithothamnion
glaciale is known from several post-glacial sediments both in Norway and Greenland. There
are also other species recent and fossil which, when compared more closely, will not unlikely
prove to be identic, although it is often impossible to make sure of the relation. Thus in
several recent species the reproductive organs are more or less superficial and get dissolved
after maturity, not leaving conceptacles which become overgrown by new-formed tissue and,
therefore, it can be impossible to decide even to which genus the specimens belong when they
are sterile. This is particularly the case in several species within e. g. the genus Lithothamnion,
whereas in e. g. the genus Lithophyllum the conceptacles mostly become overgrown, and in the
genus Archaeolithothamnion the cavities after the sporangia as well. A similar fact makes its
appearance in fossil species too. Therefore, such ones in many cases cannot be determined
with any degree of certainty, when they are destitute of overgrown conceptacles. Besides, of fossil
species frequently but fragments are in hand, and the structure often affords but an uncertain
character. ( )n the other hand, specimens with well developed overgrown conceptacles seem
mostly to be recognizable even in fossil sediments, but by no means always.

The mentioned sections, which Professor K. Martin kindly sent me for examination at
the request of Madame A. Weber-van Bosse, are from the island Soëk in Geelvink Bay on
the northern coast of New Guinea. Those "121. v. Rosenberg (6) Soëk" contain, among other
calcareous algae, several smaller or larger fragments of a plant which in my opinion without
any doubt must be ranged with Mastophora melobesioides . Professor A. Wichmann of Utrecht also
kindly sent me a number of sections for examination. Among these one is from Mineralogisch-

1 This speci includes more tlian one ipecii - as suggested by Solms-Laubach himself, at least Lithothamnion

and Goniolithon Brassii
C. VV. G Dii rgenannten Nulliporen und ihre Betheiligung an der Zusamm der Kalkgesteine. 1. —

Vïathem.-phys. Cl. vol. XI, 1S74, p. 29.

the cells of Mi pecies described by Gümbel 1. c. cp. Rothpli i/. 1. c. p. 303.

77

Geologisches Institut in Utrecht, - - "G. 13. 1897. N" 3542. Kalkstein-Hügel hinter Skroë, Neu-
Guinea"; another "Neu-Guinea Exped. 1903. N° 734. Lithothamnion Kalkstein. Kap Haharu
bei der Tami-Mündung, Neu-Guinea"; and some ones "166.
Macklot. Pulu Samauw". (Cp. p. 16). These sections too show
some fragments of the same species, which apparently is rather
commonly dispersed in tossil sediments in the said regions.
However, in the sections from Pulu Samauw the species in
question is very scantily represented in company with numerous
fragments particularly of a species, which probably is identic
with Lithothamnion eruèescens, although the latter not ascer-
tainable with certainty. Fig. 32 represents a part of the plant
in one of the sections from Soëk. It coincides in every respect
with similar sections of recent specimens. Cp. fig. 31. Among
the rather numerous fragments examined I have not seen any
one differing. Besides, in one of them occurs even a part of a conceptacle, which also in the
main agrees with those of sporangia in the species in question. Thus another species is added
to those already known living as well as fossil.

Fig. 32. Mastophora vu Us Fosl.

Part of a vertical section <>{ f.jssil

sediment from Skroë; X 72.

II.

CORALLINEAE VERAE OF THE MALAY ARCHIPELAGO

BY

A. WEB ER— van bosse.

W'ith plate XIV— XVI.

Introduction.

In Areschoug's Corallineae in J. G. Agardh's Species, Genera et Ordines Algarum,
the Corallineae verae are divided into five different genera: Amphiroa, Arthrocardia^ Cheilo-
sporum, J'ania and Corallina. The distinction between these genera is based upon the position
which the conceptacles, keramidia Aresch., occupy on the plant. The validity of these five
genera has often been disputed; "Jania and Corallina have so many characters in common that
Solms-Laubach united j'ania to Corallina in his excellent monograph "Die in Neapel's Umgebung
bis jetzt beobachteten Corallineenformen", and Schmitz and Hauptfleisch in Engler und Praxtl's
"Pflanzenfamilien" I Theil, have sunk Arthrocardia into Cheilosporum. The characters hitherto
employed to distinguish the genera are indeed very confusing, for they are very variable, and
again the characters distinguishing the species are not better defined than those of the genera.
It was therefore with much doubt as to the success of my work that I undertook to name the
Corallineae of the Siboga Expedition.

This collection contained a great number of specimens belonging to different species of
Amphiroa ■. Arthrocardia failed to be represented; of Cheilosporum there were two species, of
j'ania two and of Corallina one. Amphiroa as forming the greatest bulk of the collection was
studied first; and this study afforded me material for modifying the whole group of the Coral-
lin, ' . The following pages however treat principally of the genus Amphiroa.

\\ hile naming the Amphiroae of the Siboga Expedition I was struck by the fact that
two or three species, of which 1 had much material at my disposal, showed such differences in
their ontward appearance, that, seen apart, the different branches of one plant would have to
be ranged under different species if I were to use the ontward form of the joints or the position

79

occupied by the conceptacles as a guide, whereby to name my specimens. On the other hand
I found that specimens which looked much alike, belonged to quite different species.

I tried to find an anatomical character, that would prove a surer guide for the deter-
mination of the species of this genus and that would be less liable to variation than the form
of the separate joints, the manner of ramification, the diameter of the conceptacles or the number
and manner in which they appear on the joints; characters which certainly have a great value
and have been used till now, but which are not implicitly to be relied upon in all species.

Lamouroux *), Areschoug '•) and Yendo 3) when describing new species have made some
very interesting remarks about the node between two joints, but nowhere did I find a systematic
and careful study of the nodes. Solms-Laubach tells us only that the node of Corallina consists
of one layer of long pericline cells, that of Amphiroa of two layers of pericline cells, while
the layer of cortical cells falls off as a rule and is only to be found abnormally in some feeble
specimens of A. verrucosa.

While pondering over this question I remembered the results obtained by Miss E. S.
Barton *) while investigating the nodes of Halimeda, and it occurred to me that perhaps the
nodes of Amphiroa might also reveal an analomical character, that would be a useful help in
classifying the different species.

In order to understand the structure of the node of Amphiroa it is necessary to recall
in a few words the structure of these plants, giving first a little historical survey of all we
know about the genus Amphiroa.

Historical.

The genus Amphiroa was created by Lamouroux") in 1812 with the following diagnosis:
"Plants with filamentous, articulated tufts, branching; branches dichotomous or verticillate, joints
entirely separated by a horny substance, naked". After this short diagnosis Lamouroux cites
Corallina tribulus Eli. and C. cuspidata Eli. as examples of the new genus adding the following
words: "and many non-edited species". I wish to call attention to the fact that Lamouroux in
his first note on the genus Ampliiroa mentioned Ampliiroa (Corallina) tribulus Eli. and
Amphiroa (Corallina) cuspidata Eli.; they are therefore the types of the genus. In His "Histoire
des polypiers coralligènes flexibles 181 6" he places Amphiroa between Cymopolia and Halimeda.
Corallina is described before Cymopolia. This shows the value Lamouroux attributed to the
articulated appearance of all the above named genera, two of which belong to the Florideae
and two to the Chlorophyccae.

Exdlicher 6) 1836 — 47 divided the genus in different sections which were taken up and
worked out by Decaisxe.

1) Lamouroux. Histoire des Polypiers coralligènes flexibles. Caen 1816.

2) J. E. Areschoug in Agardh's Genera et Species Floridearum 1851, p. 506.

3) Yendo. Corallinae verae Japonicae. Journ. of the Coll. of Science, linp. Univ. Tokyo. vol. XVI, prt. 2, 1902. — Corallinae
i-erae of Port Renfrew. Minnesota Botanical Studies 1902.

4) E. S. BARTON. The Genus Halimeda. Monogr. I.X. of "Uitkomsten op zool., bot., ocean. en geol. gebied der Siboga-Expeditie".

5) Lamouroux. Essai sur les gem-es de la fam. des Tkalassiophytes non articulées. Mem. du Muséum. Paris 1S12.

6) Exdlicher. Genera plantarum cum supplementis. Vindobouae 1836 — 1847.

8o

Decaisni l), 1842, based his division of the family oi the Corallinaceae on the form and
position of the conceptacles and distinguished the genera: Corallina, Jania, Amphiroa and
Melobesia. Under the name oi Amphiroa he united all the species which bear on their joints
wartlike or conical conceptacles, and he divided the genus into tour sections according to
the joints: (1) cylindrical with wartlike conceptacula, Euamphiroa\ (2) flatténed, obcordate,
subalate, with conical conceptacula, Arthrocardia\ (3) flat and two-edged with conical concep-
tacula. Eurytio \ obcordate with acute lobes bearing conceptacles on their superior margin,
Cheilosporum. But he adds that this ditïerence alone is nol sufficiënt as a generic character, for
one observes transitions between the cylindrical branches of the typical Amphiroa and the
flatténed branches of the Arthrocardia section, just as one sees the conceptacles developing
gradually on the side of the joints and thus forming the section named Cheilosporum.

Two years later Zanardini i [844 treated of the Corallines and he was the first to
call attention to the anatomical differences between Amphiroa and Corallina. He pointed out
that in the genera Jania and Corallina the cells - - which he still called articoli and otricoli —
have all the same dimensions, but that this is not the case with Amphiroa, where short cells
are intercalated at regular intervals between long ones. This disposition of the cells seems,
according to Zanardini, to belong exclusively to the genus Amphiroa and by this internal
structure it is distinguished at once trom either Jania or Corallina . By these alternating lavers
of long and short cells the transverse zones of Amphiroa are indeed much more marked
than in all other genera; but this remark of Zanardini seems to have been overlooked by
succeeding authors.

Harvey3) (1847 — I,s49.) and Kützing4) (1849) foliowed Decaisne and characterized the
genera by the form of the fruit. Areschoug ( 1 85 1 ) used both the form of the joint and that
of the fruit but as Grav') (1867) observed : "Both these characters are good, for they afford
the means of dividing the species into very natural groups but in most of these groups there
are always species, which combine the characters in such a manner as to make it difficult to
determine to which genus they are most nearly allied". Grav created in 1867 the genus
Lithotkrix which was sunk into Amphiroa by Anderson 6) in 1891.

Solms-Laubach in his above-quoted monograph of the Corallines of the Gulf of Naples
(1881) treated of the genera Amphiroa. Melobesia, Lithothamnion, Lithophyllum and Corallina.
He pointed out the development of the fruit of Corallina, which had already been the object
of such arduous research by Thuret and Borxet') in 1878, and he also called attention to the
many points ot resemblance between Amphiroa and Lithothamnion, which will be spoken of,
when the anatomy of Amphiroa is discussed.

In "Die Natürlichen Pflanzenfamilien" of Engler (1897), the Corallinaceae are treated

1) Iim A1SNE. Essai sur une Classification des Algues et des polypiers calcifères. Ann. des Sc. Nat. 2"IC ser.. t. XVII, Paris 1S42.
1 Revista sulle Corallinee. Atti R. Istituto Veneto vol. III. Serie I. Venezia 1S44.
VEY. Nereis australis or algae of the Southern Ocean. London 1S47 — 1849. Phycologia britannica. London 1849 — 1S51.

ina. 1 Ion 1 ^46 — 1851.

4) Kützing. Phycologica generalis. Leipzig 1S43. Tabulae phycologicae 1849 — 1869. Species algarum 1S49.
1 otkrïx a new genus of Corallirjae. Journ. of lïotany. V, 33. Febr. 1867.

Zoe vol. II, [891, p. 225.
7) Tim 1 et. Études phycologiques. Paris 1S7S.

8i

by F. Schmitz and P. Hauptfleisch. The authors recognize only the genera Amphiroa,
Cheüosporum and Corallina out of the old Corallineae verae of Areschoug. When speaking
of Amphiroa they call attention to the joints as being "quergezont mit dickem Bundel dicht
zusammengeschlossener, in bestimmt alternierender W'eise gegliederten Markfasern". This seems
to point to the long and short cells described by Zanardini.

Setchell and Gardner l) in their valuable paper on the Algae of Northwestjèrn
America (1903) say: "After a long study of the forms of Amphiroa with flattened joints from
the western coast of Xorth-America, we have come to the conclusion that they are all forms
of one polymorphous species, the forms owing their very considerable differences to variations
in the environmental conditions under which they may be found growing .... We do not
understand the reasons for referring some of the forms to the genus Chcilosporum, as Yendo
has done, since that genus if separable at all from Amphiroa, applies only to a very limited
number of species, none of which occur in our territory".

In my opinion Setchell and Gardner are perfectly right in considering the algae
known as Amphiroa tuberculosa, A. Orbigniana, A. Caiiforuica as forms of only one
polymorphous species, but I think that this species belongs to the genus Arthrocardia and not
to Amphiroa. The reasons why I think so will be shown in the following pages.

Anatomical.

The fronds of Amphiroa consist of a tuft of vertical, sometimes horizontally spread,
calcified branches which, according to Solms-Laubach, probably arise from a horizontal disc,
or at all events from a small 'Keimscheibe', resulting from the first divisions of the eerminatine
spore. The branches are as a rule articulated and each joint (articulus) is separated from the
next one by a narrower or broader articulation or node (geniculum of Yendo) of non calcified cells.

The branches, whether ascending vertically or spreading horizontally, have all a central
strand of elongated cells and a cortical layer of isodiametrical or ellipsoidal cells. The cells
of the central strand have their origin in a layer of meristematic cells at the flat top of the
branches, where, in conformity to the mode of growth described by Kny 2) for Jania rnbens,
Corallina officinalis and C. granifera, they divide as a rule only with transverse walls except
at the periphery, where the cells are divided alternately by transverse and by oblique longitu-
dinal walls, these latter ones running from the inner part of the lower transverse wall to the
apicai part of the outer longitudinal wall. The cells resulting from this latter division are
pushed sideways towards the periphery where, by subsequent division, they become smaller
and smaller, are divided by transverse and longitudinal walls and form the cortex. This mode
of growth proper to Jania and Corallina, is also that of Amphiroa as Solms-Laubach has
pointed out, but he showed at the same time that a continuous layer of "Deckzellen" covers
the meristematic cells of Amphiroa even at the moment of the most intense growth. This
layer of Deckzellen, which is also to be found in Lithothamnion is only now and again renewed

1) Setchell and Gardner. Algae of northwestern America. Berkeley 1903.

2) Kny. Bot. Zeitung 1872, p. 705.

SIBOGA-EXPEDITIE LXI.

I

82

frorrt below and the okl layer thrown off. Solms-Laubach speaks also in bis paper of the
growth in thickness of the cortical layer after the manner of Lithothamnion, whereby new
conceptacles can spring forth and old ones becoming overgrown, are submersed in the tissue
of the plant. Solms-Lai :rs rather briefly to the nodes, only saying that the node

consists of two layers of cells and that the cortical layer falls off as a rule.

The anatomical structure, first noted by Zanardini and afterwards mentioned by Schmitz
and H ii proved to be a character of great systematic value and I wil! therefore

describe it at f uil length.

In Arthrocardia, Cheilosporum, j'ania and Corallina the cells in the central strand of
each joint have the same dimension throughout the whole joint; each successive layer is as a
rule as long and as broad as the preceding or succeeding one. The cells stand in vertical rows
one above another; they may grow a little longer or shorter, the difference is a gradual one,
never an abrupt one.

In Amphiroa the case is different; i, 2, 3, 4 or even more layers of long cells (articoli
Zan.) will alternate with a row of short cells (otricoli Zan.) marking a period of intra-nodal
growth; each joint may have several of these periocls of growth. The long cells in the successive
rows of one period of growth are not of exactly the same size ; as a rule the cells of the first
row are the longest, the cells of the succeeding ones becoming gradually a little shorter, but
the transition to the last row, the one with the short cells, is as a rule abrupt. The regularity
with which these rows of long and short cells alternate is liable to much variation in each
species and in each plant. Still a certain rule may be detected. Thus in the case of Amphiroa
fragilissima (L.) Lam., one of the many types of the genus, 4, 5 or 6 rows of long cells will
be foliowed by a row of short ones. This number of long-celled rows may be reduced to 2
or 3, or increase to 7 or 8, but by far the greater number of joints will have intra-nodal
periods of growth consisting of 4, 5 or 6 rows of long cells. In Amphiroa verrucosa Kütz.
we find very regularly one row of long cells alternating with one row of short cells and in
Amphiroa rigida Lam. two long ones alternating with a short one, but in the latter case the
alternation is not so regular as in A. verrucosa.

Studying the nodes of other Amphiroae I found that the nodes of these plants do not
always consist of two rows of pericline cells, as is the case with the Amphiroae of the Gulf
of Naples desenbed by Solms-Laubach but that a great number of rows may take part in
their formation (PI. XV, fig. 5); the cortical layer of the node does not drop off but persists
in many species and even grows in thickness, and in one or two species the cortical layer
remains calcified and the node is only indicated by a zone of uncalcified rows in the central
strand (PI. XV, hg. 6). Almost startling is the wealth nature exhibits in producing such a
simple thing as a node of Amphiroa. But however different the ways of forming the node
may be, one thing is invariable and can be relied upon as a fixed rule: the cells in the
node have always the same dimension o r n e a r 1 y s o as the cells in t he
whole plant. The only exceptions which I have discovered to this rule are A. rigida and A.
verrucosa, the long cells of which have an inclination to grow longer still in the node; but
even here the differences are not great.

83

When one studies a young growing branch of Amphiroa in a decalcified state and in
longitudinal sections, the nodes are indicated by the different colour of the membrane. That is
all. The whole young top presents the same disposition of cells throughout its whole length
and the rows of long and short cells succeed one another quite regularly.

On the other hand if one studies longitudinal sections of either Arthrocardia, CJieilo-
sporum, Jania^ Coralliua, Lithothrix or of our new genera Metagoniolithon and Litharihron,
the disposition of the cells at the top of the young branch is quite different in the region of the
future node, corresponding with the structure of the adult node in each of these latter genera.

It appeared to me that this difference in anatomical structure was indeed a good character
whereby to distinguish the genus Amphiroa. It seemed to me to be of great value since the
study of the development of the cystocarp offers many difficulties, arising from the smallness
of the object; and even if one succeedecl in tracing its development, it is highly probable that
it would follow the same mode as Solms-Laubach has described for the Corallinae and thus
not afford a good character whereby to distinguish Amphiroa either from Arthrocardia or
Chcilosporum. How difficult such research would be is shown clearly by the very few conceptacula
with procarpia I met with during my whole investigation. In my preparations I saw mostly
conceptacula with ripe or unripe tetraspores; next came antheridia and conceptacula full of ripe
carpospores. Rarely have I seen a conceptaculum with procarpia or cystocarp-spores in course
of development.

Since however the anatomical structure of the frond first recognized by Zanardini and
of the node proved indeed a good generic character whereby to distinguish the genus Amphiroa,
I have had to strike out of that genus a great many species that had been described - - often
not by their authors, but by succeeding ones — as belonging to this genus. The algae which
Lamoüroux first mentioned as belonging to Amphiroa, have the same structure as mentioned
by Zanardini. They constitute the types of the genus; and other species, differing from these
types even though they were also called Amphiroa by Lamoüroux in his later work of 1816,
must be taken out of the genus.

The genus Amphiroa when cleared of all species that are only like it in outward
appearance is a very well limited group. The presence of nodes distinguishes it from Litho-
thamnioneae \ the continuity of the frond - - the same anatomical structure of the central strand
throughout the whole frond — , distinguishes it from Arthrocardia, Chcilosporum or Corallina,
where long cells are intercalated between two succeeding joints. Solms-Laubach has already
called attention to the fact, that Amphiroa has very many afhnities with Lithothamnioii; and
the results I obtained during my investigations tend only to show the near relationship of
Amphiroa to LithotJiamnion. The anatomical structure of Amphiroa, as it shows itself in distinct
periods of growth, is a character which we find also in LitJiotliamnion, and it is interesting to
observe how the node is gradually developed in species of Amphiroa with broad, often irregularly
branched fronds as A. crassa or A. tribulus, the fronds of which resemble outwardly small
forms of LithotJiamnion. We see in these two species, as will be shown more fully in the following
pages, that the cells of the central strand alone contain no chalk and form a node often entirely
surrounded by a layer of calcified cells. A further stage in the development of the node is

84

marked by Amphiroa ephedraea where the node consists of the whole central strand and the
whole cortical layer, and intervening forms between these two extremes are numerous.

To the treillis Amphiroa belong those plants which have fronds built up of a central
strand of filaments having alternating rows of long and short cells communicating with one
another by primary and secondary small pores. How these rows of cells alternate is a question
of less importance and this will be considered when treating of the different species. The
have conceptacula as wartlike protuberances on cylindrical joints {Euamphiroa Dec.
pro parte) or as conical processes on flat two-edged joints [Eurytion Dec.) or half immersed
in the tissue of the frond: A. crassa, A. triöulus. Their nocles consist always of the central
strand ; the cortical layer may split off and leave the node naked, or persist and take part in
the formation of the node. The rows of cells of the central strand have with rare exceptions
ame size in the node as in the joint. All Amphiroae have a layer of "Deckzellen" covering
their growing tops.

I propose the name of Metagoniolithon for those algae which have joints consisting of
rows of cells all of the same dimension and communicating in their longitudinal walls by relatively
big thin places in these walls. The whole central strand and cortical layer take part in the
formation of the node, the cells of which are much smaller than in the joint. The ramification
takes place at the node and from the non-calcified tissue of the node spring forth successively
new branches, which are therefore no true whorls, though they surround the node like a whorl.
On sectioning the young top of Metagoniolithon graniferwm [Amphiroa granifcra Harv.) I saw
that it was covered by a thick mucilaginous layer in which slight striae were visible, reminding
me of the Membranprismen, described by Solms-Laubach for Corallina mediterranea. I saw no
layer of "Deckzellen".

Their conceptacles have the form of conical processes. These are flrst seen on the inner
side of secondary branches, but in old fronds they are scattered along the whole branch.

To Metagoniolithon belong the following three species, known as species of Amphiroa :

Metagoniolithon charoides Lamx.
Metagoniolithon granifernm Harv.
Metagoniolithon stelligerum Lamk.
Lithothrix aspcrgillns Gray {Amphiroa aspergillus Anders.) is such an aberrant alga,
that it could not be kept amongst the Amphiroae, by anyone who decalcifies and sections its
joints. It will be spoken of in the succeeding pages, as will also Amphiroa aiistralis Sond.,
which proved to be so totally different that it had to be taken out of the genus Amphiroa.
I propose for it the name of Litharthron ; the diagnosis of this genus will be found in the
following pages. Neither Metagoniolithon nor Lithothrix nor Litharthron were to be found
tiongst the collection of the Siboga; I looked at these different algae for comparison with
Siboga-material, and thus found out, that they were generically different from Amphiroa
and also from one another.

genera Arlhroeardia, Cheilosporum, Jania and Corallina are closely allied. In all

mr the central strand of filaments is built up of cells of equal dimension throughout the whole

joint. In Arthrocardia and Cheilosporum the cells communicate, both by their primary pores,

85

and by thin places in the longitudinal walls, after the manner of the cells of M,etagoniolithon\
but the thin places are smaller than in this latter genus. In yania and Corallina it appeared
to me that the cells adhered only at their primary pores and that secundary ones were scarce
or very obsolete, for the cells seemed always to be quite free from one another in transverse
direction. The nodes in all four genera consist of one row of long pericline cells, surpassing
many times (even so much as 1 1 times) the length of the cells of the central strand. I Saw
no layer of "Deckzellen" at the top of the branches.

The conceptacula grow on the cylindrical or winglike broadened joints in Arthrocardia
and at the top of the joint, immersed in its tissue in Cheilosporn/n, but I have seen many
specimens to which the words of Gray could be applied, that it was difficult to say to which
group they belonged. Many algae considered till now to belong to the genus Amphiroa must
be placed in the genus Arthrocardia :

Amphiroa tnbcrculosa (P. & R.) Setch. & Gardn.

Amphiroa cretacea (P. & R.) Aresch.

Amphiroa variabilis Harv.

Amphiroa Darwiui Harv.

Amphiroa rudis Harv.
. Amphiroa cpiphiegnoidcs Ag.

Amphiroa vertebralis (Harv.) Aresch.

Amphiroa declinata Yendo.

Amphiroa aberrans Yendo.
and perhaps others, that I have not seen. Amphiroa TJ'ardii Harv. and A. Stangeri Harv.
were already taken out of the genus Amphiroa by Areschoug.

Corallina and Jania are easily known from the preceding genera because they bear
their conceptacula immersed at the top of the branches. In Corallina the cells of the central
strand are relatively short for such sturcly plants and have the same size throughout the joint.
In Corallina officinalis, C mcditerranea and C. Cuvieri the cells of the nodes are many times
longer than the cells of the central strand in the joint.

In Jania adhaerens and J. rubens the cells of the central strand in the joint are relatively
long for such slender plants and the cells in the node are not longer, sometimes shorter, as a
rule of the same size. This is beautifully illustrated by Kützing Phycologia generalis pi. 79,
fig. I and II. Fig. I shows a node of C officinalis and part of the two neighbouring joints,
and we see that the cells of the node are much longer than the cells of the joint; in fig. II
we see a short-celled node of Jania rubens and the long cells of the joints still partially covered
by cells of the cortical layer.

In Jania longifurca the cells of the node are twice as long as those of the central
strand in the joint. Jania longifurca is a transition from the Corallinae with long-celled nodes
to such Janiae as have short-celled nodes.

This difference in length of cells of the node, being as far as I know, the only difference
between Jania and Corallina, is not a very striking character; still it may be a help in disen-
tangling the confusing group of Corallinae, and the name of Jania ma)- perhaps be retained

86

for those slender tonus, which have a central strand with cells of almost equal size in the joints
and in the nodes.

Synoptical Key

to the Genera of the Corallineac verae.

1. Fronds branched, articulated, calcified joints separated by horny pliable nodes. Joints consisting
of a central strand of filaments and of a more or less developed cortical layer. Conceptacula
either occurring as wartlike or conical processes on the joint or immersed in the tissue of
the joint.

(7. Joints cylindrical or broadened; in the central strand rows of short cells are intercalated
between rows of long cells all standing vertically one above another. Nodes consist of
two or more, — rarely of one, — rows of cells that alternate in the same way as, and
have the same size or almost the same size as, the cells in the joint. Conceptacula
on the joints Amphiroa Lamx.

b. Joints cylindrical ; in the central strand the cells have throughout the whole joint almost
the same dimension and stand vertically one above another. Nodes consist of many
rows of cells which are much smaller and have thicker walls than the cells in the joint.
Conceptacula on the joints Metagoniolithon n. g.

c. Joints flat short elliptical in outline, with a thin central strand of non-calcified, intricated
filaments, with cells not standing vertically one above another, and with a considerable
layer of calcified cortical cells. Cells of cortical layer large, near central strand full of
big grains of starch, growing smaller towards the periphery, covered by a thick cuticule.
Nodes consist of small, thick-walled cells. Conceptacula unknown. Litharthron n. g.

d. In the central strand all cells have almost the same dimension throughout the whole
joint. Nodes consist of one row of long cells.

i. Joints cylindrical or broadened winglike; conceptacles form conical protuberances
on the joint Arthrocardia Aresch.

2. Joints cylindrical or broadened winglike; conceptacles on the upper margin of the
broadened wing, immersed in the tissue of the joint . . Cheilosporum Aresch.

3. Joints cylindrical or flattened; conceptacles at the growing top of the branches,
immersed in the tissue of the joint . . . Corallina Lamx. incl. yania Lamx.

II. Fronds branched, articulated, cylindrical, calcified joints separated by calcified constrictions
of the frond. The joints consist of a central strand of long, undivided non-calcified filaments
standing in vertical rows with the calcified cells of the constrictions, and of a cortical layer
of calcified horizontally elongated cells. Conceptacula on the joints . . Lithothrix Gray.

Amphiroa Lamx.
LAMOUROl \, Mémoire du Musée t. II, Paris 1812. Polyp. flexibles 1S16.

The characters of the genera having been established I began the naming of the Amphiroqe
die Siboga collection and tried to find a character, that would help me to distinguish the

8/

often very confusing species. Whether the anatomical structure of a plant, - - the number for
instance of long-celled rows in a periocl of intra-nodal grovvth, the height of the cells, and
the dropping off or persisting of the cortical layer in the node, - - offered characters of specific
value, was a question that had to be solved. The result was at fïrst not very satisfactory,
variations in the successive number of long-celled rows in a period of intra-nodal growth occur
so often, even in the sa me plant, that one must be cautious not to jurrip to rash conclusions,
and other characters already described at length by previous authors, for instance the size of
the conceptacles, the form of the froncl etc. must be taken in consideration before naming a
species. Still if the same constant structure is found regularly in different plants, this structure
may be a good help to recognize a species, for it is highly probable that such plants will
belong to the same species even when the outward form happens to be different. The node
of Amphiroa offered a character of importance and I will therefore describe it at full length.

I began my studies on Amphiroa with specimens preserved in alcohol of A. rigida and
A. cryptarthrodia ■, these plants were authentic specimens of Solms-Laubach, kindly given to
me by Prof. Mayer at Naples. My methods of research were very simple ; the nodes, which
soon became a point of special interest to me, did not allow the use of ground sections; these
were only used now and than for comparison ; I decalcified my plants trying various liquids,
but for anatomical research Perenyi's fluid gave very good results and I used it continually.
The uncalcified material was stained, hardened, and often imbedded in celloidin. Other pieces
were sectionecl immediately as soon as they were hard enough ; the one thing I had to be
very careful of, was that my sections should go straight through the middle of the plant. If
they did not do this, I was not sure that the short cells of the branches, forming the cortical
layer, did not interfere with the succession of short and long cells such as really occur in the
middle of the plant.

A. rigida and A. cryptarthrodia both have nodes consisting, as Solms-Laubach has
already observed, of two rows of cells. The node of A. rigida has been beautifully figured
by Yendo ; it consists of two cells which have the same size, and the ends of which in the
node rest not horizontally one above the other but obliquely.

I thought this might tend to increase the power of resistance in the slender node.

The node of A. cryptarthrodia consists also of two rows of cells, but here the lower
one is much larger than the upper one. In the joint the large cells have much more chalk in
their lower part than in the middle where it seemed to me that they were less calcified than
the cells of A. rigida. This has been indicated in the hg. 14, on PI. XVI, which was drawn
at the be^innino- of my research. In A. cryptarthrodia I observed also that the cortical layer
persisted much more regularly than I had supposed it would do according to Solms-Laubach.
These two al°ae had nodes so different from one another that a look at a section taken
through the node was sufficiënt to show to which species the specimen, from which the section
was taken, belonoed. This encouraged me. to pursue my research on the node and I submittecl
the whole Siboga material to an investigation on those lines.

Amongst the collections of the Siboga were two species of which I had much material
at my disposal and naturally I began my research with these. My first impression had been

NS

that at least '>ne of these species was new to science but, owing to the kindness of Prof.
Lignier at Caen, I was allowed to study the type specimens of Lamouroux ; and this study
convinced me that my specimens belonged to species already described by Lamouroux as
A. foliacea and A. crassa. Tliis latter is an almost forgotten species described by Lamouroux
at the end of the zoological part of "Le voyage de 1'Uranie par Freycinet".

My sections of A. foliacea showed that the cortical layer participated as a rule in the
formation of the node; it grew in thickness with the thickness of the frond, it was an integral
part of the node; the joint was often broader, but otherwise the only difference between node
and joint consisted in the former not being calcified.

In A. crassa the central strand alone forms the node; its cells alone are non-calcified
and have thick membranes but the cortical layer persists; it has thin-walled calcified cells and
forms an integral part of the joint. In a small form of A. crassa the node was outwardly not
visible, and while sectioning this plant it often happened that the node feil out of the section ;
in dried plants the contact between node and cortical layer is very loose, because a layer of
cells that surrounds the node, tears very easily, the upper and under joint remaining connected
by the calcified cortical layer. In larger specimens with long joints the cortical layer splits
however horizontally leaving the node visible for a very short space.

The specimens in Lamouroux's herbarium were rather flat, broad and palmate; I had
exactly the same specimens but also long, branched, spreading specimens, described by Grunow
as A. Godeffroyi and small compact ones, and all had their nodes built up after the same type.

Ought I now to consider these specimens as different species or were they only different
forms of one species? To this question it appeared to me that my specimens of A. foliacea
cjave an answer.

Of this latter species I had specimens exactly like those in the herbarium of Lamouroux,
others again that were much narrower and finally creeping ones with much broader winged
joints. I suppose that these are identical with the Amphiroa described by Grunow from Ovalau
in his paper on the "Algen der Fidschi, Tonga und Samoa Insein". Some of the branches of
these latter fronds instead of spreading horizontally grew in a vertical direction and when growing
erect the form of the branch changed. It grew narrow, almost cylindrical with a small wing
and differed totally from the creeping joints from which it sprang and also from the type of
Lamouroux (PI. XIV, fig. i — 8); they resembled and were in fact identical with the alga
named A. nobilis by Hauck in his paper "Ueber einige von Hildexbrandt im Rothen und
Indischen Ocean g-esammelte Alg-en. V". In all these branches the nodes were the same. This
led me to conclude that since :

i 1 had been fortunate enough to collect specimens that showed in an undeniable way the
specitic identity of branches so different, that seen apart, one would think they ought to be
taken for different species, although they had the same anatomical structure of joint and node,
could show that different species showed great difference in the anatomical structure of
joint and node,
It was highly probable that species showing the same anatomical structure of joint and node

>nged to one species.

89

If this conclusion is true a great number of species have to be sunk and can only be,
according to smaller or greater deviations, retained as varieties or forms.

Working out the Amphiroae of the Siboga Expedition principally with reference to the
structure of the node, I found that the collection contained the following species :
i. A. fragilissima f. cuspidata and f. cyathifera.

2. A. anastomosans, a new species nearly related to the former.

3. A. foliacea f. typica f. nobilis.

4. A. anceps.

5. A. Boiverbankii.

6. A. ep lied ra ca .

7. A. canaliculata.

8. A. c ras sa, f. Godcffroyi and f. mini ma.

With the exception of A. crassa the cortical layer takes part in the formation of the
node of all these species. The structure of the node is in all specimens the same in each
species. In A. anceps the node shows a tendency to deviations, as did also the regularity
wherewith long and short cells alternated in the central strand of the joint. This shows that
A. anceps is a highly variable plant, vvhereas in A. ephedraea for instance the structure
of the node and the regularity wherewith long and short cells alternate is much more constant.
Since however the great variability of which the genus Amphiroa is capable, is amply shown
by A. foliacea, I resolved to unite in one species all Amphiroae the nodes of which were
essentially the same as the node of the type specimen of A. anceps.

Whether this conception of the species is right or wrong can only be made out by
careful study on the spot where these algae grow ; it may at least be a help for the present,
and the study of the node will prevent various genera from being mistaken for one another, as
was the case in some collections I have been able to study.

I will now proceed to a systematic description of each species contained in the Siboga
collection.

1. Amphiroa fragilissima (Linn.) Lamx. PI. XIV, hg. 5.

Corallina fragilissima Linné, Syst. nat. ed. 12, vol. I, p. 1305. Ellis et Solander. Zooph.

p. 123, t. 21, fig. 9.
Corallina rigens Pallas, Elench. Zooph. p. 429 fide Areschoug.
Amphiroa fragilissima Lamouroux, Pol. flex. p. 298.

Amphiroa debilis Kutzing, Spec. Alg. p. 700. Tab. Phyc. t. 8, pi. 39, 40.
Corallina cuspidata Ellis et Solander, Zooph. p. 124, t. 21, fig. 9/.
Amphiroa cuspidata Lamouroux, Pol. flex. p. 300.

Amphiroa cyathifera Lamouroux in Freycinet, Voy. de 1'Uranie, Zool. par Quoy et Gaimard p. 627.
Amphiroa fragilissima Lamouroux, Areschoug in Agardh, Spec. Gen. et Ord. Alg. t. 2, 185 1, p. 53 1.

f. fragilissima .

Amphiroa fragilissima Lamx.

Stat. 34. Labuan-Pandan.

Stat. 60 and 303. Haingsisi, Island Samau.

Stat. 71. Makassar.

SIBOGA-EXPEDIT1E LXI. 12

9o

Stat. 78. Lumu-Lumu shoal, Borneo-bank.

St.u. 93. Sulu Archipel

Stat. 172. Gisser.

St.u. 1S1. Ambon.

Stat. [93. Sula-B

Stat. 307. Endeh, [pih Bay.

Sikka, Flores. leg. A. Weber-v. Bosse 18SS.

f. cuspidata.

Amphiroa cuspidata Lamx.

60 and 303. Haingsisi, Island Samau.
St.it. 71. Makassar.
stat. Si. Islainl Sebangkatan.
Stat. 89. Kanjungan Island.
Stat. 193. Sula-Besi.
Stat. 240. Banda.
Stat. 249. Ronia Island.
Sikka and Maumeri, Flores. Padang, Sumatra. leg. A. Weber-v. Bosse 1888.

f. cyathifera.

Amphiroa cyathifera Lamx.

Stat. 7 1 . Makassar.

Stat. 93. Sulu Archipelago.

Stat. 213. Saleyer Island.

Stat. 280. Kur Island.

Fronds articulated, cylindrical, branched at the node di- or trichotomously, often with
adventitious branches. Joints long, many times longer than broad, in older specimens swollen
at top and base in the fonn of a ball or pad, by local division in longitudinal and vertical
direction of the cells of the cortical layer. Nodes swollen cushionlike in conformity with the swollen
tops and bases of joints.

Central strand built up of 4 — 8 rows of long cells, seldom 2 — 3 rows of long cells, mostly
4 rows, foliowed by one or two rows of short cells.

Height of long cells from 60, 80, 100, 104, 120 u, mostly 80 [j. ; height of short cells,
12, 16, 28 <].. These short cells mark the intra-nodal periods of growth, seen so distinctly through
the cortical layer, and visible to the naked eye as little constrictions of the frond. In the middle
of the central strand the cells are often a little compressed, towards the periphery they are a
little broader. Cells of the central strand pass rather abruptly into the cortical layer, their rows
therefore as a rule straight, horizontal, little curved.

In the node the structure of the central strand is the same as in the joint (PI. XVI,
tig. 2, the walls ol the cells are only thicker not calcitïed; cells of the cortical layer very
small, closely adhering together.

Conceptacles prominent nol immersed in the frond, slightly elliptical seen from above,
or round. Diameter of conceptaculum > 300 — 340 a.

1 he collection ol the Siboga contains a great number of slender Amphiroae with long
cylindrical joints and more or less thickened nodes. Comparing these with the authentic specimens

9i

of Lamouroux, I came to the conclusion that tny collection contained A. fragilissima, A. cus-
fiidata and A. cyathifera, but that it was impossible to tracé a sharp limit between these so-
called species. Sectioning the node of an authentic specimen of Lamouroux of A. fragilissima,
I was surprised to find that its structure was exactly identical with that of the node of A.
t vat 'hij era ; this latter one being only much more sturdy. I do not attach great importance to
this point, whether specimens are more or less slencler, if they agree on all other points, for I
have a specimen of A. cyathifera, with some fronds abnormally developed and bearing very
thin slender branches, resembling exactly those of A. fragilissima. One long specimen from
Banda looked quite like A. fragilissima at its top, like A. aispidata in the middle, and some
nodes at its base might be easily taken for nodes of A. cyathifera.

I therefore think, that it is right to give the name of A. fragilissima to all Amphiroae
with cylindrical joints, swollen at their apical and basal part and with distinct periods of intra-
nodal growth, clearly visible to the naked eye, even in dried specimens, as very slight constrictions
of the frond, with swollen nodes, — not forgetting that the horny nodes shrink much in drying -
and with prominent conceptacula.

The older names may be kept to designate the various aspects which this species can assume.

f. fragilissima syn. A. fragilissima Lamx. and A. debilis Kütz.
slender plants, joints thin, nodes not very conspicuous as a rule.

f. aispidata syn. A. aispidata Lamx.

plants stronger than the typical form, but not so strong as

f. cyathifera syn. A. cyathifera Lamx.

which is characterizeel by its strong swollen nodes.

I may perhaps be criticised for sinking the above named species into A. fragilissima,
for it seems hard to believe that the delicate plant of Lamouroux should belong to the same
species as the coarse A. cyathifera. But we must not forget, that we do not know how or when
most Amphiroae were collected.

Slender forms will sometimes spring from old, thick, creeping branches, but whether therefore
all slender fronds are young stages of plants that might become coarse and big if they had
had time to live, I do not dare to assume. It appeared to me that mam species had both
a slender and a stouter form. The plants growing in deeper water showed, as a rule, a tendency
to grow in length, while those growing in shallow water or exposed to the breaking of the
waves, dwindled into small, compact, much branching forms. I believe that these different forms
belong to only one species, but whether this difference in the form of growth is indeed the
result of external inrluences is a question that for many algae must still be settled by careful
investigation on the spot where these plants grow.

2. A. anastomosans n. sp. PI. XIV, fig. 3, 4.

Tiny plants 1 — 1,4 cm. high, forming a compact tuft on the substratum. Fronds
cylindrical, branching dichotomously, branches anastomosing frequently. Joints indistinct because

92

nodes are very inconspicuous ; where nodes are visible they are composed of central strand and
cortical layer. Intra-nodal period of growth consisting of 4, 5 rows of long cells, 36, 56, 72, 76 u.
high, short cells ! 1 2 ■>. high.

Conceptacula prominent; diameter of conceptaculum 160 — ibop.; tetraspores 4S a high.

Stat. 78. Lumu-Lumu-shoal, Borneo-bank.
Sikku, Flores. leg. A. Weber-v. Bosse 1888.

The little plant is in all its details smaller than A. fragilissima and yet the structnre
of its central strand and its prominent conceptacula are so exactly like what one sees in A.
fragilissima that I feel convinced that the two plants must be closely allied. A. anastomosans
is a dwarf-form, and its small conceptacula and the absence of pad-like swollen nodes have
induced me to distinguish it as a new species.

3. Amphiroa foliacea Lamx. PI. XIV, fig. 1 — 11.

Lamouroux in Voyage de 1'Uranie par Freycinet Zoölogie par Quoy et Gaimard, p. 628,

tab. 93, fig. 2—31.
Areschoug in J. G. Agardh, Spec. Gen. et Ordines Algarum, 1851, p. 541.
Amphiroa squarrosa Grunow, in bot. herb. zu Hamburg!

f. procumbens.

Stat. 248. Island Tiur.
Stat. 282. Eastpoint Tirnor.

f. c recta.

Stat. 71. Makassar.

Stat. 78. Lumu-Lumu-shoal, Borneo bank.

Stat. 91. Muaras-reef, Borneo bank.

Stat. 93. Sulu Archipelago.

Stat. 99. North-Ubian. Sulu Archipelago.

Stat. 129. Island Karkaralong.

Stat. 240. Island Banda.

Stat. 261. Kei-Elat, Kei Islands.

Stat. 281. Timor.

Fronds branched, di- or trichotomously, with adventitious branches springing from the
nodes, dimorphous: either creeping horizontally with broadly winged joints with undulating
border, or ascending vertically with joints almost cylindrical, more or less flattened, more or
less winged with narrow undulating border. Joints not swollen and paddike as in A. fragilissima.
Nodes consisting of central strand and cortical layer; as a rule as broad as the joint, but if
the joint is very much winged, not quite so broad. Central strand built up of 3, 4 rows of
long cells 44, 60, 76, 84, 92, 96 jul high, foliowed by a row of short cells 8, 16, 24 u.
high. Cells of the central stram! pass not abruptly at the periphery into the cells of the cortical
layer forming the wing, sometimes in broad specimens these latter cells are quite as high as
: of the central strand. Rows of cells of the central strand form curved lines, not straight
as in A. fragilissima, conceptacula half immersed in the frond, not so prominent as in
A. fragilissima. Diameter of conceptacula 240 — 400 ij. ; height of tetraspores 48 p..

93

The different aspects which this Amphiroa can assume, have already been spoken of
in the beginning of this paper. On account of its variability it is not an easily recognized
plant. lts central strand and manner of forming the node have much resemblance to the central
strand and node of young specimens of A. fragilissima, still both are easily recognized if
carefully studied. A. foliacea has a strong tendency to flatten its frond and to form an undulating
border, visible even in almost cylindrical joints. The joints are thick throughuut their whole length
and lack the pad-like swellings at top and base that characterize older joints of A. fragilissima.
The half-immersed conceptacula, not nearly so prominent as in A. fragilissima, help further
to distinguish this species.

A. foliacea is a widely spread plant-, it can grow so sturdy as to resemble outwardly
Metagoniolithon charoides (Amphiroa charoides) and I got it even under that name from the
West Indies. A longitudinal section through both plants will however show at once how widely
different they are.

The different aspects A. foliacea can assume may be characterized as follows :

f. procitmbens. PI. XIV, fig. i.

fronds creeping, joints broad, winged with undulating border.

f. erecta. PI. XIV, fig. 2—8.

fronds erect, joints slightly winged, sometimes almost cylindrical. This latter form is
identical with the alga named A. nobilis Kütz. by Hauck in: Über einige von Hildebrandt
im Rothen Meere und Indischen Ocean gesammelten Algen.

Grunow again in his paper on the ''Algen der Fidschi, Tonga und Samoa Insein" speaks
of an Amphiroa from Ovalau "bei welcher einige Aeste und Glieder sehr an A. tribulus
andere aber durch die langgliederige, oft fast stielrunde dunne Gestalt sehr an A. fragilis
erinnern". I think that this plant is A. foliacea, the more so because I found in the Herbarium
of the Hamburg Museum A. foliacea under the name of A. sq nar rosa Grunow.

4. Amphiroa anccps (Lamk.) Decne. PI. XVI, fig. 6 — 8.

Corallina anccps Lamarck, Mém. du Mus. II, 18 10.

Corallina dilatata Lamouroux, Hist. des polyp. flexibles. Caen, 1 S 16.

Amphiroa Karstalskii Postels et Ruprecht, 111. Alg. Oc. Pacifici 1840, at Bot. Mus. in herb.

Petersburg.
Amphiroa Cumingii Postels et Ruprecht, 111. Alg. Oc. Pacifici 1840.
Amphiroa anccps Decaisne, Ann. des sciences nat. 1842. Bot., v. 2.

Amphiroa nobilis Kutzing, Spec. Alg. 1849. Tab. Phye. t. 8, p. 25, pi. 51 in herb. Kützing.
Amphiroa dilatata Lamouroux, Areschoug in Agardh, Spec. Gen. et Ord. Alg. 185 1.
Amphiroa galaxauroides Sonder, Plantae Preiss. vol. 2, p. 188 in herb. Kützing.
Amphiroa ephedraea in Harvey, New South Wales Algae N°. 458!
Amphiroa ? in Ferguson, Ceylon Algae N". 119! 224! 442!

Stat. 33. Pidjot Bay, Island Lombok.
Stat. 307. Bay of Endeh, Island Flores.
Sikka, Flores. leg. A. Weber-van Bosse 188S.
Southcoast Java, leg. Teysman.

"1

Fronds branched dichotomously at top of broadened joint, rarely trichotomously, with
adventitious branches springing from the joints not trom the nodes; joints flattened, two-edged,
at base of plant sometimes cylindrical and sometimes much flattened and broadened; upper joints
almost always flattened, rarely a cylindrical joint between flat ones. The top joints as a rule
a little fan-shaped, distinctly zonate, less so if top joint is, - - as seldom happens - - almost
cylindrical or bears conceptacula.

Central strand of 2, 3, 1 or 5 rows of long cells, foliowed by a row of very short cells.
1 leieju of long cells from 36 — 140 u, height of short cells 12, 16, 20 v.. Rows of cells spreading
fan-like from the centre towards the periphery. Cortical layer relatively thin except in old joints.

When the broadened top joint is going to divide, there appear two little short cylinders
of non-calcified tissue consisting of a part only of the central strand. These short cylinders
are to be the two nodes; the cells constituting the nodes increase slightly in number, thicken
their membranes and grow till they have the size of the adult node. The remaining portion
of the central strand and the cortical layer adhering to it, remain calcified and separate
around the nodes.

Three little teeth of calcified tissue surround the base of the two nodes, one at each side
and one in the middle. The node itself is however later on surrounded by a young layer of
non-calcified cortical cells, and by their repeated divisions the node increases in thickness with
the thickness of the frond. In some specimens the two nodes touch one another at the base
and 110 little tooth of calcified tissue is observed between the nodes which are then practically
only one node at their base, which divides in apical direction into two portions.

Conceptacula appear as a rule only on one side of the joint, but sometimes on both
sides. They are slightly prominent and have a diameter from 250 — 410 u.. I have seen con-
ceptacula with tetraspores and with antheridia.

Amphiroa anceps was described by Lamarck under the name of Corallina anceps from

New Holland; a few years later Lamouroux described the same alga from the Cape under the

name oi A. dilatata. Akeschoug pointed out the great affinity between A. dilatata, anceps and

rbankii but maintained the three species. My collections give me entire justification for sinking

A. dilatata in A. anceps, which name as the older one must be maintained. The joints of A. anceps

differ so much, even on the same plant or on different plants from the same locality, that the

limits must be made very large. Specimens with long, flat, zonate joints, with nodes having

e little teeth and with conceptacula in two rows on one side of the joint wil! be found inter-

mingled with joints with obtuse border, not so compressed and without the characteristic teeth.

1 he nodes of all these plants have this in common that they consist of a part of the

central strand. The two nodes in one articulation may remain isolated each by itself or adhere

h( r al their base, and this may be observed even in the same plant. The cortical layer

ring to the nodes increases in thickness with the thickness of the frond and can in old

fronds attain a considerable dimension.

1 he central strand is built up of two, three, four or five rows of long cells foliowed by

row oi shorter cells and then a row of very short cells 8 — 12 <j. high. These two latter

tot always to be found. sometimes there is only one row of 20 — 40 n. high. The

95

conceptacula are sometimes very big, 410 u. in diameter, sometimes much smaller. I thought
their size might perhaps give a good speciiïc character but since I found on one joint concep-
tacula of 300 o. and others of 400 u, I feel doubtful as to the value of this character. I
believe that all the above given synonyms belong to A. anceps, or it might be that the plants
when studied in detail at the spot where they are growing, will reveal constant differences that
escape me now; for the present I feel inclined to explain the existing differences in form of
joint etc. by influences of the environment. A. anceps is highly variable in its colour. Dried
specimens of very different hue have passed through my hands, but this cannot be considered
as a specific character.

5. Amphiroa Bowerbankii Harv. PI. XIV, fig. 14 — 17.

Harvey. Nereis Australis p. 97, tab. XXXVII.

Kützing. Tab. Phyc. t. 8, tab. 50. (The little dots of calcareous tissue on the nodes are not

given in the figure).
Areschoug in J. G. Agardh. Spec. Gen. et Ord. Alg. t. II, 185 1.

Southcoast of Java. leg. Teysman.

Fronds branching di- or trichotomously from broadened top of last joint, with adventitious
branches springing from the joint not from the node; joints at base of plant either cylindrical
or broad and flattened, more or less canaliculate with central strand protruding below, winged
with undulating border. Upper joints flattened, border slightly undulated or two-edged, in well-
grown specimens distinctly zonate.

Central strand consisting of 2, 3, 4 rows of long cells 56 — 108 p. high, foliowed by a row
of short cells 8, 12, 40 ij, high. Cortical layer thicker than in A. anceps. Nodes appear as in
A. anceps in the broadened part of the last joint and consist of a non-calcified part of the central
strand and of the cortical layer adhering to it. The other calcified part of the central strand splits
as does also the cortical layer. From the uncalcihed cortical layer of the node spring forth 1 — 7
little dots of calcareous tissue, sometimes covering the node almost entirely. Conceptacula slightly
prominent or subimmersed in the cortical layer, diameter from 360 — 460 ij..

Amphiroa Bowerbankii is closely allied to A. anceps but it can be easily distinguished
from this latter alga. As a rule it is broader, its cortical layer thicker, its conceptacula less
prominent than in A. anceps, but the character by which it is most easily known are the nodes
which are characterized by little dots of calcareous tissue. These will sometimes fa.il, though as
a rule I believe, that on carefully searching a plant apparently without dots, one is almost
always sure to find at least a single dot, indicating the species to which the plant belongs.
The joints at the base, when they are not cylindrical, can be as broad as 1 cm. and these
broad joints have an outward likeness to the joints of A. canaliculata. I once thought that
plants with such broad joints, might be an intermediate form between A. Bowerbankii and
A. canaliculata and wrote in this sense to Mr. Vendo ') but specimens that I got later on

1) K. Vkndo. Enumeration of corallinaceous algae hithevto known from Japan. — Bot. Mag. of Tokyo, vol. XVI, n» 189,
Nov. 1902, p. 6.

96

convinced me thal plants of ./. Bowerbankii can have broad joints and stil! have nothing really
in common with A. canaliculata. A. anceps can also have broad joints at its base though not
nearly as broad as I have seen them in A. Bowerbankii.

central strand has, just like the central strand of A. anceps, 2, 3, 4 rows of long
cells from 80 — 100 u long. These are foliowed by a row of shorter cells from 20 — 40 u. long,
and dien comes a row of ver\' short cells trom cS — 12 *j. long. The latter two rows are not
always developed, sometimes one finds only one row of ' 20 p. high. The cells of the central
strand spread fan-like from the centre towards the periphery; the cortical layer is thicker than
!. anceps, the conceptacula are often more or less overgrown and appear therefore smaller
than in A. anceps but they are really not smaller. As a rule they cover one side of the joint but
1 found them frequently on both sides.

On PI. XIV are figured two nodes of Amphiroa Bowerbankii; fig. 17 shows us the
node in a young stage seen from above, fig. 16 in longitudinal section. On the same plate we
see a node of A. exilis and A. ephedraea and on PI. XVI, fig. 7 and 8 a node of A. anceps.
. I . exilis was not amongst the plants of the Siboga collection, but I thought it worth while
to have its node figured, because these four nodes show clearly how with small variations a
distinctly different node is formed. In A. anceps and A. Bowerbankii the node is formed of a
part of the central strand and a part of the cortical layer, in A. exilis the whole central strand
and a part of the cortical layer take part in the formation of the node, and in A. ephedraea
the notie is built up of the entire central strand and the entire cortical layer. In the marmer
in which the long and short cells alternate in these four species and in the fan-like spreading
of the cells from the centre towards the periphery, there is another indication of their close
relationship. All the above named four species occur on the East coast of South Africa and
at least three of them along the coast of New Holland and interveningf regions.

6. Amphiroa ephedraea Lamck. PI. XIV, fig. 13.

Corallina ephedraea Lamarck, Mem. du Mus. II, 18 10, p. 238.

Amphiroa Gailloni Lamouroux, Pol. flex. t. XI, 18 16, p. 298.

Amphiroa fusoides Lamouroux, Pol. flex. t. XI, 18 16, p. 298.

Amphiroa ventricosa Lamouroux, Sec. Decaisne.

Amphiroa ephedraea Lamarck, Harvey Nereis Austr. t. XXXIX, p. 95.

Amphiroa Gueinzii Harvey, Nereis Australis, t. XXXIX, p. 95.

Amphiroa Poppig'n Endlicher et Diesing.

Amphiroa ephedraea Lamarck, Areschoug in J. G. Agardh, Spec. Gen. et Ord. Alg. 185 1, p. 534.

Amphiroa ephedracea Decaisne, Kützing, Tab. Phyc. t. 8, tab. 47.

Benkulen, Sumatra. de Bril. leg. Snackey.

Sturdy plants, fronds branching dichotomously, branches spreading; joints at base of plant

indrical or compressed; upper joints cylindrical or a little compressed, many times longer

broad. Central strand consisting of 2, 3 rows of elongated cells from 60 — 118 tj. high,

row of shorter cells from 20 — 40 u high, which again is foliowed by a row of

from 8 — 12 <). high. Sometimes one finds only one row instead of two, which

20 — 40 u. Node consists of entire central strand and cortical layer, which

97

grows in thickness with the thickness of the frond. In the node I observed from 7 — 13 rows
of alternately long and short cells.

Conceptacula big ; slightly prominent, subimmersed in the frond. Diameter of conceptacula
from 300 rj. to 1 mm. Tetraspores from 90, 120 to 160 p. long.

Amphiroa ephedraea is on account of its size one of the most easily known Amphiroae.
Small or dwarf specimens may give some trouble, but the structure of the node is a ready
help to recognise the species. The whole central strand and the whole cortical layer take part
in its formation, and the branching does not always take place at the node, but sometimes a
little below it. The cells constituting the node are short when the node is young, but increase
in length for a long time; I found cells at the base of the plant from 160 — 190 rj. long. I do not
think however that the number of cell-rows increases in any node ; I believe these to remain
the same in each node and not to change from the moment the node is formed. The
conceptacula and tetraspores of A. ephedraea are the biggest I have seen in this genus. In
longitudinal sections the conceptacula are seen to be immersed in the frond though, on account
of their size, they rise also above the frond.

In Kützing's herbarium is a specimen called A. linearis Kützing from Senegambia
which I think is a stunted form of A. ephedraea. The nodes are like those of this alga only
rather short, and the top joints are a little flattened as in A. Beauvoisii Lamx. [exilis Harv.).
The material at my disposal is however far too little to allow me to express myself with more
certainty. It is highly desirable that naturalists should collect Amphiroa in vvell-grown specimens
to enlighten the study of these variable plants.

7. Amphiroa eanaliculata von Martens. PI. XVI, hg. 9 — 12.

Von Martens. Die Preussische Expedition nach Ost-Asien. Bot. Theil. Die Tange 1866, p. 28.
Yendo. Corallinae verae Japonicae Tokyo, 1902.

Stat. 33. Pidjot Bay. Island of Lombok.

Stat. 34. Bay of Labuan-Pandan. Island of Lombok.

Fronds articulated, spreading over the substratum, branching dichotomously, branches
much diverging; joints broad, winged ; upper side convex, under side canaliculate with a
prominent midrib. Adventitious branches rare. Froncls consisting of central strand and cortical
layer. Central strand forms the prominent midrib and also the node and is built up of 4 or 5
rows of long cells 80 — 90 ij. foliowed by a row of short cells ± 28 p. ; cortical layer strongly
developed forms also the wings of the joints which are turned downward and sometimes meet
and coalesce. Node consists of central strand and a portion of the cortical layer which grows
in thickness with the thickness of the joints ; conceptacula on both sides of the joint, overgrown
by the cortical tissue. Diameter 320 — 360 \j,. Height of tetraspores 80 p..

A. eanaliculata is a very characteristic plant with its thick broad joints, which have a
very distinct midrib along the canaliculate under side of the joint. The node is the continuation
of the central strand, that divides dichotomously at the joint, and is surrounded by many
layers of cortical tissue. The non-calcified tissue of the node has its greatest development at

SIBOGA-EXPEDITIE LXI.

>3

>

the under side of the upper joint as may 1»' seen distinctly in fig. i i on PI. XVI. The upper side
of the node is covered by overlapping layers of calcified tissue. The cortical Iayer has a very
great development in the joint; the conceptacula appear on both sides of the joint hut they are
only slightly prominent and often entirely immersed by subsequent growth of the frond.

S. Amphiroa crassa Lamx. PI. XV. fig. i — ;.

Amphiroa crassa Lamouroux. Voyage de 1'Uranie par Freycinet. Zoölogie par Quoy et

Gaimard 1S24.
Amphiroa Godeffroyi Grunow. Algen der Fidschi, Tonga und Samoa Insein. In Mus. Godef-

froyensis at Hamb. sub nom. A. robusta.

Stat. 282. East Point of Timor.
Stat. 303. Haingsisi, Island Samau.
Stat. 304. Lamakera, Island Solor.
Southcoast of Java. leg. Teysman.

Fronds articulated, branching di-trichotomously or irregularly, many joints springing from
one broad joint. Adventitious branches arise from the joint never from the node. Central strand
built up of one, two or more rows of long cells, foliowed by a row of shorter cells, very short
ones were not observed. Length of long cells from 80 — 100 ij., of short ones from 32 — 36 u..
Cortical layer much developed. Node consisting only of the central strand, many rows of cells taking
part in its formation. Calcified cortical layer surrounds the node entirely or splits horizontally
leaving the node naked for a very short space. When the cortical layer does not split, a little
constrictton of the frond is the sole indication of the node. Conceptacles superficial, often overgrown
by cortical layers and sunk deep into the frond. Diameter of conceptacles 320 — 340 ij..

I got this beautiful Amphiroa first at the East Point of Timor. It is closely allied to
A. tribulus Lamx, both having the same way of forming their nodes; but whether they belong
to one species I could not ascertain •, the material of A. tribulus that I had at my disposal
was scanty and presented quite a different aspect.

. /. crassa however is a variable plant; I have specimens 8 cm. high, erect, branching
dichotomously and spreading widely, with long, relatively thin joints and other specimens
from 2 — 6 cm. high with short, broad joints. Again other specimens connect both the former
specimens in an unmistakeable marmer. The specimens with the long joints were described by
GrüNOW as A. Godeffroyi.

I believe that a small alga, that I got on a former voyage in the Archipelago at
Sikka. Island of Mores, is a dwarf form of A. crassa. It has the same way of formine its
node and is in all respects like the type but it is not higher than 1 — 1,5 cm. It may be
distinguished as f. minuta.

When I began my study on the Amphiroae of the Siboga Expedition, I did not intend

write a monograph of the genus. For the sake however of those who would like to name

phiroac. with reference to the structure of the node, I have made the following key.

species of which I have only seen one or two specimens and for the specihc

value of which I would not be held responsible, I have marked with a f.

99

Synoptical Key

to the species of Amphiroa.

I
I. Fronds thick, cylindrical or flattened but with round edges; ramification dichotomous or

irregular; adventitious branches springing from the joint never from the node. Node consisting

only of the central strand; cortical layer of calcified tissue splitting horizontally around the

non-calcified node or not splitting at all. In this case node outwardly only visible as a narrow

constriction of the frond. Conceptacles at first a little prominent afterwards entirely immersed:

1. Amphiroa crassa Lamx.
f. Godeffroyi Grun.

f. minitta n. f.

2. Amphiroa tribulus Lamx.

II. Fronds flattened, cylindrical, thick with distinct nodes in the structure of which central strand
and cortical layer take part. Nodes sometimes larger at the under side than at the upper
sicle in A. canaliculata ; cortical calcified layer of upper joint overlaps in older joints the
node and touches the lower joint :

fronds canaliculate. 3. A. canaliculata v. Mart.
fronds not canaliculate. 4. A. involuta Kütz. (syn. A. capensis Aresch.).

III. Fronds cylindrical or flattened with sharp thin edges ; ramification dichotomous.

a. Nodes usually in pairs consisting of part of central strand and part of cortical layer and
appearing at first as two little cylinders at base of joint; when fully developed three small
teeth of calcareous tissue, one in the middle and one at each side of the node, are visible :

5. A. anceps (Lamk.) Dec.

b. Nodes like those of A. anceps but also with little dots of calcareous tissue on the nodes:

6. A. Bowerbankii Harv.

c. Node composed of entire central strand and almost entire cortical layer; with only
two little teeth of calcareous tissue one at each side of the usual single node :

7. A. Beauvoisii (Lamx.) Born.
IA. zona fa Yendo.

}A. echigoensis Yendo.
? A. pit si l la Yendo.

d. Node consisting of entire central strand and entire cortical layer. No little teeth of
calcareous tissue at base of node :

8. A. ephcdraca (Lamk.) Aresch.
f9. A . lincaris Kütz.

IV. Fronds cylindrical, joints swollen padlike at top and base or not swollen in a young state
and never in A. anastomosans ■; rows of cells in central strand almost horizontal not
curved. Nodes consisting of entire central strand and cortical layer, swollen cushionlike
to bear the joints or not swollen in a young state, and obsolete never swollen in A.
anastomosans. Conceptacles very prominent:

IOO

[O. . /. fragilissima Lamx.
f. fragilissima Lamx.
f. cuspidata Lamx.
f. cyathifera Lamx.
ii. A. anastomosans n. sp.
V. Fronds flattened winglike with undulating border, or almost cylindrical with a narrow border
or sometimes none: rows of cells in central strand curved, above all in winged specimens,
not horizontal. Nodes of entire central strand and cortical layer as broad in most cases
as the frond but not swollen cushion-like. Conceptacles not or slightly prominent, half
immersed in the tissue of the frond:

12. A. foliacea Lamx.

VI. Fronds cylindrical. Nodes consist of central strand; cortical layer splits or takes part in
the formation of the node. Conceptacles prominent, half immersed or - - in old plants -
entirely sunk in the cortical layer:

a. Node consists of one layer of cells of the central strand :

13. A valonioides Vendo.

b. Node consists of one layer of long cells and one layer of short cells of the central
strand and as a rule also of the cortical layer:

14. A. verrucosa Kütz. (syn. A. cryptai'throdia Zan.).

c. Node consists of two layers of cells of equal length, with oblique anticline walls,
cortical layer splits :

15. A. rigida Lamx.

d. Node of three layers of cells of equal length; cortical layer splits:

1 6. A nodulosa Kütz.

e. Node of three or more layers of cells, alternately one long and one short, or two
long and one short ; irregular. Cortical layer splits or partially takes part in the
formation of the node by secondary growth :

17. A. dubia Kütz.

I should like to add a few remarks for the clear understanding of some species mentioned
in this Key.

To Amphiroa Beauvoisii (Lamx.) Born., which species includes according to Ardissone l)
and BORNET ') :

A. exilis Harv.
A. pustulata Mert.
A . co Dip la na ta Kütz.
A. po ly zona Mont.
A. algericnsis Kütz.
jst also be reckoned A. brasiliana Dec. All these plants have the same structure of node.

Phycologia mediterranea 1883.
: E. Bornet. Les algues .11- 1'. K. A. SCH01 iboe. Paris 1892.

IOI

Almost the same structure occurs also in an alga in Surixgar's herbarium called A. pertianae
Aresch. I have not been able to find the diagnosis of Areschoug for A. peruanac. Perhaps it
is only a herbarium name and therefore without value. My specimen is barren but the plant
is so sturdy that I must leave it to others to decide whether this plant is a separate species
or a variety of A. Beauvoisii. It differs in the structure of its node so far as this, that it is
very distinctly semMunar.

To A. rigida Ardissone refers :

A. spina Kütz.

A. irregularis Kütz.

A. amethystina Zan.

A. cladoniacformis Men.
I have not seen A. amethystina Zan. but the three others named pseudo-species had
all the same structure of node as the typical A. rigida. This same structure of joint and node
occurs also in a specimen in the herbarium of Prof. Süringar which was collected by him in the
West Indies. It is characterized by widely spreading, dichotomous, intricate branches and may
well be distinguished as f. divaricata pi. XVI, fig. 13. I have not seen a single type specimen
of A. parthaiopca Zan. and cannot decide whether it is a synonym of A. rigida as Bornet,
or of A. complanata (exilis) as Solms-Laubach affirms.

A. verrucosa was described by Kützing in 1843 in his Phycologia generalis; he changed
this name afterwards into vcrruculosa, probably because the name of verrucosa had already been
given by Lamouroux to a species of Amphiroa. We shall however see in the following pages
that Amphiroa verrucosa Lamx. is not an Amphiroa, and therefore the old name of Kützing
may be revived. Solms-Laubach has described A. verrucosa (verruculosa) Kützing and A.
cryptarthrodia Zan. as two distinct species. They have however exactly the same structure of
joint and node and I consider them therefore as identical. A. verrucosa Kütz. as being the
older name, must be maintained.

Though I have not seen A. valonioides Yenclo, the node of this alra is so well fio-ured
by Yendo that I was able to classify it in this key. I have neither seen A. zonata Yendo, A.
cchigoensis Yendo nor A. pusilla Yendo. According to the figures they seem to belong all three
to the genus Amphiroa and to stand near to A. Beauvoisii and A. pertianae, but as Yendo
has not given a figure of the node, I do not dare to express myself with greater certainty.
Amphiroa misakiense Yendo is probably an Arthrocardia.

Metagoniolithon n. g.
PI. XV, fig. 9-14-

The genus Metagoniolithon though it was not found in the collection of the Siboga
Expedition may well be treated here, for my researches on the Amphiroac of the Siboga led
me to create this genus. It has been characterized in the Synoptical Key on p. 86. It is
distinguished from Amphiroa as we have already seen by the anatomical structure of the
joint and node. The cells are of a uniform size throughout the whole joint, only growing

102

gradually a little smaller towards the node where the cells are relatively very small and thick-
walled. In the growing apex the node is already visible as a mass of small cells (PI. XV, fig. 1 1).

From the node, which increases in thickness with the thickness of the frond, spring
forth the branches and in one species of this genus M. stelligerum the node, the cells of
which struck me as not being particularly small, increases also in length. This takes place
thnui'di the cells of the central strand of the calcified joint losing their chalk where they are
in contact with the node. At the periphery of the central strand cell division takes place
which causes the still calcified cortical layer to be thrown off, either partially or in old plants
almost entirely.

By this process the articulated appearance of the young plant disappears gradually and
a loncr pliable horny branch is seen with remnants of the calcified cortical layer adhering to it
here and there. Such forms of M. stelligerum were described by Lamouroux as Amphiroa
jubata and A. interrupta. Sectioning the horny node of such plants I was struck by the total
disappearance of what everybody is used to consider as the typical structure of the Mclobesiaceae.
The cells <>f the central strand were no longer standing in straight rows one above another as
they had done in the young plants. They now had thick walls, had been pulled in various
directions and the places where their protoplasts clung together, had grown out, giving very
different shapes to the cells, which looked in fact like any ordinary small-celled alga-tissue,
such as one meets in so many Florideac. The cortical layer consisted of rather large cells,
horizontally stretched and growing smaller towards the periphery.

I have dwelt at some length on this structure of old plants of M. stelligerum because
it explains the structure of my new genus Litharthron. M. stelligerum is the link between
M. charoides and Litharthron australis, but it bas more characters in common with M.
charoides. Young plants in fact of M. stelligerum have the same anatomical structure as M.
charoides ■. it is only with increasing age that such characters appear which we shall find
again, only still further developed, in Litharthron australis.

The three species of the genus Metagoniolithon are:
i . M. charoides.

Amphiroa charoides Lamouroux, Hist. d. Pol. flexibles, iSiö, p. 301.

Sturdy plants with branching fronds, distinctly articulate. Joints often a little flattened
from 1 to 3 cm. long and up to 2 mm. broad. Nodes rather short, entirely surrounded by pseudo-
whorls of branches, no true whorls, for the branches do not appear simultaneously. Anatomical
structure of joint consists of many layers of cells all of the same dimension, the cells of each
horizontal row communicating by thin places in the longitudinal wall. In the node the cells
are very small and have thick walls. Fructification consists of conceptacula, which appear first
on the inner side of the branches but may afterwards cover the whole branch.

5 plant is found in South Australia; the specimens which I received under the name of
mphiroa charoides from the West Indies belonged to Amphiroa foliacea. I therefore do not
know, whether specimens of M. charoides have indeed been found in that region.

2. M. graniferum.

Amphiroa granifera Harvey, Phyc. austr. pi. CCXXX. Syn. Cat. 1863, p. XXX.

Ampkiroa intermedia Harvey.

Amphiroa stellata Kutzing, Spec. Alg. 1849.

Amphiroa similis Sonder, Bot. Zeit. 1845.

Amphiroa setacea Kützing, Spec. Alg. 1849.

Plants with branching, distinctly articulate fronds. Joints cylindrical up to 0,5 cm. long and
at the utmost 1 mm. broad. Nodes surrounded by a tuft of branches like whorls, no true whorls
just as in M, charoidcs. Anatomical structure of joint consists of many layers of cells all of
the same size, the cells of each horizontal row communicating by thin places in the thin walls.

In the nodes the cells are very small and thick-walled.

Fructification consists of conceptacula, that appear first on the inner side of the branches
but afterwards cover the whole branch.

A. similis Sond. is only distinguished from A. granifera Harv. by its smaller size. Of
A. setacea Kütz. there is only a very small specimen in Kützing's herbarium. I think it probable,
that this plant belongs also to M. graniferum but I am not quite sure. It has the same anato-
mical structure as M. graniferum.

Af. graniferum is an inhabitant of the Australian seas : the plant called Amphiroa setacea
was found on the coast of Peru.

3. AI. stclligerum.

Corallina stclligera Lamarck, Mem. du Musée II, p. 239.

Amphiroa stelligera (Lamk.) Areschoug, Die Corallineae in Agardh's Spec. Gen. & Ord. Alg.

1851, p. 540.
Amphiroa jubata Lamouroux, Hist. polypiers flex. 1S16.
Amphiroa interrupta Lamouroux, Hist. polypiers flex. 18 16.
Amphiroa verrucosa Lamouroux, Hist. polypiers flex. 18 16.
Amphiroa elegans Sonder, Bot. Zeit. 1845.

Long, slender plants with branching fronds ; distinctly articulate when young, less so in
old plants. Branches whorl-like, very slender. Old plants often drop their calcified cortical lavet-
and slender branches.

Anatomical structure of the young joints consists of many rows of cells, all of the same
size. In the nodes the cells are very small, thick-walled and take various shapes above all in
old plants, where the central layer of the joint loses its chalk and where, by subsequent divisions
of a layer of cells surrounding the central strand, the calcified cortical layer is throvvn off and
a new cortical layer formed. This process gives rise to the cup-like swellings we often see at
the top and base of nodes of AI. stclligerum pi. XV, fig. 9.

Conceptacles on the branches. This species may have a very different appearance accorcling
to whether the cortical layer has fallen off or not. Plants that were rather bare or had only slender,
scattered branches were called A. jubata or A. interrupta by Lamouroux. I believe that A.
verrucosa is also a synonym of this species. Areschoug gives the name as a synonym of A. cha-
roidcs but he ackls, that he has not seen the authentic specimen. Under the name of A. verrucosa

104

we find two specimens in Lamoi roux's herbarium. One is A. stelligera Lamk.; the other is identical
with . /. similis Sond. The first specimen has "bourrelets aux deux extrémités" as Lamouroux called
the swelling at the top and at the base of the joints in his A. verrucosa, swellings which however
are not always to be found there and often disappear in old age. On account of these "bourrelets",
I think that A. verrucosa ma) be taken as a synonym of M, stelligerum. A. stellulata, which
Aki as a synonym of A. charoides, I believe to belong to M. graniferum.

M. stelligerum is like the two other species an inhabitant of Australian seas.

Litharthron n. g.

PI. XVI, fia. 16, 17.

The alga known as Amphiroa australis belongs, different as its anatomical structure may
be, to the Corallineae. This difference was first observed by Kützixg, to whose sharp eyes
little indeed escaped that concerned algae. He was so much struck by the aberrant structure
of Amphiroa australis, that he expressed his doubts as to whether it could be kept amongst
the Amphiroae.

Mv studies have taught me, that Amphiroa australis, for which I propose the name
Litharthron australis, must be kept in the family of the Corallineae and that its place is —
different as their outward appearance ma}- be, — close to Metagoniolithon stelligerum. Some of the
characters which distinguish this latter plant, only still further cleveloped, are also the characters
of Litharthron. The cells of the central strand in the joint are non-calcified, just as in old plants
of M. stelligerum, but in L. australis these cells are intricated and imbedded in a mucilaginous
layer. The cells of the cortical layer of L. australis are big, calcified, thin-walled, filled with
starch near the central strand and becoming gradually smaller towards the periphery; in M.
stelligerum the primary cortical layer of calcified cells is thrown off; the secondary cortical layer
has round, big cells growing smaller towards the periphery but they are non-calcified and thick-
walled. I do not know the nature of the mucilaginous layer in which the central strand ot
Litharthron is imbedded but after putting thin sections of this plant on a slide with a drop
of sea-water, I saw the central part of the section swell up. With the polariser numerous
brilliant spots were visible to the right and left of this central layer, which itself remained quite
dark under the crossed nicols. But it was not the whole central strand that was non-calcifiecl ;
at its periphery I saw distinctly long cells imbedded in a layer that contained chalk.

In the node the cells were of different size, and presented the aspect of an ordinary

Florideae-tissue ; here also a distinct cortical layer was visible. From the node spring the branches

as in the different species of Metagoniolithon. The flatness of the node is a character of small

mportance, but it is a great help towards recognizing L. australis at once amongst a collection

Corallineae. I fancy however that in a living state the flatness of the joint would be less

remarkeable, owing to the mucilaginous layer in the middle of the joint being then swollen.

The conceptacles have not vet been found: I can only hope that they may soon be
overed ai ui that they will coniirm my views about the genus Litharthron.

rthron, just like Metagoniolithon is an inhabitant of the Australian seas. It has

io5

only one species Litharthron australis Sonder, which has the same diagnosis as the genus.
Harvey in his Account of the Marine Botany of the Colony of Western Australia mentions
under N°. 155 an Amphiroa sp. . . . , growing with A. australis, to which it is allied. This
specimen had however become broken in travelling. It would be interesting to know, whether
this plant of Harvey is indeed another representative of the genus Litharthron.

Arthrocardia (Dec.) Aresch.

Though there were no specimens of the genus Arthrocardia in the collection of the
Siboga, I should like to say a few words about this genus as my researches on the Amphiroae
of the Siboga led me to sink many species into the genus Arthrocardia that had hitherto
been considered as belonging to the genus Amphiroa.

The genus Arthrocardia has been sunk by Schmitz and Hauptfleisch into the genus
Cheilosporum. However not only these two genera but the four genera Arthrocardia, Cheilo-
sporum, Corallina and Jania are closely allied together, much closer indeed, than either of
them to the genus Amphiroa. They are probably a separate branch of the family of the
Corallineae, just like Ai/iphiroa and Metagoniolithon, but I have not been able to tracé the
common ancestor of these four first-named genera.

Their anatomical structure is almost the same ; in Jania alone the cells have nearly the same
length in the joint and in the node, which they do not have in the other genera. But there are
species of Jania which have longer cells in the node and seem to bridge over the difference.
The position of the conceptacula however marks a difference between the genera: in Arthro-
cardia we find conceptacula on the joints as in Amphiroa ; in Cheilosporum we find them
immersed in the margin of the broadened, winglike joint, in Corallina and Jania immersed in
the frond at the top of the joints.

We see three stages, each marking a gradation of higher development and I think
that the different names may be kept to mark the different stages. Whether they must be
considered as names of genera or only of subgenera will depend upon the value one attaches
to a character depending upon the position of the conceptaculum and upon the conception
one has of a genus.

To Arthrocardia belong the following species considered by late authors on the subject
to belong to the genus Amphiroa :

1. Arthrocardia cretacea (P. et R.).

Corallina cretacea P. et R.
Amphiroa cretacea (P. et R.) Aresch.

2. Arthrocardia tubercnlosa (P. et R.).

Corallina tuberculosa P. et R.

Amphiroa tuberculosa (P. et R.) Setchell and Gardner.

3. Arthrocardia cpiphlegnoides (Ag.).

Amphiroa cpiphlegnoides Ag.
Amphiroa rudis Harvey.

SIBOGA-EXPEDITIE LXI. 14

1 1 m

1 saw this specimen at the British Museum and I think it very likely, that it is only a
form of ./. tubcrculosa as Yendo supposed in his paper on the Corallinae of Port Renfrew.

4. Arthrocardia aberrans \ en>

iphiroa aberrans Yendo.

5. Arthrocardia declinata (Yendo).

Amphiroa declinata Yendo.

6. Arthrocardia variabilis (Harvey).

Amphiroa variabilis Harvey.

7. Arthrocardia Darwini (Harvey).

Amphiroa Darwini Harvey.

S. Arthrocardia vertebralis Dec.

'Amphiroa vertebralis Aresch.

9. ; Arthrocardia breviarticulata (Aresch.).

Amphiroa breviarticulata Aresch.

I have in the herbarium of Suringar a specimen called Amphiroa breviarticulata
Areschoug which seems to be an authentic specimen. This plant is however an Arthrocardia.
I have tried in vain to see the type of A. breviarticulata and have therefore querried the species.

I have not seen a specimen of Amphiroa crassissima Yendo but I feel almost sure that
this alga belongs to the genus Arthrocardia.

All these plants are distinguished from Amphiroa by their nodes, which consist of one
row of pericline cells which, when fully developed, have a length from 200 to 350 u. and even
more. while the cells in the joints have as a rule a length of 60 a.

Yendo has given very good figures of the nodes ancl some layers of the joint of
Arthrocardia aberrans and Amphiroa rigida but he has failed to grasp the importance of
what he was illustrating-.

&■

Cheilosporum (Dec.) Aresch.

In the collection of the Siboga was one species of Cheilosporum to which was added
a specimen not vet known from the Indian Sea, and collected by Teysman, the well-known
Indian botanist.

1. Cheilosporum spectabile 1 larv.

Harvey. Friendly Islands Algae N°. 31.
Grunow. Alg. d. Fidschi u. Samoa Ins. p. 41.

Stat. jS. Lumu-Lumu-shoal, Bank of Borneo.

Stat. <Si. Island Sebangkatan.

Stat. 129. Karkaralong Islands.

Stat. [33. Lirung, Karakelang Islands.

Plants with slender or robust fronds, branching dichotomously, forming thick tufts in the
littoral region, but straggling and creeeping in deeper waters of 34 — 36 m.

107

Breadth of flattened frond from 0,5 to 3,5 mm., apex attenuated, roundish, sometimes
cylindrical in specimens brought up with the trawl.

Lobes inserted with broad basis with acute, truncate, dentate, thin apex.

Central strand much cleveloped with protruding midrib, especially in specimens from
deeper water, in which even the lobes may be entirely wanting for a fairly long distance.
Fructification consists of conceptacles at the acute top of the lobes immersed in the frond ;
either single or in pairs.

As far as I know, no diagnosis had been given by Harvey of this widely distributed
alga. It is to be found in the Indian and Pacific Oceans and adopts many forms according
to the depth in which the specimen grows. The specimens that came up with the dredge were
very slender and several of them had attached themselves to other algae on which they
were creeping.

The specimens of Harvey from the Friendly Islands are more robust than those of the
Siboga. The specimens distributed by Grunow from Upola connect the two forms and take up
an intermediate position between the two.

According to Grunow C. spectabilis Harv. much resembles Amphiroa multifida Kütz.,
whether the two are identical I do not dare to decide.

2. Cheilosporuiii jungermannioides Rupr.

Postels & Ruprecht. Illustr. Alg. Oc. Pacifïci 1840.
Southcoast of Java. Teysman.

After careful examination of the type of Cheilosporum jungermannioides, I feit convinced
that the plant from the South-coast of Java is identical with it.

Corallina Lamx.

1. Corallina pil ij era Lamx.

Lamouroux, Pol. flexibles. Caen 18 16.

Areschoug in Agardh's Spec. Gen. et Ord. Alg. 185 1.

Stat. 169. Atjatuning, New Guinea. On dead branches submersed in the sea.
This Corallina is probably only a form of the highly variable C. Cuvieri Lamx.

Jania Lamx.

1 . Jania adhaerens Lamx.

Lamouroux, Pol. flex. 18 16.

Areschoug in Agardh, Spec. Gen. et Ord. Alg, 185 1.

Kutzing, Tab. Phyc. t. 8, pi. 83^.

Stat. 78. Lumu-Lumu shoal, Borneo-bank.

Stat. 172. Gisser.

Stat. 225. Lucipara-group.

io8

2. Jania rubetu L. Aresch.

rallina rubens (Linn.) Ellis et Solander p. 120.
Areschoug in Agardh, Spec. Gen. et Ord. Alg. 185 1, p. 557.
Kui ab. Phyc. t. S. pi. 84, VIII.

Stat. 33. Bay of Pidjot. [sland Lombok.
Stat. [93. Sula Besi I-land.

3. Jania tenella Ktzg.

Kützing, Tab. Phyc. t. 8, pi. 85 II.

Grunow, Die Alg. d. Fidschi, Tonga u. Samoa Ins. erste Folge p. 42.

Stat. 155. Waigeu Island.
Stat. 193. Sula Besi Island.

( >f Corallina the Siboga brought home only one species, of the subgenus Jania three
species; but it proved easier to say that a plant belonged to the subgenus Jania, than to say
to which species of Jania it belonged. In all the above-named three species the length of the
cells in the node was fairly equal to the length of the cells in the joint. The rows of cells in
the joint were not always of such equal length as one would suppose from our fig. 16 on
pi. X\T which is not entirely correct in this respect. The rows were usually more like fig. 18
on the same plate.

I have tried hard to name the three species rightly but these slender forms are so
extremely puzzling, that I may still have been mistaken.

Lithothrix Gray.
PI. XV, fig. 15.

Lithothrix is quite an aberrant genus on account of its calcified constrictions that take
the place of the horny nodes in all the other genera of the Corallinae verae. lts only species,
known as Lithothrix aspergillum Gray, has the same diagnosis as the genus. It is an inhabitant
of the West Coast of America.

The synoptical Key of the Genera has been made with a view to facilitate for students
the determination of the Corallineae verae\ the outward characters, so far as they coincided
with internal ones, have been used as has also the anatomical structure of the frond and the
node. The order in which the different genera are enumerated in this Key, gives no idea of their
;lationship. I believe that these genera, which are outwardly so much alike, have sprung from
different members of the group of the Lithothamnioneae. The many affinities existing between
Amphiroa and Litkothamnion have been put forward by Solms-Laubach, the observations of
inent naturalist have simply been confirmed by my researches, but we have seen at the
same time that the old group of Amphiroa breaks up into different genera.

109

It seems to me that just as closely as Amphiroa is allied to Lithothamnion, so closely is
one of the new genera allied to Goniolitkon and I called it therefore Metagoniolithon. In both
we find cells of a uniform size, in Goniolitkon throughout the whole frond, in Metagoniolithon
with this difference, that the cells in the node are smaller than in the joint ; in both, the cells
communicate by rather large thin places in their longitudinal walls and in Goniolitkon frutescens
I observed slight constrictions of the frond, that seemed to indicate how easily nature might
develop this constriction into the node of Metagoniolithon.

In Goniolitkon however I observed Deckzellen, not quite as such a regular layer as in
Lithothamnion or Amphiroa, but still each apical cell had its own "Deckzelle". In young fronds
of Metagoniolithon graniferum I did not see any "Deckzellen".

It struck me as very interesting to observe how from two genera so closely allied as
Lithothamnion and Goniolitkon had probably sprung forth two genera Amphiroa and Meta-
goniolithon which resembled one another outwardly so much and were still so different.

Fossil forms.

The study of the anatomical structure of the Corallineae verae afforded the means of
recognizing at once in sections of fossil calcareous algae, whether the plants seen in the sections
belonged to Amphiroa or to either of the other genera. The tissue of the nodes, not being
calcified, has always disappeared but the nodes remain visible in the sections as an interruption
of the calcareous tissue.

In sections, that I owed to the kindness of Prof. Martin at Leiden, I found two plants
figured on pi. XVI, fig. 17 and 18. The sections were made from specimens collected on the
mountain Nona in Ambon and at Totok in Celebes. The section of Ambon bore the number 327
of Prof. Martin's collection; the section from Totok the number 44 coll. F"orster. Prof. Martin
was so kind as to teil me that the stratum at Ambon from which the section N° 327 came, is
situated 400 m. above the level of the sea and belongs probably to the pliocene. The stratum
at Totok in Celebes, section N" 44 belongs undoubtedly to the tertiary formation and is certainly
not younger than old miocene..

The plant from Totok shows us two half joints and an open space between these two,
where most probably the node was to be found when the plant was in a living state. When
we compare fig. 18, pi. XVI with fig. 15 on the same plate representing a recent Jania,
every one can see that the plant from Totok is more like a Jania than a Corallina.

The length of the node, which is almost the same as the length of the cells in the
joint, is a character that points to Jania, for the only case that I know of Amphiroa having
nodes consisting of one row of cells, is A. valonioides and here the layers of cells succeeding
one another grow smaller and alternate as they should do in an Amphiroa ; but this they do
not do in the plant from Totok.

To which species of Jania the plant from Totok stands nearest, I should not dare to
decide for the species of Jania are very difficult to know from one another, even in a living
state. The fact that the subgenus Jania occurred already in old miocene, appeared interesting

I IO

enough to me, even though I could not make out to what species this okl plant belonged
or whether it is new to scieni

rhe plant figured in PI. XVI, fig. 17 is from Ambon; it shows periods of intra-nodal
growth so exactly like what one observes in the recent Amphiroa fragilissima PI. XVI, hg. 19
that I do not hesitate to range the fossil alga near that species, even though a node is not
visible in the section. A. fragilissima is very variable, being slender or coarser in various stages
of development; the ahnost horizontal rows of cells that build np its central strand are
however so characteristic that I feel almost sure that the plant of Ambon, which has fairly
horizontal rows, belongs to A. fragilissima.

In a section from Cape Haharu near the mouth of the Tami-River, New Guinea
N . 734, that 1 received trom Prof. Wichmann at Utrecht, I found this same Amphiroa and
also another species. This last one was coarser than the former and the rows of cells in its
central strand were decidedly curved. I think it very probable, that this is A. fo/iacca. a
speeies much resembling A. fragilissima in the anatomical structure of the frond. In the absence
of nodes and conceptacles the determination of both species is however liable to controversy.

In a section from the Island Karas on the western coast of New Guinea, I detected also
an Amphiroa spec. This section was therefore interesting because, as Prof. Wichmann told me
in a letter, the layer from which it came, certainly dated from the eocene.

PLATES

All the figures plate I — XIII are from photographs in natural size.

PLATE I.

Lithothamnion siamense Fosl. f. typica. Fig. i — 2.

pjcr. |. tile specimen attached to the shell of a mollusc and in part covering a young crust of

anotlier calcareous alga, tube of a worm and the lower part of a small coral. Stat. 49 .
Pig, 2. Fertile specimen in company with a young Squamariacea, Bryozoa and corals. Stat. 64.

Lithothamnion siamense Fosl. f. pseudoramosa. Fig. 3 — 9.

3. A net quite certain specimen of this form, apparently forming in part a thin crust over

another calcareous alga {Lithothamnion sp.), attached to the shell of a mollusc. Stat. 234.
Fig. 4 — 9. Specimens in part covering other calcareous algae {Lithothamnion sp.). Stat. 258.

Lithothamnion bandanum Fosl. Fig. 10.
10. Specimen surrounding a conglomerate of sand or stone. Stat. 240.

Lithothamnion fragilissimum Fosl. Fig. 11 — 16.

Fig. 11. An almost entire and fertile specimen of the plant, rather burdened with Bryozoa and other

animals. Stat. 81.
Fig. 12 — 16. Fragmentary specimens which are more or less burdened with extraneous objects, particularly

in the part turning downwards. Stat. 81.

Lithothamnion prolifer Fosl. Fig. 17 — 20.

Fig. 17 — 18. Specimens not quite typical and somewhat fragmentary, loosely covering a conglomerate of

other calcareous algae and animals. Stat. 78.
Fig. 19 — 20. Typical specimens in part fragmentary, with irregular prolifications. Stat. 81.

Lithothamnion sitnidans Fosl. f. crispescens. Fig. 21 — 23.

21. An intermediate form between f. typica and f. crispescens, however most nearly connected

with the latter. The specimen is attached to a coral. Stat. 299.
Fig. 22. A fertile specimen of the form. Stat. 261.

27,. A somewhat stunted specimen of the form. Stat. 282.

Lithothamnion simulans Fosl. f. typica. Fig. 24 — 2;.

24. A specimen rather closely adherent to the substratum : some hard conglomerate and a young

Goniolithon Reinboldi, the latter fully covered. Lowest to the right on the figure is to be

seen a crust of Archaeolithothamnion erythraeum, which also is partly covered with the

species in question. Stat. 213.

The plant partly surrounds the branches of a coral, partly expands between the latter. Stat. 213.

Siboèa Expeditie LXI Mebe

Bosse ft, Foslie Coralünaoe

i # 0 t 4 #$ <*

o o • e

'%-*

<Sa

■f., nft

PLATE II.

Lithothamnion fruticulosum (Kütz.) Fosl. f. pteridoides Fosl. Fig. i — 4.

| 1 — 2. Specimens with emptied conceptacles of sporangia, the first one with a smal] and young

crust of another species partly covered with the plant. Stat. 240.
Fjtr. 3. A specimen partly covered with crusts of one or two other calcareous algae. Stat. 123.

Fig. 4. A fertile specimen partly covering another calcareous alga. Stat. 64.

Lithothamnion fruticulosum (Kütz.) Fosl. f. clavulata Fosl. Fig. 5 — 9.

5. A specimen apparently forming transition to f. pteridoides. Stat. 240.

Fig. 6 — 7. Typical or almost typical specimens, especially one of them much burdened however with
small crusts of other algae, corals and animals. Stat. 109.
S. A similar, but younger specimen. Stat. 250.

9. A more delicate and somewhat stunted specimen of the form. Stat. 109.

Lithothamnion australe Fosl. f. tualensis. Fig. 10 — 17.
10 — 17. Young and old specimens of the form. Stat. 258.

Lithothamnion australe Fosl. f. iilüana. Fig. 18 — 24.

fig. 18 — 19. Specimens from stat. 86.

. 20 — 22. Specimens from stat. 99.

Fig. 23. A specimen from stat. 106.

Fig. 24. A specimen from stat. 109.

Lithothamnion australe Fosl. f. brachiata. Fig. 25 — 3S.

1 i . 25 — 30. Specimens from stat. 303.

. 31 — -32. Specimens from stat. 99.

F'g- 33 — 34- Specimens from stat. 109.

35 — 38. Specimens from stat. 282.

Lithothamnion australe Fosl. f. minutula. Fig. 39 — 62.

39 — 5°- Specimens from stat. 109.
' imens from stat. 258.

Siboga Expeditie LXI WVbei-.vaii Bosse §,FosIie Coralliiiaceae .

<A * '

^k.Mpfc

•v «**

f- f- ff •£ *, jf. i * y" v'y

&• * ¥

© ©

PLATE III.

Lithothamnion erubescens Fosl. f. haiugsisiana A. Web. et Fosl. Fig. i — 19, 21 — 22.

Fig. 1 — 11. Young specimens of the plant. Stat. 303.

Fig. 12 — 16. Older specimens. Stat. 303.

Fig. ij. Two anastomosing specimens. Stat. 303.

Fig. iS — 19. Old specimens, the latter a little waterworn in the visible part. This part is turned upwards.

Stat. 303.
Fig. 21 — 22. Specimens rather burdened vvith extraneous objects and therefore rather irregularly deve-

loped. Stat. 240.

Lithothamnion erubescens Fosl. f. americana. Fig. 20.
Fig. 20. A type specimen, seen from above. Cp. text-fig. 15. Fernando Noronha, Brazil.

Lithothamnion erubescens Fosl. f. subflabellata. Fig. 23 — 25.

Fig. 2.3. A specimen forming transition to f. haingsisiana. Stat. 303.

Fig, 24 — 25. Typical specimens of the form ; one of them attached to a coral. Stat. 240.

Siboga-E

PLAÏE IV.

Lithothamnion pulclirinn A. Web. et Fosl.

Fig. i — 4. Young or rather young specimens of the plant. Stat. 315.

Fig. 5. A specimen partially dead, here and there burdened with small corals, or worms disturbing the

tissue, or Bryozoa, the latter in part covered with crustlike formations of the plant. Stat. 315.

Fig. A specimen with the shell of a mollusc between the outermost branches, the shell partly

covered with crustlike formations issuing from the said branches. Stat. 315.

Fig. 7 — 8. Specimens here and there burdened with delicate Bryozoa and other extraneous objects. Stat. 3 1 5.

Fig. 9 — 10. Old specimens of which particularly the first one is much burdened with extraneous objects.
These objects are for a great part covered with crustlike formations of the plant which
develop even new branches. Stat. 315.

Fig. 11. A specimen the lower part of which is dead. From the upper and living branches issue here

and there new layers of tissue, which expand downwards over parts of the dead branches.
In this specimen a small and young specimen of the same species has been brought into
contact with two of the outermost branches. A crustlike formation some mm. broad has
been developed, issuing from both the said branches in a semicircle around one of the
branches of the said young specimen, but only in part confluent with the latter. Stat. 96.

Siboga -Expeditie LXI Wêber.van Bosse ^Foslie Coralliuu

PLATE V.

Archaeolithothamniou erythraeum (Rothpl.) Fosl. f. dura (Heydr.) Fosl.

Fig] i. A. piece of the type specimen from El Tor (Ex coll. Rothpletz).

Fig. 2. A piece of an authentic specimen of Sporolithon ptychoides f. dura Heydr. from El Tor.

Fig. 3 — 4. Fragmentary specimens from Obock in the Gulf of Aden.

Fig. 5. A specimen surrounding some hard object, forming transition to f. mollis. Stat. 193.

Fig. 6. A typical specimen. Stat. 193.

Fig. 7. An older specimen surrounding some hard object, in the part tuining upwards rather worn by

the action of a strong current. The edges of some of the worn branches are in part covered

by new formations of tissue issuing from the undamaged parts of the plant. Stat. 193.
Fig. 8. A specimen with coarse branches, the thickness of the latter is however, at least in part, owing

to their covering up extraneous objects, or to new crustlike formations over the primary

branches. Stat. 99.
Fig. 9. A similar specimen seen from the side. Stat. 99.
Fig. 10. A specimen with less crowded branches, the thickness of the latter at least in part also owing

to their covering up extraneous objects. Stat. 234.
Fig. 11. A specimen almost covering Goniolithon Reinboldi attached to a coral. Parts of the latter species

with conceptacles are visible at the top and to the right on the figure. East-Indian Archipelago.

Stat. unknown.
Fig. 12. A specimen attached to a coral, rather hollowed by worms and much burdened with other

foreign bodies. Stat. 234.

Siboga Expeditie LXI Wobei-.vnii Bosse g„FosIie ('orallinaceae

PLATE VI.

Archaeolithothamnion erythraeum (Rothpl.) Fosl. f. mollis (Heydr.) Fosl.

Fig. i. Part of an authentic specimen of Sporolithon ptychoides f. mollis Heydr. from El Tor.

Fig. 2 — 3. Young and not typical, rather delicate specimens. Stat. 99.

Fig. 4. A young and typical specimen attached to a piece of coral. Stat. 93.

Fig. 5. An older one. Stat. 220.

Fig. 6. A rather young and somewhat stunted specimen burdened with animals and young crusts of other

calcareous algae. Stat. 81.

Fig. 7. A specimen surrounding some small object and rather burdened with animals. Stat. 109.

Fig. 8. A similar one forming transition to f. dura. Stat. 172.

Fig. 9. A typical specimen surrounding some hard object. Stat. 96.

Fig. 10. An old specimen. Stat. 71.

Fig. 11. A similar one rather attacked by animals. Stat. 213.

ttbog^ KxpoditieLXI Weber.vaii Bosse&FosIie Caraffioaoeae

PLATE VIL

Archaeolithothamnion Sibogae A. Web. et Fosl.

Fig. i — 2. Rather young and not quite entire specimens of the plant. Stat. 99.

Fig. 3 — 7,9 — 12. Partly young, partly older specimens, some of which are rather burdened with extraneous
objects, e. g. Bryozoa and sponge-like formations. Stat. 96.

Fig. 8, 13 — 14. Similar ones. Stat. 99.

Fig. 15. An old specimen rather burdened with foreign bodies, particularly Bryozoa. Stat. 96.

I . 16. A specimen partly dead, confluent with a roundish extraneous object and apparently fragments

of another species, and also in part covered with other extraneous objects. New layers of
tissue, here and there issuing from the living parts, expand partially over the dead parts
of the plant and the extraneous objects as new crustlike formations. Stat. 99.

Fig. 17. An old and vigorously developed specimen with thick branches partly owing to covering up

foreign bodies, particularly Bryozoa. Stat. 96.

Siboga Expeditie LXI Webenvan Bosse §.Foslie Coralliua

PLATE VIII.

Archaeolithothamnion timorense Fosl. Fig. i — 14.

Fig. ij 3 — 6. Young and older specimens of the plant. Stat. 315.

Fig. 2. A young one. Stat. 282.

Fig. 7,8, 13. Old specimens partly much burdened with foreign bodies. Stat. 277.

Fig. 9 — 12, 14. Similar ones partly burdened with small corals, Bryozoa and other foreign bodies. Stat. 315.

Archaeolithothamnion Schmidtii Fosl. Fig. 15- — 17.

Fig. 15. A part of the type specimen rather burdened with foreign bodies. Gulf of Siam.

Fig. 16. A specimen rather approaching to A. timorense. Stat. 96.

Fig. 17. An almost typically developed specimen. Stat. 169.

Siboga Expeditie LXI Weber-vtm Bosse §,Foslie Coralluiaoeae

PLATE IX.

Goniolithon Fosliei (Heydr.) Fosl. Fig. i — 5.

Fig. 1. An authcntic specimen from El Tor.

Fig. 2. An older and typical cystocarpic specimen attached to a coral. Stat. 78.

Fig. 3. A loss regularly developed one. Stat. 91.

Fig. 4. An old specimen irregularly developed, with smaller or larger excrescences, at least in part owing

to an irregular substratum, or numerous passages made by worms, around the orifices of wliich

new formations of tissue are by and by heaped up. Zanzibar.
Fig. 5. A typical specimen attached to a piece of coral and partially covering Peyssonnelia sp., a part

of which is visible to the right in the figure. Stat. 91.

Goniolithon myriocarpon Fosl. Fig. 6 — 7.

Fig. 6. Part of the type specimen attached to a shell. Massauah on the Red Sea.
Fig. 7. A sporangia-bearing specimen. Stat. 86.

Goniolithon megalocystum Fosl. Fig. 8 — 9.

Fig. 8. A fragmentary, sporangia-bearing specimen, representing the type of the species. Stat. 129.
Fig. 9. A supposed form of the species loosely covering Bryozoa, Squamariaceae, a young Goniolithon
frutescens and other calcareous algae attached to a piece of coral. Stat. 93.

Goniolithon laccadivicum Fosl. Fig. 10 — 13.

Fig. 10. The type specimen from the Maldives.

Fig. 11. A rather young specimen attached to a coral. Stat. 311.

Fig. i-- An old one. Stat. 301.

Fig. 13. A not quite typical specimen (approaching certain forms of G. frutescens), accompanied by

Lithophyllum Okamurai and Goniolithon Reinboldi, which are partly visible to the left and at the

lowest part of the figure. Stat. 91.

Siboga -Expeditie Lil Weber.van Bosse §,Foslie CoraHinaceae

IX

PLATE X.

Goniolithon Reinboldi A. Web. et Fosl. Fig. i — 6.

Fig. I. A young specimen of the plant. Stat. 282.

Fig. 2. An older specimen. Stat. 193.

Fig. 3 — 5. Old and typically developed specimens. Stat. resp. 282, 71, 91.

Fig. 6. A "specimen approaching Lithophyllum Okamurai in habit. Stat. 91.

Goniolithon frutescens Fosl. f. typica. Fig. 7 — 9.
Fig. 7 — 9. Rather delicate specimens of the form; fig. 9 seen from the side. Stat. 261.

Goniolithon frutescens Fosl. f. congesta. Fig. 10 — 11.
Fig. 10 — n. Specimens of this form not quite typically developed. Stat. 220.

Goniolithon frutescens Fosl. f. subtilis. Fig. 12 — 13.
Fig. 12 — 13. Typical specimens of the form. Stat. 225.

Siboga -Expeditie L XI Webeiwaii Bosse §,Foslie Corallmaeeae

F

ig-

[

F

ig-

i

F

ig-

3

F

'kr-

4

PLATE XI.

Liihophyllum Yendoi Fosl. Fig. i — 4.

The type specimen from the Pacific coast of Japan (Shimoda harbour).
The plant attached to a kind of rather hard wood. Stat. 155.
Typically developed specimen from stat. 193.
Specimens attached to a piece of Mangrove (Rhizophora). Stat. 86.

Lithophyllum oncodes Heydr. Fig. 5 — 10.

Fart of a typically developed specimen from Funafuti, in the upper corner confluent with a

piece of coral.
A fragment of the latter specimen, seen from the side ; the dark points in the surface of

fracture are passages made by worms.
An irregularly developed specimen attached to a conch. Stat. 225.
A similar one forming small cap-shaped crusts on excrescences of a coral, in company with

other algae and sponges. Stat. 91.
A rather well developed specimen attached to a coral. Stat. 181.
An old and partly waterworn specimen surrounding a piece of coral. Stat. 225.

Lithophyllum Okamurai Fosl. Fig. 11 — 19.

The type specimen of the species. Pacific coast of Japan.
The type specimen of f. angulata. Pacific coast of Japan.
The type specimen of f. japonica. Pacific coast of Japan.
Typically developed specimens of the latter form. Stat. 91.
16. A specimen anastomosing with Goniolithon Reinboldi. Some of the branches of the latter species

are marked with a cross in the figure. Stat. 91.
Fig. 17. A specimen partly dead and waterworn, anastomosing with Archaeolithothanmion erythraeuin.

Stat. 93.
Fig. [8, 19. Specimens showing rather delicate branches. Stat. 303.

Fig.

5-

Fig.

6.

Fig.

7-

Fig-

8.

1 ig.

9-

Fig.

10.

1 1.

12.

Fig-

'3-

Fig.

'4,

15

Sibo£a Expeditie LXI Weber-van Bosse §, Foslie Corallinaceae.

PLATE XII.

Lithophyllum Batnleri Heydr. Fig. i.
1. .\ rather wel] developed specimen of the plant. Stat. 282.

Lithophyllum moluccense Fosl. f. typica. Fig. 2 — 4, 6, 8.

Fig. 2. A detached part of a specimen continuing its grovvth freely at the bottom, with eftaced surface

of fracture. Stat. 91.
Fig. 3. A young specimen attached to a coral. Stat. 91.
Fig. 4. A somewhat older one. Stat. 129.

Fig. 6. An old specimen rather burdened with extraneous objects. Stat. 193.
Fig. 8. A specimen attached to a piece of dead coral. Stat. 93.

Lithophyllum moluccense Fosl. f. flabelliformis. Fig. 5, 9, 10.

Fig. 5. A rather young specimen. Stat. 91.
Fig. 9. A detached specimen. Stat. 91.
Fig. 10. An attached one. Stat. 193.

Lithophyllum moluccense Fosl. f. torquescens. Fig. 11.
Fig. 11. The type specimen of the form, probably from the Mauritius.

Lithophxllum moluccense Fosl. f. pygmaea (Heydr.) Fosl. Fig. 7, 12, 13.

Fig. 7. Specimen of the form seen from the side. Stat. 220.

Fig. 12. An old specimen intermingled with other and filiform algae. Stat. 220.

Fig. 13. A similar one attached to a coral. Stat. 220.

Sïbo£a -Expeditie LXI Weber.van Bosse §, Foslie CoraUiiiaceae,

XII

PLATE XIII.

Mastophora macrocarpa Mont.

pjg. !_3. Young specimens partly covering other algae. Stat. 5S.

Fig. 4. A similar one attached to Carpopelüs. Stat. 279.

Pig. -_6. Older specimens. Birakéké, Celebes.

Fig. 7. A similar one. Stat. 129.

Fig. 8. An old one, the greater part of which has apparently grown freely. Stat. 115.

Fig. 9. A specimen attached to different algae. Den Brill near Macassar.

Fig. 10. An old specimen, the greater part of which has apparently grown freely. Stat. 125.

Fig. 11. A similar one. Stat. 149.

Fig. 12. A similar one. Stat. 231.

Siboga -Expeditie LXI Weber_van Bosse ^losüe Corallinaceae .

Xili

PLATE XIV.

Fi^. i — ii. Atnphiroa foliacea Lamx.

Fig. I. Creeping frond.

Fig. 2 — 8. Erect fronds with broad and narrow and even cylindrical joints. The figures i and
4 show the extremes observed in A. foliacea. All these figures are drawn in nat. size.

P'ig. 9. Part of winged frond. 30/,.

Fig. 10. Young top of creeping frond. 30/,.

Fig. 11. Node betvveen the joints. The shaded part indicates the node. 39/,.
Fig. 12. Atnphiroa fragilissima Lamx. f. cyathifera nat. size.

Fig. 13. Two nodes of Amplüroa ephedraea Lamk. 3"/r

Fig. 14 — 17. Amplüroa Bowerbankii Harv.

Fig. 14. Upperside of frond. nat. size.

Fig. 15. Underside of frond. nat. size.

Fig. 16. Section through the node with little dots of calcareous tissue growing out of the
node. 30/,.

Fig. 17. Node seen from above ; one little dot of calcareous tissue is indicated by the darkest
shade. 30/,.
Fig. 18 — 19. Atnphiroa Beauvoissii (Lamx.) Born. (A. exilis Harv.).

Fig. 18. A. specimen from Naples. 30/j.

Fig. 19. A specimen from Durban. 30/,.

Siboga-Expeditie. LX1. Weber -van Bosse & Foslie. Corallineae.

XIV.

JZ'

van Eybergen del.

P W. M.Trap,imp

PLAÏE XV.

Fig. I — 7. Amphiroa crassa Lamx.

Fig. 1, 4. f. typica. nat. size.

Fig. 2. f. Godeffroyi. nat. size.

Fig. 3. f. minuta. nat. size.

Fig. 5. Node between two joints. The cortical layer has split all around the node. 30/,.

Fig. 6. Node entirely surrounded by the calcified cortical layer. 30/,.

Fig. 7. Cells of central strand, showing the small pores in the longitudinal walls. 405/,.
Fig. 8. Node of Arthrocardia cretacea (Post. et Rupr.). *'ƒ,.

Fig. 9. Node and branches of Metagonwlitlwn stelligerum. The branches spring from the node. 30/,.
Fig. 10. Section through node of Metagoniolltlion graniferum. '''/,.
Fig. il. Longitudinal section through apex of branch of Metagoniolltlion charoldes. The small cells indicate

the future node. 95/,.
Fig. 12. Young top of Metagoniolltlion graniferum showing the thick mucilaginous layer covering the top;

in this layer slight striae were visible. 47/,.
Fig. 13. Metagoniolltlion stelligerum. Node with three branches. *'/,.

Fig. 14. Cells of central strand of Metagoniolltlion, showing the big, thin places in the longitudinal walls. 405/r
Fig. 15. Lithothrix aspergillum Gray. Longitudinal section through top of frond. *'/,.
Fig. 16 — 17. Lltharthrou (Amphiroa) australls.

Fig. 16. Joints of young fronds. 3/,.

Fig. 17. Longitudinal section through frond. "5/j-

Siboga-Expeditie, LXI. Weber — van Bos.sk 8j Foslie. Corallineae,

XV,

**S

au] ^-'^

7. —

'

J. van Eybergen del.

F. \V. M. Trap.lmp.

PLATE XVI.

Fig. i, 2, 5. Amphiroa fragilissima Lamx.

Fig. 1. Large specimen from Banda. nat. size.

Fig. 2. Young node which begins to develop itself. 30/,

Fig. 5. Full-grown node.

13;

Ir

Fig. 3, 4. Amphiroa anastomosans n. sp.

Fig. 3. Fronds in nat. size.

Fig. 4. Fronds enlarged 30/,.
Fig. 6 — 8. Amphiroa anceps (Lamk.) Dec.

Fig. 6. Frond nat. size.

Fig. 7. Young node :i"/,.

Fig. 8. Older node quite developed 30/,.
Fig. 9 — 12. Amphiroa canaliculata v. Mart.

Fig. 9. Growing top of frond '°/,.

Fig. 10. Frond upperside. nat. size.

Fig. 11. Frond underside. nat. size.

Fig. 12. Three transversal sections through frond. nat. size.
Fig. 13. Amphiroa rigida f. divaricata. nat. size.

Fig. 14. Longitudinal section through joint and node of Amphiroa verrucosa Kütz. (A. cryptarthrodia Zan.j. '°/,.
Fig. 15. Corallina officinalis Linn. "/,.
Fig. 16. Jania rubens Linn. Longitudinal section through joint and node. The horizontal lines marking the

rows of cells are too straight. They ought to have been more like what one sees in fig. 18. '" ,.
Fig. 17. Part of a section of a fossil Amphiroa spec. (fragilissima?) from Totok Celebes. '5jl.
Fig. 18. Part of a section of a fossil Jania from the mountain Nona at Ambon. "° ,.
Fig. 19. Longitudinal section through a type specimen of Amphiroa fragilissima Lamx. "/,.

Siboga-Expeditie. LXI. Weber — van Bosse & Foslie. Corallineae

f

\77.

\ 7 ;-

Éi

van Eybergen del.

1'. W.M. Trap, imp.

döf^'

RÉSULTATS DES EXPLORATIONS
ZOOLOGIQUES, BOTANIQUES, OCÉANOGRAPHIQUES ET GEOLOGIQUES

ENTREPRISES AUX
INDES NÉEBLAKDAISE3 ORIENTALES CQ 1899 1900,

a bord du SIBOGA

SOUS LE COMMANDEMENT DE

G. F. TYDEMAN

PUBLIÉS PAB

MAX WEBER

Chef de l'expédilimi.

*I.

*n.

«ui.

IV.

V.

«VI.

VII.

VIII.

IX.

X.

«XI.

«XII.

«XIII.

XIV.

XV.

«XVI.

XVII.

XVIII.

XIX.

XX.

XXI.

XXII.

XX III.

XXI Va.

XXI \i.

«XXV.

XXVI.

XXVII.

XXVIII.

XXIX.

XXX.

XXXI.

XXXILï.

XXXII*.

XXXIII.

«XXXI V.

XXXV.

XXXVI.

XXXVII.

XXXVIII.

XXXIX.

XL.

XLI.

XLII.

«XLIII.

«XLIV.

*XLV.

XLVI.

«XL VII.

XLVIII.

XLIX.

L.

LI.

«Lil.

lui.

LIV.

LV.

«LVI.

LVII.

LVIII.

LIX.

«LX.

«LXI.

LXII.

LXIII.

LXIV.

LXV.

LXVI.

Introductiori et deseription de 1'expéditiou, Max Wc bei'.

Le bateau et soa équipement scientifique, R. V. Tyde man

Bésultats hydrographiques, fl. F. Tyde man.

I'oraiuinifera.

Radiolaria, M. Hartinann.

Fbrifera, R. C. J. Vosmaer et J. H. Vernhout ').

Hydropolypi, Oh. Jnlin.

Hydrocorallinae, S. J. Hickson.

Siphonophora, Miles Lens et van Riemsdijk'.

Hydromedusae, O. Maas.

Scyphomedusae, O Maas.

Ctenophora, MUe F. Mos er.

Roi'gonidae, Alcyonidae, J. Versluys ').

Pennatulidae, S. J. Hickson.

Actiuiaria, P. Mc Murrich.

Jladieporaria, A. Alcock1) et L. Dödcrlein.

Aulipatharia, P. N. van Kampen.

Turbellaria, L. von firaff et R. R. von Stummer.

Cestodes, J. W. Spengel.

Nematodes, A. A. W. Hub recht.

Chaetognatha, R. H. Fowler.

Nemei-tini, A. A. W. Hubrecbt.

Myzostomidae, R. R. von Stummer.

Polychaeta errautia, R. Horst.

Polychaeta sedeutaria, M. Caullery et F. Mes uil.

Gephyrea, C. Ph. Sluiter.

Enteropueusta, J. W. Spengel.

Brachiopoda, J. F. van Beminden.

Bryozoa, S. F. H arm er.

Copepoda, A. Scott.

Ostracoda, R. W. Muller.

Cirrhipedia, P. P. C. Hoek.

Isopoda, H. J. Hansen.

Epicaridae, J. Bon nier.

Amphipoda, J. Bonnier.

Caprellidae, P. May er.

Stomatopoda, H. J. Hansen.

('umacea, W. ï. Cal man.

Schizopoda, H. J. Hansen.

Scrgestidae, H. J. Hansen.

Decapoda, J. R. de Man.

Pantopoda, J. C. C. Loman.

Halobatidae, J. Th. Oudemans.

Crinoidea, L. Döderlein p. p.

Echinoidea, J. O. H. de Me ij e re.

Holothurioidea, C. Ph. Sluiter.

Opbiuroidea, R. Kohier1).

Asteroidea, L. Döderlein.

Solenogastres, H. F. Nierstrasz.

Chitonidae, H. F. Nierstrasz.

Prosobranchia, M. M. Schepnian.

Opisthobranchia, R. Bergh.

Heteropoda, J. J. Tesch.

Pteropoda, ,J. J. Tesch.

Lamellibranchiata. P. Pelseneer et Ph. Dautzeuberg.

Scaphopoda, MU« M. Boissevain.

Cephalopoda, L. Joubin.

Tunicata, C. Ph. Sluiter ').

Pisces, Max Weber.

Cetacea, Max Weber.

Liste des algues, Mi"e A. Weber.

Halimeda, M«e E. S. Barton. (M">c E. S. Repp).

Corallinaceae, Mme A. Weber et M. Foslie.

('odiaceae, A. et Mm« E. S. Rcpp.

Dinollagellata. Coccosphaeridae, J. P. Lotsy.

Diatomaceae, J. P. Lotsy.

Deposita marina, O. B. Böggild.

Résultats géologiques, A. Wiek manu.

Siboga-Expeditie

THE CORALLINAGEAE OF THE 10GBIPEDITI0N

'

BV

A. WEBER- van bosse and M. FOSLIE

Eerbeek

'l'i ■ 1 1 )• 1 1 1 j i 1 1 1

With XVI plates and 34 textfigures

Monographie LXI of:

UITKOMSTEN OP ZOOLOGISCH,
BOTANISCH, OCEANOGRAPHISCH EN GEOLOGISCH GEBIED

verzameld in Nederlandsen Oost-Indlë 1899 — 1900

aan boord H. M. Siboga onder commando van
Luitenant ter zee ie ld. G. F. TYDEMAN

UITGEGEVEN DOOK

Dr. MAX WEBER

Prof. in Amsterdam, Leider der Expeditie

(met medewerking van de Maatschappij ter bevordering van het Natuurkundig
onderzoek der Nederlandsche Koloniën)

BOEKHANDEL EN DRUKKERIJ
E. J. BRILL

LEIDEN

jj : 1 1 1 , .. Hiilllm::»;.il"

i:'lil')|iiii . -. :: :.ii:.. - ...»r. '_L

Publié Aoüt 1904

* Les numéros avec un astérique ont déja paru; ceux marqués i) seulemént en partie.

Voor de uitgave van de resultaten der Siboga-Expeditie hebben
bijdragen beschikbaar gesteld:

De Maatschappij ter bevordering van het Natuurkundig Onderzoek der Nederlandsche
Koloniën.

Het Ministerie van Koloniën.

Het Ministerie van Binnenlandsche Zaken.

Het Koninklijk Zoologisch Genootschap > Natura Artis Magistra" te Amsterdam.

De »Oostersche Handel en Reederij" te Amsterdam.

De Heer B. H de Waal Oud-Consul-Generaal der Nederlanden te Kaapstad.

CONDITIONS GÉNÉRALES DE VENTE.

i°. L'ouvrage du „Siboga" se composera d'une série de monographies.

2°. Ces monographies paraitront au fur et a mesure qu'elles seront prêtes.

3°. Le prix de chaque monographie sera différent, mais nous avons adopté comme base générale du prix de
vente: pour une feuille d'impression sans fig. flor. 0. 15; pour une feuille avec fig. flor. 0.20 a 0.25;
pour une planche noire flor. 0.25; pour une planche coloriée flor. 0.40 ; pour une photogravurc flor. 0.60.

4°. I) y aura deux modes de souscription :

a. La souscription a l'ouvrage complet.

b. La souscription a des monographies séparées en nombre rcstreint.
Dans ce dernier cas, le prix des monographies sera majoré de 25 °/0.

50. L'ouvrage sera réuni en volumes avec titres et index. Les souscripteurs a l'ouvrage complet recevront
ces titres et index, au fur et a mesure que chaque volume sera complet.

Déja paru:

ie Livraison. M >hie XLIV) C. Ph. Sluiter. Die Holothurien der Siboga-Expëditioh. Mit i o Tafel n.

ƒ 7-5o
Pour les souscripteurs a Vouvrage complet. „ 6. —

2« Livraison. (Monographie LX) E. S. Barton. The genus Halimeda. With 4 plates .... / 2.40

Pour les souscripteurs ir Vouvrage complet. „ 1.80

5 Livraison. (Monographie I) Max Weber. Introduction et description de l'expédition. Avec Liste des

Stations et 2 ('artes ƒ 9. —

Pour les souscripteurs ir Vouvrage complet. „ 6.75

.| Livraison. (Monographie II) G. F. Tydeman. Description of the ship and appliances used for scientific

. xploration. With 3 plates and illustrations ƒ 2.50

Pour les souscripteurs a Vouvrage complet. „ 2. —

5e Livraison. (Monographie XLVII) H. F. Nierstrasz. The Solenogastres of the Siboga-Expedition. With

six plates ƒ 4.90

Pour les souscripteurs a Vouvrage complet. , 3.90

6e Livraison. (Monographie XIII) J. Versluys. Die Gorgoniden der Siboga-Expedition.

I. Die Chrysogorgiidae. Mit 170 Figuren im Text f 3.75

Pour les souscripteurs ir Vouvrage complet. „ 3. —

7e Livraison. (Monographie XVI a) A. Alcock. Report on the Deep-Sea Madreporaria of the Siboga-

Expeditión. With five plates ƒ 5.75

Pour les souscripteurs ir Vouvrage complet. B 4.60

8« Livraison. (Monographie XXV) C. Ph. Sluiter. Die Sipunculiden und Echiuriden der Siboga-Expedition.

Mit vier Tafeln und drei Figuren im Text ƒ 3.75

Pour les souscripteurs a Vouvrage complet. „ 3. —

9e Livraison. (Monographie Vla) G. C. J. Vosmaer and J. H. Vernhout. The Porifera of the Siboga-

Expedition. I. The genus Placospongia. With five plates f 3. —

Pour les souscripteurs a Vouvrage complet. „ 2.40

I0C Livraison. (Monographie XI) OttO Maas. Die Scyphomedusen der Siboga-Expedition. Mit XII Tafeln.

/9-5Q
Pour les souscripteurs a Vouvrage complet. „ 7.50

lic Livraison. (Monographie XII) Fanny Moser. Die Ctenophoren der Siboga-Expedition. Mit 4 Tafeln.

/ 3-SO
Pour les souscripteurs a Vouvrage complet. v 2.80

12e Livraison. (Monographie XXXIV) P. Mayer. Die Caprellidae der Siboga-Expedition. Mit 10 Tafeln.

f 9-7$

Pour les souscripteurs a Vouvrage complet. „ 7.80

13e Livraison. (Monographie III) G. F. Tydeman. Hydrographic results of the Siboga-Expedition. With

24 charts and plans and 3 charts of depths f 11.25

Pour les souscripteurs a Vouvrage complet. v 9. —

14' Livraison. (Monographie XLIII) J. C. H. de Meijere. Die Echinoidea der Siboga-Expedition. Mit

23 Tafeln ƒ 18.75

Pour les souscripteurs ir Vouvrage complet. „ 15. —

15c Livraison. (Monographie XLVa) René Koehler. Ophiures de 1'Expédition du Siboga.

ie Partie. Ophiures de Mer profonde. Avec 36 Planches f 20.50

Pour les souscripteurs ir Vouvrage complet. „ 16.50

ifr Livraison. (Monographie Lil) J. J. Tesch. The Thecosomata and Gymnosomata of the Siboga-
Expedition. With six plates f 4.70

Pour les souscripteurs ir Vouvrage complet. „ 3.75

Livraison. (Monographie iNla) C. Ph. Sluiter. Die Tunicaten der Siboga-Expedition.

I. Abteilung. Die socialen und holosomen Ascidien. Mit 15 Tafeln / 9. —

Pour les souscripteurs ir Vouvrage complet. „ 6.75

1

1

■lil

a

il