1 nj

, o

: m

a

REPORT

ON THE

SCIENTIFIC RESULTS

OK THE

VOYAdE OF S.Y. "SCOTIA'

SCOTTISH XATIOXAL AXTAlUTir EXPIIDITIOX.

E E P O R T

ON" THK

SCIENTIFIC RESULTS

OF THE

VOYAGE OF S.Y. "SCOTIA"

DURING THE YEARS 1902, 1903, AND 1904,

I'NDER THE LEADERSHIP OF

WILLIAM 8. BRUCE,

LL.D., F.K.8.K.

Volume F7/.— ZOOLOG Y.

PARTS I.-XIIL— INVERTEBRATES, by M. le Professeur EMILE TOPSKNT ; J. II.
KOEPPERN; HELEN L. M. PIXELL, B.Sc., F.Z.S. ; F. F. LAIDLAW, lAI.A., F.Z.S. ;
Professor J. STANLEY GARDINER, JV1.A., F.R.S. ; F. GORDON PEAIICKV :
WALTER AV. TATTERSALL, D.Sc. ; L. N. G. RAMSAY, M.A., B.Sc. ; A. PRINCI.I:
JAMESON, B.Sc.; Professor OSKAR CARLGREN; T. J. EVANS. .M.A. (
Professor J. H. ASHWORTH, D.Sc., F. K.S.

Witli Fifteen I'latrs «n-l Twenty-three 7V.<7-/- '<(/// res.

EDINBURGH :

Ci)t grotttsii) dDctanograpintal Hafooratorp.

SOLD AT

THE SCOTTISH OCEANOGRA I'll H 'A I. LABORATORY;

.TAMES Til IX, n:, SOUTH BRIDGE, EDINBURGH:
.IAMES MAi'iJ-iKisi-: >v SONS, i;i ST VIXCKNT STREET, GLASGOW.

1920.

" \
I'ri'-i' Fifty Xli/'/liiujs.

EDITORIAL NOTE.

THE present volume constitutes the fifth series of papers on the Zoological Results of
the Scottish National Antarctic Expedition, and is the third volume dealing with the
Invertebrate life of Antarctic and adjacent seas. It contains thirteen papers ranging
through most of the great groups of invertebrate animals, from Protozoa to Crustacea
and Mollusca, and to its formation have gone the researches of distinguished scientists,
not only in Scotland and England, but in France and Sweden as well. To these con-
tributors the thanks of the Editor and of the scientific world are due, the more so since
the world at large has not yet recognised that the labourer in scientific fields is worthy
of his hire, and since these contributions one and all are the fruit of labours of love.

The collections dealt with in the present papers have not only furnished many
species new to science, but have afforded material from which considerable additions
have been made to our knowledge of the structure and distribution of previously
recorded forms, and upon which useful revisions of related forms have been based.

It is with regret that I have to record the death in action at Neuve-Cbapelle, 011
21st March 1915, of one of the contributors to this volume, Lieut. L. N. G. Ramsay,
whose papers on Polychrets in the present and on ornithology in a previous volume
gave promise of a bright career as a zoologist. To the war also must be attributed the
delay in the issue of this volume of the Scientific Results of the Scotia, for the
difficulties in the way of publication have been great, and have even now been circum-
vented only through the generosity of Sir Thomas Glen-Coats. To him, as to the
Carnegie Trust for the Universities of Scotland, which assisted by contributing to the
cost of the production of Plates and of Title-pages, the Editor offers his sincere thanks.
Similar recognition is due also to the Councils of the Royal Society of Edinburgh and
the Royal Physical Society for permission to make use of contributions to their
publications, for with a single exception the papers in this volume have appeared
in their Transactions and Proceedings ; to the former in particular his gratitude is
due, in that it bore the charges of the primary printing of the majority of the papers.
To obviate all possibility of confusion in cases of reference, the original pagination
of these papers is given at the foot of every page, and the date of first publication
is stated in every case. Through an oversight in paging, pages 233 to 236, between
two independent Parts of this volume, are non-existent. My cordial thanks are due
to Dr R. N. Rudmose Brown and Dr J. Ritchie for their valuable assistance in the
final arrangement of the volume.

WILLIAM S. ISRUCU,

Editor.

CONTENTS.

PART I. — SPONGIAIRES DE L'EXPEDITION ANTARCTIQUE NATIONALS ECOSSAISE. Par EMILK
TOPSENT, Professeur u 1'Universite de Dijon. Pn'senU par le Dr W. S. BRUCE. (Aveo
six planches) .......... 3-72

(Read December 10, 1012. Revised MS. received December 24, 1912. Issued separately August 20, 1013.)
(See also Supplementary Report, Part XIII., p. 315.)

PART II. — SlPHONOPHORA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. T>y J. H.

KOEPPERN, Zoological Department, University of Edinburgh. (With Three Text-Figures) 73-82
(Received December 13, 1912. Read December 10, 1!)12. Issued separately April 10, 1913.)

PART III. POLYCH/ETA OF THE FAMILIES SERPULIDJi AND SADELLID.E, COLLECTED BY TIID

SCOTTISH NATIONAL ANTARCTIC EXPEDITION. By HELEN L. M. PIXELL, B.Sc., F.Z.S.,
Demonstrator of Zoology and Reid Fellow, Bedford College, University of London.
Communicated 6y Dr J. H. ASHWORTH. (With One Plate) .... 83-98

(MS. received January 22, 1913. Read February 17, 1913. Issued separately June 24, 1913.)

PART IV. TURBELLARIA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. By F. F.

LAIDLAW, M.A., F.Z.S. (Second Paper) .... . 99-104

(Proof returned September 8, 1913. Issued November 9, 1914.)

PART V. — THE CORALS OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. By J. STANLEY
GARDINER, M.A., F.R.S., Professor of Zoology and Comparative Anatomy in the University
of Cambridge. Communicated by Dr J. H. ASHWORTH. (With Two Text-Figures) . 105-112

(MS. received May 14, 1913. Read July 7, 1913. Issued separately September 1, 1013.)

PART VI. — FORAMINIFEKA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. By F. GORDON

PKARCEY, Bristol Museum, late of the Challenger Expedition and Commission. Com
iitunicated by Dr J. H. HARVEY PIRIE. (With Two Plates) . . 113-170

(Read May 26, 1913. MS. received August 2, 1013. Issued separately March 30, 1014. )

PART VII. — THE SCHIZOPODA, STOMATOPODA, AND NON- ANTARCTIC ISOPODA OF THE SCOTTISH
NATIONAL ANTARCTIC EXPEDITION. By WALTER M. TATTERSALL, D.Se., Keeper of the
Manchester Museum. Communicated by Dr J. H. ASHWORTU. (With One Plate) . 171-204

(MS. received June 5, 1913. Read July 7, 1913. Issued separately November 18, 1913.)

PART VIII. — PoLYCH^ETA OF THE FAMILY NEKEIKK, COLLECTED BY THE SCOTTISH NATIONAL

ANTARCTIC EXPEDITION". By L. N. G. RAMSAY, M.A., B.Sc., Carnegie Researcli Scholar,
Christ's College, Cambridge. Commit niea/t<l % Dr J. H. ASHWORTH. (With One Plate) 205-216
(MS. received October 7. 1013. Read December 15, 1913. Issued separately Mari-li ."n, 101 I. i

.-.••

v/ ^OS *j_ < ^N

31955

PART I.
S 1> () X (I K S.

I.-THE SPONGES OF THE SCOTTISH NATIONAL
ANTARCTIC EXPEDITION.

BV

EMILE TOPSENT;

I'rofessor of Zoology in the University of Dijon.

(WITH SIX PLATES.)

Spongiaires de 1'Expedition Antarctique Nationals Ecossaise. Par Emile Topsent,
Professeur a 1'Uui.versite de Dijon. I'rf'xnitt'- j>ai- le Dr. W. S. BRUCE. (Avec
six planches.)

(Read December 16, 1912. Revised MS. received December 24, 1912. Issued separately Aiujii.-t 20, 1913.)

Au retour de sa belle campagne de 190:2 a 1904 dans 1'Antarctiijue. M. le Dr.
WILLIAM S. BRUCE m'a fait 1'honneur de m'ofi'rir IVtude des Spongiaires recueillis a
bord de la Scotia. Voici de quoi se compose cette collection qui, a certains egards, offre
un interet considerable.

HEXACTINELLIDA.
I. S.O. HEXASTEROPHORA.

Famille EUPLECTELLID.E.

Malacosaccus pedunculatus, Topsent.

,, coatsi, Topsent.

Accelocalyx brxcci, Topsent.
Docosaccus ancoratus, Topsent.

Famille CAULOPHACID.'E.

Caulophacus i>ix(«l>ilix, Topsent.
„ scotise, Topsent.

Famille ROSSELLID.K.

Batkydorus levis, F. E. Schulxc, var. ciliatus, Topsent.
Calycosoma r<t/i<lit.in, F. E. Schulze.

Famille CO.SCINOPORID.K.
Cho)id'(Hin.t.t sp.

II. S.O. AMPIIIDLSroi'HUlfA.
Famille HYALOXKM, \TID.K.

sp.

(REPRINTED FROM THE TRANSACTIONS OF THE HOY.U, SOCIKIY oi KI>INBUKOH, VOL. XLIX., rr. 579-643.)

EMILE TOPSENT:

TETRACTINELLIDA.

S.O. CHORISTIDA.
Tribu Astrophora.

Famille ASTEROSTREPTID^E.

Pachastrella monilifera, 0. Schmidt.
PoBcillastra incrustans, Sollas.

,, compressa, (Bow.) Soil., var. parvistellata, n. var.

MONAXONIDA.

1. S.O. HADROMERINA.

Famille SUBERITID^E.
Pseudosuberites exalbicans, n. sp.

II. S.O. HALICHONDRINA.

Famille AXINELLID^:.

Hymeniacidonfernandezi, Thiele.
Bubdris murrayi, n. sp.

Famille PCECILOSCLERID^;.

Sous-famille Ectyoninse.

RaspaxiUa phakellina, n. g., n. sp.
Dictyociona discreta, (Thiele), n. g.
Clathria toxipriedita, n. sp.
Stylostichon toxiferum, \\. sp.

,, nobile, var. patayonicum, (Ridley et Dendy)

'I'opsent.
Dendoryx nodaspera, n. sp.

Sous-famille Myxillinse.

Mijj'illn xfHnujiosa, Rdl. et D., var. asigmata, Topsent.
Lissodendoryx buchanani, n. sp.
loplwn pluriconn's, Topsent.

,, unicornis, Topsent.

(ROY. soc. KDIN. TRANS., VOL xux., 580.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATIONALE ECOSSAISE. 9

IOJI/HHI sjHitttlattix, Kirkpatrick.

,, flabeUo-digitatus, Kirkpatrick.
Tedania murdochi, n. sp.
„ charcoti, Topsent.

Sous-famille Mycalime.
M/fc<i/i' magellanica, (Ridley) Topsent.
,, acerata, Kirkpatrick.
,, pellita, n. sp.
Cladorhiza tfiomsoni, Topsent.
Homoeodictya verrucosa, n. sp.
,, setifera, Topsent.

Desmacidon ? sp.

Famille HAPLOSCLERID^E.

Gellius arcuarius, n. sp.
Toxochalina robitsta, Rdl. et D.
Petrosia depellens, u. sp.
Reniera cylindnca, n. sp.
„ dancoi, Topsent.
sp.

Une autre liste des especes, dressee par stations, aura 1'avantage de faire connaitre a
la fois leur distribution geographique et bathymetrique et leurs associations.

STATION 301.

13 mars 1903, 64° 48' lat. S., 44° 20' long. W.; 2485 brasses.

Docosaccus ancoratus.
Accelocalyx brucei.

STATION 313.

18 mars 1903, 62° 10' lat. S., 41° 20' long. W.; 1775 brasses.

Caulophacus instabilis.
Calycosoma validum.

STATION 325.
Avril-aout 1903, Scotia Bay, Orcadcs du Sud, GO0 43' 42" lat. S., 44° 38' 33" long. W. ;

9-10 brasses.

lophon unicornis.

,, pluricornis.
„ spatulatus.

„ jiabello-digitatus.

(ROY. soc. EUI.N. TUANS., voi,. xi.ix., 581.)

1 0 EMILE TOPSENT :

Mycale acerata.
Homveodictya setifera.
Gellius arcuarius.
Petrosia depellens.
Reniera cyhndnca.

dancoi.

sp.

STATION 346.
ler decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W. ; 56 brasses.

Rossellides roulees.
Raspaxilla pli akellina.
Clathria toxipr&dita.
Stylosticlion nobile patagonicum.
Dendoryx nodaspera.
Lissodendoryx buchanani.
Tedania charcoti,
Mycale magellanica.

,, pellita.

Homceodictya verrucosa.
Desmacidon ? sp.
Toxochalin a robusta.

STATION 118.
ler fevrier 1904, Port Stanley, lies Falklands ; fi brasses.

Tedania murdochi.
Hymeniacidon fernandezi.

STATION 417.
18 mars 1904, 71° 22' lat. S..16° 34' long. W.; 1410 brasses.

Malacosaccus pedunculatus.

„ coatsi.

Caidopli acus scotite.
Chonelasma sp.
Hyalonema sp.
Myxilla spongiosa asigmata.

STATION 420.

21 mars 1904, 69° 33' lat. S., 15° 19' long. W.; 2620 brasses.

Bathydorus levis ciliatus.

(HOY. soc. EDIN. TRANS., VOL. XLIX., 58^.)

SPONGTAIRES 1)E L'EXPEDITK )N ANTAU* TK.XM-: NATK>NALK ECOSSAISE. II

STATION 461.

22 avril 1904, lie Cough, 40° -20' hit, S., 9° 56' 30" long-. W. : 100 brasses.
Pachastrella monilifera.
Pcecillastm incrnstans.

„ comprexsa parvistellata.

Pseudosuberites exalbicans.
Biibaris murrayi.
Dictyodona discrete.
N/ ///ostichon toxiferum.

STATION 4G8.

29 avril 1904, 39° 48' lat. S., 2° 33' long. VV. ; 2770 brasses.
Cladorhiza thomson i.

HEXACTINELLIDA.

La recolte de la Scotia en Hexactinellides a <>ti'' particulieremenl riche. Ces
Eponges abondent dans 1'Antarctique et cela par des profondeurs meme mediocivs.
Cette constatation est certaiuement, en matiere de spongologie, le n'-sultat le plus
interessant de 1'exploration zoologique de cet ocean, a laquelle plusieurs nations
collaborent depuis line douzaine d'anne'es. Je 1'ai di-duite, des 1901 (26), de IVtndr
des Spongiaires recueillis par la Belgica, 1'opposant, en reponse a I'liypotlirse dc la
bipolarite des faunes, a la remarquable peuurie des mers arctiques en Hexactinellides.

II y a surtout des Rossellidse. Toutes les Hexactinellides rapporti'-es par la
Discovery et les deux tiers de celles obtenues par le 'A///** appartiennenl a c-.o.tte
faraille, qui avait rleja found la part la plus importante de la collection de la l><*/t/ir<i.
Le /''ni iii-ii/.^ n'a ricn troiivi'- qui merite d'etre cite, inais le /'<>itr</it<>/ !'«*'.', -i en juger
par un examen superficiel des materiaux qui m'out ete confies, a ivuni de Rossellides
une nouvelle et copieuse collection.

En revanche, il est de grosses families d'Hexactinellides dont les expeditions
antarctiques ne nous ont presque rien appris : les EuplectelUdx, les Caulophaddse et
les Hyalon&rnatidae.

Cela tient a ce que ces Eponges font en general partie de la faune abyssale et que la
plupart des campagues scientifiques ont ete dirigees dans des portions dc L'Antarctique
aux eaux pen profondes. La Belgica a opeiv le long des terres de (Jrahain et d'Alex-
andre Ier, par des profondeurs inferieures a 600 m. Le Fn<i\<-nis et le J'n/tn/noi Pn*.'
ont successivemeut explore ces memes pa rages. La Discovery, le long de la Terre
Victoria, a pris presque toutes ses Hexactinellides par moius de 500 m. (plusieurs
meme par inoins de 100 m.), a 1'exception d'une seule — une Rossellide quand mrnu-
recueillie par 914 in. de profondeur an large des Mts. Erebus et Terror. Le Gauss,
enfin. dans le nonl-west de sa station (('<>'< -' ;i" lat. S., 89° 38' long. E.). rencontra des

(KOY. sou. KUIN. THA.XS., VOL. XLIX., 583.)

^ /

12 EMILE TOPSENT :

profondeurs de 2450 m. a 3397 m. qui lui fournirent un Caulophacus, C. antarcticus
et un Hyalonema, H. drygalskii.

De toutes les expeditions, c'est celle de la Scotia qui, fouillant les eaux profondes
entre les Oreades du Sud et la Terre de Coats, a 1'entree de la Mer de Weddell, a le
plus contribue a faire connaitre les Hexactinellides antarctiques de families abyssales.

La Scotia a, comme le Gauss, outre quelques Rossellides de grands fonds, des
genres Batliydorus et Calycosoma, obtenu un Hyalonema. Elle a aussi dragu6 deux
Caulophacus nouveaux, dont 1'uu, C. scotiie Topsent, de taille geante, atteint pres d'un
metre de hauteur. Mais mieux, elle fut la seule a recueillir des Euplectellidse.

Ce qu'elle en prit offre un inte'ret d'autant plus grand qu'il s'agit d' ' Euplectellinte et
que, des quatre espeees qui figurent dans sa collection, deux se rattachent au genre
Malacosaccus, fort rare et jusqu'a present mal connu, tandis que les deux autres
repre"sentent chacune un genre nouveau, ce qui eleve de 3 a 5 le nombre des genres de
cette sous-famille.

Je n'ai inscrit sur la liste gen^rale des Hexactinellides de la Scotia (p. 7) que celles
dont il m'a etc" possible de reconnaitre au moins le genre : un Hyalonema et un
Chonelasma n'etaient pas determinates comme espeees et ne pouvaient servir de types
a des espeees nouvelles. J'ajouterai que, sur'le Bane de Burdwood, par la tres faible
profondeur de 56 brasses, il a ete pris plusieurs fragments d'Hexactinellides dont je
n'ai meme pas pu soupconner le genre ; ce sont des masses feutrees, compactes, sans
orifices ni canaux visibles, sans structure defiuie, composees surtout de diactines mais
avec un melange d'hexactines et de pentactines inegales et de beaucoup d'hexasters
dont les rayons se montrent quelquefois bifides, quand leur taille est tres faible. J'ai
photographic" deux de ces masses (PI. III. fig. 10). Elles m'ont rappele dans une certaine
mesure les masses feutrees de spicules de Rossellides que le Frangais a recueillies en
abondance (28, p. 11) dans un dragage aupres de 1'ile Anvers, a peu pres par la meme
profondeur ; pourtant, les spicules sout ici ordinairement en bon etat et je crois qu'il
s'agit cette fois d'Eponges roulees plutot que d'amas de spicules fagonnes par des
courants.

I. S.O. HEXASTEROPHORA.

Famille EUPLEOTELLID^E.

Malacosaccus pedunculatus, Topsent.
1910. Malacosaccus pedunculatus, Topsent (3O, p. 1).

Station 417, 18 mars 1904; 71° 22' lat. 8., 16° 34' long. W. ; profondeur, 1410
brasses.

Le type de cette espece est de tous les Malacosaccus recueillis jusqu'a present celui
dont la forme peut recevoir la description la plus complete. En raison de sa mollesse,
il n'a naturellement pas pu subir sans de graves dommages les brutalites d'un engin qui
1'a t rai IN'' sur la vase a.vant tie le ramener d'une profondeur aussi considerable ; il est

(HOY. SOC. EDIN. TRANS., VOL. XLIX., 584. )

SPONGIAIRES DE L'EXPKDTTFON ANTARCTIQUE NATIONALE ECOSSAISE. 13

arrive a la surface divise en deux morceaux plus ou moins deteriores, mais ses debris
suffisent a imaginer de 1' ensemble une restauration qui ajoute a la counaissance de la
forme des Malacosaccus : le rebord de la coupe peut avoir uu bourrelet marginal et le
corps est porte sur une colonne de triactines.

L'un des fragments se presente comme une grande plaque spongieuse, tres molle,
haute de 14 centimetres, large de 18 centimetres a uu bout et de 12 centimetres a
1'autre, epaisse de 4 a 7 millimetres ; un bourrelet, <|ui limitc son cote le plus etendn.
contraste avec les autres bords, eftiloches, et marque de toute evidence le bord superieur
de 1'Eponge. Les deux faces de la plaque s'orientent aiscment grace a la difference
de leurs orifices. En pliant la plaque la face cloacale en dedans, le bourrelet marginal
en haut, de maniere a placer ses bords lateraux en regard 1'iin de 1'autre, on lui redonne
presque ce qui semble avoir e"te sa forme naturelle, celle d'un calice, qu'un accident a
fendu et ouvert de haut en bas. Uu tel calice mesure, a cet etat, 14 centimetres
de hauteur, 9 centimetres de largeur en haut et 6 centimetres environ vers le
bas. A supposer que sa section ait ete circulaire dans des conditions normales,
son diametre exterieur, a Ten tree, devait etre a peu pres de 6 centimetres ; mais sa
mollesse est telle que, dans 1'alcool, ses moities repliees s'aplatissent 1'une sur 1'autre.
Les parois de ce long calice ont leur maximum d'epaisseur vers sa base, puis vont
doucement s'amincissant jusqu'au niveau du bourrelet marginal. Celui-ci est large de
10 a 12 mm. Dans les conditions oil je 1'observe, il apparait comme une bande
imperforee, densement veloutee, nettement limitee, et tout entiere situee en bordure
de la face cloacale ; mais on peut supposer que les bords du calice se tenaient, durant la
vie, quelque peu erases et ramenaient le bourrelet dans une position a peu pres
horizontale. Notons de suite qu'il ne porte pas le moindre vestige d'une frange ciliee.
La face externe du calice est percee de nombreux orifices, plus ou moins circulates,
pour la plupart petits (1 mm. a 1 '5 mm.) mais atteignant souvent 3 mm. et plus de
diametre. Us ont du etre tous tendus d'un reseau d'hexactines en glaive, qui, par
places, les recouvre encore. Ces hexactiues dermiques forment aussi un revetement aux
portions charnues comprises entre les orifices, et leur actine distale, bien developpee,
donne a toute la surface du corps une legere hispidation, perceptible a Frail nu. A la
face cloacale sont beants, comme chex Malacosaccus vastus, de grands orifices composes,
larges de 8 a 12 mm., dont les intervalles se parsement en outre de perforations
beaucoup plus etroites. Par en bas, une dechirure profondc entame la paroi. C'est
de la, selon toute probabilite, que le second fragment s'est detache, mais non sans une
certaine perte de substance, car il est actuellement impossible de remettre en rapport
d'une facon satisfaisante les parties disjointes.

Autant le premier fragment, qui represente le corps de 1'Eponge, ressemblait a
1'unique morceau de Malacosaccus vastus obtenu par le (Jlmllcinjcr (12, pi. xviii. tig. 1),
autant le second, qui en figure 1'appareil de fixation, rappelle un specimen de M.
floricomatus recueilli par la Princesse- Alice (27, pi. iii. fig. 10). C'est, de meme, une
colonne tres molle, relativement longue (10 cm.) pour son epaisseur (2 cm.) et

(ROY. soc. EDIN. TUANS., VOL. XLIX., 585.)

I 4 KMILE TOPSENT :

terminee par de belles soies fasciculees. La colonne est pleine ; sa surface, diverse-
meat plissee, a beaucoup souti'ert et ne montre d'orifices pareils a ceux du M.
jloricomatus en question que tout-a-fait en bas. a la naissance des soies. Celles-ci se
disposent en deux touffes bien distinctes, paralleles mais fort inegales, 1'une tres fournie,
(•[laisse de 3 mm. et longue de 95 mm., 1'autre plus maigre, n'atteignant guere que le
tiers de cette longueur.

Comme les soies ont elles-iuemes beaucoup de souplesse, on se demande comment
une Eponge aussi molle et d'aussi grandes dimensions peut s'etre tenue debout sur un
systems aussi peu resistant. Dependant, la spiculation etant ideutique, les deux
fragments appartiennent certainemerit a une meme espece du genre Malacosaccus ; or,
considerer ce corps caliciforme et cette base h'xatrice comme provenant de deux
specimens differents recueillis en meme temps, serait moins logique que les tenir pour
des parties d'un seul et meme individu et ne modifierait pas la conception qu'ils nous
suggerent de la forme generale de M. pedunculatus.

La charpeute du parenchyme se compose typiquemeut d'hexactines souples a actines
lisses, tres legerement renfiees au bout, longues de 1 mm. a 1'8 mm., epaisses seulemeut
de 0'013 mm. a la base et de 0'007 mm. vers 1'extremite ; mais, au voisinage des deux
faces comme au pourtour des canaux qui traversent le corps, elles se nioditient pour
ne determiner aucune hispidation et, perdant uue de leurs actines ou plusieurs, se
reduisent en pentactines, en triactines, plus rarement en tetractines, jamais en
diactines.

Le reseau dermique est fait, comme d'habitude, d'hexactines en glaive ; leur
actiue distale, longue de 0'3 mm., en moyenne, se renne, devient fusiforme, epaisse
de O'OOS mm. a 0'012 mm., et se couvre d'epines appliquees, relevdes vers la pointe
libre ; les actines tangentielles. faiblement epineuses, sont un peu plus longues que
la distale (0'38 a 0'4 mm.) et de moindre epaisseur qu'elle ; quant a la proximale,
lachement epineuse et souvent flexueuse, c'est de beaucoup la plus grele et la
plus lougue, car elle varie entre 075 mm. et 1 mm. Au niveau du bourrelet
marginal, ces spicules se serrent davantage, leurs actines tangentielles prenant
alors la meme longueur que la distale ; celle-ci est presque toujours surrnontee d'un
fioricorne. C'est a 1'ensemble de ces pointes plus nombreuses et ornementees, ainsi
qu'a 1'absence d'orifices entre elles, que le bourrelet marginal, en relief sur les parties
avoisinantes, doit son aspect particulier. A la base des touffes fixatrices, les hexaetines
en glaive ont leur actine distale tres developpee.

A la face cloacale aussi se trouvent des hexaetines en glaive, mais par places seule-
ment ; les megascleres de revetement de cette face sont pour la plupart des pentactines,
a actine distale rtkluite a 1'etat de tubercule court, a actines tangentielles longues de
0'22 mm. a 0'38 mm. et a proximale de 0'4 mm. Les hexaetines, la ou elles se
rencontreut, m'ont paru avoir une actine distale plus courte (0'23 mm.) que 1'actine
correspondante des hexaetines dermiques.

Malacosaccus pedunculatus ne possede pas d'hexactines hypodermiques.

(ROY. soc. EDIN. TRANS., VUL. xi.ix., r>sii.)

SPONGIAIRES DE L'EXPEDITION ANTARCTIQUE NATIONAL!-: ECOSSAISE. 15

Les microscleres sont : 1. des //"/vVo^r.s-, a rayons primaires courts, portant 8 ou 9
rayons secondaires tres arqurs, tivs Civics a la base, puis renlles et termini's par line
palette de O'Ol mm., qu'ornent G a 8 dents marginales. Us abondent sur le bourrclet
marginal et y atteignent 0'145 mm. de diami-tre. Ils demeurent un pen plus petits
(O'l 2 mm.) sur le reste de la surface de 1'Eponge. Us no sont pas rares, avec
O'l mm. a O'l 2 mm. de diametre, dans le parenchyme. Enfin, je les observe, avec un
diamHre de O'l mm., en tn>p grand nombrc a la face cloacale pour ne pas admcttre
qu'ils s'y trouvent en place.

2. Des discohexasters, a rayons primaires courts, portant cliacun quelquefois trois
mais le plus souvent deux rayons secondaires presque droits, raboteux, un pen rpaissis
(ils passent de 0'002 mm. a 0'0035 mm.) au-dessous de leur disque terminal, lequel est
large de O'OOG mm. a O'Ol 1 mm. et porte de petites dents a son pourtour. Le diametre
de ces discohexasters varie eutre O'l mm. et O'l 7 mm. Sans diminuer sensiblcment
de taille, elles ont parfois leurs rayons secondaires tres givles, lisses, un pen renll<js
quand meme au bout, au-dessous du disque, qui est alors etroit mais s'orne en revanche
de dents plus longues. II existe de ces spicules un peu partout, dans le parenchyme,
aupres de la surface et a la base des toutl'es de soies ; cependant ils s'y montrent
toujours clairsemes.

3. Des oxyhexasters, a rayons secondaires par deux et pointus. Je n'en ai pu
decouvrir que trois ou quatre, de sorte que je n'ose affirmer qu'elles appartiennent
reellement a notre Eponge.

Les touti'es fixatrices se composent, comme eh ex Malacosaccus floricomatus, de
briactines lisses derivees d'hexactines ; deux des actines qui leur restent, s'orientaut
suivant le grand axe de la touft'e, acquierent une grande longueur, taudis que la troisieme,
qui leur est perpendiculaire, demeure beaucoup plus courte ; c'est par cette dernirre
seulement «|ue les soies se tiennent entre elles. car il ne se d6veloppe nulle part de
synapticules. Ces triactines sont de taille tivs int'-gale, tous les intermediaires existant
entre celles qui, donnant aux touffes leur aspect fibrillaire, arrivent a depasser 6
centimetres de longueur sur O'l 7 mm. d'epaisseur et celles, en quantite bien
superieure, qui restent fines et relativement courtes et forment un remplissage entre
les precedentes. En outre, de leurs deux actines axiales, 1'uue prend fre'quemment
un plus grand developpement que 1'autre, de sorte que 1'actine transversale ne
se detache pas du spicule en son milieu mais plus ou moins pivs de 1'une de
ses extremities.

Des prrlrvements eil'ectiu's en di\-ers points de 1'appareil lixateur ne m'ont [ias
fourni une seule ancre.

En re.sunii'', Malacosaccus pedunculatus oft're dans sa spiculation des particularity
i[iii le caracterisent bieu en tant qu'esprce. II est le premier dont le revrtcn.rnt
cloacal comprenne des pentactines. Ses Horicomes se rapprocheni par leur forme de
ceux de M. rastits, mais ses discohexasters sont plus simples que partout ailleurs. Les
onychasters lui font defaut et peut-etre meme les oxyhexasters.

(ROT. soc. EDIX. TRANS., VOL. XLIX., 587.)

16 EMILE TOPSENT :

Malacosaccus coatsi, Topsent. (PI. II. fig. 4 et 5, et PI. VI. fig. 1.)
1910. Malacosaecus coafsi, Topsent (30, p. 3).

Station 417, 18 mars 1904; 71° 22' lat. S., 16' 34' long. \V. ; profondeur, 1410
brasses.

Les notions que le type de Malacosaccus pedunculatus nous a fournies sur la
morphologic cles Eponges du genre Malacosaccus se trouvent en quelque sorte
completees par celui de M. coatsi ; eelui-la, malgre une dechirure longitudinale, n'a
rien perdu de sa partie superieure ; celui-ci possede au complet son appareil de fixation.
Grace a eux, les Malacosaccus se revelent corame des Euplectellinse en forme de coupe
molle, portee par un pedoncule qu'une touffe d'ancres termine et fixe dans la vase. Le
plus beau des pedoncules recueillis de M. floricomatus aurait pu faire deviner un tel
mode d'attache, mais tout demontre aujourd'hui que je lui ai d'abord prete une orienta-
tion inverse de celle qui lui est naturelle (27, pi. i. fig. 1). Fort bien caracterise,
d'ailleurs, par ses beaux floricomes a rayons termiuaux en groupes de quatre, par ses
petites onychasters peripheriques et par ses hexactiues hypodermiques speciales,
M. floricomatus est, de tons, celui dout on connait le moins bien la forme. La
Princesse- Alice n'eii a obtenu que trois fragments charnus, dont deux, prolonges en
un commencement de pedoncule, representent manifestement des bases de specimens,
correspondent au fragment inferieur de M. pedunculatus et sont, comme lui, depourvus
de cavite axiale ; le troisieme est plus aplati, mais je n'ai pas cru devoir le considerer
comme un lambeau de la portion superieure d'un grand specimen tubuleux, pour ces
motifs que ses bords tournent et que les spicules se repetent sur ses deux faces.

II est bien difficile de se faire une opinion ferme sur de tels fragments et la mienne,
d'apres laquelle M. floricomatus manquerait de cavite cloacale et constituerait ainsi une
veritable exception, a besom d'etre corroboree. Je suis revenu de la surprise que m'a
causee 1'existence par tout son parenchyme de floricomes nombreux, alors que, chez les
seules especes avec lesquelles la comparaison etait alors possible (M. vastus F. E. Sch.
et M. unguiculatus F. E. Sch.), ces microscleres avaient ete notes comme exclusivement
cantonn6s a la pointe des hexactines dermiques : d'autres exeinples tres nets de leur
repartition par toute 1'epaisseur du corps nous sont, en effet, offerts maintenant par les
deux Malacosaccus de la Scotia. Ce qui paraissait etrange a ce sujet sernble desormais
naturel. Mais je me rends compte que la decoupure en gros lobes des parois de
M. coatsi pent conduire a se demander si M. floricomatus a reellernent le corps plein
ainsi que ce que j'en ai vu me 1'a fait admettre.

II est regrettable que ce Malacosaccus coatsi ait lui-meme sa partie spougieuse si
gravement endommagee. Cupuliforme dans son ensemble, celle-ci contraste avec la
partie homologue des M. vastus, M. unguiculatus et M. pedunculatus par une allure
fort irivguliere. Ses parois, des la base, se soulevent en de grosses tuberosites qui lui
donnent un aspect lobe. Entre les tuberosites s'etendent des plis profouds et meme,
sur un cote (PI. II. fig. 5, a gauche), une haute echancrure par laquelle la cavite cloacale

(ROY. soc. F.DIN. TRANS., VOL. XLIX., 588.)

SPONGIA1KES DE [/EXPEDITION ANTA lUTTgr K NATTONALK ECOSSAlsK. 17

communique largement avec Textrriciir. Kpaissrs. on grando partir. dc 1:! ;'i 15 mm.,
les parois conservent cependant UIH- grandc mollrssc ; aussi se sont elles dcchii
avec trop de facilite. On nc pent plus se faire une idee de ce quY-tait 1'ouverture
cloacale ; lY-chancrurc pri'vifec laisse seulement supposer tju'elle devait in:iiic|uer de
symetrie. De meme. il cst dillieilc de dire si le corps spongieux ctait beaucoup plus
vaste que ses debris ne le montrent ; pour ma part, je ne le crois pas, parce qu'un
lambeau at-tuellement dcrarhe. ijui complctait sans doute la coupe (1M. II. fig. 5. a droile
et en avant), haut de 6 centimetres, va s'aniiiicissant jusqii'a mesurer moins tie. '2 mm.
d'epaisseur. De ce cote, la hauteur totale, dc la coupe, en comptant les grosses tube-
rosites de sa base, ne serait gut-re que de 8 a 9 centimetres. Son plus grand diametre
est de 7 centimetres, y compris toute 1'epaisseur des parois. La surface, anfractueuse,
mais tres faihlement hispicle, a posscde un revetement reticule continu ; dans les points
seulement ou il a ete enleve, les orifices, inegaux, sans ordrc, se montrent a nu.

La cavite cloacale, etraiiLdce par le plissement des parois, est. on le concoit, fort
tourmentee ; relativement t'ti-oite. elle se divise, an niveau du sommet de la plus grosse
tuberositt'1 inferieure, en trois diverticules inegaux, dont le plus vaste sc prolonge encore
verticalement sur une longueur de 4 centimetres. La surface cloacale, unic, tres fine-
ment hispide, se perce d'orifices en general plus pet its que ceux de la face externe,
rarement larges et composrs, toujours nus et beants.

Le pedoncule a une taille peu en rapport avec ce qui reste de la portion charnue
et probablement memo avec ce que fut cette partie de 1'Eponge. C'est une colonne
solitle, un peu noueuse dans son tiers superieur, marquee en bas d'un bourrelet
annulaire, puis fortement renflee en un bulbe porteur d'unc abondante toufte de soies.
Le tout mesure. a partir du bas de la grosse tuberositt'' spongieuse et sans compter les
soies, environ 38 centimetres de longueur. Le diametre n'a pas moins de 7 centimetres
au niveau du bulbe ; il se reduit a 5 centimetres au niveau du bourrelet annulaire et
a 3 centimetres vers le milieu de la longueur du pedoncule proprement dit ; mais ce
dernier est un peu aplati dans sa region mecliane et ne mesure pas 2 centimetres
comme autre dimension. Ces diverses parties de 1'appareil fixateur ont leurs homologues
dans le pedoncule de Malacosaccus floricomatus que j'ai fait dessiner (27, pi. i. fig. 1),
par meprise, la base en haut. Le pedoncule n'a plus son revetement d'hexactines
en glaive que dans sa portion noueuse et au voisinage du bourrelet annulaire, et, par
suite, parait fibreux sur la majeure partie de son etendue. Le bourrelet annulaire est
lisse et blanc ; nous verrons qu'il est constitue par une accumulation dc petites
hexactines, drja observee au point correspondant chez M. Jioncomatus. Le bulbe,
enfin, se compose d'une masse d'hexactines semblables dans laquelle s'implantent en
outre des ancres ; un nombre considerable d'ancres brisees tout pres de sa surface et
les traces de la vase dans laquelle il s'enfoncait, le rendent hispide ct gris.-'itre.

Speculation. — Les spicules dermiques sont des hexactines en glaive (PI. VI. fig. la),
d'une ressemblance frappante avec leurs homologues de Malacosaccus pedunculatus ;
1'actine distale, claviforme, longue de 0'31 mm. a <>'38 mm., epaisse de 0'015 mm. a

(ROY. SOC. KD1N. TltANS., VOL. XI. IX., 5sO.)

18 EMILE TOPSENT:

0'022 mm. en son renHement subterminal, porte des epines retroussees, faibles, souvent
reduites a 1'etat de tubercules ; les autres actines sont fines et a peine rudes, la proximale
habituellement fort longue et souple, les tangentielles egales a la distale ou seulement
un peu plus longues qu'elle. Ces megascleres dessinent un reseau carre ou plus
frequemment rectangulaire, a mailles larges de 0'35 mm. a 0'38 mm., chaque actine
tangentielle de 1'un d'eux s'accolant a une aetine tangentielle d'un spicule voisin et se
terminant a une faible distance du pied de son actine distale. Sur le pedoncule, les
hexactiues dermiques, au lieu de dessiner un reseau, entrecroisent leurs actines
tangentielles et se serrent, par consequent, davantage les unes centre les autres.
Elles y changent, en outre, un peu de caractere (PL VI. fig. ib) : 1'actine distale. reste
plus courte (0"23 mm. a 0'24 mm.) mais devient plus grosse (0'024 mm. a 0'028 mm.) ; les
autres actines s'epaississent (0'013 mm. a 0'015 mm. a la base au lieu de O'OOG mm. a
O'OOS mm.) et aceusent leurs epines ; la proximale atteint moins de deux fois la longueur
des tangentielles.

Les spicules gastriques sont des hexactines, a actines e'pineuses, la distale longue de
0'23 mm. a 0'24 mm., plus breve, par consequent, que celle des hexactines de la surface
du corps, plus mince aussi et non claviforme mais fusiforme (PL VI. fig. le), avec
0'013 mm., par exemple, d'epaisseur en son milieu et seulement O'OOS mm. a quelque
distance de sa pointe ; les tangentielles un peu plus longues et la proximale un peu
plus encore. Par places seulement, ils se reduisent en des pentactines pareilles a
eelles de M. pedunculatus. Ils forment, comme les spicules dermiques, un reseau
regulier a mailles de 0'26 mm. environ ; mais, en outre, en bien des points, sinon
partout, ils se disposent en deux rangs superposes.

Les megascleres du parenchyme sont des hexactines (PL VI. fig. Id) a actines
longues et souples, epaisses de 0'013 mm. a 0'02 mm. a leur origine, lisses, a pointes
obtuses. Elles se reduisent frequemment en pentactines aux approches des parois
des cauaux ou des deux faces du corps. Elles subissent en outre au-dessous du
revetement dermique une importante modification qui ne s'observe pas chez Malaco-
saccus pedunculatus : elles doublent (0'03 mm. a 0'04 mm.) 1'epaisseur de leurs actines
et souvent se reduisent en pentactiues (PL VI. fig. le) ; augmentant dans une certaine
mesure la consistance de la chair au voisinage de la surface, elles remplacent evidem-
ment, a ce niveau, les hexactines hypodermiques plus hautement differencie'es de
M. floricomatus.

Ce sont les memes spicules que ceux du parenchyme qui forment le pe'doncule,
mais, a cette fin, ils se reduisent presque tous en pentactines, en tetractines ou en
triactines. Les megascleres principaux sont meme toujours ramenes a 1'etat de
triactines speciales, a actines paires placees bout-a-bout suivant le grand axe de
1'organe, ;i actine impaire beaucoup plus courte et plus grele, recourbee dans la direction
du corps spongieux et engagee entre les autres spicules pour les maintenir en place.
Certaines de ces triactines atteignent 7 centimetres de longueur et JUTS de 0'5 mm.
d'epaisseur en leur milieu.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 590.)

SPONGIAIHES DE L'EXPKDITION ANTARCTIQUK NATloNALi'. KcnssAISK. T.I

Le hourrelet annulaire consiste en un feutragc d'hexactines inegales, a actines
droites, fines, I'pineuses, longues de 0'04 mm. a 0'05 mm. seulement, scmblables a
celles qui revetent le renflement basilaire du plus beau pddoncule connu de
Malacosaccus floricomatus.

Les mf-mes bexactines, dens6ment eiirhevetrees, forment encore le bulhc, avec
addition des ancres. Celles-ci sont remarquables en ce que le croisement de leurs
trois axes s'opere mm pas, comme chez les Kn/ilt'cti-l/K, ;i. unc certaine hauteur le Ion-
de letir tige barbel(Se, mais dans I'interieur de leur grappin terminal. Cinq des actines
ont done avorte et 1'ancre a la valeur d'une monactine. F. E. SCHCL/K (13, p. G5,
pi. vi. fig. 16) a deja observe pareille particularity chez Placopegma solutwm, mais,
comme le nombre des dents qu'y porte le grappin des ancres se trouve etiv pr6cisement
de quatre, il a cru pouvoir tenir ces dents pour autant d'actines de longueur reduite.
Les ancres de Hhdm/nsuccns coal^i ne permettenfc point une telle interpretation,
les 6 dents de leur grappin conique ne correspoudant certainement pas a 4 actines.
Nous verrons de meme des ancres monactinales chez d'autres Euplectellines, Accelo-
cul i/.c bi*ucei et Docosaccus ancoratus.

Les microseleres ue different que par des details de ceux de M. pedunculatitx.

lies floricomes se reucontrent dans toutes les parties de la coupe: a la surface, en
rapport avec le rayon distal des hexactiues dermiques ; dans le parenchyme, surtout an
pourtour des canaux ; du cote cloaca], enfin, au-dessous de 1'assise limitante d'hexactines
gastriques, sans le rnoindre rapport, par consequent, avec 1'actine libre de ces spicules.
Chacun de leurs rayons primaires, court, comme d'habitude, porte 8 ou 9 rayons
secondaires beaucoup plus longs, greles a leur origine, puis doucement renhVs a mesure
qu'ils se recourbent en dehors, enfin termines par une palette rtroite dont le bord
extreme se decoupe en 4 a G dents aigues. Le diametre des floricomes n'est que de
0'105 mm. a O'll mm.

Les discohexasters se repandent aussi par tout le corps ; mais c'est an voisinage de
sa surface qu'elles se trouvent le plus nombreuses et qu'elles prennent leur plus beau
ddveloppement. Leurs rayons secondaires se groupent toujours par 3 a 5. Leur
diametre, qui varie entre 0'08 mm. et O'l mm., est g(;n(:ralement en rapport avec la grosseur
de ces rayons. Les plus belles (PI. VI. fig. 1m) se singularisent presque toutes par
une tendance a imiter les floricomes ; leurs rayons secondaires, en eff'et, ne se termincni
pas par un disque veritable, mais par une expansion oblique a bord dent/' du fot<; libre
seulement ; beaucoup de ces rayons presentent meme une courbure, parfois tn-s brusque,
destinee a rejeter leur terminaison en dehors du groupc mais, maladroitement, pro-
duisant souvent un effet contraire. La courbe, de direction encore plus hc'-sitante sur
des discohexasters moins robustes, se transi'orme frequemment en une ondulation. Les
discohexasters a rayons tivs givles que la chair renferme seules, rcssemblent plus
rarement, meme de loin, aux floricomes; leurs rayons, a courbure tres douce, ont pour
la. plupart une expansion terminate fixee en son milieu et di'^-oupre en dents si profondes
que le disipue a proprement parlor fait drfaut (PI. VI. fig. 1 n ).

(BOY. soc. EDIN. TRANS., VOL. XLIX., 591.)

20 EMILE TOPSENT:

Les oxyhexasters sont rares, greles, a rayons secondaires par trois ; elles paraissent
se cantonner aupres de la surface du corps.

En resume, independamment de ses caracteres exterieurs, Malacosaccus coat si
possede dans sa spiculation quelques traits importants qui le distinguent de
J/. pedunculatus ; sa surface est renforcee par des hexactines 6paissies; son revete-
ment gastrique se compose d'hexactiues en glaive ; ses discohexasters, abondantes et a
rayons nombreux, se rapprochent du type floricome.

J'ai attache a cette espece le nom de MM. JAMES et ANDREW COATS, dont la generosite
a beaucoup contribue a renclre possible 1'expedition antarctique nationale ecossaise.

Genre Accelocalyx, Topsent.

Euplectellinse dont le corps tres mou, creuse d'une cavite cloacale a bourrelet
marginal, raais sans profondeur, est porte sur un pedoncule qui se termine par une
touffe d'ancres. Les orifices, assez grands, se distribuent sur toute la surface et sont
reconverts d'un reseau spiculeux. Les megascleres dermiques et cloacaux sout des
hexactines en glaive. Le squelette du parenchyma se compose d'hexactines souples,
distributes d'une facon irreguliere. Elles se modifient en triactines pour constituer le
pedoncule. 11 u'existe pas de Horicomes ; les seuls microscleres presents sont des disco-
hexasters et des oxyhexasters.

Le genre Accelocalyx prend place a cote du genre Malacosaccus parce que le
squelette sans ordre, fait d'hexactines souples, laisse chez 1'un et chez 1'autre une grande
mollesse au corps, et parce que, chez tous deux, 1'appareil fixateur est semblablement
constitue. II s'en distingue par la forme de la partie spongieuse du corps ainsi que par
le manque absolu de floricomes. II possede cette deruiere particularity en commun avec
Solascus, un autre genre d'Euplectellines, mais la ressemblance de ce cote ne va pas
plus loin.

Acrdocalyx brucei, Topsent. (PI. I. fig. 1 et 2, et PI. VI. fig. 4.)
1910. Accelocalyx brucei, Topsent (30, p. 5).

Station 301, 13 mars 1903; G4° 48' lat. S., 44° 26' long. \V. ; profondeur, 2485
brasses.

Un seul specimen, mais dans un etat de conservation absolument remarquable, etant
donnf} ses dimensions et sa gracilit^ ainsi que la profondeur d'oii il fut ramene. Sa
longueur totale depasse un peu 40 centimetres, dont environ 15 pour le corps spongieux,
22 pour le pedoncule et 4 pour la toulfe fixatrice.

Le corps, excessivement mou, est de forme elancee, comprimee ; large de 23 milli-
metres un peu au-dessus de son attache et de 44 millimetres en son bord superieur, il
n'est epais que de 10 millimetres par en bas et de 7 ou 8 millimetres seulement vers
le haut. II a du cependant etre un peu plus cylindrique ; on peut supposer, en effet,
que 1'orifice, en fente large aux levres, (|ii'il presente en son sommet, etait beant pendant

(ROT. soc. EUIN. TRANS., VOL. XLIX., 592.)

SPONOTATKES DE I/KX I'KTHTTON ANTAItCTTQUE NATIONALK ECOSSAISE. 21

la vie pour permettre I'eVacuation do sa cavit4 oloacale et retroussait plus ou moins
ses bords en dehors.

Sur ces bords sV'tend uu bourrelet marginal oontinu, sans frange cilieo, largo de
4 a 5 millimetres, en relief et bien delimite, imperfor6, d'aspect veloute comine cclui de
Malacosaccus pedunculatus dont il cst rimmologue evident ei dnnt il a la constitution.
La surface du corps, intacte, sauf an voisinage du pedonculc sur les deux faces, est
tapissee par un reseau spieuleux a mailles carries, dent chaque noeud se souleve en uno
petite opine ; nous verrons que les elements constitutil's de ce reseau sont des hexaeiines
en glaive et que c'est par lour aotino distale differenciee qu'ils determinent uno hispida-
tion breve de 1'Eponge. Lo reseau passe sans se modifier au-devant des orifices pern -
dans la paroi du corps, permettant quand memo de les voir facilement par transparence.
Us se repartissent assez bien en deux categories : les uns, ovales ou polygonaux, plus
grands, n'excedant pas pourtant 1'5 mm. de diametre, distants de 2 a 3 millimotres ;
les autres, plus ou moins punctiformes, semes dans les intervalles entre les premiers.
La oavito cloacale, extraordinairement roduite par rapport a la hauteur du corps, n'est,
en somme, i|ii'une petite poche de 2 centimetres a peine de profondeur et assez ir-
roguliere, son fond manquant de limite nette ; ses parois sont percees d'orifices de meme
taille que ceux de la face externe mais nullement tendus d'uii reseau spiculeux.

Le pedoncule, presque rigide, un peu compressible toutefois, est clroit, a pen prrs
cyiindrique, avec un diametre moyen de 6 millimetres. II a evidemment portV' sur
toute son etendue un revotemeut d'hexactines dermiques qui continuait a sa surface
1'hispidation du corps ; mais des frottements Ten out en majeure partie dopouillo,
mettant a nu ses longues soies paralleles et le faisant paraitre fibreux.

La toutfe fixatrice commence par un renflement opais de 12 millimetres, ou
s'entassent parallolement des triactines et des ancres et que recouvreut encore des
hexactiues dermiques ; puis, se reduisant a des ancres, elle devient un pinceau
bieu souple.

Dans son ensemble, la spiculation offre des caracteres spocifiques tros saisissables.

Le pareuchyme a pour mogascleres des hexactines a actines lisses mesiirant
couramment 2 millimetres de longueur mais seulement de O'Ol mm. d'epaisseur a la
base et de O'OOo mm. bien avaut la pointe, fines et souplcs, par consequent, et tout
a fait impropres a lui constituer une charpento quelque peu solido. An voisinam- des
surfaces et au pourtour des canaux qui traversent la chair, elles se reduisent en
pentactines, quelquefois meme en stauractines, mais jamais en diactines.

Les hexactines en glaive du reseau dermique out 1'actine distale l'\^ereinont renHee
en massue a pointe coniijue, ornee d'epines basses, longue de O'.'lo mm. a 0'37 mm.,
epaisse de O'Ol 7 mm. a 0'02 mm. a sa naissanco ot de 0'02 mm. a (C02:! nun. an-
dessous de sa pointe ; leurs actines tangentielles sont longues de 0'45 mm. et la proximale
do 0'3 mm. a 1'4 mm., suivant leur position ; ces dernieres sont lachemenf et faililenient
epineuses. Les hexactines du bourrelet marginal, plus serroes, n'ont pas leurs actines
taugentielles plus longues quo la distale : leur actine proximale est toujours tres longue

(HOY. SOC. EH1N. TKANS., VOL. XI.IX., OU3.)

22 EMILE TOPSENT :

et grele. Celles de la surface du pt'-doncule ont, au contraire, 1'actine proximale assez
courte ; elles conservent une longueur egale a leurs autres aetines mais epaississent la
distale (0'03 mm. a 0'04 mm.) et en accusent les epines.

II existe encore des hexactines en revetement de la cavite cloacale et jusque dans
les parois des canaux larges de 1'Eponge. La longueur respective des aetines y est la
meme que sur les hexactines dermiques, soit 0'35 mm. pour la distale, 0'45 mm. pour
les tangentielles et 1'5 mm. environ pour la proximale ; mais la distale s'y differencie
moius ; on la reconnait surtout a ce qu'elle est beaucoup plus rugueuse que les autres,
car elle esfc h peine fusiforme et ne mesure que 0'009 mm. a 0'012 d'epaisseur. On
voit de temps en temps, dans les parois des canaux, cette actine reduite a un
court tubercule.

Les soies du pedoncule sont des hexactines modifiees en triactines, de grosseur
et de longueur tres inegales, car il en est qui constituent la charpente principale de
1'organe alors que les autres ne forment qu'un feutrage entre les precedentes. Les plus
belles atteignent 13 centimetres de longueur et 0'3 mm. d'epaisseur; les autres peuvent
n'avoir que quelques millimetres et ne pas depasser O'Olo mm. d'epaisseur. Deux
aetines opposees se developpent beaucoup et s'orientent suivant le grand axe du
pedoncule ; 1'une, qui se tourne du cote du corps, devieut d'habitude sensiblement plus
longue que 1'autre, dirig^e vers la touffe tixatrice ; de sorte que 1'actine impaire ne se
degage pas de la soie a egale distance de ses extremites ; cette actine impaire est
beaucoup plus courte que les autres et, sur les grands spicules, demeure beaucoup plus
grele ; elle forme d'abord un angle droit sur la soie, puis brusquement se recourbe pour
remonter vers le corps sans pourtant s'allonger parallelement a elle ; sa torsion com-
pliquee lui permet au mieux de s'insinuer entre les soies adjacentes et de les maintenir
en place. Le pedoncule acquiert de la sorte une resistance suffisante sans le secours
de synapticules.

Les ancres de la touffe fixatrice sont aussi de longues soies, pointues au bout
proximal et lisses sur une bonne partie de leur etendue, puis armees de crochets
recurves, enfin terminees par un cone a base entouree d'une demi-douzaine de dents.
De forme banale, en somme, elles ont cependant quelque chose de particulier : leur
centre reel, marqur par rentrecroisement des trois lignes du filament axial, au lieu
de se trouver plus ou moins haut le long de la tige, est situe exactement dans 1'epaisseur
du grappin (PI. VI. fig. 4c). Ces ancres ne sont done pas des diactines, comme
d'ordinaire chez les EuplecteUa et les Holascus, mais de veritables monactines C'est
une particularite qui se retrouve, ai-je dit, chez trois des Euplectellines de la Scotia.

Les microscleres sont de deux sortes, tres abondantes 1'une et 1'autre :

1. Des discohexasters robustes (PI. VI. fig. 4a), de 0'175 mm. a 0'2 mm. de
diamt-tre, a rayons principaux courts (0'005 mm.) ('mettant chacun deux ou trois rayons
terminaux divergents, assez droits ; ceux-ci, lisses, epais de 0'0025 mm. a leur naissance,
se renfient du double uu peu au-dessous du disque ; le disque est lui-meme large dc
O'OIG mm. a Q'018 mm. et porte en petit nombre (5 a 7, quelquefois 4 ou ineme 3 seule-

(ROT. soc. EDIN. TRANS., VOL. XLIX., 594.)

SPONGIAIKES DE [/EXPEDITION ANTAHCTIQUE NATIONAL!-: ECOSSAISE. 23

ineut) des dents puissantcs. Les discohexasters se distribuent par tout le corps ; dans
le parenchyme et au voisinage du court cloaque, elles ne produisent qu'exceptionnelle-
ment 3 rayons tenninaux par rayon principal ; 1'exception clevient la. regie pivs de
la surface du corps ct surtout dans 1'epaisseur du bourrelet marginal.

2. Des oxykexasters (PI. VI. fig. 4/>) d'une graeilite remarquable, ^randes quand
meme, leur diametre habituel mesurant 0'175 mm. Leurs rayons prineipaux, longs de
0'007 mm. a 0-009 mm., se continuent par deux ou trois rayons secondaires (trois
surtout dans les parties superficielles de 1'Eponge, deux dans ses parties profoudes),
epais a peine de G'0013 mm. a leur origine, puis progressivement amincis, filiform es
et flexueux.

Genre Docosaccus, Topsent.

Euplectellinse probablement saeciformes et fixees par des touffes d'ancres. Elles ont
des parois minces mais rigides, dont les perforations nombreuses ne s'eteudent pas
directement d'une face a 1'autre. Les hexactines du parenchyme y torment deux
categories distinctes : les unes souples, pas tres grandes, correspondant aux hexactines
des Malacosaccus et se reduisant frequemment en diactines ; les autres, bien moins
nombreuses, mais de faille considerable, ne developpant bien que quatre de leurs actines
suivant 1'epaisseur des parois du corps, ou elles se croisent sans ordre et auxquelles elles
constituent une charpente fondamentale depourvue de symotrie. Les spicules de
revetement de la face dermique, comme aussi ceux de la face cloacale, sont des
hexactines en glaive. II existe en fait de microscleres des floricomes, des oxyhexasters
et des discohexasters.

r(n,;>r<'(tns, Topsent. (PI. III. fig. 4 et PI. VI. fig. 6.)

1910. Docimaccm aurorn/i^, Topsent (3O, p. S).

Station 301, 13 mars 1903; 64° 48' hit. S., 44' '26' long. W. ; profond.-ur. 'J485
brasses.

Quatre fragments, quatre petites plaques irregulieres, dont 1'epaisseur ne depasse
pas •! millimetres, et qui, toutes cintrees suivant une courbure a long rayon, paraissent
provenir d'une large coupe a parois minces. Des orifices tres nombreux et sans ordre
les percent, diffdrents d'aspect sur les deux faces, ceux de la face externe simples,
generalement petits, ceux de la face cloacale plus grands et manifestement constitiif-.
par la convergence de plusieurs canaux courts. Les deux faces sunl couvertes d'une
hispidation fine, due a la projection de 1'actine distale de tous les spicules superficiels,
tin pen plus haute, cependant, en dehors qu'en dedans, l/un des fragments pivsenlait.
en plus des autres. sur sa face externe, groupees vers 1'un de ses bonls ct distantes les
unes des autres d'environ 4 ou 5 millimetres, sept ou liuit protuberances coniques dont
plusieurs portaient encore en leur sommet une touffe assex dense de soies fort longues.
La photographic, faite de face, nc dnnne qu'une idee tres imparfaite de cette interessante
particularity. Les soies en question I'-tant des ancres, leur direction indique 1'orienta-

(ROY. soc. EDIN. TRANS., VOL. xi.ix., 59

24 EMILE TOPSENT:

tion naturelle de la plaque dont elles emanent et permet de supposer que celle-ci
se trouvait plus pres que les autres de la base du corps. On ne saurait, malgre cela,
imaginer uue restauration de 1'Eponge.

A sa spiculation, on la reconnait aisement comme une Eupleetellicle. Ses tourtes de
soies semblent meme annoncer un mode de fixation d'Euplectelline plutot que de
Corbitelline. J'accorde a ce leger indice d'autant plus de confiance qu'aucune des
Corbitellines connues ne presente une forme qui se puisse deviner dans nos fragments
d'Eponge, une structure qui leur convienne, un ensemble de spicules qui soit tres
semblable au leur.

Se limitant done aux Euplectelhnie , on ne pent songer a considerer notre Eponge
comme un Holascus, puisque les especes de ce genre garden! leurs parois imperforees,
se font une charpente a mailles rectangulaires et ne produisent pas de rloricomes. Les
Euplectella, d'autre part, presentent entre elles un air de parente qui ne se retrouve
pas dans nos fragments ; du reste, elles ont toujours une charpente reguliere, presque
toujours des perforations parietales disposees avec ordre et le plus souvent des pent-
actines cloacales. Eestent les Malacosaccus, sacciformes, a parois sans perforations
etendues directement d'une face a 1'autre, a charpente sans regularity, a spiculation
comprenant des hexactines cloacales et des floricomes, C'est a cote d'eux, evidemment,
que se place le type examine, mais avee une difference appreciable dans la structure de
ses parois ; celles-ci, beaucoup plus minces que celles des Malacosaccus, ont cependant
une consistance bien plus ferme ; il s'y etablit deux categories distinctes de megascleres
du parenchyme, dont 1'une, faite d'elements de taille considerable, a pour but de con-
stituer une charpente fondamentale rigide. Cette particularite est ce qui caracterise le
nouveau genre Docosaccus en attendant que soient connus la forme generale et le mode
de fixation de ses representants.

La face convexe ou externe du corps de Docosaccus ancoratns est garnie d'un
reseau d'hexactines en forme de glaives ; leur actine distale ou poignee est renflee,
fusiforme et fortement epineuse, longue de 0'24 mm. a 0'38 mm., epaisse dc (V015 mm.
a 0'025 mm. ; leurs actines tangentielles composant la garde sont moins ornees que la
poignee, a pen pres de memo longueur qu'elle mais progressivement auiincies jusqu'a leur
extremite ; leur actine proximale ou lame, enfin, est bien plus longue que toutes les
autres (0'6 mm. a 1'2 mm.), plus grele aussi et presque lisse, sauf a sa pointe. A ces
hexactines dermiques correspondent, sur la face concave ou cloacale, d'autres hexactines
n'en differaut que par leurs proportions un peu plus faibles.

Les megascleres du parenchyme qui se placent entre ces reseaux superficiels sont de
trois sortes.

D'abord, comme pieces principales de la charpente, des hexactines de tri-s grande
taille. Elles restent incluses dans les parois, pourtant bien minces. A cet eft'et, elles
ne donnent tout leur developpement qu'a quatre de leurs actines et restreignent con-
siderablement celui des deux autres (PI. VI. fig. 6«). En outre, les actines bien
developpees preunent une disposition particulicre : longues de 10 mm. a 20 mm.,

(nov. sue. EDIN. TRANS., VOL. .\Lix., 596.)

SPONGIAIRES DE L'EXPKIMTION ANTARCTIQUE NATTONALE ECOSSAISE. 25

elles doivent a la fois conserver leur rigidite pour servir de soutien aux parties molles
et permettre quand meme I'enroulement du corps ; aussi sY-tendent-elles entre les deux
surfaces, 11011 pas tout droit, mais en s'incurvant toutes a partir de leur origine d'un
meme cote, qui correspond toujours a la face cloacale ; de la sorte, elles parvicnnoit ,i
s etendre a pen pres parallMement aux deux surfaces. Ces grandes actines, qu'on peut
qualifier de tangentielles, sout lisses sauf vers leur extn'mitr, pointue et plus ou moius
raboteuse ou bosselee ; elles mesurent environ O'l mm. d'epaisseur a la base. Les deux
actines qui leur sont perpendieulaires restent infiniment plus courtes et ne se
ressemblent meme pas : celle qui monte du cote convexe du spicule, dans la direction
de la surface externe de 1'Eponge, est notablement plus breve et plus epaisse que celle
qui se tourne vers la face cloacale. Cependant, a la base de ehaque papille setigere, il
existe une grande hexactine qui est cliargee d'en former 1'axe. Celle-ci se comporte
d'une maniere un peu speciale (PI. VI. fig. 66) : elle pousse encore quatre grandes actines
tangentielles repliees du meme cote et une actine cloacale plus grele que toutes les
autres, mais son actine dermique, robuste, s'allouge beau'coup plus que de coutume et
atteint G a 7 millimetres de longueur. La disposition des grandes hexactines est tout-
a-fait irreguliere ; leurs actines tangentielles, s'entrecroisant sans ordre, ne dessinent
pas de lignes longitudinales et transversales comme chez les Euplectella et les Holascus.

Quoique en nombre bien plus eleve, d'autres hexactines jouent, en raison de leurs
dimensions modestes, un role secondaire dans la constitution de la charpente ; on peut
les comparer aux fibres secondaires du squelette des Eponges cornees car elles forment
comme elles, en se placant bout a bout, un reseau tendu dans le parenchyme entre les
rayons des hexactines principales ; elles ont des actines lisses ou un peu rugueuses,
plus ou moins flexueuses, souvent in6gales, longues en moyenne de 0'5 mm. a 0'6 mm.
et epaisses seulement de U'Ol mm. a 0'014 mm. a la base. On peut dire encore
qu'elles correspondent aux hexactines souples du parenchyme des Malacosaccus.

Enfin, Docosaccus ctncuratnti possede des diactines, mais seulement en nombre
restreint, si bien que ce sout les moins importants des megascleres de son parenchyme.
Bien moins fortes et beaucoup plus courtes que les actines tangentielles des grandes
hexactines, elles n'en repirsentent en quelque sorte que des comitalia et se tiennent,
pour la plupart, autour d'elles par maigres faisceaux parallMes a leur 'longueur, a la
facon de ce que F. E. SCHULZE a decrit chez Holascus ridleyi. Leur epaisseur est de
0'012 mm. a 0'(H4 mm., a peu pres la meme, par consequent, que celle des actines des
petites hexactiues. Comme elles aussi, elles se montrent souvent raboteuses, si bien
qu'il serait naturel de reunir en une meme categorie diactines et hexactines. Pourtant,
les terines de passage, les triactiues, par exernple, s'observent en petit nombre. Au
contraire, presque toujours les diactines montreut en leur centre un groupe de quatre
tubercules bien marques, correspondant a quatre actines atrophiees.

Les ancres des touffes fixatrices sont, comme d'ordinaire chez les Euplectellinai, des
soies longues de plusieurs centimetres ; lisses sur la majeure partie de leur longueur,
elles s'arment, du cute distal, d'epines espacees, basses d'aburd, puis rohustes et recurvees

(ROY. soc. LUIN. 'IKANS., VOL. SLIX., ">97.)

26 EMILE TOPSENT :

vers 1'Eponge et disposees suivant une ligne spirale ; enfin, elles renflent leur extremite'
libre en nn cmie a base couronnee de quatre a cinq crochets. C'est dans 1'epaisseur de
ce cone ijue se place 1'entrecroisement des lignes du filament axial, lei done, comme
chez Acoelocatyx lirncei et Malacosaccus coatsi, 1'ancre est un spicule reduit a 1'etat
de monactine.

Les microseleres les plus abondants sont des oxyhexasters, pareilles a celles de
Holascus ridleyi (12, pi. xvii. fig. 8); les plus belles mesurent O'l 2 mm. de diametre.
Repandues par tout le corps, elles sont sujettes a des variations ; souvent chaque rayon
primaire ne porte que deux rayons terminaux ; quelquefois plusieurs rayons demeurent
simples ; enfin, il arrive qu'ils se tordent a la facon de ceux de certaines oxyhexasters
signalees chez Holascu* ridleyi (12, pi. xvii. fig. 7), ma is cela est exceptionnel.

Les floricomes, localises sur la face extern e des plaques, sont nombreux et tres
beaux, chacun de leurs rayons primaires portant de 9 a 13 rayons secondaires gracieuse-
ment recourbes en dehors pour figurer un calice bien epanoui. 11s mesurent
ge"neralement O'l mm. de diametre. Leurs rayons terminaux, lisses et non t'-largis
avant la palette, qui n'a pas elle-meme plus de 0'004 mm. de largeur, portent au bord
de celle-ci, comme chez Malacosaccus vastus, 4 ou 5 dents breves.

11 existe enfin, mais rares, des discohexasters. Je n'ai reussi a en decouvrir que
deux, fort semblables a celles de Regadrella phoenix (12, pi. xiii. fig. 3) et mesurant
O'l mm. de diametre ; mais 1'une n'avait que deux rayons terminaux par rayon principal,
tandis que 1'autre en portait quatre ou cinq.

Famille CAULOPHACID^E.
Caulophacus instdbilis, Topsent. (PI. IV. fig. 8 et PI. VI. fig. 1G.)

1910. Caulophacus in&iabilis, Topsent (3O, [>. 12).

Station 313, 18 mars 1903; 62° 10' lat. S., 41° 20' long. W. ; profondeur, 1775
brasses.

Le specimen recueilli par la Scotia a et£ d6terior6 pendant le dragage a tel point
qu'on ne saurait se faire une id4e de sa forme primitive : il n'en reste que des lambeaux
au bout d'un troneon de pedoncule. En ce piteux etat, il est difficile a determiner, s'il
appartient a une espece deja connue, comme a decrire, s'il en represents une nouvelle.
Sa spiculation offre, il est vrai, beaucoup de traits de ressemblance avec celle de
Caulophacus elegans F. E. Schulze, mais, pour oser 1'identifier a cette espece, il
faudrait supposer que son corps, en ombrelle bien plus large que celle des specimens du
Challenger, a ete arrache, a 1'exception de quelques lambeaux de sa face inferieure.
Cette hypotht'se n'est point inadmissible, un pedoncule aussi epais ayant probablement
supports une Eponge de belles dimensions ; cependant, comme il se peut aussi que
notre Caulophacus ait affeete une forme plus voisine de celle de C. lotifolium Ijima,
ses caracteres exterieurs ne peuvent guere etre pris en consideration. D'autre part, on
ne peut oublier que c'est a 1'Est du Japon que le Challenger a obtenu les seuls

(ROY. soc. EDIN. THANH., VOL. XLIX., 598.)

SPONGIAIRES DE L'EXPEDITION ANTARCTIQUE NATIONALE ECOSSAISE. 27

specimens connus de C. <•/<'</«, ts, et, quoique les Hexactinellides jouisseut souvent d'une
distribution g^ographique immense, il ivste des chances pour quo le < '(t,/tlo/>li«r>tx de la
Scotia soit le type d'une espece difterente. L'etude dr la spiculation elle-ninne ne
permet pas non plus d'affirmer qu'il s'agisse de C. e/c</«i>x : les m^gascleres superficiels
ne sont pas tout-a-fait les memes de part et d'autre, et les microscleres dont je prends
connaissance ne se trouvent ni par les details de leur configuration, ni par leur taille,
ni par leur nombre relatif, identiques a ceux que F. K. SCHULZK a emuneres et figures.
Tons ces motifs d'hesitation reunis m'ont fait penser qu'il serai t raisonnable d'6tablir,
au moins sous reserves, une nouvelle espece.

Le pedmicule creux, bien plus gros que ceux des Caulophacus elegans du Challenger,
ne se montre pas liispide comme eux. Cela tient, je pense, a ce que le rayon distal de
ses spicules autoderraiques, au lieu de se d^velopper considerablement, conserve, comme
nous le verrons, une longueur mediocre, et j'estime qu'il y a lieu de tenir compte de
cette particularity. En meme temps que le specimen photograph ie, la Scotia a encore
recueilli cinq fragments de pedoncules, longs de 2 a 6 centimetres, epais de 5 a 9 milli-
metres. Deux d'entre eux, qui peuvent etre mis bout-a-bout, semblent constituer la
suite du pedoncule de notre Eponge sur une longueur de plus de 8 centimetres, ce qui
donne a ce que Ton possede de cet organe une longueur totale de plus de 12 centimetres.
Le morceau qui continue directement la portion superieure du pedoncule est par en haut
large et souple comme elle, puis va s'amincissant et se durcissant ; 1'autre morceau,
tout-a-fait dur, s'atte'nue encore par en bas, jusqu'a ne mesurer que U'5 mm. de
diametre a son extre.mite inferieure. Les autres fragments paraissent devoir etre
attribues a des specimens differents ; Fun d'eux represents une base de pedoncule,
d'abord assez grele, puis un peu elargie et incrustee de petits cailloux noirs ; fendu en
long, il montre sa cavite axiale, qui s'etend presque jusqu'au contact du support.

Les lambeaux de chair out pour squelette surtout des diactines plus ou moins
fasciculees, pas bien longues et assez greles (O'OOG mm. a 0'012 mm.), peu ou point
renflees en leur centre et ornees d'epines en leurs deux extremites, qui sont obtuses ;
puis, des hexactines eparses, a actines longues de 1'6 mm., peu pointues et un peu
epineuses seulement vers leur terminaison. Le.s megascleres superficiels sont des
hexactines a rayon distal transforme en pinule ; ce rayon, long de O'Hi mm. a 0''24 mm.,
se couvre de crochets redresses, serres, composant une touffe large de 0'06 mm. a
0'075 mm., a sommet obtus ; les autres rayons ne portent que de petites epines, soit
sur toute leur etendue, soit seulement dans leur moitie distale ; a peu pres egaux entre
eux, ils mesurent environ O'l mm. de longueur. Au-dessous de ces hexactines viennent
de grandes pentactines hypodermiques, a actine proximale plus longue, }ilus grosse et
plus ornee que les quatre autres, chargee de fortes <;pines sur toute ou sur la majeure
partie de sa longueur a partir de son origine, tandis <|uc !c> autres n'ont que leur pointe
seule un peu ornee.

Le long du pedoncule, les diactines deviennent pour la plupart plus grosses et
beaucoup plus longues et, dans ses parties solides, contractent de fmjuentes unions

(ROY. SOI . KDIX. TRANS., VOL. XLIX., 599.)

28 EMILE TOPSENT :

entre elles au moyen de synapticules. Les pentactines hypodermiques, au contraire,
diminuent beaucoup de taille tout en conservant a leur actine proximale son caractrre
special. Quaut aux spieules autodermiques, ils se reduisent assez souvent a 1'etat de
pentactines par atrophie de leur rayon proximal ; leurs rayons tangentiels ne subissent
aucun cliangement et leur rayon distal en pinule ne devient pas sensiblement plus
long (0'18 mm. a 0'24 mm.) que celui des hexactines autodermiques de la portion
charnue du corps.

Des megascleres, les spieules autodermiques sont les seuls qui different un peu de
ceux de C. elegans. A la surface de la chair (PI. VI. fig. 16a), leur rayon en piuule
est moins epais, plus allonge, plus fusiforme, bien plus semblable a celui des m ernes
spieules de C. lotifolmm. Sur le pedoncule, ce rayon devient plus claviforme (PL VI.
fig. 166), accusant ainsi une tendance qui s'exagere chez C. lotifolium; il n'y atteint
jamais les dimensions de 075 mm. et plus notees par F. E. SCHULZE et ne produit par
suite point d'hispidation perceptible. Parmi les spieules a pinule, prepares avec des
lambeaux de chair, je n'en ai pas trouve qui parussent representer une categorie a part,
comparable aux spieules autogastriques des autres Caulophacus precites.

Les microscleres sont uniquement des discohexasters. Les plus nombreuses sont
hexastrales (PL VI. fig. IGc) : leur diametre ne depasse pas 0'14 mm.; chacun de leurs
rayons principaux, remarquablement court et large, porte 4 a 6 rayons terminaux,
droits, armes d'epines recurvees non serrees, et couronnes d'un large disque a
plusieurs dents. Ainsi, sur une discoliexaster de 0'12 mm. environ de diametre, les
rayons principaux mesurent a peu pres O'OOS mm. de longueur sur 0'012 mm. de
largeur, et les terminaux, epais de 0'003 mm., sont longs de 0'05 mm. et surmontes d'un
disque large de 0-012 mm. Ni ces dimensions ni le nombre des rayons terminaux sur
chaque rayon principal ne concordent avec les details donnes a propos de C. elegans
(12. pi. xxv. fig. 6). Contrairement aussi a ce que F. E. SCHULZE a vu chez cette
espece, les discohexasters hexactinales sont rares ici ; leur taille, superieure a celle des
discohexasters hexastrales, s'eleve a 0'155 mm. et 0'21 mm.; leurs rayons simples ont,
d'ailleurs, la meme grosseur et la meme ornementation que ceux des discohexasters
hexastrales les mieux developpees ; rares aussi se montrent les discohexasters hemi-
hexactinales, avec un diametre atteignant 0'218 mm.

La taille des discohexasters hexastrales varie beaucoup, ainsi d'ailleurs que
1'epaisseur de leurs rayons ; on en trouve, par exemple, qui, pour un diametre de
0'045 mm., ont des rayons terminaux droits, e~pais de 0'0014 mm., alors que d'autres,
qui mesurent encore 0'09 mm. de diametre, ont leurs rayons terminaux droits aussi
mais tres greles, ue depassant pas 0'0005 mm. dY>paisseur. Jusqu'a un certain degr6 de
gracilite, les rayons paraissent epineux ou tout au moins raboteux ; les plus fins seuls
sont peut-etre re"ellement lisses. Mais tous ces intermediaires m'empfichent de dis-
tinguer ici des pachydiscohexasters et des lophodiscohexasters. Le nombre des rayons
terminaux n'est jamais superieur a 5 ou 6 sur chaque rayon principal, si bien que jamais
rien ne s'offre de comparable ti la belle discoliexaster de C. elegans figuree par SCHULZE

(ROY. soc. EDIN. TRANS., VOL. XLIX., 600.)

SPONGTATItKS DR L'EXPEDITION ANTARCTIQUE NATIONALE ECoss.MSE. 29

(12, pi. xxvi. fig. 2). Enfin, si greles que soient lex rayons de certaines des
discohexasters trouvees par moi sur les pedoncules, entre les diactines synaplirulees,
c'est toujours un groupe de dents n'-curvees qui les couronne et je n'ai pas trouvt'- nne
seule onychaster.

D'une faeon generale, les microscleres du CaulopJutcnx de la Scotia atteignent des
dimensions moindres et un moindre degre de complication que ceux des C. elegaits du
Challenger.

scotiif, Topsent. (PI. II. fig. 1-3 et PI. VI. fig. 17.)
1910. Cauloplirtni* m-n/ifr, Topsent (3O, p. 10).

Station 417, 18 mars 1904 ; 71° 22' lat. S., 16° 34'long. \V. ; profondeur, 1410 brasses.

Le specimen type de Caulophacus scotue est de beaucoup le plus grand des
Caulophacus conuus. Les dimensions qu'il atteint par suite du dc'-veloppement extra
ordinaire de son pedoncule, le classent meme au nombre des Hexactinellides geantes.
Sa hauteur totale est de 93 centimetres, dont 85 pour le pedoncule. Contrairement a
ce qui, d'apres WILSON (31, p. 44), a lieu pour Caulophacus schulzei, le corps
spon gieux est done term fort au-dessus du niveau de la vase abyssale. D'ailleurs, le
support devait etre ici un objet solide car la partie inferieure de 1'Eponge c.onstitue une
plaque en cuilleron, longue de 6 centimetres, large de 4, compacte et dure, a bords
minces, a face concave unie et d'aspect vitreux. Obliquement sur cette base d'insertion
se dresse, disproportionne, le pedoncule proprement dit. II est remarquablement droit
pour sa hauteur et ne presents une ondulation un peu marquee que vers le milieu de sa
longueur. Comme chez tous les Caulophacus. grele en bas, il s'epaissit en montant.
II debute sous forme d'une colonne cylindrique de I 3 mm. de diametre. A mi-hauteur,
ce diametre est porte a 20 mm. Au-dessus cle son ondulation medians, Is pedoncule se
modifie et s'aplatit quelque peu, si bien que, a 4 ou 5 centimetres de la portion
spongieuse du corps, la oil il atteint sa plus grande epaisseur, il faut considi'-rrr a celle-ci
deux diametres, mesurant 1'un 37 millimetres et 1'autre seulement 24 millimetres.
Enfin, il decrit un coude assez brusque pour porter, de meme que chez les autres
Caulophacus, le corps dans une position inclines. Dans cette derniere region, il
conserve un certain degre de compressibilite ; il rests meme spongieux, lui aussi, en
partie, et, tout en haut, se creuse de canaux dont les orifices apparaissent brants a sa
surface. Partout ailleurs, il est rigide et ferme. Sa surface est depouillee de son
revetement, sauf sur uns toute petite etendue, au voisinage du corps, sur la partie
convexe de sa courbure superieure (PI. II. fig. 3). J'ai pratiijut'' une entaille dans sa
paroi ]>our m'assurer que son axe est crsux ; mais, nature! lenient, j'ignore ri'tcndur de
sa cavite axiale. Elle ne s'ouvre pas au fond du corps, du moiiis d'une manii'n- directs,
mais il se pent qu'elle communique avec la cavitr cloacale par 1 interm^diaire des
canaux dont le sommet du pedoncule est precis<;ment traversi'.

Le corps, mou et sans resistance, a defaut d'une charpente robustc, a pu supporter
sans dommages les vicissitudes du dragage. II affecte la furnn' d'un golielet profond

(ROY. soc. EDIN. TKANS., VOL. XMX., UOl.)

30 EMILE TOPSENT :

mais taille sans regularity, ses parois ayant plusieurs plis et .son bord ne s'61evant pas
tout autour au meme niveau ; ajoutons cependant que ce bord, effiloche, n'a nulle part
son integrity. Malgre ces imperfections naturelles et ces deteriorations, on peut voir
que notre Caulophacus scoti® ne se rapproche par la conformation de sa portion charnue
que du minuscule Caulophacus pipetta Schulze, originaire conime lui de 1'Antarctique.
L'epaisseur de ses parois, voisine de 12 millimetres vers le bas, se reduit progressive-
ment par en haut a 2 millimetres a peine ; il est probable que le bord nature! de la
coupe etait fort mince. La face externe a perdu en majeure partie son revetement
spiculeux et, de ce fait, apparait fibreuse, avec des orifices beants. Son aspect fibreux
et comme peigne est determine par les paquets de diactines paralleles qui constituent la
spiculation principale du parenchyme. Ses orifices aquiferes sont d'une inegalite
frappante ; quelques uns mesurent de 5 a 10 millimetres, tandis que la plupart
demeurent beaucoup plus petits ; leur distribution semble absolument capricieuse.
Dans un vaste pli de la paroi, le revetement dermique, a 1'abri du frottement, est
parfaitement conserve : c'est une membrane mince, unie, finement et regulierement
quadrillee, qui passe sans s'interrompre au-devant des orifices et tamise 1'eau prete a s'y
engager. La face cloacale, mieux protegee, porte au complet son revetement propre,
qui, lui non plus, ne laisse pas d'orifices a nu ; meme sans loupe, il apparait different du
revetement dermique ; il a plus d'epaisseur que lui et se couvre d'une hispidation facile
a percevoir.

Les differences d'aspect entre le revetement dermique et le revetement cloaca!
s'expliquent par celles qui existent entre leurs spicules respectifs et qui, precisement,
contribuent beaucoup a caracteriser 1'espece. Les spicules dermiques sont des hexactines,
a actine proximale longue seulement de 0'14 mm. a 0'16 mm., toujours plus courtes que les
tangentielles, lesquelles mesurent couramment 0'23 mm. a 0'24 mm. mais se montrent
sou vent inegales entre elles (PI. VI. fig. 17a); ces cinq actines s'ornent pareillement
d'l'pines medicares clout le nombre va croissant aux approches de leur extremite,
generalement obtuse ; 1'actine distale, en pinule, est de toutes la plus breve (0'07 mm.
a O'll mm.), mais elle se rentfe passablement (0'04 mm. a 0'05 mm.) et se herisse de
fortes epines relevees vers son sommet, que dissimule leur accumulation. On ne
connaissait pas encore de Caulophacus oil les hexactines dermiques eussent ainsi 1'actine
en pinule beaucoup plus courte que toutes les autres. Ces spicules, chez C. scotitv,
sont, on le comprend, dans 1'incapacite de determiner une hispidation appreciable de son
revetement dermique. lls composent un reseau regulier a mailles rectangulaires ou
carrees dont les dimensions restent un peu inferieures a la longueur des actines tangen-
tielles, puisque celles-ci, se correspoudant, ne se placent pourtant pas rigoureusement
bout a bout, et ce reseau etroit est supporte par un autre reseau a mailles elles-memes
carrees ou rectangulaires, plus larges (0'44 mm. a 0'56 mm.), dessine par les actines
tangentielles de pentactines hypodermicjues.

Les spicules cloacaux ou, comme on dit souvent, gastriques sont des hexactines, peu
dinV'rp.ntes des precedentes en ce qui concerne leurs actines proximale et tangentielles, mais

(ROY. soc. EDIN. TRANS., VOL. xux., 602.)

SPONGIAIKES DE L/EXPKblTIoN ANTARCTTQUE NATFONAIJ: K< 'OS.s.MsK. 31

tout-i'i fait remarquables par le developpement de leur actine distale, cause de 1'hispidation
signalee plus haut. Cette actine, en effet, s'allonge ici beaucoup plus que les cinq
autres (PL VI. fig. 176) et varie entre 0'4 mm. et 0'9 mm. de longueur, la graude taille
etant frequemment atteinte ; elle se renrle, au contraire, assez pen (ir02 mm. a 0'03 mm.)
et progressivement s'atti'-nue en une pointe aceree ; des epines assez fortes, incurvres
mais non appliquees, 1'ornent jusqu'a son extremite, au voisinage de laquelle, toutefois,
on les voit diminuer de hauteur. Ainsi constitutes, les hexactines cloacales de
Caulophacus scoti® sont comparables aux pentactines cloacales a piuule de C. arcticus,
C. kit us et C. agassizi, mais elles ne se reduiseut qu'assez rarement en pentactines par
atrophie de 1'actine proximale. Elles s'appuient sur un reseau de pentactines
hypogastriques.

Pentactines hypodermiques et hypogastriques sont robustes et se ressemblent.
Leur actine proximale, qui 1'emporte le plus souvent sur les autres en longueur mais
subit a cet egard les variations habituelles, s'orne assez souvent d'epines ou de tubercules,
tandis que les tangentielles, droites ou un pen recourbees en dedans, se montrent
generalement lisses ; 1'actine distale atrophied ne laisse meme pas un tubercule a sa
place ; les pointes, surtout du cote dermique, sont obtuscs, parfois meme arrondies.

Les megascleres du parenchyme sont, en grande majorite, lachement fasciculees, des
diactines longues, fines (0'007 mm. a 001 mm.), a bouts arrondis, un peu renfles et
epineux, a centrum peu ou point marque. Des hexactines lisses s'y ajouteut, surtout
dans la profondeur de la chair, mais elles sont trop solitaires, avec des actines trop
minces (O'OIS mm. a 0'02 mm. a la base) et trop souples, pour constituer a 1'Eponge une
eharpente fondamentale de quelque solidite.

Le pedoncule est fait tout entier de diactines, pour la plupart orientees parallelement
a son grand axe, mais cimentees entre elles a de courts intervalles au moyen de
synapticules ; leur taille devient bien un peu superieure a celle qu'elles prennent
communement dans le parenchyme, mais, malgre 1'importance de 1'organe, les plus
grosses d'entre elles n'ont guere plus de 0'03 mm. d'epaisseur : il suffit qu'elles soient
abondantes et solidement unies. On les. retrouve encore, moius en ordre, toutefois,
dans la plaque basilaire. Mais nous savons que la face inferieure de cette plaque
presente un aspect tout particulier ; elle le doit a ce qu'il s'y developpe, sur une
epaisseur variant entre 0'5 mm. et 1 mm., un lacis siliceux scrre, a trame mince, a
mailles etroites et arrondies. La surface libre du reseau est unie ou pourvue seulemeut
de petits tubercules epars, ce qui explique qu'elle se soit tout d'un bloc di'tachee du
support.

Les microscleres sont tous des discohexasters. Les plus nombreuses sont des
discohexasters hexactinales (PL VI. fig. 17c), pareilles a celles des Caufophacus agassizi,
C. valdiviie, etc., a actines simples, longues de 0'0'J mm. ;i O'll mm., I'paisses de
O'OOo mm. a 0'007 mm. a la base, armees d'epiues fortes et recurvees, et couronnees
d'un disque de O'Ol mm. de diami'tre, a G dents. Vers 1'exterieur, elles se modifient
un peu, devenant un peu plus trapues et acqut'-rant 8 a 9 dents au disque un peu elargi

(ROY. soc. EDIX. TKANS., VOL. XLIX., 603.)

32 EMILE TOPSENT :

(0'014 mm.). C'est seulement au-dessous du revetement dermique de la partie
superieure du pedoncule que j'en <ii vu se transformer, en diminuant de taille, en des
discohexasters hemihexastrales (PI. VI. fig. 17 d) ; leurs actines divisees ont une portion
principale lisse et des divisions au nombre de '2 a 4, plus longues, epineuses, avec un
disque a 8 ou 9 dents. La seulement encore, j'ai vu de ces discohexasters reguliere-
ment hexastrales (PI. VI. fig. I7e), portant sur leurs rayons priucipaux lisses, longs
de 0'015 mm., cinq rayons secondaires epineux, longs deO;04mm. C'est peut-etre enfin
une ramification plus riche des spicules de la meme categoric qui produit des discohexasters
hexastrales de O'l mm. de diametre, ayant sur chaque rayon principal jusqu'a dix
rayons terminaux greles.

Mais ces dernieres se rapporteraient presque aussi naturellement a la categorie de
discohexasters hexastrales que le parenchyme renferme, surtout au voisiuage de sa
paroi cloacale ; de sorte qu'elles formeut comme un terme de passage. Les plus grandes
discohexasters du parenchyme ont des rayons longs de O'l 65 mm., composees d'une
partie principale lisse, longue de 0'065 mm., epaisse de 0'004 mm., et de 6 a
9 divisions terminates, longues de O'l mm., minces, raboteuses et couronnees
d'un disque denticule. Leurs rayons principaux sont ainsi generalement un pen plus
courts que les terminaux et ces derniers, se recourbant legerement en dehors, figurent
par leur ensemble une longue coupe, etroite en bas et plus ou moins evasee vers le
haut(Pl. VI. fig. 17/0-

II existe encore une categorie de discohexasters, abondante surtout le long du
pedoncule, au-dessous de sou revetement dermique et, jusqu'en ba.s, parmi ses diactines
synapticulees. Ce sont des spicules de 0'07 mm. a 0*08 mm. de diametre, a rayons
principaux lisses, portant un verticille ou un buisson de 12 a 15 rayons terminaux
excessivement greles, avec un tout petit bouton au sommet (PL VI. fig. 17* et 17j).
II s'en trouve quelques unes, de taille superieure (0'13 mm.), dont les rayons principaux
sont, contrairement a ceux des autres, moitie plus longs que les terminaux
(PI. VI. fig. 17k).

Famille EOSSELLID^;.

Batliydorus levis, F. E. Schulze, var. ciliatus, Topsent. (PI. I. fig. 6 et 7.)
1910. Batliydorus levis, F. E. Schulze, var. ciliatus, Topsent (30, p. 15).

Station 420, 21 mars 1904; 69° 33' lat. S., 15° 19' long. W. ; profondeur, 2620
brasses.

Un beau specimen en entonnoir, haut de 8 centimetres, large de 9 centimetres en
haut, a supposer son bord entier et ses parois afiaissees jusqu'a se toucher. II est fixe
sur un petit galet par un pied court, lisse et ferme, qui s'est trouve brise pendant le
voyage au point oil il se continuait avec la partie spongieuse du corps. Ce point etait
d'autant plus fragile que la continuite ne s'operait entre le pedoncule et la paroi molle
que d'un seul cote ; une large ouverture, a bords coupes nettement, naturelle, par

(ROY. sou. EDIN. TRANS., VOL. XLIX., 604.)

SPONGIAIRES DE I/EXPEDITION ANTAHCTIQUE NATIONALE ECOSSAISE. 33

consequent, occupe la majeure partie du fond de 1'entonnoir. La surface du corps est
egale et glabre avec des orifices aquifi-res bien visibles, un peu plus grands sur la face
cloacale <|iie sur la face externe. Ce qui fait la beaute et 1'interet de ce specimen, c'est
qu'une magnifique frange de soies, haute de 10 millimetres, ornc son rebord. Les deux
Eponges du S.W. de la baie du Bengale qui ont servi de types a 1'espece Bathydorus
levis n'avaient pas de frange du tout. D'autre part, les specimens de B. levix xjiiimsus
Wilson, de la cote de Colombie, ont les deux faces munies de prostalia 6pars.

Station 417, 18 mars 1904; 71° 22' lat. S., 16° 34' long. W. ; profondeur,
1410 brasses.

Un specimen en entonnoir mou, haut d'environ 7 '5 centimetres, et probablement
aussi large, porte par tin pied court mais arrache de son support. II est beaucoup
moins bien conserve que le precedent ; notammant, sa partie superieure est detachee ;
elle presente aussi une frange mais beaucoup plus courte (3 millimetres) et faite de
soies plus fines. II existe encore (est-ce un hasard ?) une perforation a la naissance du
pedoncule, mais son contour n'est pas r6gulierement arrete et son diametre ne depasse
pas 6 millimetres.

Les stauractines dermiques ont des actines a bouts obtus, longues de O'OG mm. a
O'l mm., epaisses de 0'OU4 mm. a 0'005 mm. a la base. Elles n'out pas un centre mi
comme celles de Bathydorus lens type et leurs epines, repandues partout, sont moins
fortes que chez B. levis spinosus.

Les hexactines gastriques ont des actines finement pointues, longues de 0'08 mm. a
0'09 mm., sauf la distale qui est toujours beaucoup plus longue (O'l 8 mm. a 0-2 mm.)
et armee d'epines ayant le double de la longueur des leurs et releve'es vers sa pointe.

Les diactines du corps sont fines (O'Ol mm. de diametre moyen), a renflement median
peu marque, a bouts en forme de rnassue allonge' e et epineuse.

Les pentactines hypodermiques sont lisses, a actines tangentielles un peu recourbe'es
en dedans, longues de 0'43 mm., epaisses de 0'02 mm., a actine proximale droite, longue
de 0'66 mm. a 0'92 mm., toutes a bouts legerement renfles et un peu raboteux. Le
long du pedoncule, ces pentactines sont entitlement et assez fiuement epineuses, avec
des actines obtuses, les tangentielles longues de O'l 8 mm. a 0'25 mm.

Les soies de la frange du specimen le mieux conserve ont 0'04 mm. d'epaisseur ; elles
sont entierement lisses et pointues. Celles de 1'autre sont des diactines a centrum
apparent, pas plus longues que celles du parenchyme mais dressees cote a cote sur le
bord aminci du corps, uues sur la majeure partie de leur longueur et notablement plus
pointues que d'ordinaire.

Les microscleres, melange d'oxyhexasters et d'hemioxyhexasters, ont 0'09 mm. a
O'l 25 mm. de diametre ; leurs rayons secondaires bien divergents, se montrent raboteux,
surtout sur les hemioxyhexasters, ou leur epaisseur est toujours un peu plus forte.

L'operation du 18 mars 1904 a encore fourni une grande plaque de ce Bafl(i/>/,,rn\.
mais sans pied ni rebord reconnaissable.

(ROY. soc. EDIN. TRANS.. VOL. xi.ix., 605.)

34 EMILE TOPSENT :

Calycosoma validum, F. E. Schulze. (PL III. fig. 2 et PI. VI. fig. 5.)

Station 313, 18 mars 1903; 62° 10' lat. S., 41° 20' long. W. ; profondeur,
1775 brasses.

L'espece n'etait encore connue que d'apres un seul specimen, recueilli par Y Albatross
(14, p. 27) dans 1'Atlantique Nord, au large de la cote du Massachusetts, par 40° 34' 18"
de latitude N. et 66° 09' de longitude W.

Par ses spicules superficiels a actine distale de>eloppee et arme'e comme chez les
Caulophacidse, le genre Calycosoma a de grandes affinites avec deux autres genres de
Rossellidie, Asconema Saville Kent etHyatascus Ijima ; mais la possession de strobilo-
plumicomes au lieu de discohexasters 1'a fait ranger par F. E. SCHULZE, en derniere
analyse (15, p. 176), dans la sous-famille des Lanuginellinai, les deux autres genres
prenant place dans celle des Rossellinse.

Le specimen type de Calycosoma validum, quoique fort d^teriore, se trouvait quand
meme dans un etat qui permit d'en imaginer une restauration : c'etait une graude
Eponge en coupe, a parois graduellement amincies vers le haut, a pied court et epais,
probablement detache'e d'un support solide ; sa surface portait, a intervalles assez
grands, des papilles coniques du sommet desquelles s'e'chappait une toufFe de
longues soies.

II n'eut pas 6te possible de prendre une aussi bonne ide"e de 1'Eponge sur le materiel
de la Scotia, qui consiste, en effet, en une demi-douzaine de lambeaux irre'guliers, ne
depassant pas 3 millimetres d'epaisseur. lls proviennent probablement de la partie
superieure d'un echantillon de belles dimensions, que 1'engin a dechire puis roule
dans la vase. Leurs deux faces presentent encore en place leur spiculation super-
ficielle ; toutefois les verrues de la face externe se trouvent a pen pres degarnies de
leurs soies. Les orifices, de forme ovale, sont bien visibles des deux cot6s et la
disposition irreguliere des lignes squelettiques se distingue elle-meme fort bien, surtout
sur la face cloacale. Naturellement, la question reste toujours ouverte de savoir si la
coupe possede ou non une frange marginale.

Dans son ensemble, la spiculation du Calycosoma de 1'Antarctique correspond a
la description de celle du C. validum de 1'Atlantique Nord. Rieu a aj outer au sujet
des diactines, de leur groupement par faisceaux, de leur union si frequente au moyen de
synapticules, pas plus qu'au sujet des grandes hexactines eparses au milieu de la
charpente. Mais j'ai fait, a propos des pentactines hypodermiques, une remarque que
SCHULZE n'a pas consignee : tandis que ceux de ces megascleres qui se tiennent au-dessous
de la surface plane du corps ont leurs quatre actines tangentielles parfaitement droites,
ceux qui doublent la surface au niveau des verrues sans en depasser les contours,
re"duisent ces memes actines en longueur mais surtout les incurvent toujours d'une facon
brusque en leur milieu (PI. VI. fig. 5d) ; chacune de ces actines incurvees ne mesure
guere alors que 0'3 mm. en tout, au lieu de 0'7 mm. qu'elle atteint, d'habitude, au bout
d'une actine proximale longue de pres de 1'5 mm. Les pentactines a actines tangen-

(EOT. 8OC. EDIN. TRANS., VOL. XLIX., 606.)

SPONGIAIRES DE I/EXPEDITION ANTAROTIQTJE NATIONALE ECOSSAISE. 35

tielles droites existent du cote cloacal, plus clairsemees, toutefois, et un pen plus I'aililes
que du cote extern e'du corps.

Ce qui m'a le plus embarrasse dans la comparaison des deux Calycosoma, ce sont
leurs spicules derraiques et gastriques (ou mieux cloacaux). Je n'ai pas retrouve dans
celui de la Scotia une seule des pentactines a rayon proximal rudimentaire decrites par
SCHULZE (14, pi. iv. fig. 5). En outre, j'ai constate qu'il n'y existe point d'uniformitr
dans les hexactines. J'ai bien revu, a pen pres dans la taille indiquee, les cloacales
demeurant un peu plus greles que les dermiques, les hexactines a rayon distal en forme
de pinule et a rayon proximal plus court que tous les autres (PI. VI. fig. 5a, 5fr), mais
j'ai observe avec surprise que, par places, surtout le long de la face cloacale, elles
subissent en grand nombre une modification profonde. Leurs dimensions, depassant
alorsla moyenne, s'accroissent souvent d'une maniere considerable ; en meme temps, leur
rayon proximal devient de beaucoup le plus long; frequemment leur ornementation
s'efface, sauf sur le rayon distal et vers 1'extremite des autres rayons (PI. VI. fig. 5c).
Ces hexactines agrandies constituent, par 1'agencement de leurs rayons tangentiels,
un reseau plus ou moins regulier et assez lache.

S'il est permis d'expliquer 1'absence ici des pentactines a rayon proximal rudimen-
taire du Calycosoma de V Albatross en supposant qu'elles pouvaient, dans cette Eponge,
n'etre que des malformations resultant d'uue tendance individuelle, il est difficile de
compremlre pourquoi le Calycosoma de la Scotia produit en proportion si forte des
hexactines de taille si inegale et qui vont jusqu'a acquerir un rayon distal de 0'43 mm.
a 0'64 mm., un rayon proximal de 077 mm. a 0'9 mm. et des rayons tangentiels de
U'5 mm. de longueur.

J'ai hesite a etablir une espece nouvelle d'apres ces particularites. Les microscleres
ui'ont paru differer trop peu de ceux du type de Calycosoma ralnlnn). Ce sont aussi
des oxyhexasters, tres nombreuses partout, avec quelques oxyhexactines et interme-
diaires, et des strobiloplumicomes confines aux deux faces de 1'Eponge. Les uns et les
autres sont cependant un peu plus grands que dans le type : le diametre des strobilo-
plumicomes atteint ici O'OG mm. et 0'07 mm. et celui des oxyhexasters, qui sont souvent
un peu rugueuses, varie entre 0'15 mm. et 0'18 mm. Tout bieu consider^, je tiens ces
differences pour purement inclividuelles.

Farnille COSCINOPORID^;.
Chonelasma sp.

Station 417, 18 mars 1904. 71" 22' lat. S., 16° 34' long W.; profondeur, 1410
brasses.

Deux fragments completement macert'S : 1 ° une Ijase sans support, enroulee en cornet
com prime" a parois epaisses de 1 mm. a "2 mm., haute de 30 a 40 mm., longue d'autant,
large de 7 mm. a 15 mm., interessante seulement par une serie de quatre trous ronds,
de 3'5 mm. a 4 mm. de diametre, perfores suivant sa hauteur, a une distam-e de 2 mm.

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 607.)

36 EMILE TOPSENT:

1'un de 1'autre, le long de son bord vertical le plus etroit, au ras de sa surface, et donnant,
par un canal de meme calibre et assez profond, directement acces dans sa cavite ; 2° un
raorceau de plaque, mince et fragile.

II. S.O. AMPHIDISCOPHORA.

Famille HYALONEMATID^:.

Hyalonema sp.

Station 417, 18 mars 1904; 71° 22' lat. S., 16° 34' long W.; profondeur, 1410
brasses.

Un fragment long de 5 centimetres, compose d'une touffe axiale de grosses soies
brisees et d'un revetement de chair effilochee, epais de quelques millimetres. C'est
une portion de la base d'un individu a 1'origine de sa touffe fixatrice, avec un peu du
bourrelet annulaire encore en place.

La spiculation comprend, en fait de megascleres : des hexactines et surtout des
diactines, lisses et pointues, dans le parenchyme ; des pentactiries lisses hypodermiques ;
un melange de tt'-tractines, de triactines et de diactines a bouts renfles et t'pineux,
pareilles a celles de Hyalonema elegans (12, pi. xxxi. fig. 15), au niveau du pad ;
implantees sur le tout, des pentactines dermiques, a pinule long de 0'13 mm. a O'l 4 mm.,
;'*pais de O'OOG mm. a O'OOS mm., arme d'epines relevees, incurvees, hautes de 0'017 mm.

e

environ ; les memes pentactines ont des actines tangentielles longues de 0'05 mm. a
0-07 mm. et raboteuses clans leur moiti£ terminals ; quelques hexactines a pinule se
rencontrent parmi les pentactines, n'en differant que par la possession d'une actine
proximale de 0'07 mm. de longueur. Les soies fixatrices etaient toutes brisees et je
n'ai pas vu de pentactines gastriques.

Les microscleres sont : des oxyhexasters du parenchyme, nombreuses, d'un diametre
de 0'08 mm., a actines lisses constamment et fortement courbees dans leur dernier quart ;
des amphidisques de trois categories, a dents etroites, disposees en cloches parallelement
a la tige. Les petits amphidisques, longs de 0'023 mm. a 0'032 mm., ont une tige
noueuse au centre et des ombrelles ecartees 1'une de 1'autre ; les macramphidisques,
longs de 0'095 mm. (je n'en ai pas trouve rle plus grands), ont une tige grele (0'003 mm.
a peine), ornee d'epines en outre de sa nodositti mediane, et des ombrelles larges de
0'024 mm. a 0'026 mm., entre lesquelles reste un ecart de plus de 0'02 mm ; les
mesamphidisques, enfin, longs de 0'06 mm. a 0'07 mm., ont des ombrelles beaucoup
plus rapprochres 1'une de 1'autre.

I'ar 1'ensemble de ses spicules, ce Hyalonema se rapproche beaucoup des H. ovuli-
ferum F. E. Schulze et H. urna F. E. Schulze, surtout du premier, en raison de
1'abondance de ses oxyhexactines. De tous deux il parait ditferer par ses macramphi-
disques plus petits et par ses pinules plus courts avec epines plus longues ; cependant,
je ne puis pas affirmer avoir obtenu la taille la plus grande de ses macramphidisques et

(ROY. soc. EDIN. TRANS., VOL. XLIX., 608.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATIONALE ECOSSAISE. 37

je dois tenir compte de ce que les pentactines dermiques dont j'ai pris les mcsures,
proviennent du bourrelet annulaire et, par consequent, ne sont peut-etre pas absolument
typiques. Ne sachant rien de sa forme et nV-tant pas .sur de eonnaitre exactement sa
spiculation, j'eviterai d'encombrer la litterature en ajoutaut aux Hyalonema, deja si
nombreux, une espece (jui serait caract6risee d'une maniere insuffisante. A d'autres de
la denommer qui, en possession do specimens mieux conserves, se trouveront a menu1
de verifier et de completer cette description succincte.

TETRACTINELLIDA.

Les expeditions antarctiques out fourni 1111 nonibre remarquablement restraint de
Tetractinellida largo sensn. La Belgica a reeueilli seulement une Carnosa, que j'ai
rapportee a Placina trifaj>l>n F. E. Schulze, en appelant Fattention sur les differences
que presque tons les elements de sa spiculation presentent avec ceux de la forme
mediterranean ne (26, p. 30) ; LENDENFELD 1'a appelee depuis P. trilo/ilm F. E. Schulze,
subsp. riit/aivfica (6, p. 326). La petite collection du Frangais ne renfermait aucune
Eponge de ce groupe. Celle du (jauss (6) se composait de la Placina precitee. en de
nombreux exemplaires, d'une proche parente, Pl<n-ina. monolopha F. E. Schulze,
subsp. antarctica Lend., d'une Oscarella'f douteuse, de trois Tetillidss, dont deux
nouvelles en taut qu'especes, et d'un nouveau Tnbrachion. La Discovery (3) a obtenu
encore bien moins : en tout, quatre especes de Tetillidie deja connues, dont deux avec
des varietes nouvelles. Quant a la Scotia, elle n'a pas rencontre de TetractinellicUss du
tout dans 1'Antarctique et les trois especes suivantes, deja plus ou moins connues,
proviennent toutes de la limite septentrionale de la region subantarctique, des pa rages
de 1'ile Gough.

Famille ASTEROSTREPTID/E.
Pachastrella monilifera, (). Schmidt.

Station 461, 22 avril 1904, He Gough, 40° 20' hit. S., 9° 56' 30" long. W. ;
profondeur, 100 brasses.

Un specimen, ou mieux un fragment sans support de cette Eponge, mesurant pros
de 7 centimetres de longueur, assez uniformt'-ment 3 '5 centimi-tres de largeur et 2
centimetres au plus d'epaisseur. II a etc brise suivant son epaisseur et c'est sa face
porifere qui seule a ete bien conserves. Celle-ci est lissc, uuie, et entierement revetue
par uue Pcecillastra incrustans Sollas, epaisse de 1'5 mm. a 2 mm.

Les oxes, en faisceaux diriges pour la plupart perpendiculairement a la surface,
mesurent 2 a 3 millimetres de longueur sur 0'015 mm. au plus d'epaisseur.

Les calthropses distribues par tout le corps en abundance et sans on Ire, sont, com me
d'habitude, fort inegaux, leurs actines variant entre 0'035 mm. de longueur sur
0'002 mm. d'rpaisseur a la base et 0'6 mm. sur 0'06 mm. I.es plus petits demeurent
reguliers tandis que les plus grands presentent toutes les deformations dont SOLLAS a
donne des exemples (17, pi. xi. lig. 15-23).

(ROY. soc. EDIN. TRANS., VOL. xi.ix., ij'JD.)

38 EMILE TOPSENT:

Les microscleres sont des microstrongyles presque constamment centrotylotes, fine-
ment rugueux, longs de 0'014 mm., epais de O'OOS mm. au centre, et des spvrasters,
plutot clairsemees, souvent de forme amphiaster, a tige et a actines tres greles, et
mesurant au total de 0*012 mm. a 0'014 mm. de longueur.

Pcecillastra incrustans, Sollas. (PI. VI. fig. 2.)

Station 461, He Gough, 40° 20' lat. S., 9° 56' 30" long. W. Sur 1'Eponge precedente,

SOLLAS a signale (17, p. 105) sur les deux specimens de Pachastrella abyssi
( = P. monilifera) recueillis par le Challenger a Tristan da Cunha, par 110 brasses
de profondeur, 1'existence d'une Pcecillastra qu'il a nominee P. incrustans, sans la
decrire en detail.

L'Eponge s'etendait en croute sur la face porifere des specimens en question et
mettait ses canaux en continuity avec les leurs (17, pi. x. fig. 15). On trouvait
isolement ou par places dans les Pachastrella des microxes centrotylotes qui avaient
du lui appartenir.

II est remarquable que le troisieme specimen de P. monilifera qui provienne de ces
parages de 1'Atlantique, drague a trois degres de latitude de plus vers le Sud, et par la
meme profondeur, porte egalement ce parasite et que les rapports entre les deux
Eponges soient identiques a ceux observes par SOLLAS.

La Pcecillastra forme ici une croute epaisse de 1 -5 mm. a 2 mm. Sur une section
d'ensemble, on peut la distinguer de la Pachastrella a 1'oeil nu par sa blancheur un peu
plus terne. Son ectosome, parfaitement lisse, ne montre pas de grands orifices.
Son choanosome est riche en corbeilles vibratiles du type des Poecillastra, claires,
spheriques ou ovales et d'un diametre de 0*032 mm. a 0*038 mm.

La limite entre les deux Eponges est nette. Un decollement partiel s'opere meme
generalement sur les coupes fines. II semble aussi que lorsqu'un canal de la Pcecillastra
se continue avec un canal de la Pachastrella, chacune des Eponges tende d'un diaphragme
ou d'un crible la partie qui lui est propre. La face interne de la Pcecillastra abonde,
comme sa face externe, en microxes tangentiels et 1'ectochrote de la Pachastrella est
charge de microstrongyles centrotylotes.

Les oxes de Pcecillastra incrustans sont courbes, pointus, le plus souvent groupes en
faisceaux qui s'orientent tangentiellement, en des points varies de 1'epaisseur du corps ;
ils mesurent de 077 mm. de longueur sur O'OOS mm. d'e'paisseur a 1*4 mm. sur 0*017
mm. ; ils sont proportionellement plus gros que ceux de la Pachastrella.

Les calthropses sont enchevetres en tous sens entre les deux faces du corps, ce qui
est exceptionnel pour une Pcecillastra. Ils sont inegaux, mais leur taille s'abaisse en
general moins que pour ceux de Pachastrella monilifera ; les plus petits, qui ne sont pas
nombreux, ont des actines longues de 0'04 mm. ; les actines des plus grands atteignent
pros de 0'4 mm. de longueur sur 0*045 mm. d'epaisseur a la base. II s'en rencontre un
assez grand nombre qui sont diflformes, ce qui augmente encore la ressemblance de ces
m6gascleres avec ceux de la Pachastrella parasite.e.

(ROY. soc. EDIN. TEANS., VOL. XLIX., 610.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATION A LK KCDSSA INK. 39

Les microxes caracteristiques drs Pcecillastra einBten\ en quantitr considerable pur
toute 1'Eponge, mais surtout, comme il a ete (lit plus luiut, an niveau de sea deux faces.
Ils sont pointus, toujours courbes, toujours centrotylotes et raboteux (I'l. VI. fig. 2);
ils raesurcnt de 0'09 mm. a O'lO mm. dc longueur sur 01)03 mm. d'epaisseur. Un E
curieux est qu'il s'en retrouve un certain nombrc par endroits dans le corps de la
Pachastrella ; il est difficile de se rendre cnmpte comment ils y sout parvenus. Peut-
etre faut-il admettre quo la Pcecillastra en rejette et que ces spicules eliinines pcuvcnt
se trouver apportes par 1'eau d'inhalation dans la P>«-/ia*fi-<'Ila au meme titre que tons
autres corps etrangers, notamment que les tests de Foraminiferes, qui y aliondent.

Les •<]>/>•< i f<fers, enfin. sont nombreuses, transformees rarement en amphiasters mais
couramment en metasters ; elles ont, en moyenne, uue longueur totale de 0'017 mm.
avec des actines de 0'005 mm. et tres greles ; par exception, quelques me'tasters ont
leurs actines longues de O'Ol mm. a 0'015 mm.

Poecillastra conn>rcxx«., (Bowerbank) Sollas, var. parvistellata, n. var.

(PL IV. fig. 1 et 2.)

Station 461, 22 avril 1904, Lie Gough, 40° 20' lat. S., 9° 56' 30" W. ; profondeur,
100 brasses.

Je considere le genre Poecillastra de SOLLAS comme valable car un certain nombre
de Streptastrosa s'y laissent rapporter sans hesitation et constituent par consrquent un
groupe naturel. LENDENFELD a pivfere (4, p. 72) le noyer dans le genre Pachastn-dla,
prenant celui-ci dans uue acception si large qu'il 1'a rendu lu'-terogene, ce qui 1'a,
d ailleurs, oblige d'abord a le diviser en sous-genres, puis a reconnaitre (5, p. 237) dans
le sous-genre Pachastrella 1'existence d'une serie de formes du type <!<• /'. compressa
(Bowerbank), correspondant precisement an genre ]'<«-i//<txtr<<. Ces formes. LENDENFELD
en a passe quelques unes en revue, s'attachant a caract6riser les especc,-- /'. fom/n-i'ssa
(Bowerbank), /'. stylifera Lendenfeld, P. crassa (Bowerbauk) et tine espece nouvclle,
P. tt.'iHiijiiliisit. ]jrov(.-nant des dragages de la \'<i/</iritr. Si /'. crassiuscula Sollas cl
P. tenuilaminaris Sollas se confondent, comme il le suppose, avec P. < /-<ts.«i (liowor-
bank), il resterait encore a corapter dans le genre Pcecillctstra I'. ,>r//»/:'^ Sollas,
P. laimnaris Sollas et P. symbiotica Topsent.

La Scotia a recueilli a File Gough une Pcecilfastra que je ne pui> pas distiuguer
specitiqucment de P. compressa car, depourvue d'oxes dermiques speciaux, clle presente
en abondance des tetractines irregulieres, a eludes fortement courbes, et des asters dont
les plus faibles imitent frequemnieut plus ou moius des amphiasters. 11 me parait juste
cependant de la tenir pour une varirtt'- de cette espece en raison surtout de la petite
taille de ses microscleres.

II y en a quatre fragments, vraisemblablement dr-taclu'-s d'un meme specimen, qui
devait affecter la forme d'une coupe largement evasec. Leur epaisseur, de 8 millimetres,
diminue jusqu'u C millimetres en leur bord, i|iii est arrondi. Ils sont tout-a-fait blancs

(ROT. soc. EDIN. TRANS., VOL. XLIX , 611.)

40 EMILE TOPSENT:

et lisses. Nettement bif'aeiaux, ils portent leurs oscules suv la face concave. Oseules
et pores sont, d'ailleurs, comme le montre (PI. IV.) la photographic cle deux des
fragments, vus Tun (fig. 1) par la face inhalante et 1'autre (fig. 2) par la face exhalante,
revetus d'un ectosome continu, mince, crible d'orifices microscopiques. Les oscules
ont environ 0'7 mm. a 1 mm. de diametre ; leurs intervalles sont perces d'orifices
punctiformes. Les pores se distinguent des oscules par leur nombre plus 6leve et par
leur calibre plus faible ; les canaux qui en partent traversent quelquefois la plaque de
part en part pour aboutir a des oscules.

Spoliation. — I. Megascleres : 1. Oxes un peu courbes, longs de 1'S mm. a 2'2 mm.,
epais de 0'025 mm. a 0'03 mm. Ils composent la majeure partie de la charpente ; dis-
tribues sans ordre dans le ehoanosome, ils s'entrecroisent encore capricieusement dans
1'ectosome mais en y demeurant tangentiels au corps, ce qui laisse 1'Eponge parfaitement
lisse sur ses deux faces. Qa et la, un de ces oxes tronque 1'une de ses extremites et
1'arroudit, se transformant alors en un style veritable. LENDENFELU accorde une
certaine importance a 1'existence ou a 1'absence de ces grands styles chez les especes
qu'il a voulu caracteriser. D'apres lui (5, p. 238), Ptecillastra compressa n'en possederait
pas. En realite, j'en ai trouve de clairsemes comme ici chez des P. compressa de ROSCOFF
et je les prends pour des malformations, dont la frequence varie suivant les individus.

Parmi les oxes robustes, il s'en rencontre de beaucoup plus freles, qui ne depassent
guere 1 mm. de longueur sur O'OOS mm. d'epaisseur. Ces spicules greles ne represented
pas une categoric a part ; ils sont rares par tout le corps, et, comme ils font precise-
ment defaut dans 1'ectosome, ils ne sont nullement comparables aux spicules superficiels
de P. crassa et de P. tenuipilosa.

2. Tetractines fort variables. Leur forme typique est celle d'orthotriienes a
rhabdome n'excedant guere de plus d'un tiers la longueur des clades, soit, par exemple,
0'3 mm. pour 0'19 mm., 0'39 mm. pour 0'245 mm., sur 0'025 mm. d'epaisseur. Mais la
plupart sont difformes, reduites a Tetat de disenes ou de moneenes ; ou bieu leurs clades se
tronquent, se coudent brusquement ou meme se bifurquent ; souvent meme, sur les plus
reduites d'entre elles, le rhabdome pousse une forte protuberance cladoide quelque part
sur sa longueur. Enfin, quelques uues, dirigeant deux de leurs actines vers le haut et
les deux intermikliaires vers le bas, se transformeut en calthropses. Presque toutes se
])lacent ver.s la surface du corps, le cladome appuye contre la face profonde de recto-
some ; elles sont surtout abondautes du cote exhalant ou cloaca!

II. Microscleres : 3. Microxes droits, fusiformes, raboteux. assez souvent centro-
tylotes ; ils ont de 0'08 mm. a 0'12 mm., generalement moins de O'l mm. de longueur
sur O'OOo mm. a 0'0035 mm. d'epaisseur. Ils se distribuent en grand nombre par
tout le corps. 4. Asters. SOLLAS distinguait chez les Poecillastra des metasters
et des spirasters. LENDENFELD, en 1903 (4), cite des amphiasters et des spirasters
chez P. compressn et seulement des spirasters chez les autres especes. En 1906
(5), il recommit des amphiasters et des metasters a P. compressa et riert que des
nillcurs. ( 'Vst qu'on eprouve un certain embarras a designer d'un seul
(HOY. soc. EDIN. TRANS., VOL. XLIX., Gli.)

SPONGIAIRES |)E L'EXPEDITION ANTARCTIQUK NATION ALE KCOSSAISK. II

nom des asters aussi variables. En ce qui concerne notre Eponge, elles se tiennent en
tres grande abondance dans le revetement des deux faces sous la forme de spirasters
greles a epines nombreuses ; quand leurs epines se trouvenfc en nombre moindre, elles
ont une tendance a se disposer en deux groupes opposes rappelant ceux qui caracterisent
les amphiasters ; enfin, dans 17-paisseur clu corps, des asters se rencontrent a epines en
petit nombre mais plus fortes et dingoes en tons sens, suivant le type metaster. Oes
formes, se reliant les unes aux autres par tous les intermediaires possibles, il n'existe,
en realite, qu'une seule sorte d'asters, qu'on peut appeler des nx'hixh'rx pnisqu'elles ne
font defaut a cet i-tat chez aucune espece du genre.

II est a remarqucr quo les mi-tasters vraies de P. compressa var. jwrrix/c//n/i/ sont
relativement pen nombreuses et qu'elles demeurent faibles ; elles ri'atteignent que
O'OIG mm. a 0'023 mm. de longueur et se composent de 3 a 7 epines fines et pointues,
longues de 0'007 mm. a O'Ol mm.; les spirasters sont de beaucoup ce qui domine, mais
elles demeurent assez courtes (O'Ol mm. a O'OIS mm. de longueur pour la plupart.
rarement davantage, souveut moins) et portent des epines nombreuses, bacilliformes, si
breves que leur largeur totale varie le plus souvent entre 0'004 mm. et O'OOG mm. ;
quant aux amphiasters, elles restent clairsemees et ne se montrent qu'exceptionnellement
et comme par hasard de type assez pur.

MONAXONIDA.

La collection des Monaxonides de la Scotia comprend une trentaine d'especes, mais
il convient d'en faire trois parts d'apres leur distribution geographique : les especes
antarctiques, les especes subantarctiques et une espece recueillie seule dans une operation
hors serie.

A 1'exception de Jfi/.riU'.t- spongiosa (Ridley et Dendy) var. asigmata, draguee
avec des Hexactinellides, le 18 mars 1904, Station 417, aupn'-s de la Terre de Coats,
par 1410 brasses de profondeur, les especes antarctiques proviennent des Orcades du
Sud (Station 325) et sont des formes littorales, recueillies par 7 a 10 brasses seulement.
pendant le si'jour de huit mois fait par la Scotia, en 1903, dans la liaic qui pnrte
desormais son nom. Dans ce lot se trouvcnt quatre lopJton decrits ant^rieurement,
deux, /. pluricornis Tops, et /. unicornis Tops., d'apres le materiel de la Belgicck, et
deux, /. spatulatus Kirkp., et /. flabcllodigitatus Kirkp., d'apres eelui de la D/.^-i,,-,'!-//:
puis, une Mycule, Af. acerata Kirkp., deja rencontree par la MVo, >•/•//, une ll<>n,<i<>-
dictya du Franqais, H. setifera et une Reniera, II. <l «»<•<>}, «|ue j'ai fail commit iv en
1901 (26, p. 12) et que KIRKPATRICK a revue (3, p. 53). .!///./•///« spongi<>.«/ «*<':/i,,«t<r
etant dans lememe cas que Reniera dancoi (26, p. 18 et 3, p. 28), il n'y a a meiitidiiner.
comme Monaxonides antarctiques nouvelles, que trois espi'ces, (:,',•// n/.-i areuarius',
Petrosia depellens et Reniera cylindrical. Je ne f'ais pas entrer en ligne de eompte
des fragments sans valeur appartenant a des Reniera.

Les especes subantarctiques forment elles-memes deux lots bien distinct'-. l,e

(ROf. SOC. EDIN. THANH., VOL. XLIX., 013.)

42 EMILE TOPSENT :

premier, pris sur le bane de Burdwood, Station 346, et aux lies Falkland, Stations 1 1 8
et 349, se compose de treize especes, dont les sept suivantes sont nouvelles : Homceo-
dicti/it ri'rrucosa, Mycalepellita, Tedania murdochi, Lissodendoryx buchanani, Clathria
toxipr&dita, Dcndoryx nodaspera et Raspaxilla phakellina, la derniere, type d'un
genre nouveau, et i'avant-derniere, type d'un genre ancien a denomination changee.
Des six autres especes, quatre etaient connues du Chili (Hymeniacidon fernandezi
Thiele), de la Patagouie (Mycale mugellanica Eidley, Stylostichon nobile patagonicum
Rdl. et D.) ou de 1'Antarctique (Tedania charcoti Tops.); une cinquieme, Toxo-
c//n/iiifi rolmsta Ridley, avait ete decouverte a Port- Jackson, puis retrouvee a Bahia ;
la sixieme n'est pas de terminable meme comme genre.

Le deuxihne lot de Monaxonides subantarctiques a ete obtenu bien loin du premier,
dans un dragage par 100 brasses de profondeur, a 1'ile Gough, Station 4(51. II comprend
quatre especes, dont trois nouvelles, Stylostichon, toxiferum, Bubaris murrayi et
Pseudosuberites exalbicans, celle-ci representant seule le sous-ordre des Hadromerina
dans la collection ; la quatrienie espece, Dictyociona discreta (Thiele), avait ete signalee
sur la cote du Chili, mais je la tiens pour le type d'un genre nouveau.

II me reste, pour terminer eet expose, a citer une Cladorhiza nouvelle, C. thomsoni,
qui, prise entre 1'ile Gough et Cape Town, au retour de 1'expedition, pent etre con-
sideree comme faisant partie de la faune Sud-Africaine.

I. S.O. HADROMERINA.

Famille SUBERITID^;.
Pseudosuberites exalbicans, n. sp. (PI. IV. fig. 5.)

Station 461, 22 avril 1904, He Gough, 40° 20; lat. S., 9° 56' 30" long. W. ;
profondeur, 100 brasses.

Les six specimens obtenus permettent de se faire une idee probablement exacte
de cette Eponge qui, a certains egards, rappelle sa congenere Pseudosuberites hyalinus
(Rdl. et D.) Tops.

De dimensions inegales et de formes capricieuses, ils incorporent tous beaucoup de
corps etrangers ; mais si les uns se reduisent a une mince couche spongieuse autour
d'un conglomerat, d'autres, d'une croissance plus vigoureuse, sont devenus massifs par
eux-memes. De ces derniers la partie superieure surtout est spongieuse ; elle est meme
assez molle et facilement declarable ; elle emet des lobes arrondis et greles, en n ombre
tout-a-fait restreint dans le specimen figure, mais qui, parfois, s'allongent, se lobent a
leur tour, s'anastomosent aveo les voisins et constituent en definitive une ramification
irreguliere, a rameaux minces, bossclrs et obtus. Nulle part les orifices aquiferes ne
sont apparents.

L'ectosome est une membrane incolore, translucide, detachable par grands lambeaux
et molle, malgre ses spicules de soutien. Ceux-ci se disposent dans son rpaisseur en

(ROY. soc. EDIN. TRANS., VOL. XLIX., 614.)

SPONGTAIRES DE L'EXPEDITION ANTARCTIQUE NATloNALE ECOSSATSE. 43

des faisceaux assez fournis qui, se couchant obliquement, s'y entrecroisent .sur divers
plans, sans former un n'-seau veritable. Les faisceaux s'e'panouissent en bouquets a leur
terminaisnn, et, comrne ils ne sout pas horizontaux, ils percent ca et la la surface
gencrale du corps. II en n'sulte pour elle une hispidation lache, tres courte et oblique.

Le choanosome, jaunatre, est charnu, a charpente haliclioudrio'ide, sans ordre et
pen dense.

Les spicules sont des tylostyles a liase bien degagee, pareille a celle dcs tvlnstyles
de Tfrpiosfiujn.fi (25, pi. vi. fig. 10), c'est-a-dire le plus souvent cordiforme en coupe
optique, quelquefois globuleuse aeuminee avec bourrelet annulaire ; leur ti^e.. Icgi-rement
courbee, a peine fusiforme, ne depasse guere en epaisseur la largeur de leur tete; leur
pointe s'etrile progressivement. Ils sont inegaux et varient entre 0'13 mm. de longueur
sur 0'0045 mm. d'epaisseur et 0'34 mm. sur O'Ol mm.

La forme de ces spicules et leur disposition dans 1'ectosome servent a distinguer
Pseudosuberites exalbicans des P. sulphweus (Bean) Tops, et P. «>n/r< -/r.sV Kirk p. :
leurs dimensions empedient toute confusion avec /'. lit/nHmix (Udl. et D.) Tops.

II. S.O. HALICHONDRINA.
Famille AXINELLID^E.

Hymeiuarnlon l'<jni« inlt'-.i, Thiele. (PI. III. fig. G.)

1905. Hymeniacidon fernandezi, Tliielc (20, p. 422).
Station 118, ler f6vrier 1904, Port Stanley, iles Falklands ; profondeur, C brasses.

Un specimen sans support, en croute epaisse. Je le prendrais volontiers pour un
Hymeniacidon caruncula Bow., tant pour son aspect que pour sa structure. Mais sa
coloration a Tetat de vie n'a pas ete note's et ses styles, tout en restant de meme forme
que ceux de H. caruncula, sont plus courts qu'eux. Ils ne depassent pas 0'3 mm.
de longueur sur 0'007 mm. d'epaisseur alors que chez des //. cnrn/icula de la Manclie,
ils atteignent 0'43 mm. sur O'OOS mm. On les trouve, d'ailleurs, de taille inegale de
part et d'autre dans un meme individu.

Comme les variations des styles de H. caruncula u'ont pas rtr r-tudiees, je crois
devoir rapporter V Hymeniacidon de la Scotia a 1'espece H. fernandezi Thiele. dmit les
specimens conserves etaient decolon's aussi. Je fais cependant des I'l'-serves au sujct
de cette identification car THIELE a releve des differences de taille entre les spicules
de ses II. f<'rn<ni<l,-.i plus grandes (pie celles que je viens d'indiquer entre les styles
de \'Hi/in<'iiit«-i'<lt>ii. de la Scotia et ceux de H. caruncula. En outre, 1'un dcs //.
fernandezi possedait des styles pareils a ceux de //. t-ubiginosa Thiele qui, vivante,
rappelait H. caruncula par sa coloration.

Ne peut-on pas se demauder si tout cela n'appartient pas a une seule et
meme espece ?

(ROY. soc. KHIX. THAXS., VOL. XLIS., 615.)

44 EMILE TOPSENT :

Bubaris murrayi, n. sp. (PI. III. fig. 1.)

Station 4G1, 22 avril 1904, Ik Gough, 40° 20' lat. S., 9° 56' 30" long. W. ;
profondeur, 100 brasses.

Cette Eponge affecte des caracteres exterieurs qui la rendent aisement reconnaissable.
Les quatre specimens qui en ont ete recueillis ne different entre eux que par leur taille.
II y en a un de me me grosseur que le plus petit de la figure 1, PL III., mais le quatrieme
est le plus volumineux de tous : avee une forme aussi reguliere que les autres, il mesure
4 centimetres de hauteur, 4 centimetres d'epaisseur et 5 centimetres de largeur.
Tous sont massifs, lobes, a lobes decomposes en de nombreux tubercules vaguement
arrondis, inegaux. Tous sont libres de tout support et ne paraissent pas avoir ete
attaches ; ils n'ont done certainement pas passe par un etat encroutant. Us sont d'un
blanc pur clans 1'alcool. Leurs diverses faces se montrent plus ou moins anfractueuses.
Les tubercules superficiels demeurent lisses, mais les plus proteges par leur position se
herissent d'une hispidation lache, assez haute. Entre les tubercules, 1'ectosome se tend
pellucide et luisant, au-dessus de cavites sous-dermiques spacieuses. II n'y a pas
d'orifices aquiferes distincts. La masse est de consistance ferme, peu compressible.

La charpente squelettique n'occupe que 1'axe de chaque tubercule sous forme de
solides faisceaux de spicules qui divergent pour se terminer en divers points de la
surface. Autour de ces faisceaux et entre eux est une chair abondante, en grande partie
collenchymateuse. Les faisceaux se composent surtout d'oxes flexueux unis par de la
spongine. Ces spicules, sans dimensions fixes, mesurent comme longueur ordinaire
0'7 mm., niais il en est de plus longs comme aussi de moitie plus courts ; de menie, ils
ont pour la plupart 0'014 mm. d'epaisseur environ, mais il s'en trouve dont 1'epaisseur
atteint 0'025 mm. Des sti/Jcs lisses peu combes, non renfles a la base, a pointe mal
aceree, les accompagnent mais deviennent surtout nombreux a I'extremit^ des faisceaux,
s'y disposant en bouquets laches qui causent 1'hispidation de la surface. La longueur
de ces styles varie entre 0'5 mm. et 1 mm. ; leur epaisseur est celle des oxes les
plus gros.

Je dedie cette espece a, 1'illustre oceanographe ecossais, Sir JOHN MURRAY.

Famille PCECILOSCLERIDJE.
Sous-famille Ectyoninse.

Genre Raspaxilla, n.g.

EctyoninsB dressees, stipitees, axinellidiformes. D'un reseau spiculo-fibreux axial
rayonnent vers la surface, proches les unes des autres, des colonnes plumeuses. Des
styles longs et lisses occupent I'interieur des fibres et, la pointe tournee en haut, com-
posent aux colonnes un axe grele ; des acanthostyles particuliers, tordus a la base,
epineux tout au plus dans leur moitie apicale, herissent abondamment reseau et
colonnes, sur lesquels ils s'appuient obliqucment. Au sommet des colonnes, les styles

(ROY. soc. EDIN. TRANS., VOL. XLIX., 61G.)

SPONGIAIRES DE L' EXPEDITION ANTAKCTIQUE NATIONALE ECOSSAISE. 45

de leur axe divergent en bouquet et rendent hispide la surface du corps tandis qii'autour
d'eux se disposent en faisceau d'autres styles, longs mais tenus. Pas de niicroscleres.

Raspaxil!« /<//<//.W//,,^, n.sp. (PI. I. fig. 4 ct PI. VI. fig. 15.)

Station 346, ler decembre 1903, Bane de Burdwood, 54° '25' lat. 8., 57 32' long. \V. ;
profondeur, 56 brasses.

Un specimen.

C'est une Eponge dressee, stipitee, ramifiee, a rameaux seusiblemeut etales dans un
rat'- me plan. Son pedicelle est ferine, court, subeylindrique, pen t'-largi en son point
d'attache au support. Ses branches, au contraire, sont souples et aplaties et n'atteignenl
pas 2 millimi'tres d'epaisseur. En les regardant par transparence, on apercoit au
milieu des quatre branches primaires une bande sombre qui emaiie du pedicelle,
constitue leur axe, se divise dans les branches secondaires mais progressivement
s'y efface. L'individu a une hauteur totale de 67 millimetres et une euvergure de
1 0 centimetres ; mais un long sejour dans un Hacon trop etroit 1'avait un peu euroule
en oublie, de sorte que, sur la photographic (PI. I. fig. 4), les deux moities du corps
ne semblent pas pareilles et que les rameaux de droite, vus obliquement et mal au
point, paraissent cylindriqu.es. Les deux faces se ressemblent ; elles sont hispides et
sans orifices apparents ; une chair unie, marquee de craquelures, les revet sur la plus
grande partie de leur etendue. Le tout est blanc creme, dans 1'alcool.

Le squelette de cette Eponge est tres caracteristique. L'axe des branches est un
reseau fibro-spiculeux a niailles allougees dans le sens de la hauteur ; la spongine,
abondante et jaune, le rend tres resistant ; elle y enveloppe, solitaires et clairsemes,
des styles longs, lisses et droits ; elle y supporte, appuyes sur ses fibres par leur base,
dans une direction plus ou moins oblique, d'abondants acauthostyles de type particulier.
Sur 1'axe ainsi constitue se dressent, gagnaut toutes la periplu'rie, et fort rapprochees
les unes des autres, des colonnes plumeuses dont le centre est occupe par une file asse/
grele de styles longs, lisses et droits, la poiute invariablement tournee vers I'e.xterieur ;
les acanthostyles precites entoureut ces colonnes et, s'y appliquant par leur base,
obliquemeut, la pointe vers la surface du corps, les rendent plumeuses. Au bout de
chaque colonne, les styles forment un petit bouquet divergent et rendent la surface
hispide. Enfin, autour des styles saillants au dehors, se dispose un faisceau de styles
lisses, remarquables par leur longueur et leur gracilite.

Spicidation. — I. Megascleres : 1. ,S7///V.s lisses (PI. VI. tig. I.")"), generalement
droits ou a peine courbes, a base ui renfiee ni tordue, a pointe terminee en un mucron
bref. Les plus superficiels d'entre eux, qui sont aussi les plus longs, atteignent
1'2 mm. a 1'4 mm. sur G'018 mm. d't'-paisseur a la base. Ce sont le- -pimles principaux
de 1'Eponge puisqu'ils se trouveut dans 1'epaisseur des fibres du reseau axial et qu'ils
constituent 1'axe des colonnes plumeuses. -. A<-«ntli<>*h/li'$ (PI. VI. fig. 156), re-
marquables en ce qu'ils ne portent jamais d'epines dans leur moitit'1 basale ; ces epines,
qui sont toujours faibles mais serrees, ne inanquent jamais completement sur toute leur

(ROY. soc. EUI.N. ntA>>., VOL. xux., 017.)

46 EMILK TOPSENT :

longueur, mais elles devienuent diffieiles a voir sur les plus grands d'entre eux tandis
qu'elles sont bien apparentes sur ceux de taille moyeune, les ornaut souvent sur toute
leur moitie distale. Ces spicules sont. en outre, to uj ours eourbes a pen de distance de
leur base, souvent fortement, souvent aussi brusquement, et frequemment assez pres
de cette base pour que celle-ci forme une crosse. Enfin, il n'est pas rare que la base,
destinee a s'im planter dans la spongine, se renfie un pen en une tete de subtylostyle.
Leur taille, variable, oscille entre O'lG mm. sur O'OOS mm. et 0'35 mm. sur O'OIS mm.
3. Styles (PL VI. fig. 15e), a base tres simple, a pointe excessivement fine, re-
marquablemeut greles, puisqu'ils mesurent O'G mm. de longueur sur 0'002 mm. seule-
ment d'epaisseur. lls se disposent par douze a quinze autour dcs styles super ficiels
qui herissent le corps, en faisceaux generalemeut epanouis dans 1'epaisseur de l'ectosome.

Pas de microscleres.

Par ses megascleres epiueux, herissant ses fibres axiales et ses colonnes radiales,
1'Eponge se revele comme une Ectyonine. Ses bouquets de styles greles, par leur
nature merne comme par leur disposition, rappellent ceux des Raspaiha, L'absence
totale de microscleres accentue le rapprochement. Mais la structure est bien differente
de part et d'autre, les Raspailia ne formant pas de colonnes plumeuses et n'ayant
d'acanthostyles, de type d'ailleurs banal, qu'a titre de complement. Ici, au contraire,
les acanthostyles, de forme particuliere, constituent de beaucoup la majeure partie du
squelette. Le reseau fibro-siliceux reste confine dans 1'axe des branches principales et
la charpente se compose surtout de colonnes plumeuses. L'epanouissement en bouquet
superficiel des styles qui forment 1'axe de ces colounes est un dernier trait qui distingue
les Raspailia de 1'Eponge de la Scotia, type, pour taut de motifs, du genre nouveau
Raspaxilla.

Genre Dictyociona, n.g.

Ectyoninse a charpeute composee d'un reseau fibro-spiculeux serre, a spongine
abondante dans la profondeur, puis de plus en plus rare vers la surface. Les mega-
scleres principaux et les megascleres herissants sont des acanthostyles ne differant entre
eux que par leurs dimensions ; les premiers, plus ou moins completement enveloppes
dans la spongine des parties profondes du squelette, se disposent vers la surface en
bouquets qui se touchent ; les seconds s'implantent isolement aux nceuds du reseau.
Les spicules ectosomiques sont monactinaux. II y a des microscleres, isocheles
et toxes.

Dictyociona discreta, (Thiele). (PL III. fig. 5.)
1905. Microciona discreta, Thiele (2O, p. 447, fig. 65).

Station 461, 22 avril 1904, He Gougli, 40° 20' lat. 8., 9° 56' 30" long. W. ;
profondeur, 100 brasses.

Un specimen,blanc dans 1'alcool.

THIELE, qui a vu le premier representant de cette espece, en a fait une Microciona.

Le specimen de la Scotia se compose de 80 a 90 tigelles dressees, cylindriques ou

(ROY. soc. EDIN. TRANS., VOL. XLIX., 018.)

SFONGIAIRES DE L'KXPEDITION ANTARCTIQUE NATIONALE ECOSSAISE. 47

un pen comprimees, souvent geminees, hautes an plus de 10 millimetres, s'elevant
d'une base commune, puis independantes, a de rares anastomoses pres, sur presque
toute leur hauteur. Une membrane pellucide tres mince etait tendue entre elles a
une petite distance au-dessous de leur sommet. Nul orifice aquifere n'etait distinct.

Ces tigelles, epaisses de 1 mm. a To nun., et fennes, out nne structure absolument
difterente de celle des Microciona et caracteristique d'un genre nouveau. Leur squelette
ne se compose pas, en effet, de colonnes plumeuses independautes et paralleles mais d'un
reseau fibro-spiculeux continu.

Dans la partie inferieure des tigelles et suivant leur axe, sur les trois quarts environ
de leur hauteur, la spougine est tres developpee et peut etre largement debordante ;
mais elle diminue graduellemeut d'importance vers la peripherie et vers le sommet des
tigelles. La charpente est serree, a mailles losangiques etirees vers la surface. Elle
peut s'etre constitute au centre de chaque tigelle aux depens d'un certain nombre de
fibres de rnerne valeur que celles des Microciona, mais alors de frequentes anastomoses
fibro-spiculeuses les ont unies en un reseau veritable et leur ont fait perdre toute
individuality ; en outre, vers la surface et sur une certaine epaisseur, les bouquets
spiculeux qui terminent ces colonnes plumeuses, rayonnant tout autour de la tigelle,
sont entres en contact et ont continue le reseau, devenu surtout spiculeux ; la direction
que prend la pointe des spicules est assez significative. Les spicules herissants
s'implantent surtout aux points d'entrecroisement des fibres et aux nceuds du reseau
plu.s superficiel. Dans son ensemble, une telle structure donne aux tigelles de 1'Eponge
de la rigidite et de la tenacite.

Speculation. — I. Megascleres : I. Subtylostyles ectosomiques pen nonibreux, droits,
a pointe courte, a base un peu rentlee elliptique et couromiee de quelques epines fort
petites ; ils mesurent 0'18 mm. a 0'24 mm. de longueur et 0'0043 d'epaisseur au-dessus
de la base. 2. Acanthostyles choanosorniques longs de 0'2l mm. a 0"28 mm., epais
de O'Olo mm. a O'OIS mm. au niveau du cou ; ils ont uue base renfiee, orm-e d't'-pines
nombreuses, generalement obtuses, droites ou un peu recourbees vers la tige ; la premiere
partie de celle-ci est lisse ou ne porte que des epines clairsemees, tandis que sa seconde
moitie s'arme d'epines abondantes, robustes et recurvees ; la pointe est ac£ree et peu
epineuse. 3. Acanthostyles herissants, exactement de mo me type que les precedents,
mais touj ours sensiblement plus petits, puisqu'ils ne mesurent que 0'09 mm. a OM2 mm.
sur 0'008 mm. au niveau du cou.

II. Microscleres : 4. Tsochdles nonibreux, remarquables par leur exiguite, leur
longueur ue depassant pas O'OOS mm. 5. T».rex lisses. clairsemes, fortement arques,
a bouts pointus ; leur envergure attaint 0'07 mm. a 0'08 mm.

Cette spiculation complicjuee ressemblc a tel puini, taut par les details de conformation
ijue par les dimensions de ses cinq sortes dY-lenients, a cellc de la M«-r</c/<n/<i discrete
de THIKLE qu'on ne peut douter avoir affaire a tin second individu de 1'espece. THIELE
a remarque que les acanthostyles priucipaux ue sont pas pareils a ceux des MnTociona.
Ils rappellent les acanthostyles herissants des Rhaphidophlus et de certaines Clathr'w

(ROY. SOU. KI'I.\. TKANS., VOL. XLIX.,

48 EMILE TOPSENT :

et me conduisent a rapprocher de ces genres d'Ectyonines le nouveau genre Dictyociona ;
il s'eu distingue, d'.iilleurs, aisement par ce fait que megascleres principaux du choano-
some et megascleres herissants sont chez lui d'un meme modele.

Clathria toxiprtedita, n. sp. (PL V. fig. 4 et PL VI. fig. 12.)

Station 346, 1" decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Cette espece est nettement caracterisee par la possession simultanee d'isocheles de
deux sortes et de toxes tres abondants, d'une longueur inaccoutumee.

Elle est representee dans la collection de la Scotia par un seul specimen, blanchatre,
long de 7 centimetres, large de 6 centimetres, epais de 25 millimetres. II s'est developpe
sur un cote d'un fragment de polypier rameux, qui devait reposer en long sur le sol:
Une Tedania a gene sa croissance mais se trouve en majeure partie enclavee dans sa masse.
Quoique massif, il doit etre manie avec une certaine precaution, parce qu'il se compose
de colonnes dressees, d'inegale grosseur, plutot greles, lachement anastomosees entre
elles et fragiles. Ces colonnes, tronquees a leur extremite libre, aboutissent toutes
sensiblement a la meme hauteur et se presentent a la surface comme des lobes aplatis
ou polygonaux, limites par de profondes crevasses. L'ectosome ne forme pas de
membranes detachables.

Speculation. — I. Megascleres: 1. Subtylostyles ectosomiques, droits, a tete ovale,
assez renflee, large de 0'007 mm., ornee en son sommet d'un tout petit nombre d'epines,
difficiles a voir ; ils mesurent 0'31 mm. a 0'32 mm. de longueur sur 0'005 mm.
d'epaisseur. 2. Styles des fibres, lisses, courbes dans leur tiers basilaire, legerement
renfles en tete arrondie a un bout, termines en pointe breve a 1'autre ; relativement
gros pour leur longueur, ils mesurent 0'55 mm. a 0'65 mm. sur 0'03 mm. 3. Acantho-
styles herissant les fibres, assez clairsemes, generalement droits, entitlement epineux,
pointus, a base renflee, longs de O'l mm. a 0'155 mm., epais de 0'009 mm. au niveau de
la tete et de 0'007 mm. au-dessous d'elle.

II. Microscleres : 4. IsochMes palmes, du type des isocheles habituels des Clathria
(PL VI. fig. 12c), avec une petite pointe a chaque bout; ils sont longs de 0'0155 mm.,
larges de 0'003 mm. et tres nombreux. 5. Isocheles palmes (PL VI. fig. 126), longs
de 0'022 mm., larges de 0'007 mm. a O'OOS mm., munis d'un tubercule sur chaque dent
et remarquables surtout parce que leur tige avec ses ailes s'arrondit, tandis que leurs
deux dents, s'ecartant beaucoup en avant d'elle, viennent ou pen s'en faut, au contact
1'une de 1'autre; le spicule, vu de profil, affecte la forme d'une navette large de O'Ol
mm. en son milieu, a bouts largemeiit emousses. Ils sont abondants, sans qu'aucun
intermediaire de forme ni de taille existe entre eux et ceux de la categoric precedente.
G. Toxes lisses (PI. VI. fig. 12a), excessivement nombreux et de toutes dimensions,
depuis O'l mm. d'envergure sur 0-0003 mm. d'epaisseur jusqu'a 175 mm. sur 0'007 mm.
Leur courbure varie aussi avec leur taille ; en general, elle n'est accusee qu'en leur centre

(ROY. soc. EDIN. TKANS., VOL. XLIX., 620.)

SPONGIAIRES DE L'EXPEDITION ANTARCTIQUE NATIONALE ECOSSATSE. 49

et leurs deux moities se trouvent sur la majeure partie de leur longueur dans le prolonge-
ment 1'une de 1'autre. Les plus grands de ces spicules ne se revMent com me toxes que p.-ir
une 16gere inflexion situee en leur milieu ; leurs pointes sont aigues, mais assez breves.

Stylostichon toxiferum, n. sp. (PI. IV. fig. 7 et PI. VI. fig. 14.)

Station 461, 22 avril 1904, He Gough, 40° 20' lat. S., 9° 56' 30" long. W.;
profondeur, 100 brasses.

Un specimen massif, brun clair, tres compressible, de dimensions a peu pres e"gales
dans tous les sens, developp^ tout autour d'un caillou qui se trouve maintenant occuper
sa base. II n'a pas d'orifices distincts. Son ectosome, lisse et translucide, se detache
avec facilite. Son choanosome a une charpente de longues lignes squelettiques brunes,
peu serrees, a structure typique de Stylostichon. La coloration generale est due en
partie a de la spongine jaune fonce qui renforce les colonnes spiculeuses et en partie a
des cellules spheruleuses brunatres, tres abondantes, grosses de O'Ol mm. a O'Ol 2 mm.,
a spherules petites.

Spiculation. — I. Megascleres : 1. Tylostyles composant 1'axe des colonnes; longs
de 0'5 mm., lisses, a base assez renflee (0'015 mm.), a tige epaisse (0'012 mm.) et
courbee dans son tiers basilaire, a pointe breve, aceree, ils offrent bien 1'aspect
d'acanthostyles dont les epines se seraient effacees ; quelques uus, d'ailleurs, en con-
servent un petit nombre sur leur base. Ils se groupent par trois a cinq, de sorte que
les colonnes demeurent greles. La plupart d'entre eux conservent une direction
generale ascendante, la pointe tournee vers la surface de 1'Eponge, mais il s'en trouve
toujours un certain nombre qui s'implantent obliquement sur la fibre par leur base.
2. Acanthostyles herissants des fibres, de forme banale, droits ou peu courbes, a base
renflee sans exaggeration, entierement couverts d'epines faibles et legcrement recurvees ;
ils varient de 0'07 mm. a 0'2 mm. de longueur sur 0'004 mm. a O'OOS mm. d'epaisseur
au-dessus de la base ; ils forment un angle tres ouvert avec les fibres, sur lesquelles ils
s'implantent en grand nombre. 3. Subtylostyles ectosomiques, lisses a 1'exception de
leur base qui, de forme allongt^e, porte un groupe d'epines toutes petites en sou sommet ;
cette base est le plus souvent legerement tordue sur la tige, qui est generalemeut droite
et progressivement effilee, longue de 0'32 mm. a 0'57 mm., epaisse de 0'004 mm. a
O'OOS mm. Ces spicules, tres abondauts a la surface du corps, s'y pressent en bouquets
divergents qui s'epanouissent dans 1'ectosome et s'y couchent sans determiner la moindre
hispidation.

II. Microscleres : 4. Toxes de deux sortes, les uns lisses (PI. VI. fig. 14), a centre
dessinantune courbe ouverte, arrondie, a bouts doucement reflechis, pointus ; ils mesurent
de 0'04 mm. a 0'06 mm. d'euvergure, le plus couramment 0'05 mm. et O'Ol 4 mm.
d'epaisseur en leur milieu ; les autres plus grands (0'085 mm.) mais toujours bien plus
greles, rugueux vers leurs extremitt's. Ce sont les seuls microscleres presents, mais
ils existent, les premiers surtout, en abondance.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 621.)

50 EMILE TOPSENT :

Stylostichon nobile var. patagonicum, (Eidley et Dendy) Topsent.
1887. Myxilla noli/is var. patayonica, Eidley et Dendy (11, p. 142, pi. xxvii. fit,'. 13).

Station 346, ler decembre 1903, Bane cle Burdwood, 54° 25'lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Un specimen informe, sans support, assez volumineux, car il est long de 95 milli-
metres, large de 60 millimetres, epais de 35 millimetres. Sa surface, irre"guliere,
rugueuse, est parcourue par des sillons si nombreux qu'ils la decoupent en une infinite
de lobules rappelant ceux de Clathria toxiprsedita (PI. V. fig. 4), mais plus etroits.
Pas d'orifices distincts. Coloration gris fonce, Consistance fragile.

Les caracteres de 1'ectosome et du choanosome se retrouvent ici tels que RIDLEY et
DENDY les ont decrits. Les fibres squelettiques ne gardent d'individualite que sur une
faible longueur, mais, d'apres leur existence meme, 1'Eponge doit etre extraite du genre
Maxilla stricto sensu et rattachee au genre Stylostichon.

RIDLEY et DENDY ont cru remarquer 1'existence d'epines sur la tete des tylotes.
Leur observation se verifie aisement dans le specimen de la Scotiit : chacun des
renflenients de ses tylotes se r^treeit pour se terminer par deux toutes petites epines
pareilles a un double mucron.

Des sigmates se rencontrent dans mes preparations, mais j'ai tout lieu de supposer
qu'ils proviennent de specimens de Mycale pellita conserve's avec le Stylostichon.

Les mesures des spicules sont sensiblement les memes que dans le type, qui provenait
du S.W. de la Patagonie.

Genre Dendoryx, J. E. Gray (emend.).

GRAY, en 1867, a etabli le genre Dendoryx (2, p. 535) avec cette diagnose : " Sponge
massive, irregularly reticulated. Spicules of four kinds: 1. Fusiform or cylindrical,
pointed at each end, smooth. 2. Needle-like, spinulous all over. 3. Equibianchorate,
three-spined at each end. 4. Bihamate." II y rattachait huit especes de BOWERBANK
noyees par leur auteur dans le genre Halichondria, et H. incrustans Bow., inscrite en
tete de la liste, devenait le type du genre Dendoryx.

La diagnose ecrite par GRAY etant tellement vague qu'elle 1'avait conduit a grouper
des Eponges tres differentes les unes des autres, j'ai cherche a plusieurs reprises a la
preciser d'apres la structure et la spiculation de son type, et je suis arrive, notammemt
en 1892 (23, p. 97), a la tracer de la sorte : Esperellinx ordinairement sans formes
definies ; megascleres de 1'ectosome ordinairement diactinaux (tylotes, strongyles ou
tornotes), rarement monactinaux, normalement lisses ; megascleres du squelette, in-
variablement styles epineux.

Je faisais remarquer en meme temps que les Dendoryx se distinguaient des Myxilla
par 1'absence de spicules herissant la charpente squelettique. Au genre Dendoryx ainsi
compris se rattachaient des especes nombreuses.

Mais THIELE, en 1903 (19, p. 953), puis LUNDBECK, en 1905 (7, p. 132), ont object^
que la diagnose pre'cedente convenait au contraire au genre Myxilla Schmidt, 1862,

(ROY. soc. EDIN. TRANS., VOL. XLIX., 622.)

SPONGIAIRES DE L/EXPEDITION ANTARCTTQUE NATIONAI-K ECOSSAISE. 51

clout 1'espece type etait M. rosacea Liebk., a squelctte reticule fail d'aeanthostyles et
non herisse d'acanthostyles plus petits. LUNDBECK a conserve dans le genre Miixilln
reduit a cette expression les especes pourvues d'isancres ; il a constitue le genre
Lissodendoryx Topsent (emend.) avec les especes a isocheles.

Le genre Myxilla avait joui a un moment donne, d'une comprehension tres large.
Ainsi RIDLEY et DENDY (ll) appelaient Myxil/n des Eponges a sijuelette reticule
d'acanthostyles, non herisse, com me M . ms-</fm, d'autres a squelette reticule d'acantho-
styles, herisse, comme .17. mariana, d'autres encore, a squelette en fibres herissers,
comme M. nobilis et M. frondosa. Les premieres restent dans le genre Mi/xif/n stricto
sensu ; les dernieres sout des Stylostichon. Que faire des autres?

De la liste des especes inscrites par GRAY dans son genre Dendoryx, la seconde,
D. thompsoni, est mal connue, mais, a cause des acanthostyles qui sont cites seuls
comme megascleres de son squelette et des styles, indiques comme megascleres des
membranes, elle doit rentrer dans le genre Myxilla au meme titre que />. incrustans,
dont elle a les microscleres, " bidentate, equianchorate," c'est-a-dire les isancres. I .a
troisieme espece, D. albula, est une Grayella, a moins que, pour elle, on ne tienne, ce
que je deplorerais, pour ma part, a faire tomber le genre Grayella en synonymic du
genre Dendoryx. La quatrieme, D. irregulaiis, rentre dans la categoric des Ectyouines
a squelette reticule, telles que Mi/xiUa mariana, non encore classees. Je propose de
considerer, par consequent, cette D. irregwlaris (Bow.) Gray comme le type du genre
Dendoryx Gray, rnais pris dans une acception nouvelle ; une simple interversion de
noms donuera de la sorte satisfaction aux reclamations resultant de la priorite incontes-
table de Mi/.i-illd sur Dendoryx au sens ou je 1'avais pris jusqu'ici.

Les Dendoryx seront, d'apres cela, des Ectyoninas possedant pour megascleres
ectosomiques des spicules diactinaux (ou pouvant passer pour tels), le plus ordinairement
lisses ; pour megascleres choanosomiques des acanthostyles (peut-etre a 1'occasion des
styles lisses ou a epines rares), disposes en un reseau plus ou moins n'-gulier, aux noeuds
duquel s'implantent des acanthostyles plus courts, herissants. Les microscleres sont
des isocheles (non pas des isancres), accompagnes ou non de sigmates ou de toxes.

L'Eponge suivante devient pour ces raisons :

Dendoryx nodaspera, u. sp. (PL VI. tig. 3.)

Station 346, 1" decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. \\ .:
profondeur, 56 brasses.

Cette espece ressemble beaucoup par sa spiculation a Dt'infuri/x comjiresiia (Kdl.
et D.) (ll, p. 139); mais ses megascleres sont plus petits, ses sigmates lineaires se
contournent rarement et surtout ses isocheles affecteiit une forme toute difft'-rente, plus
semblable a celle des memes microscleres d'uu Stylostichon <|ui lui est sur plus d'un
point comparable, & repens Tops. (27, p. 170).

Le specimen unique qui la represente, embrassaut un tube de Serpule, est une petite
Eponge massive, grise, assez ferme, longue de 25 millimetres, large de 10 a 20 milli-

(ROY. soc. EDIN. TRANS., VOL. XLIX., 623.)

52 EMILE TOPSENT:

metres, epaisse de 12 millimetres. Sa surface se montre lisse sur une certaine etendue,
dans nn enfoncement qui parait etre le moule d'un corps etranger car elle est, a ce niveau,
dure et imperforee ; partout ailleurs, elle est irreguliere, marquee de petites crevasses
sinueuses ou punctiformes ; nulle part elle ne porte d'ectosome distinct. Des cassures
rnettent a nu 1'interieur, oa des canaux aquiferes etroits mais nombreux se croisent en
tous sens. L'etat du specimen ne permet pas de reconnaitre ses oscules.

La charpente choanosomique forme un reseau tres net a mailles triaugulaires, dont
les cotes, ou plusieurs acanthostyles peuvent se placer cote a cote, ne mesurent qu'une
seule longueur de spicule ; aux noeuds du reseau se dressent un acanthostyle herissant
ou deux, rarement davantage. La densite de ce reseau explique la consistance de la
masse. II ne se dessine pas de fibres, de sorte que le type de structure est tres pur.
Enfin, comme il n'existe pas de spongine d'union des megascleres, la charpente demeure
assez fragile.

Spiculation. — I. Megascleres: 1. Tornotes ectosomiques droits, peu abondants,
epars ; ils ressemblent tout-a-fait a celui qui a ete figure de Dendoryx compressa
(11, pi. xxvii. fig. 9c), car 1'un de leurs bouts se renfle et se termine par un mucron ;
ils ne mesurent que 0'145 mm. a 0'16 mm. sur 0'004 mm. 2. Acanthostyles choano-
somiques du reseau, droits ou a peine courbes, courts et epais, assez uniformes, longs de
0'16 mm. a 0'165 mm., epais de 0'009 mm. a O'Ol mm. a la base, sans compter les epines ;
ils ont, au contraire de ceux de D. compressa, une base toujours renflee, avec des epines
fortes et longues qui, avant leur terminaison, se coudent brusquement vers le bas, c'est-
a-dire dans la direction de la tige ; celle-ci est entierement epineuse, sauf en sa pointe,
qui est courte et aceree, et ses epines sont hautes, coniques et droites, pas plus serrees
que chez D. compressa. 3. Acanthostyles herissants, droits, de meme type que les
precedents et n'en differant que par leurs dimensions et par le fait que les epines de leur
tige se retroussent vers leur base ; leur taille est assez uniforme aussi, car ils mesurent
0'075 mm. a 0'083 mm. de longueur sur 0'005 mm. a O'OOG mm. d'epaisseur a la base,
sans compter les epines.

II. Microscleres : 4. Isochcles abondants, a tige fortement courbee, a ailes et a dents
larges, fort in^gaux, depuis 0'013 mm. jusqu'a 0'04 mm. de longueur; les plus grands
ont une tige epaisse de 0'004 mm. La figure 3a de la planche VI. montre a quel point
ils different des microscleres correspondants de D. compressa, dont ils ont pourtant les
dimensions. 5. Sigmates lineaires, mesurant O'OIS mm. a 0'02 mm. de corde, tres
nombreux, rarement contournes, le plus souvent arques dans un plan et tres arrondis
(PI. VI. fig. Sb).

Des caracteres distinctifs trop importants se relevent entre cette Dendoryx du Bane
de Burdwood, recueillie par une faible profondeur, et les D. compressa qui vivaient par
600 m. a 1'embouchure du Rio de la Plata, pour que je me decide a identifier ces Eponges.
Pourtant, je ne serais pas surpris que de nouveaux materiaux provenant de localites
et de profondeurs intermediaires diminuassent un jour a nos yeux la valeur de ces
differences.

(BOY. soc. EDIN. TRANS., VOL. XLIX., 624.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATIONALS ECOSSATSK. 53

Sous-famille Myxillinae.
Myxilla spongiosa, Ridley et Dendy, var. asigmata, Topsent. (PI. III. fig. 3.)

1901. Lissodendoryx xpongiosa (Rdl. et D.), var. asigmata, Topsent (26, p. 18).

Station 417, 18 mars 1904; 71° 22' lat. S., 16° 34' long. W. ; profondeur,
1410 brasses.

J'ai inscrit, en 1901, cette Eponge dans le genre Lissodendoryx a cause de ses
styles li.sses.

Depuis, LUNDBECK a montre (7, p. 153) qu'il serait plus naturel de reserver le genre
Lissodendoryx aux Eponges voisines des Myxilla qui possedent des isocheles au lieu
d'isancres ; dans les deux genres ainsi compris, il existe des especes a, styles lisses et
d'autres a acantliostyles.

La maniere de voir de LUNDBKCK me parait d autant plus acceptable que les styles
generalement lisses de certaines formes de Lissodendoryx, au sens priniitif, peuvent a
1'occasion s'orner de quelques epines. L'Eponge en question en ofi're precisement
des exemples.

II faut done la rapporter au genre Myxilla, a cause de ses microscleres, qui sont des
isancres. Mais je la tiens pour une bonne variete de 717. spongiosa oil les sigmates font
defaut. Je 1'ai rencontree, en eftet, dans les collections de la Bdgica, du Frangais et
de la Scotia, recueillie par des profondeurs tres differentes et en des points bien eloignes
les uns des autres. J'y constate des variations, mais je n'y decouvre pas la moindre
trace de sigmates. D'autre part, KIRKPATRICK (3, p. 28) a revu des specimens de
Myxilla spongiosa Rdl. et D. ; il a observe des sigmates chez tous, quoique en
proportion variable ; il ne dit pas que ces spicules s'y soient trouves tres rares.

La ressemblance des megascleres de deux sortes et des isancres de Myxil/'t xj-H>n</x»«t
et de ceux de ce que j'appelle M. spongiosa asigmata est vraiment trop grande pour
qu'on puisse regarder cette derniere comme une espece distincte.

II s'agit evidemment d'une Eponge polymorphe. Le specimen qu'en a recueilli la
Scotia est gris fonce, dresse, comme pedicelle. Son support manque. Sa surface,
irregulit're, se couvre d'un ectosome membraneux tres mince au travers duquel se voient
par transparence de larges orifices, separes par d'etroites bandes de substance, dont
['ensemble dessine un reseau. L'ectosome est tendu aussi, sur la face qui a (''ti'1
photographiee, au-dessus de longs canaux exhalants superficiels.

Spiculation. — I. Megascleres : 1. Tylotes ectosomk^ues droits, un peu fusiformes,
a bouts un pen rennYs et termines par un faisceau de courtes pointes paralleles, ;i
1'imitation de ceux de Io-phon unicornis ; ils mesurent 0'375 mm. a 0'4 mm. de
longueur sur O'Ol mm. d'epaisseur. 2. Styles choanosomiques robustes, courbes, a
pointe breve, longs de 0'87 mm. a 0'9 mm., epais de 0'02G mm. a 0'028 mm.

II. Microscleres : 3. Isancres abondantes, longues de 0'063 mm. a 0'075 mm.

Si Ton compare les dimensions de ces spicules a celles relevces sur les specimens de

(ROT. soc. EDIN. TKA.NS., VOL. XLIX., 625.)

54 EMILE TOPSENT :

la Belgica et du Francais et aussi a celles des spicules de Myxilla spongiosa, on n'y
releve que des differences pen sensibles. Les tylotes sont exactement de la taille de
ceux de Myxilla spongiosa du Challenger. Les styles sont ici un peu plus grands que
p.irtout ailleurs mais leur grosseur etait souvent atteinte dans le specimen du Frangais;
tout 1'interet en ce qui les concerne se borne a ce que leur base est quelquefois ornee
d'epines a peine distinctes et strictement localisees en son plateau. Les isancres sont
grandes a peu pres comme celles du specimen de la Belgica ; le nombre de leurs dents
reste le plus generalement de trois a chaque bout, mais frequemment il s'eleve a quatre
et parfois meme a cinq ; cette variation interessante coincide avec une disparition
absolue des renflements superieur et inferieur de la tige, qui se montraient cependant
avec tant de nettete sur les microscleres <le ce meme specimen de la Belqiea et qui
constituaient une ressemblance si intime avec les isancres de Myxilla spongiosa.
En revanche, la tige de ces microscleres, courbee, est ici epaisse de 0'006 mm. a
O'OOS mm., alors qu'elle etait plutot grele et plus droite sur ceux du specimen du
Frangais.

Lissodendoryx buchanani, n. sp. (PI. I. fig. 5 et PI. VI. fig. 7.)

Station 346, lerdecembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Un specimen.

L'espece est curieuse a la fois par sa forme et par sa spiculation, aussi me fais-je un
plaisir de la dedier a M. BUCHANAN, qui s'est tant interesse a 1'expedition de la Scotia.

Le specimen, obtenu sans son support, se compose de cinq troncs dresses qui se
ramifient irregulierement a plusieurs degres et contractent quelques anastomoses entre
eux par leurs rameaux. II est tout blanc dans 1'alcool. De consistance ferme, il a,
une surface lisse, sans orifices distincts.

La charpente choanosomique est faite de bandes spiculo-fibreuses ascendantes,
polyspiculees, oil la spongine existe en bonne proportion ; des spicules solitaires se
repandent sans ordre dans leurs intervalles. Le squelette de 1'ectosome consiste en des
bouquets dresses, equidistants, de spicules propres.

La structure est certainement un peu speciale, mais elle est en rapport avec la forme
generale du corps ; elle n'empeche pas de rapporter 1'espece an genre Lissodendortjx,
car il est admissible que la structure des Lissodendoryx preseute les memes variations que
celle des Myxilla. Ce qui ferait hesiter a placer notre Eponge parmi les Lissodendoryx
plutot que parmi les Dendoricella Lundbeck, c'est 1'aspect diactinal de ses megascleres
choauosomiques, mais je me crois fonde a les prendre pour des spicules monactinaux.

Spiculation. — I. Megascleres: 1. Tornotes ectosomiques, droits, a pointes acerees ;
un de leurs bouts est assez souvent un peu plus epais que 1'autre, mais sans la moindre
trace d'ornementation. Leurs dimensions varient entre 0'415 mm. et 0'46 mm. de
longueur sur O'OOS mm. a O'Ol mm. d'epaisseur. Les bouquets qu'ils forment, distants
les uns des autres d'environ OM5 mm., en comprennent une trentaine et s'epanouissent

(HOY. soc. EDIN. TRANS., VOL. XLIX., 626.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATTONALE ECOSSA1SK. 55

legerement du cote distal ; leurs pointes ne depassent pas ou depassent a peine la
surface generale du corps. 2. Pseudoxcs clioanosomiques (PI. VI. fig. 7«), pas plus
gros que les tornotes de 1'ectosome mais plus longs qu'eux (0'48 mm. a 0'55 mm.),
reconnaissables en outre a ce qu'ils sont ton jours courbes ou me.me fiexueux et a ce
que leurs extremite"s sont dissemblables ; 1'une d'elles reste simple, assez breve, souvent
precedee a distance d'une inflexion plus ou moins marquee de la tige ; 1'autre obtuse,
s'orne assez loin de sa terminaisou d'une epine courte ou de plusieurs, parfnis disposees
comme en un verticille ; les opines se placent frequemment aussi tout au bout de la
tige et, par leur ensemble, formeut une tete noueuse. Cette ornamentation, variable
mais constante, semble indiquer que les spicules en question sont reellement des
megascleres monactiuaux, des acanthostyles deguis6s ; cela est d'autant moins douteux
que leur courbure principals s'etablit gene"ralement au niveau de leur tiers le plus
rapproche de ce qui doit representer leur base.

II. Microscleres: 3. Isocheles (PI. VI. fig. 7b), arques, abondants, longs de
0'033 mm. A 0'037 mm., a tige epaisse de 0'0045 mm.

lophon pluricornis, Topsent. (PI. VI. fig. 9.)

1908. lopJiuii pluricornis, Topsent (28, p. 29).

Station 325, avril 1903 et juin-juillet 1903, Scotia Bay, 60° 43' 42" I at. S.,
44° 38'' 33" long. W. ; profoudeur, 9 a 10 brasses.

Deux specimens.

Ce sont deux Eponges amorphes, enlacant des Algues et des Bryozoaires. L'une
d'elles, prise d'abord, plus petite, se fait remarquer par sa teinte assez claire, 1'autre
etant, comme les specimens du Frangais, d'un bruu noiratre.

L'espece est bien reconnaissable a sa spiculation. Ses acanthostyles ont habituelle-
ment un certain nombre d'epines a leur base ; quand ils n'en portent qu'uue, elle ne se
di'-veloppe pas autant que 1'epine unique de 1. <mi<-orriis ou de I. fldbello-digitatus ; il
peut s'en trouver, au moins dans certains specimens, comme le second provenant de
Scotia Bay, qui n'en portent pas du tout et presentent alors une base reguliere de styles.
Les tylotes, courbes, fusiformes, ont des tetes bien renflees, ornees d'epines qui, generale-
ment, se clistribuent sur elles loin en arriere de leur sommet ou meme couvrent toute
leur surface (PI. VI. fig. 9h), de sorte que, des deux dispositions possibles que j'ai
decrites (28), la seconde est reellement typique ; cela est encore un bon caractere pour
distinguer /. pluricornis des autres lophon de 1'Antarctique. Les anisocheles sont de
taille inegale et atteignent 0'0:33 mm.; Fenroulement des plus petits, que j'ai note
precedemment, parait etre exceptionnel etsans interet. Ce qui est plus important, c'est
la forme des bipocilles (PI. VI. fig. 9«), qui ont les extremites en cuilleron, paralleles,
tridentees, et ne ressemblent par (.•nnsiMjuent pas aux microscleres eorrespondants de
lophon flabello-diyitatus ; ces bipocilles sont abondants dans tous les specimens que
j'ai examiues.

(ROY. sou. EDIN. TRANS., VOL. XLIX., 027.)

56 EMILE TOPSENT :

lophon unicornis, Topsent. (PL III. fig. 9.)
1908. lophon unicornis, Topsent (28, p. 27).

Station 325, mai, juin, juillet et aout 1903, Scotia Bay, 60" 43' 42" lat. S.,
44° 38' 33" long. W. ; profoadeur, 9 a 10 brasses.

Plusieurs specimens, massifs, amorphes, a surface irrtSguliere, entremeles d'Algues.

Les acanthostyles possedent constamment 1'epine unique, puissante, droite on un
peu tordue, apicale ou oblique, qui vaut sen nom a 1'espece. Les Opines des tylotes se
localisent sur unc sorte de plateau terminal retreci, de sorte que les extremites de ces
spicules ressemblent un peu a des bourses aux bords fronces ; les deux extre'mite's d'un
meme tylote sont souvent un peu inegales mais elles ne sont pas a proprement parler
dissemblables. Les anisocheles, extremement nombreux, sont de taille assez uniforms
(autour de 0'02 mm. de longueur). Je n'ai, en revanche, pas pu decouvrir de
biponilles.

Comme on doit s'y attendre, il se produit des variations individuelles dans les
dimensions des megascleres ; ainsi, les tylotes de 1'un des specimens recueillis en mai
1903 atteignent couramment 0'36 mm. de longueur sur 0'012 mm. d'e"paisseur.

lophon spatulatus, Kirkpatrick.
1908. lopJi on spatulatus, Kirkpatrick (3, p. 29).

Station 325, avril 1903, Scotia Bay; profondeur, 9 a 10 brasses.

Je rapporte a cette espece quelques fragments de forme rameuse, a branches cylin-
dracees, translucides, de couleur simplement ambree. Mais je ne considere pas ma
determination comme certaine parce que je n'ai pas reussi a voir de bipocilles. Ces
microscleres, qui, outre les caracteres exte'rieurs, doivent permettre de distinguer de
rnon lophon unicornis 1'espece decrite peu de temps apres par KIRKPATRICK sous le nom
de /. spatulatus, sont, d'ailleurs, declares rares par cet auteur. Cela rend difficile a
reconnaltre /. spa tula f us, qui pourrait, a tout prendre, n'etre pas davantage qu'une
vari6t6 de /. unicornis.

lophon flabello-digitatus, Kirkpatrick. (PI. III. fig 7 et 8.)

1908. loplwn flabello-digitatus, Kirkpatrick (3, p. 30).

Station 325, mai, juin, juillet et aout 1903, Scotia Bay, G0° 43' 42" lat. S.,
44° 38' 33" long. W. ; profondeur, 9 a 10 brasses.

Plusieurs specimens.

Au lieu de ressembler aux specimens de la Discovery d'apres lesquels 1'espece a ete
nominee, ceux de la Scotia sont amorphes et ne se distiuguent pas exterieurement des
InfiJioit iiiiicornix avec lesquels ils ont 6td recueillis. Leur coloration, ordinairement
foncee, varie d'un inclividu a 1'autre, et il s'en trouve un aussi pule que les fragments

(ROY. soc. EDIN. TRANS., VOL. xr,ix., 628.)

SPUNGIAIRliS DE [/EXPEDITION ANTAKCTIQUK NATION A 1, 1C K( '( >SSA ISK. 57

que j'ai rapportes a lophon t<i>«/ii/ittt(.'*. II existe done, chez cette espece comme chez
beaucoup de PoeciloscMndes, un polymorphisme tres accuse.

Les acauthostyles rappellent a s'y meprendre ceux dc /«/</M<// nnir<>i-ins; les
dimensions indiquees de part et d'autre ne constituent pas line diHrrciire ire lie entre
eux ; je leur ai trouve, par exeniple, 0'6 mm. sur O'Olo mm. chez un /. Jlabello-
digitatus et U'415 mm. a 0'43 mm. sur 0'U2 mm. chez uu autre, tandis que ceux d'un
/. unicornis me donnaient 0'525 mm. sur O'OIS mm.

II se produit des variations individuelles semblables dans la taille des tylotes. Ces
megascleres ectosomiques, tels que les a vus KIRKPATRICK, oii'rent une particularite 4111',
dans uue certaine mesure, pent guider la determination : leuis tetes ne portent
dVpines qu'eu leur sommet, a la faoon de celles des tylotes d' 'lophon unicornis; seule-
ment, 1'uue de ces tetes. plus etroite que 1'autre, degage du centre de sa couronne
dV-pines un fort mucrou. Or, ce caractere manque de fixite : present sur tous les
tylotes de certains specimens, il s'etface sur une partie des memes spicules de certains
autres et, chez d'autres encore, fait totalement defaut ; les tylotes ressembleut alors
beaucoup a ceux de lophon unicornis.

Des microscleres, ce sont ces bipocilles que KIRKPATRICK a dit spatules, qui sout le
plus caracteristiques ; je les ai trouves nombreux partout et c'est a eux, quaud, en
meme temps, les tylotes etaient mucrones a un bout, que j'ai d'abord reconnu 1'espece.
Les grands anisocheles doivent etre rares car je n'en ai pas vu ; les autres ont la plus
"•rande ressemblance avec ceux de /. unicornis.

o

Tedania murdochi, n. sp. (PL V. fig. 5.)

Station 118, ler fevrier 1904, Stanley, lies Falklands ; profondeur, 6 brasses.

Un magnifique specimen, debout sur une coquille, dans la position oil on le voit
figure. C'est une Eponge gris-jaunatre, peu fragile, a pcine compressible, baute de
12 '5 centimetres, large de 6 a 7 centimetres. Une perforation verticale, visible sur la
photographic, vers la gauche et un peu au-dessous du milieu de la hauteur, un rameau
qui se dirige en avant, a droite en haut, une lame qui fuit en arriere, au bord droit du
specimen, un lobe rampant du meme cote, en bas, le long du support, tout cela iudique
que cette Eponge est de nature rameuse, mais que ses rameaux, epais, se trouvent ici
en majeure partie concrescents ; ils torment ainsi une lame dressee, pliee le long d'uue
crete verticale, du sommet de laquelle se detache le rameau precite. Son epaisseur
moyenne est, comme celle de ses lobes, d'environ 15 mm.

La surface est unie et d'aspect grenu, parce que les pores, excessivement nombreux,
sont de fines perforations de Fectosome, autour de chacune desquelles se dresse uue
haie de spicules. Au cot6 gauche de la figure, 1'ectosome est comme use, ce qui permet
de voir la section de beaucoup de canaux inhalants. Des otcules beants, larges de
2 a 4 millimetres, sureleves, a margelle non membraneuse, la parsement, situ6s pour la
plupart le long des cretes ou de ce qui correspond au bord superieur des rameaux. La

OT. SOC. EDIN. TRANS., VOL. Xt,IX., 629.)

58 EMILE TOPSENT:

chair n'est pas caverneuse. Le squelette, dans le choanosome, forme un reseau dense
a mailles etroites, renforce par de la spongine ; plusieurs styles de front entrent
generalement dans sa trame et ses nceuds ne sont pas distants de la longueur d'un
spicule. Dans reetosome, une bonne partie des megascleres qui lui sont propres se
dressent par faisceaux autour des pores. II s'y mele beaucoup de microscleres, orientes
comme eux, mais ces fins spicules se rencontrent aussi en abondance dans le choanosome,
sans ordre et solitaires.

Les spicules presents sont de meme type que ceux de la plupart des Tedania
connues des mers du Sud : des styles a pointe courte et des tornotes, comme
megascleres, et, pour microscleres, de ces organites caracteristiques qu'on appelle
couramment des raphides, mais qu'il me parait preferable de designer d'un nom
particulier, car ils sont, en realite, autre chose que de simples raphides. Toujours
epineux, d'apres LUNDBECK (8, p. 5), a epines apprimees, ils ont deux bouts dis-
semblables ; 1'un commence en pointe fine et 1'autre se termine brusquement ; les epines
qui les couvrent sont recurvees vers la pointe brusque ; au niveau de cette derniere, il
y en a un bouquet dont se detache 1'une, qui devient plus longue que les autres.
Souvent un nodule s'observe a une petite distance de cette extremite abregee ; je 1'ai
considere comme un centrum dont la situation indiquerait 1'atrophie partielle d'uue des
moitit^s du spicule, mais il est a remarquer qu'au dela de cette nodosit6, les epines con-
servent la direction qu'elles ont en cleca d'elle. Ces spicules ne sont pas sans une
certaine ressemblance avec les uncinetes des Hexactinellides, et je propose de les appeler
des onychetes.

Speculation. — I. Megascleres : 1. Styles du choanosome, lisses, a pointe courte, a
base un peu courbee, longs de 0'225 mm. a 0'25 mm., epais de 0'007 mm. a O'Ol mm.
2. Tornotes a pointes courtes, OA'ales, puis brusquement mucronees, comme ceux de
T. charcoti ou de T. mucosa ; ils mesurent environ 0'2 mm. a 0'24 mm. sur 0'0045 mm.
a 0-006 mm.

II. Microscleres : 3. Onychetes de taille inegale, depuis 0'04 mm. jusqu'a 0'175 mm.
Les plus petites sont greles avec une nodosite voisine de leur bout tronque ; celles qui
ont O'll mm. de longueur sont les plus epaisses et mesurent 0'0015 mm. vers leur
milieu; quant aux plus longues, elles ne depassent guere 0'0012 mm. d'epaisseur et, a
partir de 0'14 mm., elles n'ont generalement plus de nodosite distincte. Elles se laissent
assez bien rejDartir en trois groupes, les courtes et fines, les moyennes et grosses et les
longues et minces.

Tedania charcoti, Topsent. (PI. V. fig. 3 et 7.)

15)08. Tedania charcoti, Topsent (28, p. 30).

Station 346. ler decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W. ;
profondeur, 56 brasses.

La Scotia a recueilli de cette espece plusieurs specimens qui nous la font connaitre
aomrne susceptible de varier dans une assez large mesure.

(HOY. SOC. EDIN. TRANS., VOL. XUX , 030.)

SPONGIAIRES DK L'EXPEDITION ANTARCTIQUE NATIONALK KCOSSAISK. 59

A. Un specimen massif, volumineux (PI. V. fig. 7), incomplet, d'ailleurs, sans
support et n'ayant sa surface intacte que sur une aire restreinte de 5 centimetres de
longueur sur 2'5 centimetres de lar^eur. La se retrouvent les caracteres exterieurs de

o O

1'Eponge du Fr<nn-nis ijue j'ai photographiee (28, pi. i. fig. 3), larges oscules a bords
membraneux, nervures saillantes a anastomoses frequentes. La coloration dans 1'alcool
est gris-jaunatre uniforme.

La spiculation est calquee sur celle des T. charcoti de 1'ile Booth- Wandel et n'en
differe que par la taille de certains de ses elements. Les styles lisses, doucement
courbes vers la base, a pointe courte et epaisse, mesurent 0'58 mm. a 0'625 mm. de
longueur sur O'Ol 5 mm. a 0'02 mm. de largeur. Les tor notes, legerement courbes vers
1'uu des bouts, qui est tin taut soit peu plus renfle que 1'autre, ont 0'425 mm. a 0'47 mm.
sur O'OOS mrn. a O'Ol mm. Les petites onychetes ont les dimensions liabituelles, mais
leur nodosite est rarement marquee. Quant aux grandes onychetes, autrement pareilles
a celles que j'ai decrites de T. charcoti, elles ont plus du double de leur longueur
(0'56 mm. a 0'6 mm.) avec une epaisseur proportionnelle (O'OOS mm. a 0'004 mm.).

B. Un fragment, long de 55 millimetres, large de 30 millimetres, epais de 5 a
18 millimetres, a surface irreguliere, grisatre comme le precedent, plus clair cependant
sur la cassure.

Ses styles ont les dimensions de ceux des T. charcoti typiques, soit environ
0'415 mm. sur O'Ol 3 mm. Ses tornotes sont a peu pres aussi longs que les leurs
(0'3 mm. a 0'325 mm.) mais plus minces (0'006 mm.). Ses grandes onychetes, de
meme forme, sont un peu plus grandes et mesurent 0'39 mm. a 0'425 mm. de longueur
sur 0'003 mm. d'epaisseur, soit beaucoup moius que les onychetes correspondantes du
specimen precedent. Ses autres onychetes, par centre, restent notablement plus courtes
(0'06 mm. a 0'07 mm.) et frappent surtout par leur gracilitr, leur tige devenant
progressivement lint'-aire a partir de leur base, qui atteint a peine O'OO 1 mm. d'epaisseur ;
leur minceur ne les empeche pas de paraitre rugueuses et de presenter une petite
nodosite, situee comme de coutume.

C. Un specimen assez gros, massif, tout blanc. II devait, comme le montre sa face
inferieure presque intacte (PI. V. fig. 3), n'adherer que faiblement a un support couvert
de plaques de Bryozoaires, dont elle a conserve 1'empreinte. A la face superieure, des
nervures en relief forment par leurs entrecroisernents r6petes de faibles elevures ou des
oretes basses.

Les styles ont 0'4 mm. a 0'425 mm. sur O'Oll mm. a 0'012 mm. ; les tornotes, 0'32
mm. sur 0'005 mm. a 0'007 mm.; les grandes onychetes, 0'18 mm. a 0'2 mm. sur 0'003
mm.; les petites onychetes, 0'06 mm. a 0'07 mm. sur O'OOl mm. a la base, avec une tige
lineaire. Ces demises ressemblent tout-a-fait a celles du specimen B ; les grandes
onychetes sont un peu plus courtes mais plus epaisses que celles du type.

D. Un quatrieme et dernier specimen, fragment assez gros, blanchatre, se rapproche
le plus par sa spiculation du spt'cimi.-n B ; ses grandes onychetes mesurent de 0'32 mm.
•t ()'4 mm. sur 0'0025 mm.

(J10Y. SOC. EDIN. TRANS., VOL. XL1X., 031.)

60 KMILE TOPSENT:

Sous-famille Mycalinse.

Mycale magellnnica, (Ridley). (PI. IV. fig. 4 et I'l. VI. fig. 10.)
1881. Esperia magellanicn, Riilley (9. p. 117).

Station 346, ler decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W. ;
profondeur, 56 brasses.

Quatre specimens, dont deux en assez bon etat pour presenter nettement les carac-
teres exterieurs de 1'espece, tons deux dresses, bosseles, a surface reticulee, et creuses
d'une cavite cloacale lisse qui s'ouvre en leur sommet par un large orifice. Les deux
a litres ne sont que des fragments.

L'un des specimens et Tun des fragment? sont blancs dans 1'alcool ; le reste est brun.
Dans les deux cas, il existe d'abondantes cellules spheruleuses qui, lorsqu'elles sont
arrondies, atteiguent O'Oli mm. de diametre, et dont les spherules brillantes mesurent
de 0'002 mm. a 0'003 mm.; mais celles des individus incolores sont jaune pale, tandis
que celles des individus fences sont d'un brun noiratre.

Les megascleres, subtylostyles a tete un peu plus longue, out ici de 0'32 mm. a 0'57
mm. de longueur et de 0'012 mm. a 0'016 mm. d'epaisseur.

Les anisocheles se repartissent assez bien en deux categories, les uns mesurant de
0'04 mm. a 0'045 mm. et les autres de 0'027 mm. a 0'032 mm. de longueur. Con-
trairement a ce qu'ont note RIDLEY (9), puis THIELE (20, p. 442), ici les trichodragmates
existent en abondance. Les raphides. fusiformes, varient entre 0'02 mm. et 0'05 mm.
de longueur.

Aucun auteur n'a fait mention d'une troisieme sorte de microscleres dont j'ai constate
1'existence dans tous les specimens de la Scotia ; je veux parler de sit/mates lin^aires,
droits ou legerement tordus a un bout, jamais recurves (PI. VI. fig. 10) ; leur tige,
doucement courbee, d'ordinaire, est quelquefois rectiligne comme celle des anisocheles
greles, avec lesquels il est, d'ailleurs, impossible de les confoudre ; ils mesurent de 0'035
mm. a 0'05 mm. d'envergure. J'ai trouve, en outre, epars, d'autres sigmates, de taille
et de formes diverses, mais ils proviennent sans doute de 1'espece suivante de Mycale,
qui en est excessivement riche et dont les specimens ont voyage meles a ceux de
M. magellanica. N'ayant vu, au contraire, que chez M. magellanica les sigmates
crreles en question, je suppose qu'ils lui appartienneiit bien en propre. S'ils sont con-
stants, leur gracilite expliquerait qu'ils aient pu passer inapei^us.

Mycale acerata, Kirkpatrick. (PI. V. fig. 6 et PI. VI. fig. 8.)

' 1908. Mi/i:a/e a<:emta, Kirkpatrick (3, p. 36).

Station 325, mai et juin-juillet 1903, Scotia Bay, 60° 43' 42" lat. S., 44°38'33" long. W.;
profondeur, 9 a 10 brasses.

Deux fragments assez gros, mais en mauvais etat et presque entierement depouilles
de leur ectosome. Ils ont une teinte brun clair qui parait due surtout a des cellules
spheruleuses de 0'012 mm. de diametre, a spherules petites.

(ROY. 800. EDIN. TEANS., VOL. XLIX., 632.)

SPONGIAIRES DR I/EXPEDITION ANTARCTTQITE NATTONALE ECOSSATSE. 6f

Les megascleres sotit remarquables par leur base moditirr en un mucron, qui rappelle
un pen 1'epine unique de IUJI/IOH unicornis mais qui reste droit et se place exactement
dans le prolongement de 1'axe (PI. VI. fig. 8). Us mesurent 075 mm. a 0'8 mm. de
longueur sur G'018 mm. a 0'023 mm. dYpaisseur.

Outre les anisocheles de O'l mm. et de 0'047 mm. a 0'057 mm., il y en a davantage
de 0'03 mm. settlement de longueur dans les deux specimens de la Sc«lm.

Les raphides, fascicules, n'ont pas une longueur constante mais varient entre 0'05
mm. et O'l 4 mm.

Mi/cafe i><'Uit«, n. sp. (PL V. fig. 2.)

Station 346, 1" decembre 1903, Bane de Burdwood, 54° 25' Lit. 8., 57° 32' long. W.;
profondedr, 56 brasses.

L'espece est representee par cinq fragments informes, dont 1'un, massif, atteint
presque la grosseur du poing. Us n'ont plus ni base ni surface; pourtant, de grands
lambeaux d'ectosome encore en place sur deux d'entre eux, indiquent que I'Eponge se
revet d'une peau spiculeuse, lisse, unie, mince mais comme parcheminee. Le choanosome,
ailleurs a nu, a une charpente de fibres tres apparentes ; elles sont solides et constituent
un ivseau a mailles larges et irregulieres ; elles sont grosses (il en est qui mesurent 1 mm.
de diametre), aussi tranchent-elles sur la chair par leur aspect blanc et brillant ; elles
diminuent de diametre aux approches de la surface, mais celles qui supportent 1'ectosome
sont encore polyspiculees. La chair, toute blanche chez deux des specimens, est d'un
brun plus ou moins fonce chez les autres.

Speculation. — I. Megascleres : 1. Sfi/fcs droits ou peu courbes, legerement fusi-
form es, a pointe tres breve, parfois meme atrophiee, de telle sorte que le style se trans-
forme en strongyle ; taille ordinaire, 0'45 mm. a 0'5 mm. de longueur sur 0'012 a
(V014 mm. de diametre.

II. Microscleres. 2. Anisocheles palmes, nombreux. en rosettes et solitaires, longs
de 0'07 mm., larges de 0'023 mm. ; leur dent superieure, un peu plus courte (0'035
mm.) que les ailes (0'037 mm.), est large des son point d'attache et a son bord libre
droit ou legerement echancre ; leur dent inferieure (la petite dent) a son bord
libre large et droit. 3. Anisocheles palmes plus petits et epars, lungs de 0'04 mm.,
larges de 0'014 mm., a tige etroite, a dent superieure ovale, un peu plus longue
(0'023 mm.) que les ailes (0'021 mm.), a dent inferieure portant un tout petit mucmn
au milieu de son bord libre. 4. Siymates tres abondants, tordus a un bout ou
contournes en S, longs de 0'15 mm. a 0'16 mm., epais de 0'05 mm. a 0'07 mm.
5. Sigmates nombreux aussi, tordus a un bout ou contournes en S, longs de 0'038 mm
a 0'052 mm., epais de 0'002 mm.

Une peau lisse, mince et ferine, oil les styles se serrcnt tangentiellement, s'y
croisant en tous sens ou souvent y formant de larges faisceaux, des fibres choano-
somiques robustes, des micmscliTes tres abondants, anisocheles et sigmates, dont on

(ROY. KIR'. Kl'I.V THAXS., VOL. XLIX., 633.)

fi2 EMILE TOPSENT :

peut si nettement distinguer deux categories, tout cela caractrrise suffisamment la
Mjicale nouvelle, de forme probablement massive.

Elle est voisine d'une Mycale sp. de 1'Antaretique, signalee par KIRKPATRICK
(3, p. 37), mais qui, poss6dant des microscleres encore plus grands, n'a, en revanche,
pas deux tailles de sigmates. Je lui trouve aussi une certaine ressemblance avec
Mycale porosa Ridley et Dendy (11), de Port Jackson, mais on ne connait pas la
nature de 1'ectosome de celle-ci et ses deux sortes de microscleres sont decrites chacune
comme d'une seule categorie.

Cladorhiza thomsoni, Topsent. (PI. IV. fig. 6.)

1909. CladorMza thomsoni, Topsent (29, p. 15).

Station 468, 39° 48' lat. S., 2° 33' long. E. ; profondeur 2770 brasses.

C'est en achevant son second voyage que la Scotia recueillit cette Eponge. Elle
fut envoyee par megarde avec la collection d'Alcyonaires a M. le professeur J. ARTHUR
THOMSON, qui reconnut en elle une Monaxonide, en piteux etat mais assez curieuse
pour faire de sa part 1'objet d'une courte notice (21).

Elle se compose de deux fragments inegaux, qu'une section tres nette lirnite tous
deux par en has, 1'un court et gros, mesurant 9 centimetres de longueur et 7 '5 mm.
de plus grand diametre, 1'autre plutot grele, long de 17 centimetres, epais de moins
de 4 mm. a sa partie inftirieure, puis progressivement aminci vers le liaut. Ils
apparaissent bien comme des portions d'une meme tige mais ils ne sont pas directement
superposables ; la difference de grosseur entre le sommet du fragment £pais et la base
du fragment grele prouve qu'il manque entre eux une portion probablement assez
longue de tige.

Ambres, vitreux et lisses dans 1'alcool, ils deviennent, a sec, blancs, opaques et
d'aspect fibreux, a cause de leurs spicules disposes tous en long. Le plus gros est
rigide, le plus grele tin peu flexible. Ils portent, implantes suivant plusieurs lignes
spirales, qui, de temps en temps, se confondent, des 6pines releve"es vers le haut,
toutes incompletes, vestiges sans doute d'autant de rameaux brises tres pres de leur
origine. Ce que furent ces rameaux, je ne saurais le dire ; mais je releve a leur propos
la phrase suivante de la description de THOMSON : " One of the branches is long, and
bifurcates at a distance of 7 cms. from the stem, being continued in two slender
divisions (one with a secondary twig) for about 4 cms." Rien de semblable n'existait
plus sur l'un ni sur 1'autre des fragments qui m'ont ete remis.

Au premier abord, ces fragments me parurent absolument decharnes et je m'attendais
a n'y plus trouver, comme THOMSON, que des megascleres. Cependant, me doutant
bien qu'il s'agissait non pas d'une Axinellide mais d'une Cladorhiza. ou d'une Chondro-
claclia, je me mis en devoir d'en racier la surface, surtout a 1'aisselle des Opines, avec
1'espoir d'y trouver quelques microscleres encore accroches aux styles superficiels.
J'y reussis et j'eus meme la chance de detacher un assez grand lam beau de chair a meme

(KOY. soc. EDIN. TBANS., VOL. XLIX., 634.)

SPONGIATRES DE L'EXPEDITION ANTARCTIQUE NATION A LK KCOSSAISE. (53

lequel je fis plusieurs preparations sans le sacrifier tout entier : j'avais attain' a
Cladorhiza, et mieux, dans ce genre, a une espece nouvelle, dont j'ai fait hominage a
M. le professeur J. A. THOMSON, qui s'y est inte"resse le premier.

Speculation. — 1. Megascleres. Ce sont tous des xti/les. Le tronc est solidc surtout
a cause de leur abondance et de la spongine qui les cimente, cur ils n'atteignent gum-
plus de 1 millimetre de longueur. Ils s'y disposent, comnie THOMSON 1'a rcmarqiir,
en plusieurs bandes concentriques rappelant sur la coupe les zones d'accroissement du
bois dans un tronc d'arbre. Ils s'y placent parallelement entre eux, suivant le grand
axe de 1'organe mais sans donner a leur poirite uue direction constante.

Ils sont droits, un pen fusiformes, t-pais de 0'028 mm. en leur centre; leur pointe
est breve, obtuse ; leur base, an contraire, longuement amincie, se termiue sans
renflement et ne mesure que 0'013 mm. a O'OIS mm. de diametre.

Dans la chair, les styles decroissent de taille jusqu'a n'avoir plus que 0-5 mm. de
longueur sur 0'015 mm., quelquefois moins, de largeur au milieu et que 0'012 mm. a
0'007 mm. de base.

Independamment de ces spicules, qui se trouvent en immense majorite et <jui,
quelles que soient leurs dimensions, appartiennent manifestement an mrrne type que
ceux du tronc, il en existe d'autres qu'ou serait tente de considerer comme formant
une categorie de styles a part, tant ils sont greles et tant leur pointe s'effile ; celle-ci
se montre meme souvent flexueuse. Pour une longueur de 0'64 mm. a 0'88 mm., ils
n'ont que (V005 mm. de largeur au milieu et seulement 0'002 mm. au niveau de la base.
Un role special leur est peut-etre devolu, mais je 1'ignore car leur distribution dans
1'Eponge m'echappe.

II. Microscleres. Les anisancres unguiferes, comme d'habitude excessivement
abondantes, ont leur tige ailee seulement sur un tiers de sa longueur et portent cinq
dents a chaque extremite. Elles sont longues de 0'031 mm. ii 0'034 mm. ; leur gros
bout, vu du face, est large de O'Oll mm. ; leur tige, au-dessous de sa portion ailee,
mesure 0'004 mm. d'epaisseur.

De grands sigmates, longs de 0-15 mm. environ, epais de U'008 mm., parsement
la chair en assez grand nornbre ; leurs bouts, non tranchants, se recourbent generale-
ment dans un meme plan ; je n'en ai pu trouver qu'un seul qui fut tordu ; leur tige,
fortement convexe, presente, surtout dans sa portion mt'xliaue, uue voussure tivs
accentu<;e.

Quelques microscleres a peu pres de meme longueur mais plus rtroits (0'005 mm.)
et minces, a bouts pointus et peu recourbes, contenant de Fair dans leur axe, repiv-
sentent soit des ancistres incomplets, soit des sigmates mal conformes de la categorie
precedente.

En outre, il y a des sigmates petits, a peu pres en meme quantitr que les grands,
mais plus diniciles a trouver en raison de leur taille. Ils mesurent O'Oo mm. dc
longueur et se montrent tantut droits et tantot tordus sur leur axe. Leur
est reguliere et leur concavite me parait sans encoche.

(ROY. soc'. EDIN. TRANS., VOL. XLIX., (j.35.)

64 EMILE TOPSENT :

Je n'ai pas pu reconnaitre avec certitude ties sigmancistres et je n'ose attacher
d'importance a deux grands ancistres (?) qui m'ont semble crochus seulement a un bout,
redresses, au contraire, a 1'autre, dans le prolongement de leur axe, en forme de point
d'interrogation, mais qui, intriques parmi d'autres spicules, se pretaient tres mal ;\
1'observation et u'etaient peut-etre pas du tout tels que je me le figure.

Homceodictya verrucosa, n. sp. (PI. V. fig. 1 et PI. VI. fig. 13.)

Station 346, 1" decembre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Un specimen.

Cette Eponge, qui appartient au genre Homceodictya au sens de LUNDBECK (7, p. 117),
est proche parente de H. conulosa (Rdl. et D.), de H. sctifera (Tops.) et de //. spini-
gera (Kirkp.), toutes les trois australes, tant par son allure generale que par sa spicula-
tion et, plus particulierement, par ses isocheles palmes.

Le specimen type, par malheur en mauvais etat, a ete arrache de son support,
declare et en majeure partie depouille de sa peau. II se dressait evidemment en une
grosse colonne cylindro-conique. Son epaisseur est de 5 a 6 centimetres vers le bas et
sa hauteur de 1 1 centimetres environ. Les lambeaux de peau qui lui restent revelent
un caractere qui distingue de ses congeneres precites 1'espece a laquelle il appartient :
les lignes ascendantes qui atteignent 1'ectosome le depassent fort peu, sous forme de
verrucosites etroites, de 0'5 mm. a 1 mm. a peine de hauteur. II y a done ici une
membrane superficielle mince, translucide, semee de petites verrucosites. La oil
1'ectosome est enleve", on voit la charpente fibreuse et la chair retractee. Les fibres
aont grosses dans la profondeur et depassent souvent 0'5 mm. de diametre ; les plus
superficielles atteignent encore 0'14 mm. d'epaisseur. Malgre leur force, elles laissent
beaucoup de souplesse au corps parce qu'elles ne se ramifient qu'a d'assez longs inter-
valles et parce qu'elles ne sont pas serrees. L'Bponge est, en effet, irriguee par de
uombreux canaux de 2 a 3 millimetres de diametre. Elle presente, en outre, en son
sommet, deux orifices de 5 a 7 millimetres, deux oscules, sans doute, dont les conduits
pe.netrent profondement dans sa masse.

Spiculation. — I. M^gascleres : 1. Oxes, doucement courbes, longs de 0'52 mm.,
epais de 0'017 mm. II. Microscleres : 2. Isocheles palmes (PI. VI. fig. 13), longs de
0-027 mm. a 0-037 mm., larges de 0'006 mm., abondants.

Des especes precitees, c'est de H. conulosa (11, p. 106) que //. verrucosa se rapproche
le plus par la facon dont ses fibres se terminent a la surface, mais c'est a celle de
//. spiniyera (3, p. 39) que sa spiculation ressemble surtout. Ses oxes, un peu plus
pointus, sont cependaut sensiblement plus faibles, et ses cheles, dont la taille oscille le
plus souvent entre 0'03 mm. a 0'035 mm., peuvent etre de pres d'un tiers plus grands.
Dans les deux especes, les cheles vus de profil montrent une pointe a chaque bout, a
1'uninn dc la dent et des ailes ; mais la dent de ceux de //. verrucosa forme une courbe

(ROY. HOC. BDIN. TRANS., VOL. XLIX., 636.)

SPONGIAIRES DE L'EXPEDITION ANTARCTIQUE NATION ALE ECOSSAISE. 65

plus prononcee vers la tige avant de rejetcr sa pointe an dchors et ellc cst notablement
plus courte (de 0'003 mm. sur les isocheles les plus grands) que les ailes.

Cos details semblent corroborer 1'opiuion suggere"e par les differences exterieures, A
savoir que nous avons affaire a deux especes distinctes.

Homceodictya setifera, Topsent.
1901. Desmacidon setifer, Topsent (26, p. 17).

Station 325, avril 1903, Scotia Bay, 60° 43' 42" lat. S., 44° 38' 33" long. W.; pro-
fondeur, 9 a 10 brasses.

Fragments rneles a des Algues.

Comnie ceux de la collection du Frangais, ils ont une hispidation bien moins longue
mais beaucoup plus serree que celle des specimens de la Belgica. L'un d'eux est brun
fonce, rude au toucher, solide ; une tubulure osculaire cylindrique se dresse en son
milieu, haute et large de 7 millimetres, avec un canal exhalant de 3 millimetres de
diametre. Un autre, sans support, est inforrne, massif, plus clair et plus mou. S'il
s'agit, comme je le pense, toujours de la meme espece, elle est tres polymorphe. Je ue
saisis pas dans la spiculation de caracteres suffisants pour distinguer plusieurs especes ;
megascleres et microscleres ne different que par la taille. Les oxes ont ici 0'75 mm. a
0'8 mm. de longueur sur 0'023 mm. a 0'026 mm. d'epaisseur ; les isocheles palmes ont
0'06 mm. a 0'063 mm. de longueur. Les dimensions des spicules sont done sensible-
ment les mernes que dans les specimens du Frangais, inferieures a celles des specimens
de la Belgica. Mais partout les oxes ont la meme forme et partout les isocheles palinrs
offrent les memes details de conformation : leurs bouts peuvent etre dissemblables, ce
qui tient sans doute a une legere torsion de leur tige, qui pr^sente 1'un de face et 1'autre
plus ou moius de profil. Le degre de frequence comme aussi le degre d'amplitude de
cette torsion, varient d'un specimen a 1'autre; elle m'a toujours paru tres facile a
coustater.

Desmacidon^ sp. ? (PI. IV. fig. 3.)

Station 346, ler decernbre 1903, Bane de Burdwood, 54° 25' lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Un fragment, presque entierement reduit au squelette.

Contrairement a mon espoir de decouvrir ses microscleres propres dans les lambeaux
rnembraneux qu'on voit tcndus entre les terminaisons de quelques unes des ligues de
sa charpente, je n'y ai trou\c que des microscleres etrangers, surtout, en abondance,
des anisocheles et des sigmates des Mi/cale pellita conservees dans le merne bocal. II
t-tait impossible de tirer parti d'un tel deliris d'Eponge. Le genre meme en est
douteux. Les megascleres sont des oxes courts et gros (0'2(i mm. sur 0'015 mm.),
a pointes breves; ils composent des fibres solides qui peuvent mesurer 0'13 min.
de diametre.

(ROY. .so<:. KHIN. TRANS., VOL. XLIX., 037.)

66 EMILE TOPSENT :

Famille HAPLOSCLERID^E.

Gellius arcuarius, n. sp. (PI. VI. fig. 11.)

Station 325, juin-juillet 1903, Scotia Bay, 60° 43' 42" lat. S., 44° 38' 33" long. W.;
profondeur, 9 a 10 brasses.

Un fragment d'Eponge sans support, informe, long de 20 millimetres, large de 12
millimetres, epais de 3 millimetres, de teinte sombre, verdatre, compact, fragile, a
orifices indistiucts, a surface tres finenient e'pineuse, tel est le specimen, a premiere vue
banal, type de cette espece.

Pourtant, il est remarquable, d'abord, par sa coloration, due en majeure partie a
d'abondantes cellules spheruleuses spheriques, de 0'017 mm. a 0'023 mm. de diametre, a
spherules nombreuses, puis par sa spiculation composee exclusivement d'oxes et de toxes.

J'ai deja fait connaitre (24, p. 470) un Gellius, G. toxius, ne possedant que des toxes
en fait de microscleres. THIELE, qui 1'a retrouve aux Celebes, en a figure les spicules
(18, pi. v. fig. 16). G. arcuarius s'en distingue nettement a la fois par ses oxes, qui ont
les pointes beaucoup plus allongees et qui mesurent 0'35 mm. a 0'4 mm. de longueur
sur 0'012 mm. a 0'015 mm. d'epaisseur, et par ses toxes, qui sont beaucoup plus ouverts,
relativernent plus givles, a bouts a peine recurves, et qui, variant entre 0'07 mm. et
O'll mm. de longueur sur O'OOl mm. a 0'002 mm. d'epaisseur, atteignent communement
les plus fortes de ces dimensions.

Toxochalina robusta, Ridley. (PI. I. fig. 3.)

1884. Toxochalina robust a, Ridley (10, p. 403).

Station 346, ler decembre 1903, Bane de Burdwoocl, 54° 25' lat. S., 57° 32' long. W.;
profondeur, 56 brasses.

Cette espece a d'abord ete recueillie par 1' Alert ;i Port-Jackson, par la faible pro-
fondeur de 0 a 5 brasses. RIDLEY et DENDY ont pens£ en avoir retrouve" dans les collec-
tions du Challenger (11, p. 50) un jeune specimen, provenant des environs de Babia,
par 7 a 20 brasses.

La Scotia en a pris deux de belle taille sur le bane de Burdwood, ce qui confirme la
vaste dispersion de 1'espece. Leurs caracteres de forme et de structure ne laissent pas
de doute sur 1'exactitude de la determination. Les plus beaux de leurs oxes tornoto'ides
mesurent 0'8 mm. de longueur sur O'OOfi mm. de largeur ; ils sont done un peu plus courts
mais plus gros que ceux du type. Leurs toxes atteignent 0'07 mm. de longueur sur
0'002 mm. d'epaisseur.

Petrosia depellens, n. sp.

Station 325, avril-aout 1903, Scotia Bay, 60° 43' 42" lat. S., 44° 38' 33" long. W.;
profondeur, 9 a 10 brasses.

Cette Eponge, dont il a fob reeueilli a plusieurs reprises, attaches a des Algues, des
fragments gros eomme des noix, presente des caracteres exterieurs qui la rendent assez

(ROY. soc. EDIN. TRANS., VOL. XLIX., 638.)

SPONGIAIRES DE I/EXPEDITION ANTARCTIQUE NATIONALE ECOSSATSE. 67

f'acilement reconnaissable. De forme massive, lobee, a surface unie, elle a une coloration
brune plus ou moins foncee. Son cctosome, epais de 0'5 mm., forme une ecorcc a
charpente reticulee ; il se detache du clioanosome par plaques, avec une facilite tellc quo
le nom specifique depellens lui convient fort bien. Get ectosome manque en majeure
partie sur les specimens de la collection, et, la ou il se trouve encore en place, il ne se
perce d'aucun orifice aquifere. Le clioanosome mis a nu montre pourtant a sa limite
des orifices nombreux, parfois punctiformes mais pour la plupart assez grands, variant
de 0'5 mm. a 1'5 mm. de diamHre, distants de 07 mm. a I mm. et irreguliers de
contour.

Le reseau ectosomique est a trame unispiculee assez serree et sans lignes directrices ;
les oxes superficiels depassent souvent la surface sous un angle variable et lui donnent
une fine hispidation, appreciable seulement a la longue.

Le squelette choanosomique esfc dense, irregulier, fait d'oxes entrecroises en tous
sens, frequemment et sans lien de spongine ; des canaux nombreux et plus larges qu'on
ne s'attend a les voir d'apres leurs orifices, rendent 1'Eponge assez caverneuse ; aussi,
quoique compacte entre ses voies aquiferes, celle-ci est fragile et friable a la faeon de
Petrosia friabilis Topsent (23, p. 69) de 1'Atlantique Nord.

II y a, d'ailleurs, une assez grande ressemblance entre ces especes. Mais, indepen-
damment de sa couleur, P. depellens se distingue de sa congenere par ses oxes de
dimensions plus fortes : de forme banale, doucement fusiformes, a pointes acerees, ni
longues ni courtes, ces spicules mesureut, en eJf'et, 0'35 mm. a 0'375 mm. de
longueur et 0'013 mm. a 0'015 mm. d'epaisseur, tant dans 1'ectosome que dans
le clioanosome.

J'ai constate 1'existence de cellules spheruleuses brunes, assez grosses, a spherules
de taille mediocre. Enfin, dans un specimen recueilli au mois d'aotit, c'est-a-dive en
plein hiver antarcfcique, j'ai trouve des poches ovariennes bounces d'oeufs en segmenta-
tion assez avancee, et rappelant celles que j'ai decrites chez Reniera simulans eu 1887
(22, p. 103). 11 ne s'y etait encore developpe qu'un tres petit nombre de larves sur le
point de s'echapper ; je les ai reconnues a leur forme allongee avec arriere retreci et
saillant ; leur ciliatiou n'etait pas visible et 1'alcool avait dissous leur pigment, mais un
faisceau de spicules etait facile a mettre en evidence daus leur partie posterieure.

Reniera cylindrica, n. sp. (I'l. II. fig. 6.)

Station 325, mai et juin-juillct 1903, Scotia Bay, (50° 43' 42" hit. S., 44° 38' 33"
long. W.; profondeur, 9 a 10 brasses.

II a ete recueilli trois specimens de cette Eponge. Les deux premiers, gris clair
dans 1'alcool, se dressaient cote a cote, relies 1'un a 1'autre seulement par des rameaux dr
ces Algues dont les entrelacements leur servaient de support. Us sont simples, malgre
leur hauteur de 57 a 75 millimetres, et subcylindriques avec uu diametre d'environ 15
millimetres dans leur tiers suprrinir, plus ou moins effiles par le bas, legerement retricis

(ROY. soc. EDIN. TRANS., VOL. XLIX., 639.)

68 EMILE TOPSENT :

vers le haut avant cle s'y terminer par un bord nettement tronque. Ce bord forme la
marge d'un orifice cloacal circulaire, b6ant, large dc plus de G millimetres, dormant
acces dans une cavit£ axiale de meme calibre ou peu s'en faut et profonde, car elle
depasse la moitie de la hauteur totale.

La surface du corps, tres finement hispide, se perce de pores punctiformes, nombreux
et tres apparents. Le corps est mou, compressible, fragile meme, d'oii la cassure que la
photographic montre sur chacun des iudividus en question.

Le troisieme specimen n'est que la partie tubuleuse, haute de 34 millimetres, d'un
individu difterant des precedents seulement par sa coloration brunatre.

Le reseau squelettique est faible, lache et souple, car ses spicules, relativement longs,
composent des lignes primaires qui ne comptent pas plus de deux elements de front,
sauf en leur terminaison oil ils s'6panouissent en un pinceau grele, et des lignes
secondaires, horizontales, discontinues, unispiculees. Dans chaque direction, ces lignes
se tiennent en outre distantes les unes des autres d'environ la longueur d'un oxe, souvent
meme davantage. Enfin, elles n'ont en leurs noeuds que des liens faibles de spongine
incolore.

Les corbeilles vibratiles mesurent 0'05 mm. de grand axe. II existe dans tous les
specimens des cellules spheruleuses nombreuses, a spherules petites ; mais ces cellules,
incolores et de O'OOS mm. a O'Ol mm. de diametre dans les individus gris, sont brun
clair et grosses de O'Ol 3 mm. dans le fragment brunatre; toute la chair cle ce dernier
s'impregne, d'ailleurs, d'un pigment assez fonce.

Les oxes ont de 0'53 mm. a 0'6 mm. de longueur, sur O'Ol 7 mm. d'epaisseur au
centre. Ils sont legeremeut fusi formes, un peu courbes, a pointes brievement ac6rees, ce
qui leur fait souvent comme une sorte de mucron conique.

Reniera dancoi, Topsent.
1901. Reniera dancoi, Topsent (26, p. 12).

Station 325, avril-aout 1903, Scotia Bay, 60° 43' 42" lat. S., 44° 38' 33" long. W.;
profondeur, 9 a 10 brasses.

Plusieurs petits specimens massifs, blancs, mous, lobes, pourvus d'osoules qui peuvent
avoir jusqu'a 3 millimetres de diametre. Ils sont attaches a des Algues ou a des
Bryozoaires.

Les oxes mesurent 0'61 mm. de longueur sur 0'02 mm. d'epaisseur.

Reniera, sp.

Station 325, avril et juin-juillet 1903, Scotia Bay, 60° 43' 42" lat. S., 44° 38' 33"
long. W. ; profondeur, 9 a 10 brasses.

II en a <-te recueilli cinq ou six fragments que je n'ai pas re"ussi a determiner et qui
sont trop petits, en trop mauvais etat et trop denues de caracteres marquants pour servir
de types a de nouvelles especes.

(ROY. soc. EDIN. TRA.NK., VOL. XLIX., (i40.)

SPONUIAIRES DE L/KXPEDTTION ANTAllCTTQUE NATIONALE ECOSSATSE. G9

INDEX BIBLIOGRAPHIQUE.

(1) BOWERBANK, .1. S., A Monograph of the liritixh Kpongiada:, vol. i.-iv. (Ray Society, London, 1st'. I

1882).

(2) GRAY, J. E., Notes on Ike Arrung: nn-nt <>j Sp»n<j<'*, u-ith the Description of some new Genera (Proc. Zool.

Soc. London, 1867, p. 492).

(3) KIRKPATRICK, R., Porifera, Tetraxonida (Nat. Antarctic Exped., Nut. Hist., vol. iv., 1908).

(4) LENDENFELD, R. VON, Tetraxonia (Das Tierreich, Porifera, 19. Lief., Berlin, 1903).
(5) Die Tetraxonia (Deutsche Tiefsee-Exped., 1898-99, Bd. xi., Jena, 1'JOG).

(6) - - Tetraxonia (Deutsche Sudpolar-Expcd., 1901-03, L'd. ix., Zool., i., Berlin).

(7) LUNDBECK, W., Porifera, part ii., Desmacidonidx (pars) (Danish Ingolf-Exped., vol. vi., Copenhagen,

1905).

(8) Porifera, part iii., Desmacidonidx (pars) (Danish Ingolf-Exped., vol. vi., Copenhagen, 1910).

(9) RIDLEY, S. O., Account of the Zoological Collections niaile dnriny the Surrey of JJ.M.S. "Alert" in the

Straits of Magellan and on the Con*/ of I'tttaijonia : Sponyiila (Proc. Zool. Soc. London, 1881,
p. 107)."

(10) - _ Spongida: Report on the Zoological Collectio7is made in the Indo-l'acAfic Ocean during the Voyage

of H. M.S. "Alert," 1881-82, pp. 36C, 582 (British Museum, London, 1884).

(11) RIDLEY, S. 0., and DENDY, A., Report on the Mouaxonida collected l»j H.M.S. " Challenger" during the

years 1873-76 ("Challenge" Reports, Zoology, vol. xx., Edinburgh, 1887).

(12) SCHULZE, F. E., Report on the Hexactinellida collected bij H.M.S. " Challenger" during the years 1873-

76 (" Challenger " Reports, Zoology, vol. xxi., Edinburgh, 1887).

(13) Hejeactinelliden des Indischen Uceanes, ii. Theil, die tlexasterophora (Abh. Konigl. Preuss. Akad.

Wiss., Berlin, 1895).
— Ameri/t anise he He.mrtine/liden nach dem Material e der " Albatross "-Expedition, Jena, 1899.

(15) Hexactinellida (Wiss. Ergebn. Deutschen Tiefsee-Exped. "Valdivia," 1898-99, Bd. iv., Jena,

1904).

(16) and KIRKPATRICK, R., Die Hejeactinelliden (Deutsche Siidpolar-Exped., 1901-03, Bd. xii., Zool.,

iv., Berlin).

(17) SOLLAS, W. J., Report on the Tetractinellidse collected by H.M.S. " < 'ha/I, n,/cr " during the yrars 1873-

76 ("Challenger" Reports, Zoology, vol. xxv., Edinburgh, 1888).

(18) TIIIBLE, J., Studien iiber pazifische Spongien, ii. Theil (Zoologica, Heft 24, ii., Stuttgart, 1899).

(19) Kieselschu-amme con Ternate, ii., Kukenthal II. Reise-ergebnisse, Bd. iii. (Abh. Senckenb. naturf.

Ges., Bd. xxv., Heft 4, Frankfurt-a.-M., 1903).

(20) • Die Kiesel- und Hornschwamme der Sammlung Plate (Zool. Jahrbuch, Suppl., Bd. vi. (Fauna

Chilensis, Bd. iii.), Heft 3, Jena, 1905).

(21) THOMSON, J. ARTHUR, and FIDDES, J. D , Note on a rare Sponge from the "Scotia" Collection (Proc.

Physical Soc., vol. xvi., p. 231, Edinburgh, 1906).

(22) TOPSENT, E., Contribution a V etude des Clionides (Arch. Zool. cxp<'r. (srr. 2), v. bis, 1887).

(23) Contribution a Vvtude des Spongiaires de /'Atlantiijne Xord (Result, oainpagnes scient. Albert Ier,

Prince de Monaco, fasc. ii., Monaco, 1892).

(24) - - Spongiaires de la bale d'Amboine: Voyage de M. Bedot et C. I ictet dans 1'Archipel Malais (Revue

Suisse de Zool., vol. iv., fasc. 3, p. 421-487, pi. xviii.-xxi., Geneve, 1897).

(25) Etude monographique des Spongiaires de France, III. Monaxonida (Had mine /•/««) (Arch. Zool.

exper. (s&. 3), viii., p. 1-332, pi. i.-viii., 1900).

(26) Expedition antarctique beige: Result. Voyage du S.V. "Belgica" en 1897-1898-1899, Zool.,

Spongiaires, Anvers, 1901.

(27) Spongiaires des Acore* (Result, campagnes scient. Albert 1"', Prince de Monaco, fasc. xxv.,

Monaco, 1904).

(28) Expedition antarctique franchise (1903-05) commanded par !<• Dr -loan Charcot, Spongiaires,

Paris, 1908.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 641.)

70 EMILE TOPSENT:

(29) TOPSENT, E., Etude sur quelques Cladorliiza et sur Euchelipluma pristina n.g. et n. up. (Bull. Inst.

Oceauogr., No. 151, Monaco, lcr sept. 1909).

(30) Les Hexasterophora recueillies par la "Scotia" dans I'Antardique (Bull. Inst. Oc&nogr., No. 166,

Monaco, 20 avril 1910).

(31) WILSON, H. V., Reports on an Exploration off the West Coasts of Mexico, Centr/tl and South America,

and off the Galapagos Islands, by the Steamer "Albatross" during 1891 : The Sponges (Memoirs
Mus. Comp. Zool. Harvard, vol. xxx., No. 1, Cambridge, U.S.A., 1904).

EXPLICATION DBS PLANCHES.

PLANCHE I.

1 et 2. Acodw.alyx brucei, Topsent, p. 20. Le specimen type, reduit de moitie, vu des deux cotes.

3. Toxochalina robusta, Ridley, p. 66. L'un des specimens. Gr. nat.

4. Raspaxilla phaltdlina, n. sp., p. 45. Le type. Tres Idgerement reduit.

5. Lissodendoryx buchanani, n. sp., p. 54. Le type. Gr. nat.

6 et 7. Bathijdorus Icvis, F. E. Schulze, var. ciliatus, Topsent, p. 32. Photographic en grandeur naturelle

et sous ses deux faces du specimen a belle frange marginals.

PLANCHE II.

1 et 2. Caulophacus scotiee, Topsent, p. 29. Silhouettes tres reduites du specimen type.

3. Caulophacus scotix. L'Eponge, photographies en deux portions et reduite de plus de moitie.
4 et 5. Malacosaccus coatsi, Topsent, p. 16. Le specimen type, vu des deux cotes, reduit de pres
des £.

6. Reniera cylindrica, n. sp., p. 67. Deux specimens. Gr. nat.

PLANCHE III.

1. Bubaris murrayi, n. sp., p. 44. Deux specimens. Gr. nat.

2. Calycusoma validiim, F. E. Schulze, p. 34. Deux des lambeaux obteuus. Gr. nat.

3. Myxilla spongiosn, Ridley et Deudy, var. asiymata, Topsent, p. 53. L'Eponge, en grandeur

naturelle.

4. Docosaccus ancoratus, Topsent, p. 23. Trois des fragments obtenus. Gr. nat.

5. Didyodona discreta (Thiele), Topsent, p. 46. Le specimen de la collection. Legerement reduit.

6. Hymeniacidon fernandeei, Thiele, p. 43. Le specimen recueilli. Legerement reduit.

7 et 8. lophonfiabello-diijitatus, Kirkpatrick, p. 56. Deux specimens. Gr. nat.

9. lophon unicornit, Topsent, p. 56. Specimen, de grandeur naturelle.
10. Deux des Rossellides roulees du bane de Burdwood, p. 12. Grandeur naturelle.

PLANCHE IV.

1 et 2. J'cecillastra compressa (Bow.), Sollas, var. parvistellata, n. var., p. 39. Deux des fragments
obtenus. Gr. nat.

3. Desmacidonl sp. ? p. 65. Le specimen, en grandeur naturelle.

4. Mycale magellanica (Ridley), p. 60. Le specimen le mieux conserve1. Gr. nat.

5. Pseudosuberites endbimns, n. sp., p. 42. L'un des specimens massifs. Gr. nat.

6. Cladorhiza thomsoni, Topsent, p. 62. Les deux fragments decharnes du type. Gr. nat.

7. Stylostichoii toxiferum, n. sp., p. 49. Le specimen type. Gr. nat.

8. Caulophacus instabilis, Topsent, p. 26. Le fragment qui sert de type. Gr. nat.

(ROY. soc. EDIN. TKANS., VOL. XLIX., 642.)

SPONGTAIRES DE L'EXPEDITION ANTAUCTIQUE RATIONALE ECOSSAlsli. 71

PLANT i n; V.

1. Hom<nodicti/a ri'rrwoxa, n. s[i., p. (14. Le type, eu p.irtie macciv. (Ir. nat.

2. Mi/rale peltitti, n. sp., p. 01. Deux fragments de spdcimens, Tun blam-, 1'autiv bnm. <lr. nat.

3. Ti;/iuiiii i-1/ttn-tifi, Topscnt, p. 58. Un specimen massif, blane, vu par sa face infeiieure. Or. nat.

4. Clnthria tu.ri/n-.-n/i/ii, u. s[i., p. 48. Le specimen type, nn pen reduit.

5. Ti'iluiiia iiiiii-ilui-lti, n. sp., p. ">7. Le specimen type, legerement ivduit.

(i. .]///<•«/<• ai-vra/<t, Kii-kpatriek, p. 60. L'uu des fragments obtenus. ( !r. nat.
7. Tedituia cliarcuti, Topscnt, p. 58. Specimen massif, yris-jaunutre. Ci-. nat.

PLANCHB VI.

1. .Mii/i<:-n.<ii.i;-n>! coatyi, Topscnt, p. 1(J. la, hexactine tlennique ; l/>, liexartinc tilt peilonctile ; If,

hexaetine gastvique ; lil, liexactine du parenchynie ; \c, pentactine hypodermique. x INO.
In/, raynn de yrande disoohexaster ; 1?;., rayon de petite discohexaster. x 310.

2. Piacil/iitti'K iiK-i'iittaiit, Sullns, p. 38. Microxe. x 540.

3. l)<'H'l»ri/s it<ji<uifi>?nt, n. sp., p. 51. 3«. deux isoeheles ; Zl, deux sigmates. x 540.

4. Artrlw.-alij.i- In'Hfvi, Topscnt, p. ^0. 4i/, discohexaster ; 46, oxyhexastcr, x 310. 4e, base d'nne

aucre, x 500.

5. L'liJi/'-'iX'nit'i ritli'luiit, F. E. Scliulsse, p. 34. 5a, hexactine denni(|ne ; 5/>, hexactine cloacale ; 5f,

grande hexactine cloacale ; 5it, pentaetine hypodermique prise au niveau d'unc verrue. x 91.
G. Docosaccus aitfurafux, Tnpsent, |>. ii3. 6a, hexactine de soutien, x 33 ; 6/<, hexactine de soiitien,
a lu base d'unc papille si'ligci-c, gr. nat.

7. Lissodendoryx luic/m/ni///, n. sp., p. 54. la, liases de pseudoxes ; 11, isoclirlc. x 540.

8. Mycale acerata, Kirkpatriek, p. GO. Li\s deux bouts d'un megasclere. x ,"> 40.

9. lujihoii 'pluricornix, Topsent, p. 55. 9«, bipocilles; 9i, extrdmite d'un tylote. x 540.

10. Mycale maijellanica (Ridley), p. 00. Sigmates lineaires. x 540.

11. Gellius arcuarius, n. sp., p. 68. UK. luxe; 116, pointe d'un oxe. x 540.

12. Clatltri/i toxiprxdita, n. sp., p. IS. \'lu, centre d'un toxe ; 12/<, iL'c, isodn'-lcs des deux categories.

x540.

l:;. ll<,ni'i'n,li,-lya verrucosa, n. sp., p. G4. Isuchele vu de profil. x 540.
14. S/i//u*/if]i<tit tij.rij'i-i-iin,, n. sp., p. -1'J. Toxcs des deux sortes. x 540.
1"). Ilatpa.rilla )iltal;i-lliiin, n. sp., p. 45. \:xi, liase de style de la cliarpente ; I't/i, deux acanthostyles ;

15f, style grele des faisceaux. x 310.
1G. Caulopkacus iuttnlnti*, Topsenl, p. I'll. llJK, hexactine dermique ; 1G/', liexactine du peduncule ;

IGc, rayon d'unc discohexaster. x 180.
17, Caulophacus N.-y//,-r, Topsont, p. 2'J. 1 7«, hexactine dermique; 17/<, liexactine cloacale; x 180.

17f, discoliexaster hexaetinale ; 17«/, discohexaster hdmihexastrale ; 17e, discohcxa.^tcr

hexastiale ; 17//, discubexaster du parenchyme; 17/, \lj, 17A, discohexasters du pi'doncule.

x 310.

(HOY. SOC. l DIN. JliANS., Vol.. XL1X., 643.)

SCOT. XAT. ANT. K\r.

To PS K NT : Sl-oNiilAIKKs HE I, A " SCOTIA " I'l. \\cllE I.

Vol. VII

Fio 2

i: T..|..,.,it. phot.

M I'.ii me \ ri-kn.. ••

Sn IT. NAT. ANT. Kxr.

Tol'SKNT: Sl'dNciAIKKS UK I. A " SCOTIA " I'l. \\rili: II.

Vol. VII.

F,G 2

i: Topsi in i hoi

•I i i -kin.' K.lii

SCOT. NAT. ANT. Exi-.

Vol. VII.

TOPSENT : Sl'o\t;[All;KS UK I. A '' SCOTIA " I'l.AMIIK III.

F.G I

" • ,

• ,

V 1**-

V-i- —*- «T

'•, T

FIG. 7

Fie 8

FIG 2

FIG

\

Fic IO

*" •

FIG 3

.

FIG 5

-;,-.

FIG 6

FIG 9

K. T..p-'-iil I'll' '

SCOT. NAT. ANT. K\i'.

TorsKNT

Vol. VII.

F,c I

*-.

Fie 1-

Fie 7

I'l.AM III. IV.

F,GO

FIG 5

Fie 3

FIG 2

F,c 6

M'1-..rl.in.- 4 Kr-kinr. E.lui

SCOT. N \T. ANT. Kxr.

Vol. \ II.

ToPSENT: SruM.iAiKi.s DK I.A " SCOTIA " I'l.. \MIII; V.

F.c 2

• -. •

F,c 5.

FIG 3

F,c

Jw

Wm
$2v

••$'•••
'

M 1 '

SCOT. NAT. ANT. Ex p.

Vol. VII.

TOPSENT : SPONGIAIRES J>K LA " SCOTIA " -PLANCHE VI.

.

1 i npn

f,d i

' • i I

:

- -

E. TopatnL del.

M Farlam £ Knklne. I.itl... Edin

PART II.
SIPHONOPHOKES.

II.- -THE SIPHONOPHORA OF THE SCOTTISH
NATIONAL ANTARCTIC EXPEDITION.

BY

J. II. KOEPPERN,

Zoological Department, University of Edinburgh.
(WITH THREE TEXT lUJ'^TRA T10NS.)

Siphonophora of Scottish National Antarctic Expedition. 77

Scotia Collections — Siphonophora of the Scottish National
Antarctic Expedition. I!y J. H. Koeppern, Xoological Depart-
ment, University of Edinburgh.

(Received 13tli December 1912. Read 16tli !><•,•, iih.T 1912.)

THE Siplioiiophora collected by the Scottish National Antarctic Expedition
(1902-1904) were kindly entrusted to me for examination by Dr W. S.
Bruce, to whom my thanks are also due for the loan of literature and
charts, and for the permission to reproduce three photographs previously
published in the Zoological Log of the expedition. I am grateful to Dr
J. H. Ashworth for various suggestions during the course of my work.

The notorious difficulty of satisfactorily preserving such fragile forms of
animal life has made the specific determination of a few of the specimens
somewhat difficult. Some of the Physalia' and Velella', however, are
excellently preserved in formalin.

The structures described by Professor J. Arthur Thomson1 as separated
gouostyles of " Scotia " Siphonophora from the South Orkneys, probably
belong to the species Mi/riotheca austrogeorgia? whereas the detached
tentacles referred to the Siphonophora by Dr J. Rennie3 must be considered
as parts of Desmonema*

Family PORPITIDjE, Brandt, 1835.

( It-nils Porpita, Lamarck, ISl'i.
Porpita umbella, O. F. Miiller, 1776.

Eleven specimens referable to this species were collected from shore
pools at Scotia Bay, South Orkneys, 2nd February 1904, the temperature
being a little below freezing-point. Their diameters in the preserved state
range from 7 to 12 mm. This is the most southern station from
which the species has been recorded.

The macerated pneumatophore of a Porpitid, with a diameter of 3 mm.,
was caught in the tow-net at the surface, at Station 30, 11° 15' N., 25C
20' W., temp. 79° F. Small isolated tentacles closely resembling those of
the Scotia Bay Porpita' were preserved along with some Velelhe from
Station 53, St Paul Rocks.

1 Proc. Roy. Phys. S<«: Kilinljiinjh, vol. xvi. (1904), pp. 19-22, 1 pi.

-' J. Ritchie, "Supplementary Report on the Hydroids of tin- Srotti.-h National Antarctic-
Expedition," Trim*, llmj. ,SV. /-,',/. (1909), vol. xlvii., pt. i., No. 1, ].. i;;i

3 Proc. Roy. /%--. Soc. !•:<!., vol. xvi. (1004), pp. -2'>--21, 1 pi.

4 E. T. Browne, " Medus;r," National Antarctic Expedition (1910), v., p. :>i.

(HEPIUXTED FIIOM THI: IMIOCKEIHXIIS OF nir. IIOYAI. I-HYSHAI. MU-IMY, vm. \i\.. IT. It! •_'].)

78 J. H. Koeppern on the

Family VELELLID^E, Brandt, 1835.
Genus Velella, Lamarck, 1816.
Velella velella, Linnseus, 1758.

Tehlla sjiimns (Forskal, 1775).

In accordance with the strict application of the rules of priority,
V. velella seems to be the correct name of this species. Specimens were
obtained at Station 51 (1° 27' N., 27° 56' W., temp. 80-1°, Dec. 8th, 1902),
and at Station 53 (0° 55' N., 29° 22' W., temp. 79'8°, Dec. 10th, 1902). One
well-preserved specimen is unlabelled, but lias possibly been caught at
Station 52 (1° 22' N., 28° 10' W., temp. 80°, Dec. 9th, 1902). According to
the Zoological Log J Velella; were also seen during the voyage at —

Stat. 61. 3° 38' S. 33° 20' W. Temp. 79° Dec. 13th, 1902.

Stat. 95. 32° 15' S. 47° 30' W. Temp. 74'9° Dec. 27th, 1902.

Stat. 360. 40° 59' S. 55° 04' W. Temp. 61-6° Jan. 25th, 1904.

Stat. 362. 43° 33' S. 55° 07' W. Temp. 60C Jan. 27th, 1904.

If the specimens are regarded from their longer side, the sail or crest runs
from N.E. to S.W., as Agassiz found to be invariably the case in hundreds
of Velellffi which he observed. Out of seventy-seven specimens examined
by Chun,2 seventy-one were of the type described, but in six the crest ran
from S.E. to N.W. Lens and van Riemsdijk B state that in all five specimens
of V. pacifica, collected by the Siboga expedition, the direction of the crests
was S.E. to N.W.

The three Velella; from the " Scotia," Station 51, were partly macerated ;
they measure respectively 42, 66 and 68 mm. in length, 16, 26 and 36 mm.
in breadth, and the crest is 16, 29 and 23 mm. in height. Nine specimens,
all of which are macerated, so as to leave only the chitinous4 supporting
substance, were taken at Station 53. The average dimensions of the
" Scotia " Velella3 are, approximately, length 54 mm., breadth 20 mm., and
height of crest 20 mm. On four of the specimens from St Paul Rocks,
numerous white spiral coils, measuring from 2 to 9 mm. in diameter and
being of a beady appearance, were observed. A microscopic examination
showed them to be segmented eggs, probably those of a gastropod. As
regards the distribution of the genus Velella in the Atlantic, the Flannan

1 Scott. Nat. Antitrr. EJ-JKI!., " Sci. Results Voyage ' Scotia,'" vol. iv., pt. i. (1908).

2 C. Chun, Die Siphonoplioren der IJl«nl /.,//-AVyi. ilitiuii, Kiel and Leip/.ig (1897).

3 A. D. Lens and T. van Riemsdijk, Tlie tiiphonopliora of the Sibot/a Expedition,
Leiden (1908).

4 M. Henze, "Notiz iiber die chemische Zusammensetzung der. Geriistsubstanz von
Velella spirans," Hoppe Seyler's Xs. physiol. Chemie., L.V. (1908).

(ROY. PHY. soc. FKOC., VOL. xix., 17.)

Siphonophora of Scottish National Antarctic Expedition.

Isles (\V. E. Clarke) and lat. 43° 33' S. (Bruce) seem to be the present
known northern and southern limits.

Family PHYSALID^E, Brandt, is.;:,.
Genus Physalia, Lamarck, 1816.
Physalia arethusa, Browne, 1750.

Chun1 distinguishes only two species of Physalia, the Atlantic-
Mediterranean P. iircthusa and the Indo-Pacific P. utriculm, since the

FIG. 1.

numerous species previously described have either been discovered to be
young specimens or slight local variations. Specimens were collected at
the following stations : —

Stat. 30. 11° 15' N. 25° 20' W. Temp. 79°

Stat. 81. 18° 24' S. 37° 58' W. Temp. 79-9°

Stat. 478. Table Bay. May 1904.

Stat. 527. 11° 32' K 20° 30' W. Temp. 78'1C

Dec. 4th, 1902.
Dec. 20th, 1902.

June 19th, 1904.

1 C. Chun, Die .V////i.///»/i/i<:/'. n <l r I'lankton-Expedition, Kid and Leipzig (1897).

(UOY. I'HY. SOf. PIIOC., VOL. XIX., 18.)

so

J. H. Koeppern on the

The occurrence of Physaliie has been recorded in the Zoological Log
at Stations 42, 53, 57, 60, 72, 78, 82, 83, 95, 97, 508, 509, 525, 529, 530
and 531.

The five young Physalue taken in the tow-net at Station 30 have well
developed pneumatophores 3, 3'5, 4, 6 and 12 mm. in length respectively,
that of the largest specimen exhibiting a comparatively large polythalamic

FIG. 2.

crest. Although not well preserved, several small siphons and a larger
central one could be distinguished. The two specimens from Station 81
are very well preserved. The length of their pneumatophores measures
roughly 9 and 10'5 cm., the breadth 4 and 4 cm., and the thickness, including
crest, 5 and 4'5 cm. respectively. The crest and cormidia are well developed.
Each specimen has about ten tentacles, the longest of which is in one
case 30 cm., and in the other 24 cm. The example recorded from Table
Bay (Station 478) is not in a very good state of preservation, the remnants
(ROY. PHY. soc. PROC., VOL. xix., 19.)

Siphonophora of Scottish National Antarctic Expedition. 81

only of a few tentacles being present. The pneumatophore is 8 cm. long ;
the gastrozociids and gonozooids, though well developed, are much shrunk.
The three well-preserved specimens from Station 527 differ from those
of Station 81, in having one tentacle more highly developed than the
others.

A faint tinge of pink is still noticeable in parts of their pneumatophores

FIG. 3.

and in some of the zooids. A coloured sketch of a /V///W/", which was
caught at Station 78, was made at the time and shows the edge of the
crest as being of a delicate pink, whereas the aboral pole of the pneumato-
phore, the zooids and the tentacles, are depicted as being of a vivid dark
blue. The three photographs show the turning over on its side of a Physalia,
which thereby presents a smaller surface to the wind.

(KOV. my. BOO. I'll'"., VCIL. six., '.20.)

82 Siphonophora of Scottish National Antarctic Expedition.

Family DIPHYIDJE, Eschscholtz, 1829.

Apart from a few not identifiable pieces of jelly, altogether eight
nectophores of Diphyids, ranging from 24 to 40 mm. in length, were
taken by the trawl (2300 f. to surface), at —

Stat. 53. St Paul Eocks. Temp. 79° Dec. 10th, 1902.

Stat. 286. 68° 11' S. 34° 17' W. Temp. 29° March 5th, 1903.

Stat. 396. 67° 53' S. 27° 20' W. Temp. 29'2° Feb. 29th, 1904.

Stat. 416. 71° 22' S. 18° 15' W. Temp. 29'6° March 17th, 1904.

Stat, 450. 48° 00' S. 9° 50' W. Temp. 40° April 12th, 1904.

Three Eudoxiie must be mentioned here, two from Station 396, and one
from Station 450. The unsatisfactory state of their preservation makes all
but the reference to the above family possible as regards classification.
Their habitat, however, is of interest, as extending the range from which
they have previously been recorded. The discovery of Diphyids as far
south as 71° 22' corresponds to their occurrence in Arctic seas.

(Issued separately, 10th April 1913.)

ROY. PHY. SOC. PllOC., VOL. XIX., 21.)

PART III.

POLYCH^TA.

(Families SERPULID.K and SARELMD^E.)

III.— THE POLYCH^ETA (Families SERPULID^, and SABELLID^)

OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

BY HELEN L. M. PIXELL, B.Sc., F.Z.S.,

Demonstrator of Zoology anil Reid Fellow, Bedford College, University of London.

(WITH ONE PLATE.)

Polychaeta of the Families Serpulidae and Sabellidae, collected by the Scottish
National Antarctic Expedition. By Helen L. M. Pixell, B.Sc., F.Z.S.,
Demonstrator of Zoology and Reid Fellow, Bedford College, University
of London. Communicated by Dr J. II. ASHWORTH. (With One Plate.)

(MS. receive! January 22, 1913. Read February 17, 1913. Issued separately June 24, 1913.)

For the opportunity of examining these specimens I am indebted to Dr W. S.
BRUCE, the leader of the Scottish National Antarctic Expedition, 1902-1904. I should
like to take this opportunity of expressing my thanks to him, and also to Dr MARETT
TIMS for his assistance during the preparation of the paper.

The following is a list of the ten species, of which four are new. Eight different
genera are represented :—

Family SERPULIDAE.

Sci'pula vermicularis Linmeus, from various stations (see p. 88).

Apomatus brotrnii n. sp., Station 417, lat. 71° 22' S., long. 16° 34' W., 1410 fathoms.

Salmacina dysteri (Huxley), Station 461, Gough Island, lat. 40° 20' S., long.

9° 56' 30" W., 100 fathoms.
Spirorbis antarcticus n. sp., Station 325, Scotia Bay, South Orkneys, lat.

60° 43' 42" S., long. 44° 38' 33" W., 10 fathoms.
,, patagonicus Caullery and Mesnil, Station 478, Table Bay, Cape Town ;

also Station 118, Port Stanley, 4 to 5 fathoms.

,, falklandicus n. sp., Station 118, Port Stanley, Falkland Islands, 4 to 5
fathoms.

Family SABELLID.E.

1. Dasychone violacea (Schmarda), Station 478, Table Bay, S. Africa, shallow water.

2. Eurato melanostigma (Schmarda), Station 24, Cape Verde Islands, shore.

3. PotamillaantarcticaGi&vieT, Station 349, Tussock Island, Falkland Islands,shore.

4. Potamis scotiw n. sp., Station 417, lat. 71° 22' S., long. 16° 34' W., 1410
fathoms.

Family SERPULIDAE.

Genus Serpula.
Generic characteristics :

1. Collar-setae of bayonet-shape, with spines at base of blade.

2. Operculum funnel-shaped, with numerous radii ending in serrations on margin.

3. Uncini with only a few teeth.

(KEI'RINTEU FROM THE TRANSACTIONS OI-' THE I1OYAL SOCIETY OF EDINBURGH, VOL. XLIX., 1>1\ 347-3."i8.)

88 HELEN L. M. PIXELL ON

Serpula vennicularis Linnaeus, 17G7.

For synonyms see MORCH (14, p. 381), SAINT- JOSEPH (16, p. 328), EHLERS (5, p. 219)
and FAUVEL (6, p. 691).

Specific characteristics :—

1. Collar-setae with generally two large blunt processes at base of blade.

2. Uncini generally have five teeth — there may be four, six, or seven.

3. Branchias (twenty to thirty-two pairs) long, with numerous pinnse and bare

filamentous extremities.

4. Serrations on operculum vary very much in number — there may be as

many as a hundred.

5. Maximum length recorded 50 mm., with 157 segments.

Localities.— Station 325, Scotia Bay, South Orkneys, lat. 60° 42' 43" S., long.
44° 38' 33" W.; Station 346, Burdwood Bank, lat. 54" 25' S., long. 57° 32' W.,
56 fathoms ; Station 478, Table Bay, shore; Station 461, Gough Island. Part of a
colony, with several specimens having intertwined tubes, was taken in a trap in
MacDougal Bay, South Orkneys.

These tubes are cylindrical with trumpet-like ends, as described by M'!NTOSH
(12, p. 516) for S. narconensis Baird. I fully agree with EHLEKS (5, p. 219) that this
species is synonymous with S. vermicularis ; the specimens show a considerable
amount of variation, as is usual in this species, but there seem to be no points of
difference having specific value.

There is one specimen only from Station 478, Table Bay, May 1904, with its tube
adhering for its whole length to the shell of Mytilus. This tube is triangular in
section, with a very distinct median ridge ending in a point overhanging the mouth.

Three tubes from Burdwood Bank, taken in the trawl in 56 fathoms, December 1905,
Station 346, are solitary, almost cylindrical, but enlarge very gradually towards the
anterior end, where there is a distinctly everted peristome. The tubes are ringed with
former peristomes at intervals, showing successive zones of growth. They have no
doubt been attached to some substratum, for a part of their length, and are more or less
overgrown with a bryozoon. The largest, though incomplete posteriorly, is 70 mm.
long and 4 mm. in diameter across the peristome. One of the smaller ones contained an
animal 25 mm. long, having an operculum with thirty-two serrations, and gills with bare
terminal ends intermediate in length between those from Scotia Bay and Table Bay.
Fig. 1, a and b, shows the variation amongst the collar-seta; of a single fascicle : the
former having the typical two processes ; the hitter, four almost equally large ones.

All the specimens are between 2 and 3 cm. in length, and have seventy to eighty
segments. Their opercula have thirty-two to forty-one serrations, and their uncini
generally six or seven teeth. The branchise (twenty-two to twenty-eight pairs) have
in some cases very long bare terminal ends measuring quite 1 mm., instead of the more
usual '48 mm. according to SAINT-JOSEPH (16, p. 330).

(ROY. soc. JSDIN. TRANS., VOL. XLIX., 348.)

POLYCH^BTA OF THE FAMILIES SERPULID^E AND SAP. KLUDGE. 89

The few specimens from Gough Isl-nul have less tlian thirty-two serrations to their
opercula, but it seems that they are quite young.

Genus Apomatus Philippi, 1814.

Generic characteristics :— -

1. Operculum globular, terminating a gill retaining its pinnfe.

2. Some of the thoracic setae are bladed sickles (seta; of Apomatus).

3. Terminal dorsal gland present.

Apomatus brownii n. sp.
Specific characteristics : —

1. Collar-setse with small fin at base of blade ; fig. 2, a.

2. Uncini begin on the second thoracic segment, and have only seven or eight

teeth ; fig. 2, d and c.

3. Abdominal setae sickle-shaped ; fig. 2, c.

Five specimens, more or less incomplete, from Station 417, lat. 71° 22' S., long.
1C" 34' W., trawled at a depth of 1410 fathoms, 18th March 1904.

This form differs from all other previously described species of the genus in the
characteristics 1 and 2 above, i.e. in the shape of the collar-setae and the uncini.
There is, however, complete agreement with the two main generic characteristics (1 and
2 above) as given by SAINT-JOSEPH (16, p. 373). It is to be noted that this species
comes from very deep waters.

Tubes almost cylindrical and smooth, except for inconspicuous growth lines. The
diameter of the oral aperture of the largest tube is 3-5 mm. ; 50 mm. from this is the
other aperture 175 mm., but this is probably not the real end of the tube, which
may have been attached to some substratum by a further narrower portion. Most
of the tubes have obviously been longer, but only one, the smallest, shows a scar of
attachment.

The animals were preserved in their tubes. The total length of the longest is
27 mm., and of the shortest 16 mm. ; the greatest width is 2 mm. The fifteen pairs
of branchiae are spirally coiled. They measure from 6 to 10 mm. in length, and are thin,
with long fine pinnae extending almost to the tops of the rachises.

One animal has its branchial crowns projecting from the tube, and has lost its
operculum. Two others have lost their branchial crowns altogether. In the remaining
two, the second dorsal gill on the left or right terminates in a transparent globular
operculum. In one there is a secondary rudimentary operculum on the other side.

The collar is I mm. deep, entire ventrally, but notched laterally, and very much
folded. There are only five or six collar-setae of the characteristic form, which, with
about half a dozen ordinary capillary set;e with narrow blades, make only a small
fascicle. The other thoracic fascicles contain numerous setae, comprising ordinary
narrow blades and setae of the usual Apomatus type (fig. 2, b).

(HOY. KOC. EDIN. TRANS., VOL. XLIX., 319.)

90 HELEN L. M. FIX ELL ON

The abdomen has only a very short asetigerous region, followed by 80 to 110 narrow
segments with sickle-shaped setae (fig. 2, c). Many of these sickles have, however,
become much straightened, owing, no doubt, to their prolonged immersion in alcohol.
The posterior setae are very elongated. The dorsal gland, generally present in the
young, is very little developed, and in two specimens appears to be wanting. There
is a deep ventral faecal groove which turns to the right on reaching the thorax.

The uncini have only seven or eight sharp teeth (fig. 2, d and e] ; consequently, if
SAINT- JOSEPH'S classification (16, p. 263) according to the characteristics of the uncini
were rigorously adhered to, this species would have to be placed in a new genus in
Group L, with Serpula, etc., instead of in the genus Apomatus, which SAINT- JOSEPH
includes in his Group V., which is made up of forms whose uncini have very numerous
and fine teeth.

Genus Filograna Oken, 1815.

Generic characteristics : —

1. Tube very slender, filiform, colonial.

2. Branchiae eight.

3. Thorax with seven to nine segments.

4. Collar-setae with a large fin-like expansion at base of blade.

5. Other thoracic setfe sickle-shaped (setae of Salmacina), or ordinary bladed

ones.

6. Abdominal setae more or less geniculate and serrated.

7. Hermaphrodite.

Sub-genus Salmacina Claparede, 1870 (2, p. 176).

1. No operculum.

2. The ends of the branchiae may or may not have spathulate enlargements due

to the presence of large granular cells.

Salmacina dystcri (Huxley), 1855.

Protula dysteri Huxley, lS.r>5 (10, p. 11.°.).
Filopora filograna Dalyell, 1853 (3, p. 250).

Specific characteristics : —

1. Branchiae with spathulate enlargements containing granular masses at their

ends. Similar granules occur at the ends of the pinnae, and just in front
of their bases along the gill rachises.

2. Spermatozoa developed in segments anterior to those producing ova.
Locality.— Station 461, Gough Island, lat. 40° 20' S.; long. 9° 56' 30" W.,

100 fathoms. Several fairly large masses brought up by the trawl with Scrpv/a
vermi<'i//<trix, coral, etc. No specimens in process of budding were observed.

(HOY. soc. KDIN. TRANS., VOL. XLIX., 3.r>0.)

POLYCH^ETA OF THE FAMILIES SERPULID^E AND SA DELUDE. 91

Genus Spirorbis Lamarck, 1801.

Characteristics of genus and sub-genera have been given in full (15, pp.
792-799).

Sub-genus Paralseospira Caullery and Mesnil (1, p. 202), emend. Pixell (15, pp. 795
and 799).

ix (intarc^icus n. sp.

Specific characteristics :—

1. Collar-setae — a few small finely striated simple blades, fig. 3, c.

2. Third thoracic fascicle contains some bladed sickle-shaped seise.

3. Abdominal setae, flattened trumpets with one side produced.

4. Tube triangular in section and generally very regularly coiled (fig. 3, «).
Locality. — Several specimens growing on fucus, dredged at Station 325, April 1903,

Scotia Bay, South Orkneys; lat. 60° 43' 42" S., long. 44° 38' 33" W., in 9 to 10
fathoms.

The large thick tubes are almost perfectly triangular in section and very regularly
coiled to form a disc, with concave upper surface, and measuring up to 4 mm. across
(fig. 3, «). There are nearly always young tubes attached to them, one specimen bearing
as many as twelve of different sizes. In this they resemble CAULLERY and MESNIL'S
(l, p. 203) Spirorbis aggregates.

The opercular plate may be flat, convex, or conical, but has in every case a thin
flattened talon. Small circular perforations occur in the calcareous plate, and through
these project small membranous projections, which are generally thorn-shaped. Though
the opercula vary so much, there is no doubt that all the variations should be included in
the same species, for there are intermediate forms between the thin, flattened plate,
which seems to be characteristic of young specimens, and the conical form. It is also
very questionable whether details as to the shape of the operculum are ever sufficiently
constant to be of much value in specific determination.

Many of the specimens are of very large size, some measuring over a centimetre in
length. There are nine long branchiae : the second on the left is transformed into the
pedicle of the operculum, and is slightly larger than the others.

This species could only be included in the sub-genus Paralneospira, provided that
the modified characteristics as suggested in a previous paper (15) be adopted for it,
owing to the fact that it differs from all previously described Paralaeospira in
having simple blades for the collar-setae (fig. 3, c). These are remarkably small, and
there are only seven on each side, so that the whole fascicle is very insignificant.
The other thoracic fascicles are much larger, and the uncini are about 40 mm.
long. There are about twenty abdominal segments following a rather long aseti-

gerous region.

(ROY. soc. EDIN. TRANS., VOL. XLLX., :!:>!.)

92 HELEN L. M. FIX ELL ON

Spirorbis patagonicus (Caullery and Mesnil), 1897 (1, p. 205); Ehlers, 1901

(5, p. 224).
Specific characteristics : —

1. Talon of operculum, a prolongation of the terminal shallow funnel.

2. Collar-se-tJB small, with a fin-like expansion at the base of finely serrated

blade (fig. 4, a).

3. Numerous sickle-shaped setae to 3rd thoracic segment.

4. Abdominal setae large, flattened, and trumpet-shaped.

5. Large somewhat irregularly coiled tubes.

Localities. — Several specimens on the shell of Mytilus, from Station 478, Table Bay,
Cape Town Docks, May 1904. Other specimens, with Spirorbis falklandicus, on a
stone from the shore at Station 118, Port Stanley, Falkland Islands, January 1903.

Most of the tubes are rather smaller than those described by CAULLERY and MESNIL.
They are generally loosely and irregularly coiled. All ages seem to be represented, and
both the young with three thoracic segments only and the adults agree with the
descriptions given by these authors.

There are apparently ten branchiae, including the opercular pedicle, and about
twenty abdominal segments. The largest specimen obtained whole from its tube
measured 6 mm.

Paradexiospira Caullery and Mesnil (l, p. 195), emend. Pixell (15, p. 793).

Spirorbis falklandicus, n. sp.

1. Last thoracic segment without dorsal setae.

2. Collar-setae small, very finely serrated blades with large fin at base.

3. Operculum very similar to that of S. vitreus.

4. The 3rd thoracic fascicle contains bladed sickles (setae of Apomatus) (fig. 5, r).
Several specimens, with S. patagonicus, on a stone from the shore at Station 118,

Port Stanley, Falkland Islands, January 1903.

This is the first time that a species of the sub-genus Paradexiospira has been
recorded from the Southern Hemisphere. CAULLERY and MESNIL (l,p. 195) state that
the three species known to them, namely S. cancellation Fabr., S. vitreus Fabr., and
S. violciceuK Lev., inhabit Greenland and Iceland. S. vitreus has since been found in
the North Pacific (15, p. 793).

The tubes of Sjiirorbix falklandieus generally have three longitudinal ridges, the
median one sometimes ending in a projection overhanging the mouth of the tube. The
terminal part may be more or less ascending, and is rather more than 1 mm. in dia-
meter. Only about one and a half turns of the spiral can be seen from the top, and the
tube measures 3 mm. across.

Branchiae ten, the second on the right bearing the operculum, as usual. In young
specimens the operculum is almost flattened on the top, and has a massive, sometimes

(ROY. soc. KDIN. TRANS., VOL. XLIX., 352.)

POLYCH.KTA <)K THE KAMI I, IKS SERPULID^E AND SABELLID^S. 93

bifid, talon (fig. 5, <t). In older specimens this is replaced, as in S. vitreus, by a thin
funnel-shaped one without a talon : in one animal examined, both were present, one
inside the other, the thin funnel-shaped one having been produced before the loss of
the smaller massive one.

There is a deep collar, and the species differs from N. vitreus and all other previously
described Paradexiospira in having collar-seta; with such fine serrations that it is
almost impossible to recognise any at all. I have already suggested that the character-
isation of CAULLERY and MESNIL'S sub-genera should be made more general (15,
p. 793), and this is another instance of the necessity of such a step, for S. fa/k-
/midicus is undoubtedly closely related to the three forms known to CAULLERY and
MESNIL and placed by them in the sub-genus Pannl<'.rn>K^n\i. Of these collar-seta;
there are about nine, making with a fe\v of the usual fine capillary ones a bundle con-
siderably smaller than the other thoracic fascicles.

The uncini of the thorax are 40 to 50 mm. long, and contain about seventeen teeth :
those of the abdomen are much smaller, being only 20 mm long.

The abdominal seta3 are flattened and trumpet-shaped, with one side produced.

There seem to be twenty or more abdominal segments, but I did not succeed in
obtaining a complete specimen from a tube, the state of preservation not being eood.

Family SABELLID.E.

Genus Dasychone Sars, 1861.

1. Thoracic and abdominal tori with single rows of avicular crotchets.

2. Thoracic seta; of one kind only.

3. Branchiae with dorsal appendices.

Dasychone violacea (Schmarda), 18G1.

Sabella riolawa Sclnnurda (17, p. 34-).
DtMjrltone tsitilacea M'Intosh (12, ]>. 504).

1. Dorsal appendices large and somewhat spathulate, covering an ocellus mi

either side of the rachis.

2. A dark pigment spot on each side of every segment, bet. ween the fascicles of

set* and the tori.

3. Thorax of eight setigerous segments.

4. Crotchets with very fine serrations above the large fang.

Locality. — One specimen from Station 478, Table Bay, May 11)04. Total length
•J(i mm., of which the branchiae form 8 mm. ; the width is about 4 mm., except close to
the posterior end, where it decreases rapidly. There are seventy abdominal segments.

The seta; and crotchets agree entirely with those so clearly described and figured
by M'INTOSH.

(HOY. M)C. KLIN. TRANS., VOL. XL1X., 3f>3.)

94 HELEN L. M. PIXELL ON

It will be noticed that this specimen is smaller than those described before : it
agrees more nearly with SCHMARDA'S specimens in having only twenty branchuL-. Each
of these has seven pairs of well-formed dorsal appendices, with rudiments of others towards
the tips. The specimen was by no means fully developed, nor was it sexually mature.

No trace of colour remains, except on the gills and the pigment spots already alluded
to. The tube is also missing.

Genus Eurato Saint- Joseph, 1894 (16, p. 249).

Generic characteristics : —

1. Thoracic and abdominal tori with a single row of avicular crotchets.

2. Branchiae without dorsal appendices and arranged in a single series.

3. Thoracic setae all simple blades.

Eurato melanostigma (Schmarda).
Sabella melanostigma Schmarda, 1861 (17, p. 36), from Jamaica.

1. Branchiae about half the length of the body, few in number, with short pinnae

and an interbranchial membrane.

2. A pigment spot on each side of every segment, between the fascicles of setae

and the tori.

3. Thorax of five setigerous segments (? always).

Several specimens from Station 24, Porte Grande, St Vincent, Cape Verde Islands.

These specimens seem to be immature, and are much smaller than SCHMARDA'S ; but
from the latter's brief description they would seem to be the same species.

They are 5 mm. or less in length, the branchiae when intact making up nearly 2 mm.
of this ; the greatest width is about 1 mm., and the largest number of abdominal seg-
ments counted was thirty-five.

The collar has two pointed ventral lobes, but is otherwise not very high, and stops
short some distance from the mid-dorsal line.

The collar-setae have long narrow blades (fig. 6, a) ; there are only four other setiger-
ous thoracic segments, and these have slightly wider blades in addition to the long
narrow ones (fig. 6, b). The abdominal segments have the wider blades only. The
thoracic tori have about twelve avicular crotchets (fig. 6, c) ; the abdominal only seven.

The anal segment is asetigerous and very distinctly bi-lobed.

There are only seven or eight pairs of gills.

In no specimen were gonads apparent ; this immature state may account for the
small number of branchiae and thoracic segments.

One very young specimen was 2 mm. long and I mm. wide, only very slightly
tapered at each end, and containing eighteen setigerous segments ; all of these seemed
to have the setae ventral to the tori, i.e. to be abdominal segments, except one or two

(HOY. SOC. EDIN. TRANS., VOL. XLIX., 354.)

POLYCHyETA OF THE FAMILIES SERPUL1D.-E AND SABELLID^E. 95

very small crowded segments anteriorly, in which no tori could be distinguished. Tin;
branchial crown was represented by two small buds with four finger-like projections.

No tube was present, but to one specimen was adhering a minute portion of a
rather thin membrane, to which very small sand-grains were fixed, and this may have
been a portion of a tube.

With these also was a larger specimen (8 mm. total length), in which no pigment
spots were present, but which otherwise seemed identical with this species.

The pigment spots in Eurato (X«l><ll«) m«n'icata (9, p. 255) are differently
arranged.

This is apparently the first time that this species has been recorded since SCHMARDA
described it from Jamaica in 18(51.

Genus Potamilla Malmgren, 1865 (13, pp. 401 and 402).

Generic characteristics : —

1. Thoracic tori, double rows, consisting of avicular crotchets and pennoued

setse respectively.

2. Abdominal tori, single rows of avicular crotchets.

3. Collar present.

4. Thoracic setse of two forms — dorsally a few with simple blades, ventrally

spathulo-mucronate.

5. Abdominal setse simple blades.

tj. Branchial crown not markedly spiral, and without sub-terminal eyes on
rachises.

Potamilla unturctica Gravier, 190(5 (7, p. 59).

1. Branchire seventeen pairs, with no trace of eyes, about one-fifth the length

of the body.

2. Avicular crotchets of thorax and abdomen of characteristic shape (see Gravier

(8) and (7) for figures).

Localities. — One specimen from shore pool at Station 349, Tussock Island, Falkland
Islands ; seven incomplete specimens, all of which had lost their branchial crowns, from
the shore at Station 118, Cape Pembroke, Falkland Islands.

The tubes of the latter specimens are much overgrown by a colonial tunicate.

The anterior part of the tube of the Tussock Island specimen contained numerous
larvte, 5 mm. long, and having five or six segments. The specimen in the tube was
well preserved, and was 31 mm. in length, of which the branchiae made up (5 mm.

The other specimens are, without their branchial crowns, 27 mm. long and 3 mm.
wide, and have nearly parallel sides — only a few of the terminal segments being
narrower. The thorax consists of eight setigerous segments, and the abdomen of
about fifty. The setse were in nearly every segment broken off below the blade, and

(ROY. soc. EUIX. TRANS., VOL. XLIX., :',.~>.~j.)

96 HELEN L. M. P1XELL ON

it was only with considerable difficulty that any perfect ones were obtained. The
collar has small dorsal lobes, and becomes gradually higher towards the mid-ventral
line, where there is a deep fissure.

The ventral gland-shields are very distinct, separated by deep segmental fissures,
and divided in the abdominal region by a deep and narrow faecal groove, which crosses
obliquely to the right on reaching the thorax, and is continued as a very shallow groove
along the mid-dorsal surface.

No pigment spots could be made out at the posterior end, and there is practically
no colour in the preserved animals ; only faint transverse markings could lie made
out on the branchiae. In all other points there seems to be exact similarity with
GRAVIER'S widely distributed species. At the same time, I cannot help thinking that
the variation, which he points out (8, p. 145) as existing in two of the characteristics
on which he has based his species in differentiating it from P. neglecta Malmgren and
P. incerta Lang, tends to do away with its distinctness.

Genus Potamis Ehlers, 1887 (4, p. 278).
Generic characteristics :—

1. Thoracic tori double rows : avicular crotchets with long bases and pennoned

setae respectively.

2. Abdominal tori single rows of avicular crotchets.

3. Collar present.

4. Thoracic setae of two forms — superior ones with rather wide blades, inferior

sub-spathulate.

5. Abdominal setae of two kinds — long narrow- bladed ones and short forms

with wider blades (lanceolate) (fig. 7,f).

6. Gill bases sub-involute. Branchiae without ocelli.

Potamis scottie n. sp.

1. Collar with high ventro- lateral lobes and deeply divided in the middle line.

2. Seta? and crotchets of characteristic shapes (fig. 7, a-b), differing in many

ways from those of P. spatliifenis Ehlers.

3. Ventral gland-shields quite distinct, divided in the abdominal region by a

wide and shallow faecal groove.

Locality. — One specimen, without its tube, taken in the trawl at Station 417,
lat. 71° 22' S., long. 16° 34' AV., from a depth of 1410 fathoms.

Total length 133 mm., of which 63 mm. is made up by the branchial crown, 10 mm.
by the thorax, and 60 mm. by the abdomen. Greatest width 5 mm.

After being preserved in dilute formalin the general colour of the specimen is
greenish-brown. The segments are clearly marked off from one another by inter-
segmental constrictions, and the setae protrude to a considerable distance.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 3f><;.)

POLYCHyETA OF THE FAMILIES SERPULID^E AND SABELLID^E. 97

The thorax has eight set igrrous segments, the abdomen 52 ; the posterior abdominal
segments are much shorter antero-posteriorly, and rather wider than the front ones.
The gill rachises are nearly 1 mm. in diameter and very nearly cylindrical ; in some
cases they end in a small finger-like process, but the thin straggling pinnae extend at
irregular intervals nearly up to this end. There are sixteen on the left and fifteen
on the right, the four ventral ones on each side being much smaller. To the most
ventral one on each side i.s attached the membrane which, just- below, enlarges greatly
and forms a kind of pocket on each side ; the membranes then approach one another
and join together at the posterior end of the fissure between the ventro-lateral lobes
of the collar.

From the dorsal aspect there is great resemblance to EHLERS' P. spiithi ferns (4.
p. 278), the collars of both species being similar.

The first segment has no torus, and its setse are of one kind only, similar to the
superior ones of the other segments (fig. 7, y).

December 23, 1912.

REFERENCES.

(1) GAULLBRV, M., et MESXIL, F., "Etudes stir la morphologic comparee et la phylogenie des especes chez

les Spirorbis," Bull. Sci. Frmn-e et Bvl<jiqni>, xxx. pp. 185-233, Paris, 1897.

(2) CLAPAREDE, E., " Annelides che'topodes du Golfe de Naples (Supplement)," .Mem. So<-. Phij*. G

xx., 1870.

(3) DALYELL, J. G., The Power of the Creator displayed in the Creation, etc., vol. ii. p. 250, 1853.

(4) EHI.ERS, E., "Florida Annelida," M-'in. Mus. Comp. ZooL lLirr<tr<L \,<\. xv., 1887.

^5) _ -"Die Polychseten des ma^ellanischen und chilenischen Strmdes," Fettsthr. K. Ge.--. ]IVw.
Gvttiwjen, 1901.

(6) FAUVEL, P., " Sur quelques Serpuliuns de la Manche et de la Mediterranee (Serputii r>-rini'-nl<iri* el

Praiula tubuluria Mont.)," Compt. Rend. Asa. Fr. Avanc. Sci., Paris, 1909.

(7) GRAVIER, C., Expedition Au/airfique Francaine, 1903-05, 'Annelides I'oh/'-li., p. •')'.', Paris, 1906.

(8) Deux ihne Expedition Antarctique Fran^aise (1908-10) : Anti'-lid^ Poll/chiles, Paris, 1911.

(9) GRUBE, A. E., " Annulata Semperiana," M,'m. Avail. Imp>''r. Sci. St Petersb., 1" sdrie, xxv., 1878.

(10) HUXLEY, T. II., "On a Hermaplinidite and Fissiparous Species of Tulii-'ular Anii.'lid," F.dut. X»w Pint.

Journ., vol. i., No. 1, 1855.

(11) LAXGEIIHANS, P., "Die \Vurmfauna v..n Madeira,'' Z»it.<. wiss. tool., xxxii.-, x\iv.. Leipzig, 188U.

(12) M'lNTOSH, W. C., " Polychseta," " ChaJlemjer " Reports, ZooL, xii., L.Hid..n, 1885.

(13) MALMI.KEX, A. J., " Nordiska HatVAnnulater," Ofrer* «t K. i}<: Vet.-Akad. /•'.-/•/(., pp. 401, 4U2,

Stiickholm, 1865.

(14) MORCH, OTTO, A. L., "llcvisi,, crili<-a Serp., etc./' Xatnrliist. Tid*gl;,-ift, i. pp. .'U7-470, Copenhagen,

1863.

(15) PIXBLL, H. L. M., " PolychuHu frc.iii the IV-ilic Coast of North Amorica : 1.. S,-i-pulid;e, with a lu-vised

Classification of Genus Spirorbis," Proc. ZooL Soc. Land., pp. 781-805, 1912.
(10) SAINT-JOSEPH, BARON- DE, "Les Annelides Pnlychetes des Cotes de Dinard," Ann. Sci, S«t. ZooL (7),

xvii., Paris, 1894.
(17) SCHMARDA, LUDWIG K., Neu>- 'iri i-l » 1 1 use Thiere, I. ii., Leipzig, lM.il

(ROT. soc. EDIN. TRASS., VOL. XLIX., 3.">7.)

98 POLYCH^ETA OF THE FAMILIES SERPULlDvE AND SABELLID^E.

EXPLANATION OF THE PLATE.

Fi^. 1. Serpiila cermicularis L. Collar-setre from a single specimen to show variation: la, with the
nnrmul two processes at base of blade ; I/', with four nearly eijual processes, x 300.

Fi'_'. 2. Apoinatas brownii n. sp. 2a, collar-seta; 26, bladed sickle from thorax; 2'-, sickle-shaped seta
from abdomen ; 2</, thoracic uncinus ; 2e, abdominal uucinus. x 375.

Fig. 3. Spirorbis antar<-ti<:us n. sp. 3a, tube with three small ones towards its centre, x!2; 3b,
operculum, x 48 ; 3c, cullar-seta, x 375.

Fi«. 4. Sfirurliis patagonicus Caullery and Mesnil. Collar-seta, x 375.

Fig. 5. Spirorbis falMan dints n. sp. 5a, operculum of young specimen, x 36 ; 56, collar-seta, x 375 ;
5i; seta of bladed sickle type, from third thoracic fascicle, x 375.

Fig. 6. Eurato melanostigma (Sehmarda). 6« and 66, thoracic setae, x 220 ; 6f, thoracic crotchet,
x220.

Fig. 7. Potanris scotix n. sp. 7«, thoracic avicular crotchet, x 48 ; 76, head of same, x 220 ; 7 c, thoracic
pennoned seta, x 48 ; 7(7, head of same, x 220 ; 7e, abdominal avicular crotchet, x220; If, abdominal
lanceolate seta, x 150; 7g, superior thoracic seta, x 150; 7h, inferior thoracic seta, x 150.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 358.)

.•<.(. Nat. Ant, Exp.

Vol VII.

PIXELL : "SCOTIA" POLYCHAKTA.

H L M P.xcU del

PART IV.

TURBELLAR1A (ID.

IV -THE TURBELLARIA OF THE SCOTTISH NATIONAL
ANTARCTIC EXPEDITION. (2™ PAPER.)

BY F. F. LAIDIAW, M.A., F.Z.S.

( 103 )

Turbellaria of the Scottish National Antarctic Expedition.— II.*
By F. F. Laidlaw, M.A., K./.S.

The two Polyclad Turbellaria recorded in this note were collected at the surface of
the sea at Station 53, St Paul's Rocks, 0° 55' N., 29° 22' \V., on December 10, 1902.

They are perfectly typical Planocera, which I cannot distinguish specifically from
P. i>ellucida (Mertens). One of the specimens has the following dimensions:
length, 10 mm. ; breadth, 5 '5 mm.

This individual differs from that figured by A7on Graff in possessing decidedly
fewer eyes in the groups lying over the brain ; otherwise the resemblance it bears
to von Graff's example is exceedingly close.

The second specimen, which is not so well preserved as the first, is, however,
specifically identical with it

Planocera pellucida is well known as a holopelagic organism, and has frequently
been recorded from the Atlantic. For details of the structure of this interesting animal,
see von Graff's account in Zeit. f. wiss. Zobl., Bd. Iv. (1892), pp. 195-200, Taf. vii.

* For an account of llie other Turbellaria collected by the Expedition, see J. F. Geminill and R. T. Leiper, in
Tram. Unit. .W. E<lin., vol. xlv. (1907), pp. 819-827, reprinted in Report on Sti> //'. Itfsultg )'./;/.r,,.- ,5.1'. "Scotia,"
vol. v. (1909), pp. 129-138.

PART V.
CORALS.

V.-THE CORALS OF THE SCOTTISH NATIONAL
ANTARCTIC EXPEDITION.

BY J. STANLEY GARDINER, M.A., F.R.S.,

Professor of Zoology and Comparative Anatomy in the University of Cambridge.

(WITH TWO TEXT- FIGURES.)

The Corals of the Scottish National Antarctic Expedition. By J. Stanley
Gardiner, M.A., F. U.S., Professor of Zoology and Comparative Anatomy in the
University of Cambridge. Communicated by Dr J. H. ASHWORTH.

(MS. received May 14, 1913. Read July 7, 1913. Issued separately September 1, 1913.)

The collection of Scotia corals is naturally small, since the main work of the Scottish
National Antarctic Expedition was in high Southern latitudes, but, during the outward
and homeward journeys of the Scotia, collections were made, when opportunity pre-
sented itself.

The species are interesting distributionally, but their soft parts are largely decayed.
There are five species, of which one is a known species ; three dead masses, generically
determinable ; and the last, a new species, which may in future prove of even greater
interest than at the present time.

Madracis scoti&, n. sp. (figs. 1 and 2).

Colony branching and anastomosing, stems 2-5 mm. in thickness.

Calicles studding the surface, with no determinable arrangement, rising as distinct
flattened cones, the largest varying from '5 to T5 mm. high. The surface between is
conspicuous, as the cones are generally at least their diameter from one another, and in
places two or three times this distance (1-3 mm.) ; it is finely granular.

FIG. ]. FIG. 2.

Mat/rncissfotiir, n.s[>. Kig. 1, Corallum, natural sizr ; fig. 2, Calicle, siilarged.

The calicular opening varies up to 1*2 mm. in diameter, and is round. Above it
project the slightly exsert upper edges of the septa ; costae are not distinguishable.
The septa are 8 in number, and extend as straight partitions to the centre of the
calicle, where they fuse to the columella, which consists of a single upstanding rod.

In section, horizontal partitions are seen extending across the interseptal chambers.

(REPRINTED FROM THE TRANSACTIONS OF THE ROYAL SOCIETY OF EDINBURGH, VOL. XLIX., IT. 687-689.)

110 PROFESSOR J. STANLEY GARDINER ON

Such sections also indicate that the corallum thickens by deposition of carbonate of
lime on the surface, not within the chambers.

Locality. — Station 81, lat. 18° 24' S., long. 37° 58' W., Abrolhos Bank, 36 fathoms,
20th Dec. 1902 ; two pieces of probably the same colony. The genus is recorded by
MILNE EDWARDS arid HAIME from Reunion, 25 fathoms (Madracispellana, Valenciennes).

The specimen is very different from any previously described form, both in its raised
calicles and in the considerable distance which separates them. I refer it to Madracis,
but I cannot regard this identification as satisfactory without an examination of the
polyps ; indeed, I suspect its polyps will prove it to be the type of a new genus.

Desmophyllnm sp. ?

A small horn-shaped corallite, bent in the middle, over 20 mm. high by 5-6 mm.
across the calicle, cannot be referred to any known species. There is only one system
of the calicle complete, the rest having been broken into. The septa number 48,
of which 12 deep down in the corallite have trabeculae fusing to some slight degree
with one another. The specimen is stained brown on the outside.

Locality.— Station 542, Princesse Alice Bank, lat. 37° 56' N., long. 29° 11' W., 350
fathoms, 4th July 1904.

Caryophyllia prqfunda, Moseley.

MOSELEY, Report on the Corals, " Challenger" Expedition, p. 138, pi. I. fig. 6, a, b (1881).

Three complete specimens 30-34 mm. high, three further specimens with broken
stems but calicles complete, and broken fragments. Calicles varying from 16x14 mm.
up to 24 x 21 mm. in diameter.

I have compared the specimens with the types of CaryophyUia profunda, Moseley,
in the British Museum ; the latter consist of about twenty specimens, including two large
masses grown together as represented in the excellent figure on plate I. referred to above.
The present specimens agree with the type in shape, and some of them, as are some of
the types, are coloured brown on the outside, with perhaps a darker ring near the edge
of the calicle. The costse are rather more marked in the larger specimens ; all have a
peculiar shiny appearance on the outside of the corallite, and in no specimens do the
costse extend to the bases of attachment. The septa vary from 80 to 96 in number,
and the pali in correspondence vary from 20 to 24 plates. The columella consists
of twisted ribbons freely anastomosing underneath, the whole forming an oval-shaped
mass varying up to one-third the length of the calicle, and generally being rather
more than twice as long as broad. The regular arrangement of the ribbons in a line,
as represented in MOSELEY'S fig. 6a, is not typical either of the type specimen or of
the forms under consideration.

The type specimens, taken as a whole, differ from the present specimens in being
rather larger and more robust, viz. in having thicker septa and pali, coarser ribbons
to the columella, and slightly more exsert septa. The type specimens show great
variation, but individual specimens of the two collections agree in all particulars.

(HOY. soc. EDIN. TKANS., VOL. XLIX., 688.)

THE CORALS OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. Ill

Locality.— Station 461, off Gough Island, lat. 40° 20' S., long. 9' 56' 30" W., 100
fathoms, 23rd April 1904. MARENZELLER, "Valdivia" Report, p. 298, records a
number of specimens from Station 167, New Amsterdam, 49G in. deep. The Challenger
specimens were obtained off Tristan da Cunha, 100-150 fathoms.

The species, in its number of septa and pali and in the characters of its columella,
resembles both Caryophyllia cyathus (Ell. & Sol.) and Caryophyllia clavus, Sacchi.
MOSELEY remarks that it differs from C. davits in having fewer and less equally promi-
nent costs?, and in the greater abundance of its epitheca. C. clavus generally has a broad
base of attachment, while C. ci/atJtus is horn-shaped. Authentically named specimens
of C. cyathus and C. clavus in the British Museum scarcely help in elucidating the
differences ; they are smaller and less robust in the sense given above. If the three
really form one species, it seems certain that different localities have their special
growth- forms, such as var. smtthii (Stokes) off Devon and Cornwall, the Antarctic
form under consideration, etc.

A large specimen reputed to be C. clavus, in the British Museum, from the Caribbean,
is of interest as its pali converge towards each other in pairs over the tertiary septa,
a character which I do not recollect having seen in any other Caryophyllia. It was
almost certainly so named by DUNCAN himself. The fragments mentioned by MOSELEY,
p. 139, from Cape Verde Islands, are much closer to the ordinary forms of cyathus than
of profunda.

Amphihelia sp. ?

MARENZELLER ("Valdivia" Expedition, Bd. vii., T. xiv., 1904) and DUNCAN (Trans.
Zool. *S'oc., pi. 45, fig. 1, 1873) give admirable figures of the mode of growth of Amphi-
helia oculata (Linnaeus). The specimen before me is the basal anastomosing part of
such a colony as represented in the above figures, with similar tubular prolongations.
It was dead when obtained, and it is so much corroded that the species must be
regarded as uncertain.

The specimen comes from Station 542, Priucesse Alice Bank, lat, 37° 56' N., long.
29° 11' W., 350 fathoms, 4th July 1904. The genus has been recorded many times
from Northern oceans, but it was not obtained either by the Challenger or V«1<Hci«
Expeditions in Southern seas. The present record indicates that it is a cosmopolitan
genus.

Solenosmilia sp. ?

A mass of dead material from Station 542, Princesse Alice Bank, lat. 37° 56' N.,
long. 29° 11' W., 350 fathoms, 4th July 1904, belongs to this genus, but 1 am not sure
whether one piece is not Lop/iohelia ; the whole is very badly corroded.

The species is probably the cosmopolitan S. variabilis, Duncan, originally described

from N. Atlantic material. The ChaUcmji'i- records it from Tristan da Cuuha, 1000

fathoms, Prince Edward Island, 310 fathoms, and Ascension Island, 42 fathoms. The

Valdivia obtained it off the west coast of Portugal, 2200 in., and off the entrance

to the Mediterranean, 1300 m.

(ROT. soc. KLIN. TRANS., VOL. MIX.. fi8!).)

PART VI.
FURAMINIFERA.

VI-THE FORAMINIFERA OF THE SCOTTISH
NATIONAL ANTARCTIC EXPEDITION.

BY F. GORDON PEARCEY,

Bristol Museum ; late of the Challenger Expedition and Commission.

(WITH TWO PLATES.)

VI. - Forarninifera of the Scottish National Antarctic Expedition. By F. Gordon
Pearcey, Bristol Museum ; late of the Challenger Expedition and Commission.
Communicated by Dr J. H. HARVEY PIRIE. (With Two Plates.)

(Read May 26, 1913. MS. received August 2, 1913. Issued separately March 30, 1914.)

The fauna of the Polar regions is of deep interest to zoologists generally, that, of
the Antarctic specially so, and in this the Rhizopodist can justly claim his share.

Since the return of the Challenger Expedition in 1876, the later British expeditions
to this area, with the exception of the Scottish National Antarctic Expedition, did but
little in the way of sounding and trawling in the deeper waters of the Antarctic.
Although much additional work has been carried out, and many new and rare species of
the higher forms of marine life from this region have been brought to light, compara-
tively little has been added to the Rhizopod fauna since the results of the Challi-ityer
Expedition were published. The following pages on the Foraminifera are due to the
energy and enthusiasm of Dr W. S. BRUCE, F.R.S.E., and his colleagues. The genera
and species here enumerated and described have been obtained from samples of deposits
sent to me at intervals by Dr BRUCE and Dr J. H. HARVEY PIRIE, collected by them,
during the S.Y. Scotia Expedition in 1903-4, chicHy from the area of the Wcddell Sea,
where the ocean floor is covered with terrigenous deposits of Blue Mud, or (as Dr PIRIE
calls them) Glacial Muds and Clays. A few samples were also obtained from the
Southern Atlantic Ocean, over the areas where the deposits form the typical Globigerina
and Diatom Oozes. With the exception of two or three samples of washings of the
material taken by the trawl, the quantity available for examination was too meagre
to enable one to tabulate a comprehensive distribution chart, or a complete list of
the various genera and species that one might expect to find at the stations from
which the samples were obtained. These samples, however, pointed to there being a
rich Foraminiferal fauna.

Whenever the trawl or dredge was worked and a sufficient quantity of the deposit
was brought up and washed, a considerable number of Foraminifera was obtained,
more especially of the larger arenaceous types, which goes to prove that the Fora-
miniferal fauna is rich all over the Blue Mud area, and that the conditions of life
are congenial on these deposits — close up to the Antarctic Continent. It has also
been shown that similar conditions prevail in the Arctic seas and North Atlantic.
Mure systematic work with the trawl, dredge, and sounding machine, and a larger
quantity of material for examination, are necessary before one can speak definitely on
the forms to be met with in any part of the area in question, or on any specially

(REPRINTED FROM THE TRANSACTIONS OF THE ROYAL SOCIETY OF EDINBURGH, VOL. XLIX., pp. 991-1044.)

118

MR F. GORDON PEARCEY ON

defined deposit, except in the case of the pelagic species, which are known to diminish
size and numbers the farther south research is made. It is to the bottom-

in

living forms we must specially look, to furnish us with definite information on the
important questions relating to the geographical distribution of the Foraminifera. From
present information, however, it would appear, with the exception of two or three
genera and species, that the Foraminiferal fauna of both polar regions is strikingly
similar — a fact which is probably connected with the generally uniform conditions of
temperature extending over the bottom of the deep sea from pole to pole.

List of Stations at wJiich Foraminifera were obtained from the Deposits,
•with their Position, Depth, and Nature of the Bottom.

Station.

Latitude.

Longitude.

Depth in
Fathoms.

Bottom.

No. of
Deposit.

118

Stanley
Falklanc

Harbour,
1 Islands

»1

Blue mud

2

286

08 11 S.

34 17 W.

2488

Glacial clay

13

291

67 33 S.

36 35 W.

2500

,, mud or clay and boulders

Ua

295

66 40 S.

40 35 W.

2425

,, clay

15

300

65 29 S.

44 06 W.

2500

i) )>

16

301

64 48 S.

44 26 W.

2485

,, mud and boulders

17

313

62 10 S.

41 20 W.

1775

„ ,, or sand

21

337a

59 46 S.

48 02 W.

2110

,, mud — Diatom ooze

26«

338

59 23 S.

49 OS W.

2180

Diatom ooze — volcanic sand

266

342

56 54 S.

56 24 W.

1946

Globigerina ooze

28

346

54 25 S.

57 32 W.

56

)!

29

Burdwood Hank

387

65 59 S.

33 06 W-

2625

Glacial clay

30

416

71 22 S.

18 15 W.

2370

mud or clay

37

417

71 22 S.

16 34 W.

1410

,, and stones

38

418

71 32 S.

17 15 W.

1221

„ and rocks

39

420

69 33 S.

15 19 W.

2620

)5

40

421

68 32 S.

10 52 W.

2487

clay

41

438

56 58 S.

10 03 W.

2518

Diatom ooze — volcanic sand

45

447

51 07 S.

9 31 W.

2103

)>

46

451

48 06 S.

10 05 W.

1742

I.liatom-GVlobigeriua ooze

47

459

41 30 S.

9 55 W.

1998

Globigerina ooze

49

467

40 OS S.

1 50 E.

2645

j»

50

468

39 48 S.

2 33 E.

circa 2645*

»»

51

81

18 24 S.

37° 5S' W.

36

Coral rock and sand

1

Abrolhos Bank

1

The examination of the Hcotia deposits with regard to Foraminifera goes to prove,
as did the Challenger work, that Foraminifera are more abundant and attain a higher
development in size, especially the arenaceous forms, the nearer we approach the
slopes of continental land between depths of 500 and 1000 fathoms, even though the

* This sample was taken by the trawl between two soundings, viz. that at Station 467 and that at Station 469,
the latter depth being 2900 fathoms, in 39° i27' S., 5° 50' E. The probable depth is therefore about 2700 fathoms.

(HOY. Si)C. EDIN. TRANS., VOL. XLIX., 992.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 119

temperature be below 40° F. Such results were found over the Glacial Clay area in
the Weddell Sea (see Scotia Station 4 '20, where a fair sample of the deposit was
obtained). This high development and abundance of individual species is probably
due to vegetable and other organic material deposited m these areas, where the pelagic
life, both animal and vegetable, accumulates in vast quantities.

Those few samples of the deposits obtained by the Scotia Expedition which
I have examined for Foraminifera have enabled me to identify no less than 267
species ; 1 1 of these are new, and are described and figured in this report. It is
worthy of note that no less than 107 of them have been recorded from the warm
and cold areas of the Faroe Channel in the North Atlantic.

In the systematic list I have thought it advisable, for the advantage of future workers
in this comparatively unknown region, to denote briefly where each species has hitherto
been recorded, with its range of depth. I find as a result of my examination that the
Foraminifera] fauna of the Antarctic is richer south of 70° than north of it, and it
appears to me that a line of demarcation should be drawn at this point, although
the deposits are of a similar composition. This line will probably be found to apply
in a like manner to the higher forms of marine life.

The classification adopted in this paper is that of Mr J. A. CUSHMAN as far as he
has published a scheme, i.e. for the Astrorhizidae, Lituolidse, and Textulariidse, and
that of the late Dr H. B. BRADY for the remaining families.

SYSTEMATIC LIST, WITH NOTES ON DISTRIBUTION, AND
DESCRIPTIONS OF NEW SPECIES.

Family MILIOLID.E.

Sub-family NUBECULARIN^E.
Genus NUBECULARIA, Defrance.

Nubecularia bra<fyi, Millet.

Obtained in fair numbers at Station 346, 56 fathoms. This irregular, wild-growing
variety has its natural habitat in the shallow, warmer waters of the tropical and sub-
tropical seas ; it was, however, taken by the (Jli«ll''ntjer Expedition as far south as off
Kerguelen, in 20 to 60 fathoms.

Sub-family MILIOLININ.K.
Genus BILOCULINA, d'Orbigny.

Biloculina elongatn. <!'( irbigny.
This cosmopolitan species was obtained (few) at three stations, 346, 459, and 467,

(KOY. SOC. KHIN. THANK., Veil,. XI. IX., {,)'.>3.)

120 MR F. GORDON PEARCEY ON

Biloculina ringens (Lamarck).
Found (rare) at three stations, 346, 420, and 459. It occurs in all seas and at

all depths.

Biloculina depressa, d'Orbigny.

Taken sparingly at six stations, 342, 420, 447, 459, 467, and 468. It has a world-
wide distribution and a great bathymetrical range.

Biloculina serrata, H. B. Brady.
This characteristic species has been found in material from two stations, 342

and 459 (rare).

Biloculina murrhyna, Sch wager.

This species was also taken (rare) at Stations 342 and 459. Both this and the pre-
ceding species are characteristic of a Globigerina ooze-deposit.

Biloculina tubulosa, Costa.
A few specimens of this species were obtained from Station 459, 1998 fathoms.

Genus SPIROLOCULINA, d'Orbigny.

Spiroloculina tennis (Czjzek).
Taken (rare) in the Scotia material at Station 459, 1998 fathoms.

Spiroloculina limbata, d'Orbigny.
Obtained at Station 342, 1946 fathoms.

Genus MILIOLINA, Williamson.
Miliolina dentistoma, sp. nov. (Plate II. figs. 17-19.)

Test triloculine, broad, compressed ; peripheral margins thick and rounded ; visible
chambers, three or four ; the last chamber appears always imperfectly formed, with a
deep transverse suture.

Aperture a wide, irregularly curved fissure, with a thick lip-like margin from which
rise one or more short tooth-like projections. Two strong, prominent teeth situated a
short distance inside the mouth, at the base of the apical opening, are not observable
in a lateral aspect (see figs. 17-18, Plate II.).

Chamber walls smooth, and glazed externally, at times wavy. The length of the
figured specimen is about -^- inch (0'95 mm.).

Miliolina dentistoma was obtained in moderate numbers at Station 346, 56 fathoms.

Miliolina venusta (Karrer).

This distinct variety was found at five stations, 342, 346, 447, 459, and 467 (few or
rare). Miliolina venusta is essentially a deep-sea form ; so far as I am aware, it has

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 904.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 121

hitherto only been recorded from the Pacific Ocean ; its range of depth being from
56 (Scotia Station 346) to 2700 fathoms.

Miliolina cin-tifnrix (Bornemann).

A few specimens were found at Station 420, 2620 fathoms. These were all of
• normal character, and showed no signs of their tests being diminished in thickness, as
might be expected at such a depth, where the deposit gave such a low percentage
of CaCOj.

Miliolina oblonga (Montagu).

Obtained sparingly at Stations 346 and 459.

MilioUna hiifcitletita, H. B. Brady.

This doubtful Triloculine form was found at three stations, 291, 313, and 420, but
was not common. This species is hitherto known only from the Arctic Seas and
North Atlantic, so far as I am aware. I have found it moderately common in the
Faroe Channel in 630 fathoms. It is therefore of interest to find it fairly well dis-
tributed in the Weddell Sea.

Miliolina bucculenta, var. placentiformis, H. B. Brady.

This variety I obtained (rare) in the material from Station 420, 2620 fathoms.
BRADY records it in his report on the Challenger Foraminifera from two widely
separate localities — off Culebra Island, West Indies, 390 fathoms ; and Balfour Bay,
Kerguelen, 20 to 50 fathoms.

Miliolina seminulum (Linne).

This, the most common species of the genus Miliolina, has been found in all
localities from the most extreme point in the Arctic Seas to as far south as the edge
of the Antarctic ice, and from the tidal zone clown to 3000 fathoms. It was,
however, obtained only at one Scotia station, viz. 346, 56 fathoms.

Miliolina, fichteliana (d'Orbigny).
Obtained very sparingly at Station 468, 2645 fathoms.

Miliolina trigonula (Lamarck).
Taken at Station 346, 56 fathoms (rare).

Miliolina terquemiana, H. B. Brady.

This triangular striated form, which is very rare, was obtained from the material
taken at Station 346, 56 fathoms ; the only other known localities for this species are
from shallow water off Ceylon, and in littoral sand on the east coast of Madagascar.

(ROY. soc. EUIN. TRANS., VOL. XLIX., 995.)

122 MR F. GORDON PEAKCEY ON

Miliolina ferussadi (d'Orbigny).
Also obtained at Scotia Station 346.

Miliolina tricannata (d'Orbigny).
This cosmopolitan form was taken also at Station 346.

Sub-family HAUERINlNx*.

Genus ARTICULINA, d'Orbigny.

Articulina funalis, H. B. Brady.

This beautiful striate cylindrical species, in its typical form, was obtained (few) in
the Scotia material from Station 346. It is one of the commonest southern and
Antarctic shallow-water Foraminifera ; its other known localities are off Kerguelen,
20 to 120 fathoms, and in Humboldt Bay, Papua, 37 fathoms, where it was taken
by the CJiallenger Expedition in fair quantities.

Genus PLANISPIRINA, Seguenza.
Planispirina contraria (d'Orbigny).

Obtained (rare) at Station 346. This species has a wide distribution, and a range
of depth from 40 to 2160 fathoms. I have collected this species off the Shetland
Islands, among the Hebrides, and in the Faroe Channel.

Sub-family PENEROPLIDIKE.
Genus CORNUSPIRA, Schultze.

Cornuspira involvens, Reuss.

Taken (rare) at Station 346, 56 fathoms. It has previously been taken in the
North and South Atlantic, North and South Pacific, and as far north as 83° 19' N.,
ranging in depth from 7 to 1900 fathoms.

Cornuspira foliacea (Philippi).

Obtained (rare) at Station 346. This species has a wide distribution, but attains
its fullest development in the North Atlantic and Arctic Sea. I believe the Scotia
record to be the farthest south that this species has been found.

Sub-family KERAMOSPH^RIN^.
Genus KERAMOSPH.ERA, H. B. Brady.

Keramosphsera murrayi, H. B. Brady.

This exceedingly rare species, figured and described by the late Dr H. B. BRADY,
represents perhaps one of the most interesting living Foraminifera, both in form and
structure. Keramosplisem murrayi is peculiar to the Antarctic Ocean ; it w;is obtained

(ROY. soc. EDIN. TKANS., VOL. XLIX., 996.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 123

by the naturalists of the Challenger Expedition in lat. 53° 55' S., long. 108° 35' E., 1950
fathoms, in the Diatom ooze area ; but only five specimens were found after examination
of a large mass of the ooze, three of these being found after publication of the
Challenger Report. It is therefore of interest to have obtained a perfect specimen
among the material from Station 420, 2620 fathoms, in the Weddell Sea, outside the
diatomaceous zone, in a terrigenous deposit of glacial mud containing but a trace of
carbonate of lime ; while it is worthy of note that 19 '29 per cent of carbonate of lime
was found in the Diatom ooze at the Challenger station (157) in which the Keramo-
sphsera was first discovered.

Family ASTRORHIZID^.
Sub-family ASTRORHIZIN^.

Genus ASTRORHIZA, Sandahl.

Astrorhiza crassatina, H. B. Brady.

The Scotia specimens were obtained from the washings of trawled material, and
reached me more or less damaged. Had they been placed in spirit when first taken, the
general character and structure of the whole organism might have been studied, as they
were present in fair quantity at Station 291.

Distribution. — Station 291, glacial mud, 2500 fathoms, common; Station 295,
glacial mud, 2425 fathoms, few ; Station 313, glacial mud, 1775 fathoms, rare. It
has also been recorded from four stations in the North Pacific : near Japan, in 36 to
943 fathoms; off James Island, Galapagos, in 885 fathoms, from a Globigerina ooze;
and from the Faroe Channel, by the Porcupine, at a depth of 640 fathoms.

Astrorhiza arenaria, Norman.

This interesting species was obtained by the Scotia, both in the character of
branching cervicorn masses and the stellate form, at Station 291, 2500 fathoms, glacial
mud, and Station 420, 2620 fathoms, glacial mud. It has been recorded from the
Faroe Channel by H. B. BRADY and by myself at several points, ranging in depth
from 530 to 650 fathoms ; also from the coast of Norway (M. and G. 0. SARS) ; from
Ears Fjord, 180 fathoms (NORMAN) ; and at the Challenger Station 142, off the Cape
of Good Hope, 150 fathoms.

Genus SYRINGAMMINA, H. B. Brady.
Syringammina minuta, sp. nov. (Plate II. figs. 1-2.)

Test adherent, consisting of a broad oval mass of branching tubes radiating from a
common centre, and arranged in the form of a network. Walls arenaceous, made up of
practically uniform fine grains of sand with a small amount of inorganic cement.

(HOY. soc. EDIX. TKANS., VOL. XLIX., 997.)

1:24 MR F. GORDON PEARCEY ON

Apertures minute, situated at the peripheral ends, closed in by very fine argillaceous
material. Colour, light gray with a yellowish tint. Diameter, one-sixth of an inch
(4 mm.).

The general contour and minute character of the test of this species are, with the
exception of the size of the arenaceous particles of which .the walls of the test are built
up, similar to S. fragilissima (BRADY).

Two specimens of Syringammina minuta were obtained from Station 420, 2620
fathoms, one being badly damaged owing to the excessively fragile nature of the test,
which consists of a network of branching tubes more or less contorted and irregularly
disposed (fig. 2). These are generally filled with dark olive or yellowish-green granu-
lated sarcode similar to that found in the tests of the larger arenaceous Foraminifera.

The external diameter of the tubes varies somewhat, with a maximum of -^ inch
(0'5 mm.). Their external surface is fairly smooth ; the interior is quite smooth and
well finished. The walls are very thin, averaging about ^^ inch (0'08 mm.).

It is worthy of note that similar arenaceous forms were collected by the Challenger
at a depth of 1000 fathoms off the Azores, but they were so fragile that, in washing the
mud through sieves, they were broken to fragments and lost sight of, their character at
that time not being known. Careful searching and preservation in future deep-sea
dredging and trawling will probably bring additional specimens to light for further
investigation.

Genus RHABDAMMINA, M. Sars.

Rhaljdammina discreta, H. B. Brady.

Fine, well-developed specimens, in their typical form, were obtained in plenty at
Station 420, 2620 fathoms, glacial mud; and a few at Station 468, 2700 fathoms,
Globigerina ooze. This species has a wide range of distribution, having been recorded
from the North and South Atlantic, from the North, South, and Western Pacific, in
depths of from 350 to 2475 fathoms; off Kerguelen, 120 fathoms; and off the coast
of Greenland, at a depth of 20 fathoms.

Rhabdammina cornuta, H. B. Brady.

This species was represented very rarely in the material examined from Station 420,
2620 fathoms.

Rliabdammina abyssorum, M. Sars.

This interesting species, in the typical form bearing the three radiating arms, with
the walls of the test coarsely built exteriorly, and of a darkish blue-grey colour, was
obtained at Stations 313, 337a., and 417 — more abundant at 337c< than at the other two.
It has a world-wide distribution, and is found off the coast of Spitsbergen in great
quantities ; its range of depth is from 71 to 2439 fathoms.

(ROT. soc. EDIN. TRANS., VOL. XLIX., 998.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 125

Genus MARSIPELLA. Norman.
Marsipetta cylindrica, H. B. Brady.

Fragmentary portions of both the straight and the curved tests of this species were
obtained amongst the material from Station 342, 1946 fathoms, in a deposit of
Globigerina ooze. They are of a more robust character than the type specimen. This
species has been recorded from the Faroe Channel, both in the warm and in the cold
area, in 530 to 640 fathoms; in the South Atlantic, 1900 fathoms; South Pacific, 210
fathoms; North Pacific, 139 to 1875 fathoms; and oft" Juan Fernandez, in 1375
fathoms.

Genus BATHYSIPHON, G. 0. Sars.
Bathy siphon filiformis, G. 0. Sars.

Obtained at Stations 301, 418, and 420, but nowhere abundant. The Scotia
specimens, for the most part, show a considerable number of mineral particles built into
the outer portion of the walls of the tests, giving them a more robust appearance than
that of the type. This is due, doubtless, to the nature of the glacial deposit on which
they were taken. Before the volume on the Challenger Foraminifera was published,
Bathysiphon was looked upon as an exclusively northern type ; it is therefore of
interest to have found it so far south. Since the Challenger publication it has been
recorded from six localities in the North Pacific, in depths from 617 to 1201 fathoms.
It was taken for the first time in considerable abundance by G. 0. SARS in the
Hardanger Fjord, and occurs in deep water in the Bay of Biscay ; off Banda, Amboyna,
in 1425 fathoms ; and in the warm area of the Faroe Channel, 570 fathoms.

Genus RHIZAMMINA, H. B. Brady.
Ehizammina indivisa, H. B. Brady.

Represented in the Scotia material by three specimens and a few fragments from
Station 420, 2620 fathoms. Known from six stations in the North Pacific (CUSHMAN),
210 to 2086 fathoms ; off the Cape of Good Hope (BRADY), 150 fathoms ; and from the
Faroe Channel (BRADY and PEARCEY), 540 fathoms.

Rhizammina algsefonnis, H. B. Brady.

This peculiar wild-growing species was obtained in a fragmentary condition, but
typical in structure (dichotomously branching flexible tubes) at Stations 338 and 447,
both Diatom ooze. The species is so delicate in construction that it is almost impossible
to preserve it intact in a dry condition. This is the most southern record for the species.

Distribution. — South Pacific, oft' Fiji, in '210 fathoms ; close to the equator, in 2425
fathoms ; in abundance between Juan Fernandez and the coast of Chili, 2160 fathoms
(BRADY). North Pacific, five stations, 1875 to 2900 fathoms (CUSHMAN).

(ROl'. SOC. EDIN. IKAN8., VOL. XLIX., HO'J.)

126 MR F. GORDON PEARCEY ON

Sub-family SACGAMMININ^.

Genus P.SAMMOSPH^RA, F. E. Schulze.

Psammosphterafusca, F. E. Schulze.

This would appear to be one of the commonest species of the Antarctic Seas, being
obtained by the Scotia at no less than nine stations. 286, 313, 337a, 342, 416, 417,
418, 420, and 468. It attains a larger size and is more abundant in the far south
stations on glacial deposits than in the Globigerina oozes. Psammosph&ra fusca has a
very wide distribution. It has been taken in shallow water oft" the coast of Norway
(120 fathoms), and in 40 to 60 fathoms off Loch Scavaig, Skye ; but its natural habitat is
evidently the abyssal depths of the great oceans, where it attains its highest development,
being known from several stations in the North Pacific, 1495 to 3026 fathoms; seven
stations in the South Atlantic, 150 to 2800 fathoms; and ten stations in the North
Atlantic, 440 to 2750 fathoms.

Genus SOROSPH^RA, H. B. Brady.
Sorosphsera confusa, H. B. Brady.

Several specimens of this colony- building form were obtained at Station 420. Hitherto
it has been recorded from the Rockall Bank, 630 fathoms ; off the Azores, 900 fathoms ;
the Faroe Channel, 542 fathoms ; and the middle of the North Pacific, 2900 fathoms.
It is therefore a wide extension of its range to have found it in the Weddell Sea.

Genus SACCAMMINA, M. Bars.
Saccammina socialis, H. B. Brady.

Several specimens of this rare species were obtained at Stations 313 and 420.
Known from the North Pacific, North Atlantic, and Faroe Channel.

Saccammina sphserica, M. Sars.

This species was obtained at five stations, 291, 301, 313, 417, and at 420, in con-
siderable numbers, of large size, and typical in all its characters. Recorded from the
North Atlantic (common) ; fjords of Norway ; Faroe Channel ; off the shores of Franz
Josef Land; North Pacific; and near the Antarctic Continent (Challenger], where
BRADY describes it as indistinguishable from the North Atlantic type.

Genus PROTEONINA, Williamson.
Proteonina (Reophax) difflugiformis, H. B. Brady.

This apparently world-wide species was taken at eleven of the Scotia's stations, viz.
118, 300, 337ci, 338, 342, 387, 447, 451, 459, 467, and 468. These vary in depth
from 2^ to 2645 fathoms, and include all the varieties of deposit obtained by the Scotia ;
but it is noteworthy that it was not obtained at any of the stations south of the circle.

(HOY. sue. EDIN. TRANS., VOL. XLIX., 1000.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 127

Genus PELOSINA, H. B. Brady.
Pi'Iofsino. arborescens, sp. nov. (Plate I. figs. 1-5.)

Test vase-shaped, elongate, sub-cylindrical, erect, smooth, and unctuous to the touch,
more or less flexible in the living state, rounded at the base, gradually narrowing towards
the superior extremity, which is drawn out into a slender main tubular chamber, some-
what dome-shaped at its base.

From about one-third of the test upwards a number of dichotomous tubular branches
extend at irregular intervals with graceful curves. These branches open out into the
main chamber (Plate I. fig. 2). The wall at the base of eacli outgrowth is thicker and
somewhat swollen, but after a short distance becomes more uniform in diameter. The
walls of the main chamber are thick, composed of fine mud deposited on a slender
chitinous envelope extending to the terminal apertures of the branching tubes, where it
becomes quite thin, and consists of little more than a membrane, so thin and soft that
it readily collapses on drying.

Colour, vandyke brown. Height of test, 1J to 2 inches (30 to 48 mm.) or more.
A few minute filamentous outgrowths come off from the outer wall of the extremity of
the basal portion, and appear in some cases to be tubular, but they are so fragile that
they break off with the slightest manipulation. It is probable that these filaments serve
to fix the test in an upright position in the deposit on the ocean floor.

Dr H. B. BRADY, in describing Petoxina variabilis, says specimens of this (mean-
ing P. variabilis) or a closely allied organism have been dredged off the west coast of
Scotland. The specimens were nearly an inch in length, with walls somewhat flexible,
the shape tolerably regular, long and tapering. The superior extremity, which is tubular
and much drawn out, is divided into a number of minute branches, each terminating in
an aperture exactly as shown in the larger arms of Astrorhiza limicola. It is therefore
very probable that Dr BRADY had seen a damaged dried specimen of P. arborescens
sent to him by the late Dr DAVID ROBERTSON of Cumbrae, who also drew my attention
to the fact that Dr BRADY had never seen a living specimen of P. arborescens when he
published his description of P. variabilis.

Fig. 1, PI. I., represents a typical example of Pelosiria arborescens taken by the trawl
on board Sir JOHN MURRAY'S yacht Medusa, and kept alive by me for five months in a
tank at Millport, Cumbrae. During this time it did not move from the position in which
it was first placed in the tank, but repaired several of the damaged terminal tubular
branches, and displayed the granulated sarcode and ramifying pseudopodia beautifully
extended from the terminal apertures. It was sketched in this condition, and afterwards
preserved in glycerine (fig. 1, Plate I.), and has been reproduced from the original draw-
ing made by my son, F. J. M. C. PEARCEY. Figs. 4-5, Plate I., represent a dried broken
specimen from the \\Vddcll Sea. Although much danuigcd, it shows all the essential
characters of the specimen from the west coast of Scotland.

Distribution.- Si-n/in Si a i ion 420, 2620 fathoms, on a deposit of glacial mud or clay.

(HOY. SOC. KIiIN. TRANS., VOL. XLIX., 1001.)

128 MR F. GORDON PEAROEY ON

The test is of a lighter colour and rather coarser texture than the specimen figured in
fig. 1, PI. I. Off the west coast of Scotland, 50 to 90 fathoms, and sparingly in the
deep-water area of Loch Fyne. Off Skate Island, 100 to 107 fathoms. It has also been
taken off the coast of Norway (NORMAN), and doubtful specimens, probably belonging to
this species, were obtained on the Triton, 1882, in the Faroe Channel, 640 fathoms.

Pelosina cylindrica, H. B. Brady.

Specimens with tests more delicate in the structure of their walls and rather
smaller than the type specimen were obtained sparingly at Stations 313, 417, and 420.
Previously known from the North Atlantic, North and Western Pacific, off New Zealand,
and off the Antarctic Continent (Challenger}.

Genus TECHNITELLA, Norman.

Technitella raplianus, H. B. Brady.

Two or three specimens answering to the description and figures given by BRADY
of this rare form were obtained from the Scotia material from Station 420. Obtained
by the CJiallenger off Kandavu, Fiji Islands, in 210 fathoms.

Technitella asciformis, sp. nov. (Plate II. figs. 3-4.)
(Greek asJtos, a leather bottle.)

Test free, flask-shaped, consisting of a single chamber, ovately curved at the apical
end ; produced to a short thick neck ; terminating in an everted phial-like lip grooved
on the inside ; aperture a small circular opening. Walls comparatively thick, composed
of fine spicular material and fine argillaceous cement, with a strong chitinous lining
which permits of the whole test becoming more or less flexible.

Colour, gray. Length, •§- inch (3 mm.).

Its beautiful symmetry of contour is that of an old-time leather bottle. There are
several fine slightly projecting ridges descending from beneath the everted lip to below
the shoulder, where they merge into the general structure of the wall.

One specimen only was found, among the material from Station 420, 2620 fathoms.

Genus THOLOSINA, Ehumbler.
Tholosina (Placopsilina) bulla, H. B. Brady.

Several specimens of this peculiar little species were obtained from Station 420,
attached to Rhaltdammina abyssorum and Rhizammina mdivisa, the light-coloured
test showing up prominently against the darker-coloured walls of its hosts. It has
hitherto been recorded from the North and South Atlantic (BRADY) ; North Pacific
(BRADY, CUSHMAN) ; and taken in the Faroe Channel by the Triton, although not
recorded, in 5] 5 fathoms.

(ROY. 8OC. EDIN. TKANS., VOL. XLIX., 10U2.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 129

(irnus WKBBINELLA, Rhumbler.

Webbinella (Webbina) hemisphserica, Jones, Parker, and Brady.

This simple little form was found, but very sparingly, in the washed material from
Station 420, attached to rock- fragments. Known from the English coasts, Faroe
Channel, and North- West Pacific, at depths varying from 25 to 630 fathoms.

Genus CRITHIONINA, Goes.
Critliionina pisum, var. hispida, Flint.

A fine specimen of this simple-celled variety, considerably larger than the type, was
found attached to a rock-fragment from Station 420. The chamber walls are built of
sponge spicules arranged perpendicularly in an amorphous siliceous cement, and lined
interiorly with a delicate chitinous membrane.

CUSHMAN records this species from two stations in the North Pacific, 1259 and
1342 fathoms.

It was taken by the Triton and Knight Errant Expeditions in the Faroe Channel,
570 to 640 fathoms, and recorded by me under the generic name of Sorosphsera sp.,
with several others of a similar character.

Genus THURAMMINA, H. B. Brady.
Thurammina papillata, H. B. Brady.

Obtained (rarely) from Stations 291, 420, and 459. This species has a world-wide
distribution, but appears to prefer the deep waters of the great oceans, where it has
been found from depths of 540 to 3125 fathoms.

Thurammina albicans, H. B. Brady.

Two specimens of this rare species were found in the material from Station 342, in
a deposit of Globigerina ooze. Recorded by BRADY from Challenger Stations 323 and
246, off South America and in the North Pacific respectively, 1900 and 2050 fathoms.

Thurammina, favosa, var. reticulata, var. nov. (Plate I. figs. 11-12.)

Test free, consisting of a single chamber ; walls thin, composed of minute angular
mineral particles firmly compacted in a siliceous cement of a bright ferric-red colour ;
apertures, one or more, round, minute (non-papillate or tubular), situated at irregular
intervals over the surface of the test, and do not project beyond the exterior of the
chamber wall.

The test is spherical or nearly so ; the exterior is marked by an irregular network
of raised ridges, which easily distinguish this form from the type T. favosa, Flint;
the ridges are robust and irregular, of a much lighter colour than that of the wall

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 1003.)

130 MR F. GORDON PEARCEY ON

generally. The colour of the latter, however, shows through, and makes this species
conspicuous by its colouring alone. The exterior aspect is illustrated in the drawings.

Diameter from T\> to -.,;a inch (0'5 to 1'5 mm.).

Obtained in moderate numbers at Station 420, 2620 fathoms.

Sub-family HYPERAMMININ^.

Genus HYPERAMMINA, H. B. Brady.

Hyperammina subnodosa, H. B. Brady.

Obtained at Station 420, moderately common. Known from the North and South
Atlantic and North Pacific (BRADY, and CUSHMAN), 20 to 2384 fathoms. Found
abundantly as far north as the coast of Greenland, 20 to 450 fathoms.

Hyperammina elongata, H. B. Brady.

From Stations 291 (few) and 313 (rare). Walls built of coarser material than is
seen in typical specimens. Only one specimen complete, with the proloculum perfect.
Distribution cosmopolitan, in depths of from 110 to 3125 fathoms.

Hyperammina friabilis, H. B. Brady.

A few typical specimens from Station 420, in a more or less fragmentary condition,
but with the proloculum perfect. This species also has a wide distribution — off the coast
of Greenland, Faroe Channel, North Atlantic, South Atlantic, North Pacific, 20 to 3800
fathoms.

Genus SACCORHIZA, Eimer and Fickert.
Saccorhiza (Hyperammina) ramosa (H. B. Brady).

Obtained at Stations 291, 313, 337a, 342, 417, and 420, although nowhere
numerous. Known from as far north as Franz Josef Land, and appears to be present in
all seas wherever arenaceous organisms are found, at depths of from 39 to 3125 fathoms.

Genus TOLYPAMMINA, Rhumbler.
Tolypammina (Hyperammina) vagans (H. B. Brady).

Fine typical specimens, but of a dark grayish-brown colour, from Stations 291, 417,
and 420. World-wide distribution. Bathymetrical range, 20 to 3800 fathoms.

Genus AMMOLAGENA, Eimer and Fickert.

Ammolagena ( Webbina) clavata (Jones and Parker).

From Stations 291 (common), 313, 342, and 420 (rare), attached to Rhabdammina,
Psammo»ph&ra, Saccammina, and on rock-fragments. A widely distributed form-
off the coast of Norway, Faroe Channel, North Atlantic, South Atlantic, North Pacific.
90 to 2000 fathoms.

(ROT. soc. EDIN. TRANS., VOL. XLIX., 1004.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 131

Sub-family AMMO-DISCING.

Genus AMMODISCUS, Reuss.
Ammodisous incertus (d'Orbigny).

Found sparingly at Station 337 a. A species of very wide distribution and
extensive range in depth. 15 to 2110 fathoms.

Genus GORDIAMMINA, Rhumbler.

Gordiammina (Ammodiscus} ckaroides (Jones and Parker).

This brilliant and beautifully coiled little species has been taken at Stations 342,
416, 447, 451, and 459, although not in any abundance. Widely distributed, from as
far north as the Faroe Channel to as far south as off Kerguelen, although the present
record extends its southern limits. Bathymetrical range, 90 to 2575 fathoms.

Gordiammina (Ammodiscus] gordialis (Jones and Parker).

Sparingly at Station 342 only. Known from as far north as off Franz Josef Land
to as far south as off Kerguelen. 55 to 3125 fathoms.

Family LITUOLID^.

Sub-family ASCHEMONE LLIN.E.

Genus ASCHEMONELLA, H. B. Brady.
Aschemonella catenata (Norman).

A few specimens from Station 420. A widely distributed species; in the Atlantic
and Pacific from depths of from 390 to 2900 fathoms.

Aschemonella ramuliformis, H. B. Brady.

One or two specimens of this rare species also obtained from Station 420. Hitherto
known only from the South Atlantic and North Pacific (Challenger}, from depths of
1900 to 3125 fathoms.

Sub-family REOPHACINyE.

Genus REOPHAX, Montfort.

Reophax adunca, H. B. Brady.

This rare species was obtained sparingly at Station 420 only. Known from the
Faroe Channel, North Atlantic, South Atlantic, southern Indian Ocean, and Pacific.
Range of depth, 540 to 2'JOO fathoms

(ROY. HOC. EDIN. TUAN8., VOL. XLIX., 1005.)

132 MR F. GORDON PEARCEY ON

Reophax nodulosa, H. B. Brady.

This beautifully built form, one of the largest species amongst recent arenaceous
Foraminifera, is represented in the Scotia collection from seven stations, viz. 301, 313,
416, 417, 418, 420, and 447. At Station 417 it was in great numbers and of very
large dimensions, several specimens measuring more than one inch in length, and girth
in proportion, the finest I have ever seen. This species has practically a world-wide
distribution and a great range in depth, from comparatively shallow water to
3950 fathoms.

Reophax robustus, sp. nov. (Plate I. figs. 6-10.)

Test straight ; composed of few (two to four) chambers or segments ; chambers
spherical, last chamber generally somewhat elongated, each much larger than its pre-
decessor ; with a well-formed and firmly constructed tubular orifice ; walls composed of
very coarse sand-grains, compactly built, and beautifully ornamented externally with
large angular mineral particles, such as quartz, felspar, magnetite, garnet, zircon, and
hornblende, giving a beautiful but very robust appearance to the whole structure.
Length, y1^- to ^ inch.

Reophax robnstum may be noted as one of the most beautiful members of the genus ;
it is readily recognised by its robust appearance and body-colour, which is of a yellowish
gray. This is broken up, however, by the variously coloured minerals which stand out
prominently from the walls, in which they are firmly cemented by a light yellow-brown
siliceous cement.

The interior of the chamber walls is smooth or nearly so, but appears to possess
little or no chitinous lining. The simple tubular apertures are of a bright ferric-brown
colour, which stands out conspicuously at the distal end of each chamber, where they
terminate in a drawn-out tubular neck, giving the test a beautifully finished appearance.

Reophax robustum is one of the most abundant species obtained at Station 420,
2620 fathoms, but it was not found in the material from any of the other localities.

Reophax dentalinifot'mis, H. B. Brady.

Obtained in its typical form at Stations 313 and 447, but rare. This species has a
fairly wide distribution in the depths of most of the great ocean basins.

Reophax membranacea, H. B. Brady.

This little, delicate, semi-chitinous form was found, rare, at Station 313. It is
exclusively a deep-sea species, and comparatively rare, ranging in depth from 1775 to
3950 fathoms.

Reophax scorpiurus, Montfort.

A very common cosmopolitan species, but apparently sparing!}' distributed in
Antarctic Seas. Found only at Station 342. Obtained by the Challenger off Heard
Island, in 75 fathoms.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 100G.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

t.r ilistans, H. B. Brady.

This rare species was obtained very sparingly and only in imperfect condition in the
material from Station 459.

Distribution. — Faroe Channel, 355 fathoms; off Kerguelen, 1570 fathoms; North
Pacific, where it is moderately common in the deepest waters.

Iteophax pilulifera, H. B. Brady.

Obtained at Stations 291, 301, 313, and 420 ; not common. A comparatively rare
species, peculiar to deep water. Known from off the south of Ireland, South Atlantic,
and Pacific. Depth, 437 to 2(JOO fathoms.

Genus HORMOSINA, H. B. Brady.

Hormosina normani, H. B. Brady.

This large species, characterised by its fine texture, smooth surface of its chamber
walls, and dark brown cement, was obtained in fair abundance and of very large size at
Station 417. Several specimens from here measure -j-V mcn in length, with the final
segment fully -x% inch across. It was not uncommon at Station 291, and rare at
Station 313.

The species is rather an uncommon one, and is exclusively a deep-water form.
Wherever taken it is usually found in fair numbers, but owing to its delicate structure
it is seldom in perfect condition.

Habitat. — North Atlantic, 1380 and 1750 fathoms; Pacific (more abundant),
1100 to 2900 fathoms (Cliallenger).

Hormosina globulifera, H. B. Brady.

Stations 295 and 420, rare. Known from all the great ocean basins, in depths
between 440 fathoms and 2900 fathoms.

Hormosina irregular is, sp. nov. (Plate I. figs. 13-15.)

Test irregular ; composed of from two to five subspherical chambers, of which those
at the commencement are the smallest. They are all closely embracing. Apertures
small, each at the end of a definite tubular, somewhat cone-shaped neck (which is often
of a bright ferric-brown colour from the tip of the neck downwards) situated at the
apex or distal end of each chamber. Successive chambers may be placed at right
angles or obliquely to their predecessors. Walls finely arenaceous, thin ; surface smooth
both within and without. Colour, light pearly gray. Length, -J- inch (3 mm.) or more.

This interesting form was taken sparingly at Station 420 only.

(ROY. soc. ED IN. TRANS., VOL. XLIS., 1007.)

134 MR F. GORDON PEARCEY ON

Sub-family TROCHAMMININyE.

Genus HAPLOPHRAGMOIDES, Cushman.

Haplophragmoides (Haplophragmium} scitulum (H. B. Brady).

This nautiloid form was obtained at Stations 291, 313, 416. and 417, but rare.
Recorded from both Atlantics and from the Pacific, ranging in depth from 66 to
2900 fathoms.

Haplophragmoides (Trochammina) trullissatum (II. B. Brady).

From Stations 300, 337«, 416, and 420, rare. World-wide distribution, ranging in
depth from 570 to 3950 fathoms.

Haplophragmoides (Haplophragmium} glomeratum (H. B. Brady).
This minute spiral form was obtained at Station 420 only (rare). A common
species in Arctic Seas and in all the ocean basins, ranging in depth from 23 to
2675 fathoms.

Haplophragmoides (Haplophragmium) canariense (d'Orbigny).
At Station 337a only (rare). Cosmopolitan ; at depths of from 2 to 3950 fathoms.

Haplophragmoides (Haplophragmium) rotulatum (H. B. Brady).

This rare and exclusively deep-sea form was taken at Station 337a (few) and
Station 342 (rare). Previous records: — North Atlantic, BRADY; Pacific, CDSHMAN.

Haplophragmoides (Lituola) subglobosum (G. 0. Sars).

At-Stations 300, 313, 416, 417, 418, 420, and 447, but nowhere common. Cosmo-
politan, and from quite shallow water down to 3950 fathoms.

Haplophragmoides umbilicatum, sp. nov. (Plate II. figs. 8-10.)

Test free ; roughly nautiloid ; piano-spiral; composed of three coils. Walls compara-
tively thin (unequally) ; deeply umbilicated on both sides ; very rough and robust
externally, almost smooth internally ; generally with a thin chitihous lining ; chambers
broad and arched ; peripheral margin roughly involute ; first and second coils often
exposed in the umbilical region. Six to seven chambers in the last-formed coil ;
aperture a crescentic slit at the base of the apertural face. Colour a light yellowish
brown. Diameter T]r inch (2'L mm.) or less.

This species is a coarse sandy form resembling in some respects Haplophragmoides
rotulatum, but is readily distinguished from that species by the character of its
chambers. These are larger, not so compactly built, and more coarsely arenaceous
(Plate II. fig. 10), also the last chamber is subspherical.

(ROY. SOC. EDIN. TRANS., VOL. XL1X., 1008.)

FORAMTNIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 135

Haploph/ragmoides umbilicatum was found in abundance in the deposit from
Station 420, 2620 fathoms, glacial mud, but not from any other.

Genus CRIBROSTOMOIDES, Cushman.

Cribrostomoides bmdyi, Cushman.
(Haplophraymium latidorsatum (Bornemann)).

Obtained from Stations 313, 418, 420, 459, and 467 (rare). One of the commonest
of arenaceous deep-sea Foraminifera. Range, from Franz Josef Land to the Antarctic
Continent, and in all oceans, at depths of from 113 to 3950 fathoms.

Genus CYCLAMMINA, H. B. Brady.

Cyclammina cancellata, H. B. Brady.

Good specimens of this species, the largest of its genus, with its wonderful
labyrinthine structure, the function of which has yet to be discovered, were obtained,
but rarely, at Stations 420 and 447. North and South Atlantic and Pacific Oceans,
75 to 2900 fathoms.

Cyclammina pusilla, H. B. Brady.

This handsome little species has been found at seven stations, viz. 300, 313, 338,
416, 417, 420, and 447, but nowhere common. Not known, so far as I am aware, from
the North Atlantic, and from bub one station in the South Atlantic and one in the
Antarctic (Challenger}. Common in the Pacific (CUSHMAN). 93 to 2879 fathoms.

Cyclammina contorta, sp. nov. (Plate II. figs. 5-7.)

Test nautiloid, compressed, slightly urnbilicate, convex towards the peripheral edge ;
composed of from three to four convolutions, the last two completely enveloping the
others ; segments numerous, usually fifteen in number. Walls of test finely arenaceous,
and thin externally; body of walls and throughout the labyrinthine structure with
larger angular mineral particles incorporated. The chambers of the first convolution
and the earlier ones of the second are thin-walled and but slightly labyrinthine : all the
succeeding chambers become partly labyrinthine, and the last three or four completely
so (Plate II. fig. 7). Suture lines slightly depressed and contorted anteriorly ; exterior
smooth, slightly iridescent ; imperforate in the perfect state, but when the outer wall is
rubbed or broken away the open tubules of the labyrinthine communications with the
main chambers become exposed to view ; aperture usually a curved slit, but in some
specimens this is altered into three or more circular openings or pores situated at the
base of the apertural face.

Colour vandyke brown near the area of the umbilicus, becoming much lighter at the
periphery ; suture lines very dark, almost black.

Diameter, 2 '5 to 4 '2 mm.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 1009.)

136 MR F. GORDON PEARCEY ON

This species in its earlier stages of development closely resembles Cyclammina
pusilla, and this resemblance can be seen when the later convolutions are broken away,
leaving the earlier chambers exposed (Plate II. tig. 7). All the specimens I have
examined are of the microspheric form.

Obtained at Stations 417 and 420, but not abundant.

Genus AMMOBACDLITES, Cushman.
Ammobaculites (Haplophragmium) agglutmans (d'Orbigny).

This, perhaps the most common species of all arenaceous Foraminifera, was obtained
from Stations 337a, 338, 342, 447, 459, and 467, but nowhere very abundantly. Its
distribution is world-wide, and its range of depth from quite shallow water down to over
3000 fathoms.

Ammobaculites americanus, Cushman.

From Station 342 only (rare). BRADY records this species, under the name of
IlapIopJt ragmium fontinense, from the South Atlantic, 1900 fathoms; from the South
Pacific, 1375 fathoms; and amongst the islands off the coast of Patagonia, 40 to 140
fathoms. CUSHMAN also records it from the Mexican coast, 772 fathoms.

Ammobaculites (Haplophragmium) tenuimargo (H. B. Brady).
This interesting species was taken (rare) at two stations, 313 and 420. It is a rare
form ; only six other localities, very wide apart, are known where this species has been
obtained : in the Faroe Channel, warm area, 530 fathoms ; south-west of the Canaries, in
2740 fathoms; east of Buenos Aires, 1900 fathoms; north of Papua, 1070 fathoms;
east of New Zealand, 1100 fathoms; and at one station in the North Pacific, 3950
fathoms.

Genus PLACOPSILINA, d'Orbigny.

Placop>silina vesicularis, H. B. Brady.

Obtained (rare) at one station only, 420, 2620 fathoms, attached to rock-fragments.
This species, which is generally associated with a rich arenaceous fauna, has hitherto
been taken in the Faroe Channel, 560 fathoms; between the north-west of Ireland
and the Rockall Bank, 630 to 1443 fathoms; from the South Atlantic east of Buenos
Aires, 1900 fathoms. These are the only localities known; it is therefore worthy of
note that this species has been found in the Weddell Sea.

Genus TROCHAMMINA, Parker and Jones.

Trochammina (Haplophragmium} nana (H. B. Brady).

Taken at two Scotia stations, 118 and 337a (rare). This minute species appears to
be at home either in deep or shallow water ; it has a wide distribution. In the Arctic

(ROY. soc. KDIN. TRANS., VOL. xux., 1010.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 137

Seas it is found abundantly, in depths ranging from 89 to 145 fathoms; and it
has been taken in the North and South Atlantic, in 1900 to 2350 fathoms ; also in the
Pacific Ocean, in depths varying from 245 to 3125 fathoms.

Trochammina (Haplophragmium) globigeriniformis (Parker and Jones).

This species, one of the most beautifully constructed forms of the genus, has been
taken at six stations, viz. 313, 337a, 342, 417, 447, and 459, but nowhere common.
Trochammina globigeriniformis has a world-wide distribution and a great range of
depth, from Franz Josef Land, in the Arctic Seas, to the Antarctic Continent, in
depths ranging from 15 to 3950 fathoms, and is one of the commonest arenaceous
Foraminifera in the deeper waters of all the great ocean basins.

Trochammina nitida, H. B. Brady.

This rotaliform species is somewhat rare, but is easily identified from its congeners
by the complanate superior surface of the test and its even margin. Found at Stations
118 and 346, but not common. Known from as far north as Franz Josef Land to as
far south as off Prince Edward Island, in depths of from 50 to 1070 fathoms.

Trochammina (Haplophragmium} turbinata (H. B. Brady).

From Stations 300 and 447 (rare). Known from one station in the South Atlantic,
off Amboyna, and from eight stations in the Pacific (BRADY and CUSHMAN). Range
of depth, 66 to 2050 fathoms.

Genus GLOBOTEXTULARIA, Eimer and Fiokert.

Globotextulama (Haplophragmium} anceps (H. B. Brady).

This characteristic wild-growing species was found at Station 342 only (rare). It is
known from Davis Straits, off Culebra Island, off the west coast of South America, and
from two stations in the North Pacific. Range of depth, 196 to 2200 fathoms.

Genus AMMOCHILO.STOMA, Eimer and Fickert.

Ammoclnlostoma (Trochammina) galeata (II. B. Brady).

This rare, inconspicuous, but handsome form was found at one station only, No. 459
(rare). It is exclusively a deep-sea species, being known from a few localities only in
the North Atlantic, South Atlantic, and Pacific, at depths of from 1825 to 2750 fathoms.

Ammochilostoma (Trochammina} pauciloculata (H. B. Brady).

A few specimens of this brilliant little species were obtained from Stations 447 and
459 (rare). Recorded from all the great ocean basins, mainly from deep water in mid-
ocean, although its range is from 173 to 2395 fathoms.

(ROY. SOC. KDIN. TRANS., VOL. XLIX., 1011.)

138 MR F. GORDON PEARCEY ON

Family TEXTULARIID^E.

Sub-family SPIROPLEOTIN^E.
Genus SPIROPLECTA.

Spiroplecta americana, Ehrenberg.

Obtained (rare) at one station by the Scotia, 459, 1995 fathoms; the only other
record for this species, so far as I am aware, is by the Challenger Expedition, off Raine
Island, North Australia, 155 fathoms.

Spiroplecta biformis, Parker and Jones.

This minute species in its typical form was obtained (rare) at Station 337a, 2110
fathoms. S. biformis is readily distinguished from its allies by the uniform manner in
which its test is built, consisting of very fine arenaceous material and brown cement.
It is to be found more frequently in the Arctic Seas, as far north as 80° lat., in depths
varying from 27 to 145 fathoms; in the South Atlantic east of Buenos Aires, 1900
fathoms ; and in the South Pacific, 2375 fathoms.

Sub-family TEXTULARIIN.E.

Genus TEXTULARIA, Defrance.
Textularia conica, d'Orbigny.

Obtained at Station 342 (rare). T. conica is a very generally diffused species,
and is found most abundantly in the coral seas of the tropical and subtropical regions.

Textularia concava (Karrer).

This species was taken (rare) at Station 342, 1946 fathoms. It is recorded from
several stations in the North Atlantic, from 173 to 2750 fathoms; from the South
Atlantic, in mid-ocean, 2475 fathoms ; and at a number of stations in the Pacific Ocean,
from shallow water to 2600 fathoms.

Textularia agglutinans, d'Orbigny.

Found in the material from Station 459, 1998 fathoms. T. agglutinans is perhaps
the commonest species of the genus Textularia, and the most typical arenaceous variety ;
it has a cosmopolitan distribution, both geographical and bathymetrical, and has been
recorded in depths ranging from 5 to 3125 fathoms.

Textularia aspera, H. B. Brady.

Obtained (rare) from one station only, 467. 2645 fathoms. This species has a geo-
graphical range from the Faroe Channel to the North Pacific, and has been collected in
depths from 210 to 2900 fathoms.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 1012.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 139

Genus BOLIVINA, d'Orbigny.

Bolivina reticulata, Hantken.

A few specimens of this species were found from Station 451, 1741 fathoms. It
has not been recorded in the North Atlantic ; the Challenger Expedition obtained this
species as far south as off Kerguelen, in 1570 fathoms.

Bolivina nobilis, Hantken.

Obtained (rare) at Station 451. This species has only been recorded hitherto from
the South Pacific Ocean, from shallow water to 1355 fathoms.

Bolivina punctata, d'Orbigny.

Taken at three stations, 346 (few), 459 (few), and 467 (rare). This species has a
cosmopolitan distribution, and a correspondingly wide range of depth, from 2 to 2750
fathoms.

Bolivina textilarioides, Reuss.

Found (rare) in the material from Station 346, 56 fathoms.

Genus PLEUROSTOMELLA, Reuss.

Pleurostomella alternans, d'Orbigny.

This interesting species was obtained (rare) at Station 459, 1998 fathoms. The
only other records for this species are off the Ki Islands, south-west of Papua, 129
fathoms ; south of the Low Archipelago, 2075 fathoms ; and off the Galapagos Islands,
in 1379 fathoms.

Sub-family VERNEUILININJv
Genus VERNEUILINA, d'Orbigny.

Verneuilina bradyi, Cushman.

Taken sparingly at Station 342, 1946 fathoms. V. bradyi is a common deep-
water form, and has a wide distribution, from about lat. 60° N., south as far as the
Antarctic Continent, with a range of depth from 420 fathoms in the North Atlantic
to 3125 fathoms in the North Pacific.

Genus GAUDRYINA, d'Orbigny.

Gaudryina pseudofiliformis, Cushman.

This new species was created by (V<n\i AN for the reception of those recent forms
figured and described by BRADY and by him referred to the cretaceous species

(HOY. SOC. EDIN. TRANS., VOL. XLIX., 1013.)

140 MR F. GORDON PEARCEY ON

G. jiliformis, Berthelin. From the examination of Challenger specimens and others,
I consider CUSHMAN is quite correct, and have therefore followed his example.

I have obtained good representatives of G. pseudqfiliformis from two Scotia stations,
313, 1775 fathoms, and 338, 2180 fathoms, and these are the most southern records for
this species.

Gaudryina bradyi, Cushman.

This species is a common deep-sea form, which Dr H. B. BRADY, in his Challenger
Report, erroneously records under the name of G. pupoides. Although I believe
CUSHMAN to be correct in his reference to this species, it would have been more to the
point had he given it a name which conveyed some definite specific character.

G. bradyi was taken at three stations, 342, 459, and 467. It has been hitherto
collected as far north as lat. 60° N., and is generally met with in most samples of deep-
sea deposits. The Scotia's records are the most southerly known.

Genus CLAVDLINA, d'Orbigny.
Clavulina communis, d'Orbigny.

This species is a comparatively common form in tropical and subtropical areas ; it
has been collected in from moderate depths down to 2300 fathoms, and is recorded from
the North Atlantic to as far south as the Antarctic Continent.

I have found it in Scotia material from six stations, 313 (rare), 337a (few), 342
(rare), 417, 418, and 447 (rare).

Sub-family BULIMININ^E.

Genus BULIMINA, d'Orbigny.

Bulimina aculeata, d'Orbigny.

Obtained at two stations, 118 (rare) and 451 (few). This species has been
collected as far north as the North Cape on the coast of Norway, and as far south as
the Antarctic Continent.

Bulimina buchiana, d'Orbigny.

Taken at one station only, 451 (few) ; it is a common species in the North Atlantic
from lat. 60° N. to the equator, and has been collected from depths of 150 to 1675
fathoms. The Scotia record is the most southerly point at which this species has been
obtained.

Bulimina ovata, d'Orbigny.

This species was obtained (rare) at Station 459, 1998 fathoms, in a deposit of
Globigerina ooze.

(HOY. SOC. EDIN. TRANS., VOL. XLIX., 1014.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 141

Bulimina elegantissima, d'Orbigny.

This species was obtained in abundance at Station 1 18, 24 fathoms. B. elegantissima
has been taken as far north as the west coast of Novaya Zemlya. It is also found on
the shores of the British Islands, Belgium, and France. It was taken by the Challenger
Expedition off the Falkland Islands in 6 fathoms; east coast of Australia, 2 to 410
fathoms ; South Pacific, G to G 1 0 fathoms ; off the coast of South America, from Peru to
Cape Horn.

(ieiuis VIKGULINA, d'Orbigny.

Virgulina pauciloculata , H. B. Brady.

This minute species was obtained (sparingly) at Station 459, 1998 fathoms ; the only
other records are off the shores of New Guinea and the adjacent islands, in depths from
11 to 129 fathoms.

Virgulina schreibersiana, Czjzek.

This somewhat doubtful species has been taken at four Scotia stations, 417, 418,
467, and 468. Nowhere abundant.

It has a very extensive distribution, and although it has been recorded from
shallow water, its natural habitat would appear to be in the deeper waters of the great
ocean basins ; its range of depth is from 11 to 3125 fathoms. The Scotia's records are
the farthest south from which this species has been obtained.

Sub-family CASSIDULININ^E.
Genus CASSIDULINA, d'Orbigny.

Cassidulina crassa, d'Orbigny.

Obtained at Station 346, 56 fathoms (common) ; its distribution is world-wide,
from lat. 83° ll'N. to as far south as the Antarctic Continent, ranging in depth
from comparatively shallow water to the deepest water in mid-ocean.

Cassidulina subglobosa, H. B. Brady.

This fine species was obtained at no less than nine stations, 118, 286, 313,
342, 417, 447, 459, 467, and 468, although it is not common at any of them. C.
subglobosa is, generally speaking, a deep-sea type, and has been recorded from all the
great ocean basins.

Cassidulina Isevigata, d'Orbigny.

This species was obtained (few) at two stations, 118, 2j fathoms, and 451, 1741
fathoms. C. l&viyatrr- is one of the commonest Arctic Foraminifera, and has a world-
wide distribution ; it was found as far south as the Antarctic Continent by the
Challenger Expedition. Its range of depth is from very shallow water, round the
British coast, to 1741 fathoms, in the Southern Ocean.

(KOY. soc. KDIN. TRANS., VOL. XLIX., 1015.)

142 MR F. GORDON PEARCEY ON

Cassidulina calabra (Seguenza).

Obtained sparingly at Station 346, 56 fathoms. 'The only other records for this
species are off Raine Island, Torres Straits, 155 fathoms, and off Kandavu, Fiji,
610 fathoms.

Cassidulina parkeriana, H. B. Brady.

Two specimens of this species were obtained at Station 346, 56 fathom?.

Genus EHRENBERGINA, Reuss.

Ehrenbergina serrata, Reuss.

This interesting species was obtained (rare) at Station 342, 1946 fathoms. It has
hitherto been recorded from the North and South Atlantic, and is comparatively
common in the North and South Pacific, generally being found in the area of the
Globigerina ooze. The Scotia station is the most southerly record.

Ehrenbergina pupa (d'Orbigny).

Taken at Station 342, 1946 fathoms. It has hitherto only been collected off the
Azores, in 450 fathoms ; off the Falkland Islands, in 1035 fathoms ; off the mouth
of Rio de la Plata, 13 fathoms; and off the coast of Patagonia, 120 and 175
fathoms.

Family LAGENID^.

Sub-family LAGEN1N/F,

Genus LAGENA, Walker and Boys.
Lagena globosa (Montagu).

Obtained at Stations 346 (few), 417 (rare), 459 (few), and 467 (rare). This species
is common in all seas and at all depths, wherever marine deposits have been examined
by the rhizopodist.

Lagena apiculata (Reuss).

Obtained sparingly at Stations 342, 346, and 459.

Lagena longispina, H. B. Brady.
This interesting deep-sea variety was obtained sparingly at Station 459, 1998 fathoms.

Lagena Icevis (Montagu).

This simple form has been obtained (few) at Stations 118 and 459. It is a very
common species, and has been found living in every sea from within 10 degrees of the
North Pole to the Antarctic Continent, and at every depth from the tidal zone down
to 2435 fathoms.

(KOY. soc. EDIN. TRANS., VOL. XLIX., 1016.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 143

Lagena vulgaris, var. pertuso-marginata, Rymer Jones.
Obtained sparingly at Station 346.

Lagena gracillima (Seguenza).
Taken (rare) at Station 459, 1998 fathoms ; a cosmopolitan species.

Lagena truncata, H. B. Brady.
This species was found (rare) at Stations 459 and 467.

Lagena hispida, Reuss.
One or two specimens of this species in its typical form were taken at Station 346.

Lagena striata (d'Orbigny).
Obtained (few) at Station 346, 56 fathoms.

Lagena lineata (Williamson).

A few specimens of this common and widely distributed species were found from
Station 346.

Lagena distoma, Parker and Jones.
A few specimens of this species were found at Station 467.

Lagena variata, H. B. Brady.
This rare variety also was obtained at Station 346.

Lagena sulcata (Walker and Jacob).

Obtained at Stations 346 (few), 459 (few), and 467 (rare). It is one of the most
abundant and most generally distributed of all the members of the genus ; it has
hitherto been recorded in every latitude, from Baffin's Hay in the north to Heard Island
in the Southern Ocean. Its bathymetrical range extends from the littoral zone to a
depth of 2750 fathoms.

Lagena acuticosta, Reuss.

Found (rare) at Station 342, 1946 fathoms.

is, Williamson.

This species was found in the deposits taken from Stations 346 (common), 459 (few),
and 467 (rare). It is a very common form, and found all over the world, from the
shallowest water down to 2775 fathoms.

(ROT. SOC. EDIN. TRANS., VOL XLIX., 1017.)

144 MR F. GORDON PEARCEY ON

Lagena semistnata, Williamson.
Obtained at Stations 346 (few), 417 (rare), and 467 (rare).

Lagena multicosta, Karrer.
Obtained sparingly at Stations 342 and 459.

Lagena stelligera, H. B. Brady.
Obtained (rare) at Station 459, 1998 fathoms.

Lagena exsculpta, H. B. Brady.

This species was found at Stations 342 and 447 (rare). Tt is a somewhat rare
form, apparently confined to the deep areas of mid-ocean. .

Lagena desmophora, Rymer Jones.
Taken at Station 459, 1998 fathoms, very sparingly.

Lagena torquata, H. B. Brady.

Taken (rare) at Station 342, 1946 fathoms. This species was taken by the
Challenger Expedition in the South Pacific, north of Juan Fernandez Island, in
1375 fathoms.

Lagena feildeniana, H. B. Brady.

This beautiful striato-punctate variety was found (rare) at Stations 447, 459, and
467. It is by no means a common species ; it has been collected from as far north as
79° N., in 80 fathoms ; and in the Southern Ocean to the west of Marion (Prince Edward)
Island, in 1570 fathoms; and in comparatively shallow water in the Pacific Ocean.

Lagena squamosa (Montagu).

Lagena squamosa, one of the commonest species, was obtained at Stations 118, 346,
417, and 459, but not in any numbers.

Lagena seminuda, H. B. Brady.
Another deep-sea variety ; obtained (rare) at Station 467, 2645 fathoms.

Lagena Isevigata (Reuss).
Was moderately common at Station 346.

Lagena millettii, Chaster.

This species was also found (rare) at Station 467.

(BOY. soc. EDIN. TRANB., vor, xi.ix , 1018.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 145

Lagena lucida, Williamson.

A few specimens of this compressed variety of L. Lvvigata were obtained from
Station 346, 56 fathoms.

Lagena acuta (Reuss).

Taken at Stations 342 (rare) and 346 (few). Lagena acuta, like the preceding
species, is found all over the world, ranging in depth from 2 to 3125 fathoms.

Lagena staphyttearia (Schwager).

Obtained (rare) at Station 447 only. This species appears to be confined to the
deeper waters of the Nortli and South Atlantic, whence it is recorded by the Challenger
as being obtained in 2200 to 2750 fathoms; but in the Southern Ocean and South
Pacific it occurs in shallow water near the coast-line.

Lagena marginata (Walker and Boys).

This cosmopolitan species was found sparingly at Stations 342, 346, 459, and 468.
Its bathymetrical distribution is also very extensive ; it has been taken from the littoral
zone down to 3125 fathoms.

Lagena semimargmata, Reuss.
Obtained at Stations 346. 459, and 468, sparingly.

Lagena lagenoides, var. tenuistriata, H. B. Brady.
Also found (rare) at Station 346.

Lagena formosa, Schwager.

Obtained (very rare) at Station 417, 1410 fathoms. L. formosa is a typical deep-
sea species ; it was also collected uear the Antarctic Continent by the Challenger
Expedition, in 1300 fathoms; and in the North Atlantic Ocean in 2750 fathoms.

Lagena trigonomarginata, Parker and Jones.

Several specimens of this somewhat rare species were also obtained from Station 346 ;
it has been collected in the North Pacific at a depth of 2300 fathoms.

Lagena quinquelatera, H. B. Brady.

Found (rare) at Station 342, 1946 fathoms. This species is recorded from the
Challenger Expedition in the South Pacific, 2350 fathoms ; and from off Marion (Prince
Edward) Island, Southern Ocean, in 50 to 150 fathoms.

Lagena orbignyana (Seguenza).

Obtained (rare) at Stations 1 18, 346, 417, 447, 459, 467, and 468.

(HOT. soc. EDIN. TRANS., vol.. XLIX., 1019.)

146 MR F. GORDON PEARCEY ON

Lagena bicarinata (Terquem).
Found (rare) at Station 346.

Lagena castrensis, Schwager.
Rare at Station 346, 56 fathoms.

*

Lagena quadricostulata, Reuss.

Taken at Stations 346 and 459. L. quadricostulata, by no means a common
form, was obtained by the Challenger Expedition in Balfour Bay, Kerguelen, 20
to 50 fathoms ; and off Sydney, Australia, in 410 fathoms.

Lagena fimbriata, H. B. Brady.

Taken (very rare) at Station 467, 2645 fathoms. L. fimbriata is a rare variety ; its
natural habitat is in the deeper waters of mid-ocean.

Lagena auriculata, H. B. Brady.
Taken at Stations 342 and 346 (rare).

Lagena alveolata, H. B. Brady.

Obtained (rare) at Station 459 only. This species has been collected only from the
deep waters of the great ocean basins ; it has been recorded from the North Atlantic,
2750 fathoms; South Atlantic, 2200 fathoms; Southern Ocean, 2600 fathoms ; and from
South and North Pacific, 2300 fathoms.

Lagena hispidipliolus, sp. nov. (Plate II. figs. 11-13.)

Test of variable contour ; oblong, compressed, ovate, elongate ; sometimes subangular.
Walls semi-transparent ; superior surface covered with short, pointed spines (hispid),
thickest at the margin of the inferior surface, where they form a distinct thickened rim
all round. Inferior surface comparatively thin, sole-like ; glassy, with raised ridges ;
diffused with minute foramina, and minute spines which may be altogether absent.
Aperture a short, tubular neck, projecting abruptly from the apical end of the test ;
ornamented with a number (six or more) of minute parallel rings, which occupy the
entire neck from its junction with the body to the free end. Diameter -£% inch
(1 mm.).

This variety of the genus Lagena would appear from the flattened sole-like inferior
surface to have been an adherent form, although I have not seen a specimen actually
in situ. Lagena hispidipholus was obtained (rare) from Station 346, 56 fathoms, from
the broken-down colonies of Cephalodiscus agglutinans, and has therefore most probably
become an adherent form by being enclosed between the chitinous tubes built up
by the Cephalodiscus during the enlargement of their colonies.

(ROY. soc. EDIN. TRANS., VOL. XLIX , 1020.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 147

Sub-family NODOSARINA
Genus NODOSARIA, Lamarck.

Nodosaria roemeri, Neugeboren.

This variety was obtained (rare) at Stations 342 and 346. N. roemeri is by no
means a common form ; I have collected this species in the warm area of the Faroe
Channel.

Nodosaria perversa, Schwager.

Obtained (rare) at Station 342. This species has only been recorded hitherto from
off the Ki Islands, south-west of Papua, in 120 fathoms.

Nodosaria calomorpha, Reuss.

Obtained at Stations 118 (rare) and 459 (rare). This minute species has been
collected from the North Atlantic, in 1215 fathoms; the South Atlantic, 1990 and
2200 fathoms; off the Falkland Islands, in 6 fathoms; off Kerguelen, 120 fathoms;
off Ki Islands, 129 to 580 fathoms ; and in 95 fathoms off the Philippine Islands.

Nodosaria mucronata, Neugeboren.

A moderately common form in the North Atlantic, and off the British coast, from
shallow water to 1750 fathoms; in the South Atlantic, to 2350 fathoms; and in the
Pacific Ocean, from 345 to 2600 fathoms. I have found this species (rare) at Stations
417 and 467 ; these are the most southerly records known for N. mucronata.

Nodosaria consobrina, var. emaciata, Reuss.

Obtained (rare) at Station 467. I have taken this species in the warm area of the
Faroe Channel.

Nodosaria scalaris, var. separans, H. B. Brady.

Several specimens of this variety were obtained at Station 346, 56 fathoms. This
is a common form off the coast of New Zealand, in 275 fathoms.

Nodosaria proxima, Silvestri.

Obtained (rare) at Station 346. N. proxima appears to prefer the warmer waters
of the coral seas. BRADY records it from Torres Straits, in 155 fathoms; off the
Philippines, 95 fathoms ; off the reefs at Honolulu, 40 fathoms ; and off the Azores, in
450 fathoms.

Nodosaria setosa, Schwager.

Found (rare) at Station 346.

(BOY. soc. EDIN. TRANS., VOL. XLIX., 1021.)

148 MR F. GORDON PEARCEY ON

Nodosariafiliformis, d'Orbigny.
This species was also obtained sparingly at Station 346.

Genus CRISTELLARIA, Lamarck.
Cristellaria convergens, Bornemann.

This species was found at Stations 337a (rare), 342 (few), 418 (rare), and 447 (rare).
The natural habitat of this species is in the deeper waters of the great ocean basins. It
has been collected in the North Atlantic, in 2740 fathoms ; in the South Atlantic, 2350
fathoms; and in the Pacific Ocean, from 16 to 1850 fathoms.

Cristellaria cultrata, var. (Montfort).

A dentate variety was obtained (rare) at one station only, 451, 1741 fathoms. This
species has been recorded from as far north as the Arctic Circle, and as far south
as the shores of Patagonia, but in most cases from comparatively shallow water.

Cristellaria articulata, Reuss.

Obtained frequently at Station 346, 56 fathoms. Previously taken off Culebra
Island, West Indies, 390 fathoms, and off Tristan da Cunha, 100 to 150 (abundant).

Cristellaria crepidula (Fichtel and Moll).

This species has a wide distribution, and is chiefly confined to comparatively
shallow water. It was obtained in fair numbers at Station 346, 56 fathoms. C. crepi-
dula has been collected from as far north as Davis Straits (lat. 68° 50'N.), and is a
common form in the North Sea and Atlantic ; its range of depth is from 6 fathoms
in the Pacific to 2350 fathoms in the tropical part of the South Atlantic.

Cristellaria schloenbachi, Reuss.

Obtained (rare) at Station 346. Taken by the Challenger Expedition off Bermuda,
435 fathoms, and Culebra Island, West Indies, 390 fathoms; also off Raine Island, North
Australia, 1 1 5 fathoms.

Cristellaria acutauricularis (Fichtel and Moll).

This thick ovoid variety was also taken (rare) at Station 346. It is not a common
form. It has been found in the North Atlantic, in from 390 to 2750 fathoms ; off the
Cape of Good Hope, 150 fathoms ; and from six stations in the South Pacific, in depths
ranging from 155 to 2350 fathoms ; but at all times rare.

Genus AMPHICORYNE, Schlumberger.
Amphicoryne falx (Jones and Parker).

This interesting dimorphous striated variety was obtained (rare) at Station 346.
56 fathoms; this species is also found in the warm area of the Faroe Channel, in 516
fathoms.

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 1022.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 149

Sub-family FULYMORPHININ^.

Genus POLYMORPHINA, d'Orbigny.

Polymorphina inflata, sp. nov. (Plate II. figs. 14-16.)

Test elongate, forming a somewhat obscure triserial spire, having five visible
chambers tapering to a point at the inferior extremity ; oblong and more or less inflated ;
the last chamber much the largest and more inflated than the rest ; compressed trans-
versely ; each chamber separated by excavated sutures. Aperture a round opening
encircled by a number of radiating grooves near the centre of the final chamber ;
furnished with a short entosolenian tube extending into the cavity of the last chamber.
Length, T^ inch (17 mm.).

The erect position of the chambers, inflated contour of the test, and separated
excavated suture lines are sufficient to distinguish this species from its near allies ;
the chamber walls are thin, smooth, and beautifully transparent. Polymorphina
inflata was found sparingly at Station 346, 56 fathoms.

Polymorphina angusta, Egger.

This species (rare) was taken at four stations, 342, 346, 447, and 459. It is
a widely distributed species, and has been found in the North and South Atlantic
and the North and South Pacific, but is for the most part a deep-sea form, with a range
of depth in these oceans of 1000 to 2400 fathoms, although it has been collected on
several occasions in shallow water.

Polymorphina gibba, d'Orbigny.
Obtained (rare) at one station only, 342, 1946 fathoms.

Polymorphina seguenzana, H. B. Brady.
Taken at Station 459, 1998 fathoms (rare).

Polymorphina oblonga, d'Orbiguy.

Obtained at Station 346, 56 fathoms. This species has been taken in the North
Pacific as low down as 2050 fathoms.

Genus UVIGERINA, d'Orbigny.
Uvigerina brunnensis, Karrer.

I have obtained this form (rare) in the material from Station 342, 1946 fathoms.
U. brunnensis is a rare species in the living state ; its only known records are
from Christmas Harbour, Kerguelen, 120 fathoms; and off the western shores of
Patagonia, 245 fathoms.

(ROY. SOU. EDIN. TRANS., VOL. XLIX., lO'2'.i.)

150 MR F. GORDON PEARCEY ON

Uvigerina aculeata, d'Orbigiiy.
Obtained at Stations 342 and 459 (not abundant).

Uvigerina pygmsea, d'Orbiguy.

This is the most common species of the genus ; it has been collected in all seas,
and at all depths from 2 to 2600 fathoms. I have obtained it in fair numbers from
Stations 451 (1741 fathoms) and 346 (56 fathoms).

Uvigerina asperula, Czjzek.

A few specimens of this species were taken at Station 459, 1998 fathoms.
U. asperula is a comparatively common species, and has been found from as far north
as lat. 56° and as far south as lat. 52°. It was obtained by the Challenger Expedition
in the Antarctic at 2600 fathoms.

Uvigerina angulosa, Williamson.

This characteristic form was obtained in fair abundance at Station 346, 56 fathoms.
It is a widely distributed form, and has been found as far north as lat. 70°, and as
far south as the Antarctic Continent, in depths ranging from 8 to 1375 fathoms.

Family GLOBIGEKINIDyE, Ehrenberg.

Genus GLOBIGERINA, d'Orbiguy.

Globigerina dutertrei, d'Orbigny.

This robust dwarfed variety apparently takes the place of Globigerina bulloides in
the Antarctic Seas. I have found this species in smaller or greater numbers in most
of the deposits obtained by the Scotia Expedition, at no less than fifteen of the twenty-
four stations from which I have samples of the deposits, viz. Stations 286, 300, 313,
338, 342, 346, 387, 417, 418, 420, 421, 447, 451, 459, and 467.

Globigerina packyderma (Ehrenberg).

Dr H. B. BRADY, in describing this species, states that it has not been taken
south of the Faroe Channel. I have examined a great number of specimens from the
Faroe Channel, Arctic and Antarctic Seas, and Southern Ocean, and can find no specific
characters differentiating those taken by the Scotia from those found in the Faroe
Channel and Arctic Sea. I have obtained Globigerina pachyderma more or less
abundantly from nine stations, viz. 300, 313, 338, 342, 346, 387, 417, 420, and 451.

Globigerina bulloides, d'Orbigny.

Obtained at eight stations, viz. 338, 342, 346, 417, 421, 451, 459, and 467.
Globigerina bulloides is the most common species of the genus. It has been taken in

(HOY. SOO. EDIN. TKANS., VOL. XLIX., 1024.)

FORAMTNIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 151

the living state by the plankton net wherever Foraminifera have been collected, all over
the world, and their tests form marine deposits from almost every latitude, down to
depths of nearly 3000 fathoms.

Globigerina ittftata, d'Orbigny.
Obtained (fairly common) at Stations 342, 346, 451, 450, and 4f>7.

Globigerina triloba, Reuss.
Obtained (few) at Stations 342, 459, and 467.

Globigerina digitata, H. B. Brady.

This delicate, beautifully formed variety was found (rare) at one station only, 459,
1998 fathoms. G. digitata is a rare species; its previous records are, three stations
in the South Atlantic, and six in the South Pacific. The Scotia station is the most
southern point at which this species has been collected ; it is not known, so far as I am
aware, from the North Atlantic.

Globigerina conglobata, H. B. Brady.
A few specimens of this species were found at Station 459.

Globigerina sacculifera, H. B. Brady.

A number of specimens of this large tropical form were obtained at Station 459.
[It was also taken in plenty at Station 81, 18° 24' S., 37° 58' W., 36 fathoms, coral and
sand.]

Globigerina dubia, Egger.

This thick-shelled variety was obtained (rare) at Station 421, 2487 fathoms; this is
the most southerly record for the species.

Globigerina (Orbulina) universal, d'Orbigny.

A few small specimens of this common form were obtained from Station 459, 1998
fathoms, and in moderate numbers from Stations 118, 2^, and 346, 56 fathoms.

Genus POLLENIA, Parker and Jones.

Piil/ciiiit </iuii<jii<'/<>hf(, Reuss.

This widely distributed species has been obtained from as far north as lat. 62° 6' N.,
and southward to the Antarctic Continent. Its range ol' depth extends from 30 to
2750 fathoms. It is found in abundance in the Faroe Channel, and has been taken
sparingly off the shores of the British Islands. I obtained I'nUcnin tjnn/(ji«'/uba in
moderate numbers from the deposit taken at Stations 459 and 467.

(HOY. SOC. KDIN. TRANS., VOL. XLTX., 1025.)

152 MR F. GORDON PEARCEY ON

Pullenia sphseroides (d'Orbigny).

This cosmopolitan species was obtained (sparingly) from Stations 342, 420, 421,
447, 459, 467.

Pullenia obliquiloculata, Parker and Jones.

This species was found (rare) at Stations 447, 451, 459, and 467. P. obliquiloculata
is found in all the deep-sea deposits, but attains its fullest development in the tropical
and subtropical waters of mid-ocean. It has a comparatively wide distribution, but
the Scotia records are the most southerly from which it has been collected.

Family ROTALIDM.

Sub-family SPIRILLININ^E.

Genus SPIRILLINA, Ehrenberg.

Spirillina obconiea, H. B. Brady.

This pretty little species was found at Station 346, 56 fathoms, in moderate numbers.
S. obconiea has hitherto been collected at three other localities only : off Prince
Edward Island, 50 to 150 fathoms; in Christmas Harbour, Kerguelen. 120 fathoms; and
in Nares Harbour, Admiralty Islands, 17 fathoms.

Spirillina tuberculata, H. B. Brady.

A few specimens of this species were obtained at Station 346. Like the former
species, S. tuberculata appears to flourish best in the Southern Ocean. I have collected
it in the Faroe Channel ; it is also recorded from one or two points off the British coast.

Sub-family KOTALIN^E.

Genus PATELLINA, Williamson.
Patellina corrugata, Williamson.

Obtained (rare) at Station 34G, 56 fathoms. This species has been found as far
as lat. 83° 19' N., in 72 fathoms, and as far south as off Heard Island, 150 fathoms. I
have taken it in the cold area of the Faroe Channel, in 365 fathoms.

Genus DISCORBINA, Parker and Jones.

Discorbina orbicularis (Terquem).
Found moderately common at Station 346, 56 fathoms.

(BOY. SOC. EDIN. TRANS., VOL. XLIX., 102G.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 153

Discorbina rarescens, H. 15. Brady.
A few specimens also obtained at Station 346.

Discorbina concinna, H. B. Brady.
This species was also taken (rare) at Station 346.

Discorbina parisiensis (d'Orbigny).

Obtained sparingly at Station 346. It was collected in abundance by the
Challenger off Kerguelen, 20 to 50 fathoms. D. parisiensis is also recorded from
off the Irish coast, and the Atlantic shores of France.

Genus TRUNCATULINA, d'Orbigny.

Truncatulina tenera, H. B. Brady.

This rare little form was obtained from the deposits taken at Stations 342, 417,
451, 459, and 467. T. tenera would appear to prefer the deep water of the great ocean
basins. Its other known localities are one station in the North Atlantic, off the
Canaries, 620 fathoms ; and three in the South Pacific, near the coast of Chili and
Patagonia, 166 to L375 fathoms.

Truncatulina pygm&a, Hantken.

Obtained (few) at Stations 300, 342, 420, 447, 459, and 467. T. pygmaea
is a typical deep-sea species ; it has been taken in the deposits from all the great ocean
basins, whenever they have been examined for Foraminifera, down to 3125 fathoms.

The Scotia records are the most southerly points from which this species has been
collected.

Truncatulina wuellerstorfi (Sch wager).

This species occurred sparingly at Stations 300, 417, 420, and 421. T. wuellers-
torf, is a common form in the deep-sea oozes all over the great oceans.

Truncatulina ungeriana (d'Orbigny).

This species was found sparingly at Stations 313, 1775 fathoms, and 467,
2645 fathoms.

Truncatulina lobatula (Walker and Jacob).

This most common and widely distributed species I have obtained sparingly at
Stations 342, 346, 447, 451, 459, and 467.

Truncatulina dutemplei (d'Orbigny).
Taken sparingly at Stations 342, 1946 fathoms, and 459, 1995 fathoms.

(ROT. SOC. EDIN. TRANS., VOL. XLIX., 1027.)

154 MR F. GORDON PEARCEY ON

Truncatulina robertsoniana, H. B. Brady.
This handsome species (rare) was obtained at Station 451, 1741 fathoms.

Truncatulina akoncriana (d'Orbigny).

Obtained (rare) at Station 459, 1998 fathoms. This is a world- wide species, and is
often taken in very deep water.

Truncatulina tumidula, H. B. Brady.

Taken (rare) at Station 468, 2700 fathoms. This is a rare species, and has only
been collected at one other locality — in the red clay south of the Canaries, from
2740 fathoms, where it was taken in fair numbers.

Truncatulina haidingerii (d'Orbigny).
A few specimens of this species were found at Station 468.

Truncatulina tenuimargo, H. B. Brady.

Obtained sparingly at Station 417, 1410 fathoms, the most southerly record for
this species.

Truncatulina variabilis, d'Orbigny.

Taken (common) at Station 346, 56 fathoms. This wild-growing form is a common
species in the warmer waters of the sub-tropical and temperate seas, in shallow water ;
it has. however, been obtained in depths above 2000 fathoms. I have taken this species
as far north as the Faroe Channel.

Truncatulina rostrata, H. B. Brady.

A few specimens of this species were also obtained from Station 346. Its natural
habitat is in the shallow water in the vicinity of coral reefs and islands ; it has not
been obtained north of the tropics.

Genus ANOMALINA, d'Orbigny.

Anomalina polymorpha, Costa.

Obtained (rare) at Stations 420, 2620 fathoms, and (few) 346, 56 fathoms. This
species was found by the Challenger off Marion (Prince Edward) Island, 50 to 150
fathoms ; at three stations in the North Atlantic, 390 to 450 fathoms ; and at three
points in the South Pacific, 210 to 410 fathoms. The Scotia records are therefore
the most southerly points known for this species.

Anomalina grosserttgosa (Gumbel).

Taken (rare) at Station 346, 56 fathoms.

(ROY. soc. EDIN. TRANS., VOL. xux., 1028.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 155

Genus PULVINULINA, Parker and Jones.
Pulvinulina crassa (d'Orbigny).

This species was obtained at Stations 342 (few), 451 (rare), and 459 (few).
Pulvinulina crassa has been taken as far north as lat. 54° 53' N., in depths ranging
from 725 to 2740 fathoms ; at seven or more points in the South Atlantic, 420 to 2350
fathoms; and at many localities in the Pacific, from 150 to 2335 fathoms. The Scotia
records are the most southerly from which the species has been obtained.

Pulvinulina canariensis (d'Orbigny).

This common and widely distributed species was found at Stations 342, 451, 459,
467, and 468 in fair numbers. P. canariensis is a pelagic form in the living state.
Its distribution, however, is better seen from the dead tests or shells to be found in the
deposits of the ocean floor.

Pulvinulina truncatulinoides (d'Orbigny).

Obtained at Stations 342, 459, 467, and 468 (not common). P. micheliniana
is a pelagic species ; it attains its fullest development in the warmer waters of the
tropics, becoming dwarfed and stunted the farther north or south of the tropics it is
found. The Scotia records are the most southerly known.

Pulvinulina exigua, H. B. Brady.

This minute species, whose natural habitat is in the deeper waters of mid-ocean,
is never taken in any great numbers. It was found in the deposits from Stations 342
(few), 417 (common), 447 (few), 451, 459, 467, and 468 (few). The Scotia records are
the farthest south that P. exigua has ever been collected.

Pulvinulina karstem (Reuss).

Obtained (rare) at Station 447 only. The best examples of P. karsteni have
been collected in the Arctic Seas, in shallow water ; the Scotia station is the
greatest depth that this species has been recorded from, and also the most southerly
point.

Pulvinulina etegans (d'Orbigny).

Obtained (few) at Station 342.

Pulvinulina umbonata, Reuss.
Taken very sparingly at Station 346, 56 fathoms.

Pulvinulina f aims, H. B. Brady.

This remarkable and comparatively rare species was taken at Station 459, 1998
fathoms, in its typical form, although somewhat dwarfed, with its test beautifully

(KOY. soc. EDIN. TRANS., VOL. XLIX., 102i>.)

156 MR F. GORDON PEARCEY ON

ornamented with " honeycomb "-like reticulations. P. favus is a typical deep- water
species, and is found most abundantly in the South Pacific, between the equator
and lat. 42° 43' S. The farthest north record of its occurrence is the Challenger's,
about eight degrees north of the equator. The Scotia station is therefore the most
southerly. Its bathymetrical range is from 1375 to 2600 fathoms.

Genus ROTALIA, Lamarck.

Rotalia soldanii, d'Orbigny.

Obtained at Stations 342 (common), 447 (rare), 467, and 468 (few). R. soldanii
is a typical deep-sea form, and has a world-wide distribution, extending from lat. 60° N.
to as far south as the Antarctic Continent.

Family NUMMULINID^E.

Sub-family POLYSTOMELLIN^.

Genus NONIONINA, d'Orbigny.
Nonionina pompilioides (Fichtel and Moll).

This typical deep-water form was found sparingly at Stations 286, 342, 447, and
467. N. pompilioides has been collected from most of the deep-water areas all over
the world.

Nonionina umbilicatula (Montagu).

This cosmopolitan form was taken at Stations 447, 451, 459, 467, and 468, although
not in any numbers ; its bathymetrical range extends from 30 to 3125 fathoms or
deeper.

Nonionina stelligera, d'Orbigny.

A few specimens of this somewhat rare species were obtained from Station 417,
1410 fathoms; this is the farthest south record, and the greatest depth at which the
species has been collected.

Nonionina boueana, d'Orbigny.

This shallow-water form was taken at Station 346, 56 fathoms, in moderate
abundance.

Genus POLYSTOMELLA, Lamarck.
Polystomella striatopunctata (Fichtel and Moll).
This very common species was taken in quantities at Station 118, 2^ fathoms.

Polystomella macella (Fichtel and Moll).
Obtained (common) at Station 118.

(ROY. SOC. UDIN. TRANS., VOL. XLIX., 1030.)

FORAMIN1FERA OK Till'; .SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 157

Polystomella imperatrix, II. B. Brady.

Taken (rare) at Station 118. /•*. imperatrix is by no means a common species,
but is the largest and most handsome of all the Polystomella. So far as 1 am aware,
it has not been found north of the equator.

Polystomella crispa (Linne).

This common cosmopolitan species was obtained in abundance at Station 346,
56 fathoms.

Polystomella verriculata, H. B. Brady.

This rare species was also taken very sparingly at Station 346. P. verriculata
has hitherto only been found at two points on the west coast of Australia, namely, off
East Moncoeur Island, 38 fathoms ; and in Curtis Strait, Queensland, by the Challi'n</>:r
Expedition.

COMPARATIVE VIEW OF THE MORE PROMINENT FORAMINIFERA
OBTAINED FROM THE VARIOUS SAMPLES OF DEPOSITS EXAMINED.

As the material in most cases was insufficient to give a full representation of the
Foraminiferal fauna, reference is made only to the main features of the groups
actually found.

In the case of Stations 313 and 420, where the material was abundant, full lists are
given, as these stations (with certain exceptions, referred to under the individual
stations) may be taken as representative of the northern and southern parts of the
glacial deposit area in the Weddell Sea.

STATION 118, Stanley Harbour, Falkland Islands.

Depth 2j fathoms ; bottom temperature 5'2° F. A dark brown mud with a
greenish tint, containing 27 per cent, of lime from calcareous organisms of various kinds.
The chief Foraminifera are Bulimina elegantissima, Rotalia karsteni, and Polystome/ln
striatopunctata. Proteonina difflugiformis, Trochammina nitida, and T. nana
represent the arenaceous forms.

STATION 286, lat. 68° 11' S., long. 34° 17' W.

Depth 2488 fathoms; bottom temperature 31°'3 F. A greenish-gray unctuous
glacial clay, gritty to the touch, containing 5 per cent, of lime.

An insufficient quantity of the deposit was available for a complete estimation of
the Foraminiferal fauna. The only species found in the sample examined are

(KOY. soc. EDIN. TRANS., VOL. XLIX., 1031.)

c

' L I a r:

158 MR F. GOKDON PEARCEY ON

Una sulHi/olioxu, Globigerina dutertrei, G. pachyderma, and Nonionina pompilioides.
This station is not far from Station 291, which yielded a more extensive collection ; and
it is probable that, if a larger quantity of material had been obtained, it would have
been found to be rich in the arenaceous types.

STATION 291, lat. 67° 33' S., long. 36° 35' W.

Depth 2500 fathoms. A greenish-gray gritty glacial mud with rock-fragments.
The sample examined consisted of washings from the mud taken in the trawl, the
finest mud and everything under ^ inch in size being washed away.

The Foraminifera belong almost entirely to the Arenacea. Astrorhiza crassatina
is present in abundance, together with A. arenana, Saccammina, Thurammina,
Reophax, Hyperammina, Hormosina, Haplophragmoides, and Ammologena. The
only calcareous form found was one or two specimens of Miliolina bucculenta.

STATION 295, lat. 66° 40' S., long. 40° 35' W.

Depth '2425 fathoms; bottom temperature 31°'3 F. A bluish-gray gritty glacial
clay. Only a very small quantity came to me for examination, so that nothing definite
can be said about the general facies of the Forammiferal fauna. The only two species
found were Astrorhiza crassatina, fragments, common, and Hormosina globulifera, a
few specimens.

STATION 300, lat. 65° 29' S., long. 44" 06' W.

Depth 2500 fathoms; bottom temperature 32°'l F. A fairly tough gray glacial
clay with gritty particles, containing 2 per cent of lime.

A large amount of material was not available for examination, but the following
Foraminifera were found : — Proteonina difflugiformis, Haplophragmoides subglobosum,
H. tndissatum, Cyclammina pus-ilia, Globigerina dutertrei, G. pachyderma, and Trun-
catidina pusilla.

STATION 301, lat, 64° 48' S., long. 44° 26' W.

Depth 2485 fathoms; bottom temperature 31°'2 F. A dark grey glacial mud
with rock-fragments. A trace of lime. Only a very small amount available for
examination.

The Foraminifera found were Saccammina sphserica (attached to a rock-fragment),
Bath i/siphon Jiliformis, Reophax nodulosa, and R. pilulifera.

STATION 313, lat. 62° 10' S., long. 41° 20' W.

Depth 1775 fathoms; bottom temperature 31°'0 F. A bluish-gray, only slightly
coherent glacial muddy sand, containing 2 per cent, of lime. The material obtained
here from the trawl was large in amount, and, although a little of the finest material

(ROY. soc. EDIN. TRANS., VOL. XLIX., 1032.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

may have been washed away, the following list of the Foraminifera obtained is probably
fairly representative:—

Miliolina Imcculenta, Br.
Astrorh/ga crassa/ii/a, l!r.
Pelosina cylindrica, Br.
Saccammina socialis, Br.

,, sphxrica, Sars.

Psammosphxra fusca, Sch.
RhaliiJammina abyssorum, Sars.
Saccorhiza ramosa (Br.).

„ elongata (Br.).
Reophctf noilulosa, Br.

„ pilulifeni, Br.

„ lientaUnij'ontiis, Br.

„ membranacea, Br.

Haplophragmoides scitulum (Br.).

,, rotulatnm (Br.).

(.'rihrostomoii/fs lirndiji, Cush.
Trochammina globigenniformis (P. ami J.).
Horinosina normani, l!r.
Arnmo/aijena clavata (P. and J.).
Cyclammina pnsilla, Br.
Gaudryina pseudoflliformis, Cush.
Clavulina communis, d'Orb.
Cassidulina subglobosa, Br.
Globigerina dutertrei, d'Orb.

,, pachyilerma (Ehrng.).
Truncatitlina ungeriana, d'Orb.

Ammobaculites tenuimargo (Br.).

The Hormosina normani from this station are exceptionally large, several specimens
measuring fully ^ inch across the final chamber. These were obtained from the wash-
ings of the mud taken by the trawl.

STATION 337 a, lat. 59° 46' S., long. 48° 02' W.

Depth 2110 fathoms. A brownish-gray sandy mud, on the border-line between a
glacial deposit and a Diatom ooze, with 1 per cent, of lime. Only a small amount
available for examination. The Foraminifera found, with the exception of one species,
all belong to the Arenacea: — Pscemmosphaera, Rhabdammina, Saccorhiza, Ammodiscus,
Huplo/ili roi/inoides, Proteonina, Trochammina, Ammobaculites, Clavulina, Spiroplecta,
and Cristellaria.

STATION 338, lat. 59° 23' S., long. 49" 08' W.

Depth 2180 fathoms. A Diatom ooze or volcanic sand (there being a large amount
of volcanic mineral particles mixed with the siliceous organisms). There is merely
a trace of lime. The Foraminifera found in the small amount available would indicate
a fairly rich Rhizopod fauna. The arenaceous forms are represented by Rhizanvrnina,
Proteonina, Cyclammina, Ammobaculites, and Gaudryina ; the pelagic forms by
dwarfed Globigerina buUoides, G. dutertrei, and G. pachyderma.

STATION 342, lat. 56° 54' S., long. 56° 24' W.

Depth 1946 fathoms. A fine-grained Globigerina ooze containing 24 per cent, of
lime. There are numerous volcanic mineral particles.

This deposit gave a rich Foramini feral fauna, no less than twenty -nine genera and
sixty species being obtained. The Miliolidaj are represented by three genera,
Biloculina, Miliolina, and Spiroloculina. Among the arenaceous types, which are
represented by twelve genera, the more rare species M<trsij>('//« cylindrica,
Proteonina difftugiformis, Thurammina n/lii<-itn.«, and Globote.rtularia anceps are

(ROT. ROC. KDIN. TKANS., Vol.. XI. IX., 11)33.)

160 MR F. GORDON PEARCEY ON

noteworthy. Textularidae, by Textularia, Gaudryina, Verneuilina, Ehrenbergina,
Clavulina, and Cassidulina, 9 species of Lagena, 1 Nodosaria, 1 Cristellaria,
2 Polymorphina, 2 Uvigerina, 5 Globigerina, with 1 Pullenia. Amongst the Rotalidse,
4 Truncatulina, 5 Pulvinulina, 1 Rotalia. Lastly, 1 Nonionina — i.e. practically all
the types peculiar to a Globigerina ooze, of the South Atlantic and Southern Ocean.

STATION 346, lat. 54° 25' S., long. 57° 32' W.

On the Burdwood Bank, depth 56 fathoms. The trawl brought up a large number
of masses of calcareous, shelly, and Foraminiferal sand with large colonies of Cephalo-
discus agglutinans ; and from these no fewer than eighty species belonging to twenty-
five different genera of Foraminifera were obtained. With a few exceptions, they are
all of typical Antarctic shallow- water character, but arenaceous types are conspicuous
by their absence, being represented by Trochammina nitida only.

The Miliolinidse are well represented by six genera and fourteen species, of which
Nubeculina bradyi, Miliolina ferussacii, and Articulina funahs are typical Antarctic
species ; while it is of interest to have found here also Cornuspira fohacea, although
of comparatively small size, which occurs widely but reaches its best development in
the North Atlantic and North Sea.

The Textularidse are well represented by several species of Bulimina, Cassidulina,
and Ehrenbergina. The Lagense alone yield twenty-four species, the most common
being L. gracilis. Of the rarer species may be mentioned L. trigonomarginata, L.
liispida, and L, castrensis. Nodosaria gave five species; Cristellaria four species;
Polymorphism two, Uvigerina two, and Globigerina five.

The most abundant species observed in the sample are G. bulloides (typical),
G. dutertrei, Truncatulina variabilis, and Polystomella crispa. Three new species
are described, viz. Lagena hispidipholus, Polymorpliina inflata, and Miliolina
dentistoma.

STATION 387, lat. 65° 59' S., long. 33° 06' W.

Depth 2625 fathoms. A brown tenacious glacial clay, with a trace of lime. In
the small sample the only Foraminifera found are Proteonina difflugiformis,
Globigerina dutertrei, and G. pachyderma.

STATION 416, lat. 71° 22' S., long. 18° 15' W.

Depth 2370 fathoms; bottom temperature 31°'5 F. A greenish-gray unctuous
glacial clay, with a trace of lime. Some two ounces of deposit yielded five genera of
arenaceous Foraminifera, viz. Psammosphsera, Gordiammina, Reophax, Haploplirag-
moides, and Cyclammina. No pelagic or porcellanous forms were found.

STATION 417, lat. 71° 22' S., long. 16° 34' W.

Depth 1410 fathoms; bottom temperature 31°'9 F. A greenish-gray glacial mud,
with rock-fragments and glauconite grains, containing 4 per cent, of lime.

(ROY. SOtJ. KDIN. TRANS., VOL. XLIX., 1034.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 161

Foraminifera fairly well represented, especially the larger arenaceous types, such
as Rhabdammina, Psammosphaera, Saccammina, Pelosina, Saccorhiza, Reopha.<\
Hormosina, T<>////><niiii,iiitt. Haplophragmoides, Trocliammina, Cydammina, several
Textularidte, several Lageniv. Globigerinse, tolerably abundant, especially the small,
rounded variety G. bnll<>i<les and G. dutertrei, together with Truncatulina ivuellerstorfi,
T. tenuimargo, T. tenera, Pulvinulina exigua, and Nonionina stelligera.

STATION 418, lat. 71° 32' S., long. 17° 15' W.

Depth 1221 fathoms ; bottom temperature 3l°'9 F. A glacial clay similar to that
from Station 417. Here also the quantity of material was insufficient to give a full
idea of the Foraminiferal fauna which may be expected to exist in this locality.
Those obtained &?Q : - - Bathysiphon filiformis, Psammosphaera fusca, Reophax
nodufasa, Haplophragmoides rotulatum, Cribrostomoides bradyi, Clavulina com-
munis, Vinjnlina schreibersiana, Globigerina dutertrei, and Cristellaria convergens.

STATION 420, lat. 69° 33' S., long. 15° 19' \V.

Depth 2620 fathoms; bottom temperature 31°'5 F. A brownish-gray gritty
glacial mud or clay. No lime in the sounding-tube sample. Some two pounds weight
of this deposit were prepared for examination, including trawl washings. The washing
away of the finest material from these may account for the absence of the more minute
Textularidse and Lagenid.u which one might expect to obtain from this locality ; but
the richness in arenaceous species makes it desirable to print in extenso the species
obtained from this locality and extremely far south station. Several species show a
high development in form and size, especially in Reophax, Psammosphaera, and
Hormosina. Most of the more prominent genera of the Astrorhizidae are represented,
and no fewer than seven new species and a new variety are described.

List of Foraminifera from Station 420.

Technitella asciformis, sp. nov.
Rhizammina /»-//v/.w, Br.
Rhabdammina discreta, Br.

Biloculina ringens (Laiuk.).
„ de/jressn, d'Orli.
Miliolina hueculenta, Br.

„ circularis (Born.).

,, bucculenta, v;ir. placentiformis, Br.
Keramosp/uera murrayi, Br.
Saccammina upli.-rrica, M. Sars.

„ socialif, I'.r.

Sorosphxra conf/isa, Br.
Psammosphsera fusca, Sc-li.
Astrorhiza arenaria, Norm.
Pelosina arborescens, sp. imv.

„ cylindrica, Dr.
Syringammina minnta, sp. nov.
Tlinlimiin I'/'Jla (Br.).
Teelmitella raphanu?, Hr.

,, cornuta, J!r.

Bathysiphon filiformis, M. Sars.
Hyperammina frialilis, Br.

„ xtibnoilosa, Br.

Saccorhiza ramosu (Br.).
Achomonella catenata, Norm.

,, i-niiiii/ii'iiriniit, Br.

Reopltax nodulosa, Br.
„ pilulifera, Br.
„ ailunea, Br.
„ roliitfium, sp. nov.
i/liiim/i/Pi-a, l!r.
irregularis, .sp. nov.

(ROY. soc. KDIN. TRANS., VOL. xux., 1035.)

162 MR F. GORDON PEARCEY ON

List of Foraminifera from Station 420 — continued.

Webbinella hemisphxrica (P. and J.).
Crithionina pisum, var. hispida, Flint.
Thnrammina papillata, Br.

,, favosa, var. reticulata, var. nov.

Tolypammina vagans (Br.).
Ammolaijena clavata (P. and J.).
Haplojihragmoides tndlissata (Br. ).

„ glomeratum (Br.).

,, rotulatum (Br.).

„ umbitieatum, sp. nov.

,, subglobosum (G. 0. S.).

Cribrostomoides bradyi, Cusbn.

Cyclammina cancellata, Br.

,, pusilla, Br.

,, contorta, sp. nov.

Ammobuculites tenuimargo (Br.).
Placopsilina vesicular is, Br.
Globigerina dutertrei, d'Orb.

,, pachyderma (Ehrng.).

Pullenia sphssroides (d'Orb.).
Truncatulina wucllerstorfi (Sell.).

,, pygmxa, Hantk.

Anomalina polymorpha, Costa.

STATION 421, lat. 68° 32' S., long. 10° 52' W.

Depth 2487 fathoms ; bottom temperature 31°-5 F. A glacial clay similar to that
from Station 420, bub containing 5 per cent of lime. Only a small quantity was
received for examination, in which the following were observed : — Globigerina bulloides,
G. dutertrei, G. dubia, Pullenia sphseroides, and Truncatulina wuellerstorfi.

STATION 438, lat. 56° 58' S., long. 10° 03' W.

Depth 2518 fathoms. A Diatom ooze with a large amount of volcanic mineral
particles. Slight traces of lime.

Only a very small quantity was sent me, and the only Foraminifera found were one
specimen of Cyclammina pusilla and a few Globigerina dutertrei.

STATION 447, lat. 51° 07' S., long. 9° 31' W.

Depth 2103 fathoms. A typical Diatom ooze containing 9 per cent of lime.

Foraminifera well represented, including : 2 Miliolidse, 11 genera of the arenaceous
types, 3 Textularidse, 4 species of Lagena, including the rare forms L. staphyllearia
and L. feildeniana ; Globigerina dutertrei moderately abundant ; 3 species of Pullenia,
2 of Truncatulina, 2 of Pulvinulina, 1 Rotalia, and 2 of Nonionina.

STATION 451, lat. 48° 06' S., long. 10° 05' W.

Depth 1741 fathoms. A deposit transitional in type between Diatom ooze and
Globigerina ooze, containing 55 per cent, of lime and 30 per cent, of siliceous
organisms.

The Grlobigerinidse are the predominating types present, the only two arenaceous
forms found being Proteonina diffluyiformis and Gordiammina charoides. The
characteristic feature of the Foraminifera from this station is the presence of the more
northern types, such as Bolivina nobilis, B. reticulata, Bulimina aculeata, B. buchiana,

(ROY. soc. BDIN. TRANS., VOL. XLIX., 1036.)

FORAMINIFERA OK THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 163

Uvigerina pygrn&a, Truncatul ina lobatula, Pulvinulina canariensis, and P. crassa,
with Nonioniiia n>nbilic«t«l</.

STATION 459, lat. 41° 30' S., long. 9° 55' \V.

Depth 1998 fathoms. A typical South Atlantic fi lie-grained Globigerina ooze,
containing 71 per cent of lime.

( >nly a small quantity available for examination, but this yielded a somewhat long
and interesting list of Fonuninifera. Amongst the most noteworthy of these are as
follows: — The strictly arenaceous types are but sparingly represented, Reophax distans,
Proteonina difflugiformis, Cribrostomoides bradyi, Ammochilostoma galeata, and
S/n'roplecta americuna ; 9 species of Miliolininse — amongst these are Biloculina
serrata, B. murrhi/na, and B. tubulosa, all deep-sea types. Textularidse are represented
by 9 species, and Lagenidse by no less than 23 species, among which may be noted as
the rarer forms L. stelligera, L.feildeniana, and L. truncata. The Globigerinidse, which
make up the greater portion of this deposit, are represented by 1 1 species ; the small
varieties of G. bulloides and G. dutertrei are by far the most abundant. G. sacculifera,
(7. conglobata, and P idle ma obliquiloculata, with Pulvinulina micheliniana, and
P. canariensis, types of the warmer-water areas of tropical seas, are met with here in
a more or less dwarfed and stunted condition.

STATION 467, lat. 40° 08' S., long. 1° 50' E.

Depth 2645 fathoms. A Globigerina ooze containing 46 per cent of lime. A
considerable variety of Foraminifera was obtained from this deposit, though the
specimens are few in numbers, and of small dimensions individually. The genera
represented are Biloculina, Miliolina, Proteonina, Cribrostomoides, Cyclammina,
Ammobaculites ; Textidaria, Gaudryina, Bolivina, Virgulina, Cassidulina, and
Layena, which are represented by ten species chiefly of the smaller deep-water types ;
also Nodosaria and Cristellaria, Globigerina by four species, with Pidlenia, Trun-
catidina, Pulvinulina, Rotalia, and Nonionina.

STATION 468, lat. 39° 48' S., long. 2° 33' E.

Depth circa 2645 fathoms. This deposit is somewhat similar in composition to that
obtained at Station 467, but with only 40 per cent of lime. The Foraminifera
obtained here are fewer, with regard to the bottom-living forms. Globigerina
bulloides and G. inflata, however, are more abundant, while Truncatulina would
appear to take the place of Lagena, so far as numbers of species go. The arenaceous
types are represented by Psainmospluvrafusca, Rhabdammina discreta, and Proteonutu
difflugiformis , while the higher forms are absent.

(ROY. 8OC. EDIN. TRANS., VOL. XLIX., 1037.)

164 MR F. GORDON PEARCEY ON

APPENDIX.

LlST OF FORAMINIFERA OBTAINED FROM STATION 81, LiAT. 18° 24' S., LONG. 37° 58' W.,
ON THE ABROLHOS BANK, OFF THE COAST OF BRAZIL.

The sounding-tube went down on hard ground and came up empty. From the
deposit taken in the. dredge, which was very much torn, and from the attached swabs,
the following Foraminifera were obtained, along with other organisms :—

Family MILIOLID^K

Sub-family MILIOLININ^E.
Genus BILOCULINA, d'Orbigny.
Biloculina ringens (Lamarck).
A few of this cosmopolitan species. Also found at three other Scotia stations.

Biloculina l&vis (Defrance).
A few specimens of this somewhat rare form.

Genus SPIROLOCULINA, d'Orbigny.

Spiroloculina limbata, d'Orbigny.
Rare. Also found at Station 342.

Genus MILIOLINA, Williamson.

Miliolina separans, H. B. Brady.
Rare.

Miliolina sclerotica (Karrer).
Rare.

Miliolina seminulum (Linne).
A few examples of this very common species. Also found at Station 346.

Sub-family HAUERININ^.

Genus ARTICULINA, d'Orbigny.

Articulina sulcata, Reuss.

A few specimens of this form, whose natural habitat is in comparatively shallow
water in the vicinity of coral islands in tropical and sub-tropical seas.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 1038.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 165

Sub-family PENEROPLIDIN^E.

Genus PENEROPLIS, Montfort.

Peneroplis pertusus (Forskal).

Specimens of the narrow compressed variety (Peneroplis arietinus (Batsch)) were
found in moderate numbers. This species is one of the most common Foramiuiiera in
the shallow waters of tropical seas, and is found as far north as the Mediterranean.

Family ASTRORHIZID&.

Sub-family HYPBRAMMININ^E.

Genus SAGENINA, Chapman.

Sagenina frondescens (H. B. Brady).

A few specimens of this interesting parasitic species were found attached to
calcareous algse and shell-fragments. It is a more or less common form in the shallow
waters of the eastern tropical seas, more especially in the areas of coral reefs and islands,
but is worthy of note to have been obtained off the coast of Brazil.

Family TEXTULARIID.E.

Sub-family TEXTULAPJBLE.
Genus TEXTULARIA, Defrance.

Textularia conica, d'Orbigny.

A few specimens. This species is widely distributed, especially in the coral seas of
tropical and sub-tropical regions. It was also obtained at Station 342.

Sub-family VERNEUILININ.E.

Genus VERNEUILINA, d'Orbigny.

I I'nieuilina spinulosa, Reuss.
Rare. This species is well known as a tropical or sub-tropical shallow-water form.

Genus CLAVULINA, d'Orbigny.

Clavulina parisiewris, d'Orbigny.

A few specimens. This species is a common form in West Indian seas, in compara-
tively shallow water. A fine-grained smooth variety was obtained by the Challenger
in the North Pacific in 3125 fathoms.

(KiiV. SOC. KDIN. TRANS., VOL. XLIX., 1039.)

166 MR F. GORDON PEARCKY ON

Family LAGENIDsE.
Sub-family NODOSARIN./E.

Genus CRISTELLARIA, Lamarck.
CristeEaria rotulata (Lamarck).

A few specimens of this fine species, which has a wide distribution, extending from
as far north as lat. 79° 45' to as far south as Tierra del Fuego. Its range of depth is
from the littoral zone to 2200 fathoms. I have taken the species in its fullest develop-
ment in the Faroe Channel, in 555 fathoms.

Cristellaria tennis (Bornemann).
Rare.

Family G LOBIGERINIDsE.

*

Genus GLOBIGERINA, d'Orbigny.

Globigerina bulloides, d'Orbigny.
Common.

Globigerina sacculifera, H. B. Brady.
Common.

Globigerina rubra, d'Orbigny.
Common.

Globigerina (Orbulina) universa, d'Orbigny.
A few.

Family ROTALID^E.

Sub-family ROTALIN^.

Genus CYMBALOPORA, Hagenow.

Cymbalopora (Tretomphalus) bulloides (d'Orbigny).

A few specimens of this curious form. Not known outside the tropical and sub-
tropical seas ; as a pelagic form it has been taken in great numbers in the regions of
coral reefs and islands.

Genus TRUNCATULINA, d'Orbigny.
Truncatulina reticulata (Czjzek).
In fair numbers.

Genus CARPENTERIA, Gray.

Carpenteria utricularis, Carter.

Rare. A tropical shallow-water form, always found adhering to various marine
organisms.

(ROY. soc. EDIN. TRANS., VOL. xi.ix., 1040.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 1 C.7

Genus PULVINUUNA, Parker and Jones.

Pu/riiin/ina, canariensis (d'Orbigny).
A few. Found at several other >Scoti<t stations.

I'ul fuiuli an oblonga (\Villiamson).
A few.

I'idvinidina repandu (Fichtel and iMoll).
A few specimens of this common shallow-water form.

Pulvinulina elegans (d'Orbigny).
A few. Also found at Station 342.

Pulvinulina umbonata, Reuss.
Also at Station 346.

Sub-family TINOPORIN^.
Genus GYPSINA, Carter.

Gypsina globulus (Reuss).
Small numbers of this interesting shallow-water form were obtained.

Genus POLYTREMA, Kisso.

Poli/treittd mi iii<n:'i-'n'iii (Pallas).

Common attached to shell-fragments, etc. This species is a common shallow-water
form in temperate and tropical seas. Although mainly found in shallow water, it was
collected by the Challenger naturalists in the Atlantic at a depth of 1000 fathoms.
With other adherent Foraminifera, it plays an important part in building up the lime-
stone rocks of coral reefs and islands.

Genus HOMOTREMA, Hicksnn.

Homotrema rul>nim (Lamarck).
Rare.

Genus SPORADOTREMA.

Sporadotrema cylindricnin (Carter).

Fragments of the basal and distal portions of this large parasitic form are of interest
from this locality.

(Itnv. SIM'. i:ii[\. 'I HANS., VOL. X1.IV, Mill.)

168 MR F. GORDON PEAKCEY ON

^1 Family NUMMULINID.E.

Sub-family POLYSTOMELLIN^E.

Genus NONTONINA, d'Orbigny.
Nonionina umbilicatula (Montagu).
Common. Also found at several other Scotia stations.

Sub-family NUMMULITIN.E.

Genus AMPHISTEGINA, d'Orbigny.

Amphistegina lessonii, d'Orbigny.
Common.

LITERATURE.

BALKWILL, T. P., and WEIGHT J., " Foraminifera of the Coast of Dublin and the Irish Sea," Trans. Irish

Acad., vol. xxviii., 1886.

BRADY, H. B., "Foraminifera," Challenger Exped. Zool, vol. is., 1884.
BRADY, H. B., "Foraminifera of the North-Polar Expedition, 1875-76," A nn. and Mag. Nat. Hist., Series 5,

vol. i. pp. 425-440(2 pi.), 1878.

BRADY, H. B., " On Some Arctic Foraminifera from Soundings obtained on the Austro-Hungarian North-
Polar Expedition, 1872-74," Ann. and Mag. Nat. Hist., Series 5, vol. viii. pp. 393-418, 1881.
BRADY, H. B., "Rhizopod Fauna of the Shetland Islands," Trans. Linn. Soc., vol. xxiv., 1864.
BRADY, H. B., PARKER, W. K., and JONES, T. R., "A Monograph of the Genus Polymorphina," Trans. Linn.

Soc., vol. xxvii., 1869.

CARPENTER, W. B., "Introduction to the Study of Foraminifera," Ray Soc. Publication, 1862.
CHAPMAN, F., "Foraminifera of the Funafuti Atoll, Ellice Island," Journ. Linn. Soc., vol. xxviii., 1900-03.

Ibid., "Foraminifera from the Lagoon of Funafuti," pp. 161-240.

Ibid., "Foraminifera collected round the Funafuti Atoll," pp. 372-417.

Ibid., "Foraminifera (chiefly deep-water) collected round Funafuti by H.M.S. Penguin," Journ. Linn.

Soc., vol. xxx., 1907-10.
CUSHMAN, J. A., "Monograph of the Foramiuifera of the North Pacific Ocean," Bulletin 71, United States

Museum, Smithsonian Institute, 1910-11.
EARLAND, A., "Foraminifera of the Shore Sand at liogiior, Sussex," Queld. Micro. Club, London (Series 2),

vol. vi., 1897.

FLINT, J. M., "Recent Foraminifera," United States National Museum, Smithsonian Institution, 1899.
GOES, A., "The Reticularian Rhizopoda of the Caribbean Sea," Kongl. Svenska Vetewlcaps-Akademiens

Hcmdlingar, Bandet xix., No. 4, 1882.
HERON-ALLEN, E., and EARLAND, A., "On the Recent and Fossil Foraminifera of the Shore Sands of Selsey Bill,

Sussex," Journ. Roy. Micr. Soc., 1908-11.
HICKSON, S. J., "On Polytrema and .some Allied Genera," Trans. Linn. Soc. (2nd Series), Zool., vol. xiv.,

1911.

JONES, T. W. O. RYMER, "On the Lagense from the Java Sea," Trans. Linn. Sue., vol. xxx., 1872.
MURRAY, Dr JOHN, "On the Deep and Shallow Water Fauna of the Kerguelen Region and the Great

Southern Ocean," Trans. Roy. Soc., Edin., vol. xxxviii. pp. 343-500, with map, 1869.

(HOY. soc. EDIN. TRANS., VOL. XLIX., 1042.)

FORAMINIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 169

NORMAN, H. N., Report Brit. Assoc. Swansea Meetimj, p. 335.

PAKKER, W. K., and T. RUPERT JONES, "On sonu-. Foraiuinifera from the North Atlantic and Arctic Oceans,

including Davis Straits and Builin liny," Phil. Trans., vol. civ. pp. 325-442, 1865.
PEARCEY, F. G., "Foraiuinifera of the Faroe Channel and AVyville Thomson Ridge," Soc, Nat. Hixf. Trans.

Glasgow (New Series), vol. ii. pp. 163-179, pi. i., 1890.
REUSS, A. E., "Die Foraminiferen-Familie der Lagenideen," Monoyraphisch daryestellt, 36 pp., 7 pi., 8vo,

WiL-n, 1863.

SAES, G. O., Vidensk-Selsk. ForhamU., 1871, p. 251.
WILLIAMSON, W. C., "Recent Foraminifera," Ray Soc. Publication, 1858.

EXPLANATION OF PLATES.

PLATE I.
Figs. 1-3, Pelosina arborescens, n. sp.

Fi<*. 1. General aspect of the test of a living specimen, showing the oral apertures of seven of the
arborescent tubes surrounded by sarcode and expanded pseudopodia. From Loch Fyne, west coast of
Scotland, 107 fathoms. x 2.

Fig. 2. Portion of test from near the base of the arborescent tubes, laid open to show the thickness of
the walls. x 8.

Fig. 3. Transverse section across the tubular neck of test, filled with granulated sarcode, viewed by
transmitted light. x 15.

Figs. 4-5, Pelosina arborescens, n. sp.

Fig. 4. A broken specimen with fragments in position ; the spaces between indicate where fracture
took place. From Station 420, Weddell Sea, 2620 fathoms. x 5.

Fig. 5. A portion of the test from near the base of the branching tubes, open transversely, showing the
thickness and character of the wall. x 6.

Figs. 6-10, Reoplta.c robustus, sp. nov.

Figs. 6, 7, 8 represent three stages of development, and show the robust external character of test, and
selective power shown by the organism of the building material. x 25.

Fig. 9. Younger specimen with test laid open longitudinally to show interior and comparative smooth-
ness of the chambers. x 25.

Fig. 10. End view of last chamber to show aperture. x 25.

Figs. 11-12, Thwammina favosa, var. reticulata, var. nov.

Fig. 11. Showing general aspect of test. x 25.

Fig. 12. A specimen laid open to show thickness of wall and the smooth interior. x 25.

Figs. 13-15, Hormosina irregularis, n. sp.

Fig. 1 3. Specimen showing the lateral aspect of test and the irregularity in forming the chambers, x 25.
Fig. 14. Younger specimen composed of two chambers only. x 25.

Fig. 15. Specimen with one chamber laid open to show the thickness of wall, and aperture from the
interior side. x 25.

PLATE II.
Figs. 1-2, Syringammina minuta, sp. nov.

Fig. 1. Specimen showing general aspect, attached to a crystal of quartz (sercone), with tubular offshoots,
adding to the rigidity of the test. x 30.

Fig. 2. Portion of a radial section, showing at a a the concentric reticulated portions. x 50.

(ROT. soc. EDIN. TRANS., VOL. XLIX., 1043.)

170 FORAMLNIFERA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

Figs. 3-4, Technitella ascifonnis, sp. nov.

Fig. 3. Specimen showing lateral aspect. x 30.

Fig. 4. Oral end of the test, showing grooved neck and everted lip. x 35.

Figs. 5-7, Cyclammina contortum, sp. nov.

Fig. 5. Typical specimen, showing lateral aspect. x 25.

Fig. 6. Showing peripheral aspect. x 25.

Fig. 7. Specimen laid open horizontally to show the internal structure, x 30.

Figs. 8-10, Haplophragmoides umbilicalum, sp. nov.

Fig. 8. Typical specimen, showing lateral aspect, x 30.

Fig. 9. Peripheral aspect. x 25.

Fig. 10. Specimen laid open horizontally to show internal structure. x 30.

Figs. 11-13, Lagena hispid-ipholus, sp. nov.

Fig. 11. Specimen showing superior aspect, x 35.

Fig. 12. Specimen showing inferior aspect, x 40.

Fig. 13. A subangular variet}r with keeled margin. x 35.

Figs. 14-16, Polymorpliina inflata, sp. nov.

Figs. 14, 15. Lateral aspect. x 30.
Fig. 16. Oral aspect. x 30.

Figs. 17-19, Miliolina, dentistoma, sp. nov.

Figs. 17, 18. Lateral aspect. x 30.
Fig. 19. Oral aspect. x 30.

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 1044.)

Scot. Nat. Ant. Exp.

Vol. VII.

I'KAKIT.Y : " SCOTIA " FOKAMIXIFEKA — PLATE I

5. x t;

4. x 4

2. x 8

3. - IS

6. x 25

»*a

1. x 2

10. x 25

14. * 25

i'-

m

7. x 25

8.x 25

L.^v-r^r-

11. x 25

12. x 25

y

1.5. x 25

F. .1 M.C.Pearccj. ''>•<"' -Vi/f

MTarlanc t Erekine. Lilh . Edni

Scot. Nat. Ant. Exp.

PEAHCEY : " SCOTIA" FORAMINIFERA — PLATE TI.

2. x 50

Vol. VII.

7. x 30

5. x 25

6. x 25

3. x 30

11. x 35

8. * 30

). x 25

10. x 30

16. x 30

13. '- 35

17.

12. x 40

M tarUiic & Erbkiue. Litb.. Ellin.

PART Vll.

SCHIZOPODA, STOMATOPODA, AND
NON-ANTARCTIC 1SOPODA.

VII.-THE SCHIZOPODA, STOMATOPODA, AND

NON-ANTARCTIC ISOPODA OF THE
SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

BY WALTER M. TATTERSALL. D.Sc..

Keeper of the Manchester Museum.

(WITH ONE PLATE.)

VII. — The Schizopoda, Stomatopoda, and non- Antarctic Isopoda of the Scottish
National Antarctic Expedition. By Walter M. Tattersall, D.Sc., Keeper
of the Manchester Museum. Communicated by Dr J. H. ASHWORTH.
(With One Plate.)

(MS. received June 5, 1913. Read July 7, 1913. Issued separately November 18, 1913.)

The present report deals with the whole of the Schizopoda and Stomatopoda in the
Scotia collections, and with those Isopoda which were taken in localities outside the limits
of Antarctica. I am indebted to the courtesy of Dr W. S. BRUCE for the opportunity of
examining and reporting on these collections. The report on the Antarctic Isopoda is
being prepared by Mr T. V. HODGSON.

I also include a few notes on a small collection of Schizopoda which were taken by
the Discovery on its outward journey to the South Pole, for the opportunity of examining
which I am indebted to the kindness of Dr W. T. CALMAN, of the British Museum.
These latter records are most appropriately included here along with those of the species
which the Scotia captured on her outward journey, over very nearly the same ground. A
comparison is thus possible between the captures of the two expeditions over the same
ground and in two different years, and may be brought out in the following lists :—

Scotia. Discovery.

Nov. 1902-Jan. 1903. Sept. and Oct. 1901.

Siriella Thompsonii. Siriella Thompsonii.

Thysanopoda tricuspidata. Thysanopoda tricuspidata.

Euphausia americana. Euphausia Krohnii.

„ brevis. „ americana.

,, tent') -a. ,, brevis.

,, hemigibba. ,, recurva.

„ pseudogibba. ,, tenera.

T/tysanoessa gregaria. ,, hemigibba.

Nematoscelis microps. ,, gibboides.

,, lucens.
„ longirostris.
Thysanoessa gregaria.
Stylocheiron carinatum .

No fewer than seven of the species are common to the two lists, and these represent
the most abundant species in the tropical Atlantic Ocean.

(KKL'KINTEI) FKO.M TI1K TKANSACTIoX.s ol HIE IUA.U. So'.'IKTY UK KDIKHURUU, VOL. XLIX., I'I>. 865-894.)

176

DR WALTER M. TATTERSALL ON

Turning to the Antarctic Schizopoda, the results of the Scotia may be compared
with those of other expeditious in the form of a table, as follows : —

Species.

French

Expedition.

Belyica.

Discovery.

Swedish
Expedition.

Scotia.

Euphausia superba, Dana

X

X

X

X

X

longirostris, Hansen

X

X

triacantha, H. and T.

X

X

frigida, Hanseu

X

X

crystattorophias, H. and T.

X

X

Vallmtini, Stebb.

X

X

Thysanoessa macrura, Sars

X

X

X

X

X

,, vicina, Hansen

X

X

Eucopia australis, Dana .

X

X

Hansenomysis antarctica, H. and T. .

X

X

Antarctomysis maxima, H. and T.

X

X

X

X

X

„ Oldinii, Hansen .

X

X

Pseudomma Belgicx, H. and T.

X

X

X

Dactylamblyops Hodysoni, H. and T.

X

Mysidetes posthon, H. and T.

X

X

Boreomysis distinguenda, Hansen

X

Brucei, W. M. T. .

X

I have taken the average limits of free and floating ice as the boundaries of the
Antarctic Eegions. This limit for the South Polar waters was laid down by Dr BRUCE*
in 1894, and is shown on a chart of the South Polar regions published by the British
Admiralty.

The French Expedition (francais) had its headquarters at Graham Land, a little
to the west of the Weddell Sea. The Schizopoda were described by COUTIERE (1906).

The Belgica worked a little further to the west of the Francais. The Schizopoda
were described by HANSEN (1908).

The Discovery had its winter quarters at the opposite side of the Pole to the Scotia,
at Victoria Land. The Schizopoda were described by HOLT and TATTERSALL (1906),
and TATTERSALL (1908).

The Swedish Antarctic Expedition (Antarctic), under NORDENSKJOLD, explored the
region to the east of Graham Laud, between the ground worked by the Scotia and the
Francais. The results, as far as the Schizopoda are concerned, have not yet been
published ; but HANSEN, in several of his recent papers, has recorded various species
belonging to the material of that expedition, and I have abstracted such records for
the purposes of the above table. (See note, p. 204.)

It will be seen, therefore, that the recent expeditions to the Antarctic have
explored that ocean between the Weddell Sea westwards to Victoria Land, about half
the South Polar Ocean, and we have a fair knowledge of the Schizopod fauna of that
area. The German Expedition (Gauss) had its headquarters half way between those of
the Scotia and Discovery, in about lat. 90° E., in the centre of the unexplored eastern

'• "Antarctic I'.irds," Knowledge, September 1, 1894.
(ROY. soc. EDIN. TRANS., VOL. XLIX., 866.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 177

part of the Antarctic Ocean, but I am not aware that any account of the Schizopods
has yet been published.

The above table includes seventeen species of Schizopoda — a complete list of the
known purely Antarctic forms, as far as I am aware. (See note, p. 204.)

Of these seventeen, three are more or less circmnpolar — Eupliausia superlx,
Thysanoessa macrura, and Antarctomysis maxima — having been recorded from the
collections of most of the recent expeditions. Of the remainder, the following species,
captured by the Discovery and the Swedish Expedition at almost opposite sides of the
South Polar Ocean, will probably be found ultimately to be circumpolar in their

distribution, viz. :—

Euphausia triacantha.

frigida.

„ crystallorophias.

„ Vallentini.

Thysanoessa vicina.
Hansenomysis antarctica.
A iitarctomi/sis OJilinii.
Pseudomma Belgicse.

Eucopia australis, captured by the Scotia and the Swedish Expedition, was not
taken by the Discovery ; but the type locality is quite near to Victoria Land, and the
species is in all probability circumpolar. This gives a total of twelve species, out of
seventeen known from the Antarctic Ocean, with a circumpolar distribution in that
ocean, leaving only five whose present known distribution is limited.

The Schizopoda, therefore, present very clear evidence of a group which is, as a
whole, circumpolar in its distribution in Antarctic waters.

The Scotia discovered only one new species, Boreomysis Brucei.

The Isopoda entrusted to me for identification include those taken by the Scotia
on the outward and homeward journeys, mainly at the Falkland Islands and at the
Cape, together with three species of parasitic Isopoda found on pelagic Decapoda and
Schizopoda captured in tow-nets in the open ocean. The collection as a whole calls for
very little comment, the recent work of STEBBING on the Crustacea of the Falkland
Islands and South Africa covering nearly all the ground. In common with most workers
called upon to identify isolated specimens of Sphseromidge from distant localities, I
have experienced considerable difficulty with this group. I have been obliged, as a
result of my work, to establish two new species, which I hope will not add to the state
of chaos in which the group remains at present. Dr BRUCE was also fortunate in re-
discovering Exosphseroma tristense, of LEACH, a species which has remained in obscurity
since its discovery nearly a century ago. I have redescribed the species, and figured, for
the first time, the adult male. A third new species among the Isopoda is established
for a specimen of the family Arcturidse found at the Cape, which appears to be very
distinct from any hitherto known form.

(HOY. soc. EDIN. TRANS., VOL. XLIX., 867.)

178 DR WALTER M. TATTERSALL ON

Finally, I desire to express my thanks to Dr BRUCE for entrusting me with this
collection, for much help with the literature, and for many courtesies in other ways ; and
to Dr W. T. CALMAN for allowing me to examine the collection of extra-Antarctic
Schizopoda made by the Discovery, and for permission to include the records here.

Order MYSIDACEA, Boas.

Family LOPHOGASTRID^E, G. 0. Sars.

Genus Gnathophausia, W. Suhm.

Gnathophausia gigcis, W. Suhm.
Scotia.

Station 450, lat. 48° 00' S., long. 9° 50' W., South Atlantic, 1332 fathoms,
trawl. — One female, 160mm.

This specimen, which is in a rather poor state of preservation, is the largest
recorded one belonging to the species, the previous largest, the Challenger specimen,
measuring 142 mm. It differs from SARS' description and figures mainly in the less
pronounced teeth on the antennal scale and in the less produced infero-posterior
corners of the carapace. Both of these slight differences are due to age, and the
specimen otherwise agrees with G. gigas so closely that I have no hesitation in
ascribing it to this species.

Family EUCOPIID^;, G. 0. Sars.
Genus Eucopia, Dana.

Eucopia australis, Dana.
Scotia.

Station 398, lat. 68° 25' S., long. 27° 10' W., Antarctic Ocean, vertical
tow-net, 0-1000 fathoms.— One female, 50 mm.

Station 414, lat. 71° 50' S., long. 23° 30' W., Weddell Sea, 2102 fathoms,
vertical tow-net, 0-1000 fathoms. — One female, 35 mm. ; one frag-
mentary specimen, tail end only.

Station 450, lat. 48° 00' S., long. 9° 50' W., South Atlantic, 1332 fathoms,
trawl. — One female, 45 mm.

Station 468, lat. 39° 48' S., long. 2° 33' E., South Atlantic, 2645 fathoms,
trawl. — One fragmentary specimen, head end only.

These specimens are in poor condition, but appear to belong, with very little doubt,
to E. australis as redefined by HANSEN (1905c).

(ROY. soc. EDIN. TRANS., VOL. XLIX., 868.)

SCHT/oPoKA. STOMATOPOPA, AND NON-ANTARCTIC ISOPODA. 179

Family MYSID.E, Dana.
Sub-family BORKOMYSIN.K, Holt and Tattersall.

Genus Boreomysis, G. 0. Sars.
Boreomysis distinguenda, Hansen.

B. scijphops, G. 0. Sars, 1885«.
li. diitingitenda, Hansen, 1908«.
nee. B. tcyphops, G. 0. Sars, 1885A.

Scotia.

Station 301, lat. 64° 48' S., long. 44° 26' W., Weddell Sea, 2485 fathoms,
trawl. — One female, 30 mm.

HANSEN, in describing the Crustacea collected by the Inc/olf Expedition in northern
seas, took the opportunity afforded by the capture of large numbers of the true
B. scyphops to separate the present southern species from its northern ally, with which
it had been confused by SARS. The distinguishing characters are to be found in the
shape of the eye and of the antennal scale. No full description of B. distinguenda has
yet appeared, but it is not possible to draw one up from the present specimen, which,
besides being only about half-grown, is in a very poor state of preservation. Suffice it
to say that, in respect of the eyes and antennal scale, its characters are in agreement
with those given by HANSEN for distinguishing B. distinguenda from B. scyphops.
Until HANSEN separated the two species, the latter was the last surviving instance
of a bipolar Schizopod. Now there is no species common to the Arctic and Antarctic
Oceans, though several genera are recognised as occurring at both Poles,

Boreomysis Brucei, sp. nov. (Plate, figs. 11-13.)
Scotia.

Station 414, lat, 71° 50' S., long. 23° 30' W., Weddell Sea, 2102 fathoms,
vertical net, 0-1000 fathoms. — One immature male, 25 mm.

Station 416, lat. 71° 22' S., long. IS" 15' W., Weddell Sea, 2370 fathoms,
trawl. — One immature male, 28 mm.

Specific Characters. — The frontal or rostral plate (Plate, tig. 11) is produced
almost to the anterior level of the eyes. The lateral margins are slightly convex, and
meet in an angle of less than a right angle, while the apex is produced into a short acute
spine. The eyes are moderately large, broader than deep, with the pigment very pale
brown. The eye-stalks have the distal tubercle well developed.

The antennular peduncle exhibits a light oblique ridge on the dorsal surface of
the basal joint, and has the second joint narrow, but much deeper dorso-ventrally than
either the first or third joints, so that it is almost circular in section.

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 8G9.)

180 DR WALTER M. TATTERSALL ON

The antennal peduncle extends anteriorly about half way up the distal joint of
the anteunular peduncle.

The antennal scale (Plate, fig. 12) overreaches the antennular peduncle con-
siderably, and is three times as long as broad, the broadest portion before the middle.
The outer margin is almost straight and ends in a prominent spine, beyond which the
apex of the scale is slightly produced.

The telson (Plate, fig. 13) is three times as long as broad "at its base, and has the
terminal cleft one-quarter of the entire length. The lateral margins are armed with
about thirty-six small spines, the first of which occurs about one-third of the way
down the margin. The spines are proximally arranged in small series of two and
three, but distally the last ten or a dozen spines are of equal length and not arranged
in series of shorter and longer spines. The lateral lobes of the apex of the telson are
armed with three strong spines, the outermost one of which is the longest.

The inner uropods (Plate, fig. 13) are about one-sixth as long again as the telson,
without spines in the region of the otocyst. The outer uropods (Plate, fig. 13) are
longer than the inner. The proximal portion of the outer margin, without spines or
setae, is about two-ninths of the length of the entire margin, and has two spines at its
distal corner.

This species comes remarkably near to B. siltogse, Hansen (1910), but differs in
the more produced rostrum and in the shape of the antennal scale.

In B. sibogai the rostral plate does not extend nearly as far forward as the
anterior level of the eyes, and its margins meet in an obtuse angle. In B. Brucei the
rostrum extends almost as far forward as the anterior level of the eyes, and its margins
meet in an acute angle.

In B. sibogse the spine on the outer margin of the antennal scale overreaches
the apex of the scale. In B. Brucei the reverse obtains. B. sibogiv is a deep-water
tropical species, B. Brucei a definitely Antarctic form.

B. Brucei also comes very near to B. rostrata, Illig (1906) ; but, so far as the latter
has been described, it differs from B. Brucei in the different form of the antennal
scale and the different armature of the telson.

I dedicate this new form to the leader of the Scottish National Antarctic Expedition,
whose intrepid researches have brought it to light.

Sub-family SIRIELLIN^E, Norman.
Genus Siriellct, Dana.

Siriella Thompsonii, H. Milne-Edw.
Scotia.

Station 11, lat. 23° 50' N., long. 21° 34' W., tow-net.— One.
Station 12, lat. 22° 19' N., long. 22° 07' W., tow-net.— Five.
Station 13, lat. 21° 58' N., long. 22° 26' W., tow-net.— One.

(ROY. SOC. EDIN. TRANS., VOL. XLIX., 870.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 181

Station 14, lat. 21° 28' N., long. 22° 40' \V., tow-net.— Ten.
Station 15, lat. 20° 34' N., long. 23° 12' W., tow-net.— Two.
Station 18, lat. 19° 59' N., long. 23° 34' W., tow-net.— One.
Station 21, lat. 18° 28' N., long. 24° 28' W., tow-net.— Seven.
Station 32, lat. 10° 46' N., long. 25° 21' W., tow-net.— One.
Station 33, lat. 9° 40' N., long. 25° 28' W., tow-net.— One.
Station 36, lat. 8° 42' N., long. 25° 28' W., tow-net.— Twenty-five.
Station 37, lat. 7° 50' N., long. 25° 31' W., tow-net.— Three.
Station 39, lat. 6° 43' N., long. 25° 48' W., tow-net,— Two.
Station 46, Int. 3° 13' N., long. 26° 30' W., tow-net.— Five.
Station 56, lat, 0° 42' S., long. 31° 20' W., tow-net.— Thirty-eight.
Station 58, lat. 2° 13' S., long. 32° 23' W., tow-net.— Forty-eight.
Station 59, lat. 2° 30' S., long. 32° 42' W., tow-net.— Eight.
Station 61, lat. 3° 38' S., long. 33° 20' W., tow-net.— One.
Station 62, lat. 4° 15' S., long. 33° 38' W., tow-net— Eighteen.
Station 515, lat. 2° 32' N., long. 19° 32' W., tow-net— Nine.

Discovery.

Funchal Bay, Madeira. — Four.

Lat 13° 59' S., long. 34° 35' W.— Two.

Lat. 17° 15' S., long. 32° 05' W.— Three.

Lat. 30° 43' S., long. 21° 36' W.— One.

Lat. 33° 53' S., long. 17° 38£' W.— Two.

All the stations listed above are in the Atlantic Ocean, and all the specimens were
captured at the surface. These records indicate that S. Thompsonii is an abundant
species in the tropical Atlantic ; and a correspondingly Jong list of captures given by
HANSEN (1 912) shows that it is likewise equally common in the Eastern Pacific. Its dis-
tribution is, in short, circumtropical, bounded, roughly speaking, by the lines of latitude
40° N. and 40° S. Several specimens in the Scotia collections were found to have the
Epicarid, Dajus sirtellie, G. 0. Sars, in their marsupial pouches. This parasite, first
found by SARS in the same host, collected by the Challenger, has only been recorded
once since its discovery, namely, by HANSEN (1912), who also found it in the present host.

Siriella denticulata, G. M. Thomson.

N. denticulata, Thomson, 1900.
Discovery.

Laurie Harbour, Auckland Isles. — One female, 6 mm., immature.

I refer this small and immature specimen to S. denticulata, Thomson, with some
little doubt, and add a few notes supplementing THOMSON'S description. The rostrum,
in my specimen, can hardly be described as spiniform. The two lateral margins meet
in almost a right angle with the apex hardly produced. There is, however, a promi-

(ROY. soc. KDIN. TKANS., VOL. XLIX., 871.)

182 DR WALTER M. TATTEKSALL ON

nent pseudo-rostrum, as in Macromysis inermis, and it is possible that THOMSON may
have mistaken this structure for the true rostrum. On the other hand, the rostrum
may become more produced and spiniform with age, as it is known to do in certain
Euphausians, e.g. Thysanoessa macrura. The telson of the present specimen has three
lateral spines on each margin, anterior to the constriction characteristic of the telson
in this genus, and between the three small spines at the apex of the telson there are
finer and longer setaj. THOMSON does not mention or figure either of these features.

o

The inner uropods have seventeen spines on their inner margins, the spines com-
mencing at the statocyst and extending the whole way to the distal extremity of the
appendage, increasing in size. They are not arranged in series. There are five spines
on the outer margin of the proximal joint of the outer uropods.

Sub-family MYSIN.*:.
Genus Antarctomysis, Coutiere.

Antarctomysis, Coutiere, 1906.
Antarctomysis, Hausen, 19086.
Antarctomysis, Tattersall, 1908.

Antarctomysis maxima, Hausen.

My sis maxima, Holt and Tattersall, 100G.
Antarctomysis maxima, Coutiere, 1906.
A. maxima, Hausen, 1908^.
A. maxima, Tattersall, 1908.

Scotia.

Station 325, lat. 60° 43' 42" S., long. 44° 38' 33" W., Scotia Bay, South Orkneys.
—One immature female, 30 mm.

This species has a circumpolar distribution, having been captured by the Discovery,
the Uelgica, the Charcot Expedition, and now by the Scotia, at the four points of the
compass in the Antarctic Ocean.

Order EUPHAUSIACEA.
Genus Thysanopoda, Milne- Ed w.

«

Thysanopoda cornuta, Illig.

T. cornuta, Illig, 1905.
T. insignis, Hansen, 19056.
T. cornuta, Hansen, 1912.
Scotia.

Station 407, lat. 40° 08' S., long. 1° 50' E., 2645 fathoms, trawl.— One
female, 79 mm.

This magnificent specimen agrees well with both ILLIG'S and HANSEN'S descriptions.
It was captured at very nearly the same place as the type-specimen, in the Benguela

(UOY. SOC. EL>IN. TRANS., VOL. XLIX., 87i!.)

SUHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC 1SOPODA. 183

Current, but rather further south. It is only known from five other specimens, three
recorded by HANSEN from the North Atlantic, one by HANSEN from the East Pacific,
and the type, captured by the V<iltln-i<i in the South Atlantic. It is one of the
largest known Euphausians.

Thysanopoda tricuspidata, Milne-Edw.
Scotia.

Station 29, lat. 12° 31' N., long. 25° 9' W., tow- net. —Two.
Station 42, lat. 5° 25' N., long. 26° 7' W., tow-net— One, larval.

Discovery.

Lat. 12° 27' S., long. 33° 33' W., tow-net.— One.

Lat. 13° 59' S., long. 34° 35' W., tow-net— One large female, 22 mm. ;

three larvae, 5-8 mm.
Lat 17° 15' S., long. 32° 05' W., tow-net— Five.

All these specimens were caught at the surface, and, with the exception of the
large female, 22 mm., captured by the Discovery, are all larval or post-larval in
development.

(ienus Euphausia, Dana.

Euphausia Krohnii (Brandt).
Discovery.

Off Madeira, tow-net — Fifteen.

This species seems at last to have found a name which may be considered more
or less a permanent one. It has been known during the last ten years successively as
Euphausia pellucida, bidentata, and Mullen; but HANSEN (1910) has definitely
established that Thysanopoda Krohnii, Brandt, an earlier name than any of the
above, was applied to specimens identical with those of the later species. By this
name, therefore, the species must henceforth be known.

h'ti-pltausia americana, Hansen.

E. americana, Hauseu, 1911.

Scotia.

Station 14, lat 21° 28' N., long. 22° 40' W., tow-net— Five.
Station 18, lat 19° 59' N., long. 23° 34' W., tow-net. -Four.
Station 26, lat. 14° 33' N., long. 25° 9' W., tow-net— Two.
Station 29, lat 12° HI' N., long. 25° 9' W., tow-net— Thirteen.
(ROY. soc. KDIN. TRANS., vol.. XLIX., sTUj

184 DR WALTER M. TATTERSALL ON

Station 36, lat. 8° 42' N., long. 25° 28' W., tow-net.— One.

Station 39, lat. 6° 43' N., long. 25° 48' W., tow-net.— One hundred and

thirty-six.

Station 42, lat. 5° 25' N., long. 26° 7' W., tow-net.— Ten.
Station 59, lat. 2° 30' S., long. 32° 42' W., tow-net.— One.
Station 512, lat. 0° 22' N., long. 18° 43' W., tow-net.— Four.

Discovery.

Lat. 7° 23' S., long. 30° 23' W.— Two.

Euphausia americana has only lately been instituted by HANSEN, and appears to
have been confused hitherto with E. Krohnii, to which it bears a considerable resem-
blance. It seems clear that some of the Cliallenger specimens referred to E. pellucida
by SARS in reality belong to the present species. HANSEN gives as the locality for
this species, West Atlantic, Cape Verde ; but the above list of captures shows that the
species has a very general distribution in the tropical parts of the Atlantic Ocean.

Euphausia recurva, Hansen.

E. recurva, Hansen, 1905c.
E. recurva, Hansen, 1912.

Discovery.

Lat. 30° 43' S., long. 21° 36' W.— Thirty-one.
Lat. 33° 53' S., long. 17° 38^' W.— Five.
Lat. 35° 10' S.: long. 13° 40' W.— Sixty-three.
Lat. 36° 27£' S., long. 8° 20' W.— One.
Lat. 37° 33f S., long. 6° 09' E.— Four.
Lat. 37° 12' S., long. 9° 30' E.— Three.

All the specimens were taken in surface tow-nettings. The majority are post-larval
in development, but appear to belong to this species.

E. recurva is known from the South Atlantic, Indian Ocean, and from two or three
localities in the Pacific (HANSEN, 1912).

Euphausia brevis, Hansen.

K. bretiis, Hansen, 1905c.
E. brevis, Hansen, 1912.
Discovery.

Lat. 12° 27' S., long. 33° 33' W.— One.

Lat. 13° 59' S., long. 34° 35' W.— One.

Lat. 17° 15' S., long. 32° 05' W.— Seven.

L:it. 36° 27i' S.,long. 8° 20' W.— One.

(HOY. SOU. KDIN. TRANS., VOL. XL1X., 874.)

SCHIZOPODA, STOMATOl'OhA. AND NON-ANTARCTIC ISOl'ODA. 185

Scotia.

Station 14, lat. 21° 28' N., long. 22° 40' W., tow-net— Twenty-three.
E. brevis has a general distribution in the tropical parts of the Atlantic Ocean, in
the Indian Ocean, and in the Eastern Pacific. It has also been recorded from the
Mediterranean (TATTERSALL and HANSKN). HANSEN notes that most of the specimens
recorded have been taken at the surface. The present records are no exception.

Euphausia tenera, Hanson.

K. tenera, 1 1 arisen, 1905c.
K. tenera, Hunsen, 1910.
E. fenera, Hanseu, 1911.

Scotia.

Station 14, lat. 21° 28' N., long. 22° 40' W., tow-net.— Three.

Station 18, lat. 19° 59' N., long. 23° 34' W., tow-net.— Two.

Station 29, lat. 12° 31' N., long. 25° 9' W., tow-net.— Twenty-two.

Station 36, lat. 8° 42' i\., long. 25° 28' W., tow-net.— Fifty.

Station 39, lat. 6° 43' N., long. 25° 48' W., tow-net.— Twelve.

Discovery.

Lat. 7° 23' S., long. 30° 23' W.— Five.
Lat. 13° 59' S., long. 34° 35' W.— Three.

All the specimens here recorded were captured at the surface.

Euphausia superba, Dana.

E. superba, Tattersall, 1908.

Scotia.

Station 152, lat. 60° 32' S.,long. 43° 40' W., February 2nd, 1903, stomach of

Lobodon carcinophaga. — Many.
Station 156, oft' Saddle Island, South Orkneys, February 3rd, 1903, edge of

the ice-floes. — Twenty.
Station 159, lat. 61° 20' S., long. 43° 23' \V., February 4th, 1903, stomach

of penguin. — Many.
Station 203, lat. 59° 38' S., long. 29° 55' W., February 13th, 1903, edge of

ice-floes, tow-net. — Seven.
Station 325, lat. 60° 43' 42" S.,long. 44° 38' 33" W., Scotia Bay, South Orkneys,

April to July 1903. — Fifty-one in nets, and many from the stomachs of fish.
Station 411, lat. 74° 01' S., long. 22° 00' W., off Coats Land, 161 fathoms,

traps. — Eighteen.

Station 414, lat. 71° 50' S., long. 23° 30' W., vertical net, 0-1000 fathoms.
—Five.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 875.)

18(5 1)11 WALTER M. TATTEESALL <>N

Station 417, lat. 71° 22' S., long. 16° 34' W., 1410 fathoms, trawl— One

large female.
Station 418, lat. 71° 32' S., long. 17° 15' W., 1221 fathoms, trawl (not on

bottom). — One male.
Station 422, lat. 68° 32' S., long. 12° 49' W., vertical net, 0-800 fathoms.

-Three.

Dr BRUCE has furnished me with three coloured sketches of Euphausians, all of
which refer to this species. One of the sketches, of a specimen caught in February
1903 at the edge of the ice-floes, agrees almost perfectly with the account of the colour
of this species as noted by Dr G. KACOVITZA during the expedition of the Belgica, and
published by HANSEN (19086). In the other two sketches, of specimens captured in
February 1903 and March 1904, there is considerably more red pigment shown on the
dorsal surface of the carapace and abdomen. The distribution of the pigments is the
same in all three sketches, but in the two latter ones the red is intensified. This
difference, it seems probable to me, may be accounted for by the supposition that, in
the animal from which the first sketch mentioned above was made, the red chromato-
phores were in a contracted condition, and in the other two specimens they were in
an expanded condition at the time they were painted.

Eupliausia superba is the Euphausian par excellence of the Antarctic Ocean. It
is circumpolar in its distribution, and has been recorded by all the recent expeditions
which have visited those waters. It likewise forms the major part of the food of the
crab-eating seal, Lobodon carcinophaga, and of certain of the penguins.

Euphausia htcens, Hansen.

E. lucens, Hansen, 1905e.
E. lucens, Hansen, 1911.
Discovery.

Lat. 36° 27^' S., long. 8° 20' W.— Two.
Lat. 37° 47' S., long. 3° 59' E.— Two.
Lat. 37° 33f S., long. 6° 09' E.— Fifteen.

This species of the genus is one of the rarest, and has not been captured by any
of the expeditions since the Challenger. HANSEN mentions specimens from three
localities in the South-East Atlantic, very much in the same neighbourhood as the
present records, and from one locality between Tasmania and New Zealand.

Euphausia hemigibba, Hansen.
E. he.migibba, Hansen, 1910.

Scotia.

Station 14, lat. 21° 28' N., long. 22° 40' W., tow-net.— Thirteen.
Station 18, lat. 19° 59' N., long. 23° 34' W., tow-net.— One.
Station 21, lat. 18° 28' N., long. 24° 28' W., tow-net.— Two.
(ROY. soc. EDIN. .TRANS., VOL. XLIX., 876.)

SCHTZOPOPA, STOMATOPOPA. AND NON-ANTAHCTTC ISOl'OPA. 187

Discovery.

Lat. 30° 43' S., long. 21° 36' W.— Six.
Lat. 35° 10' S., long. 13° 40' W.— Three.
Lat, 30° 27 -V S., long. 8° 2u' W.— One.

At present, this species is known only from the tropical Atlantic (HANSEN), Indian
Ocean, and Mediterranean (TATTKRSALL).

Euphausia pseudogibba, < >rtmann.

E. pseudogibba, Ortmaun, 1893.
E. pseudogibba, Hansen, 1910.
E. pseudogibba, Hansen, 1912.

Scotia.

Station 29, lat. 12° 31' N., long. 25° 9' W., tow-net.— One.

This species is known from the tropical Atlantic, Indian Ocean, and the Pacific,
from which the types were recorded. It is most generally distributed in the Indian
Ocean and the Atlantic, and decidedly rarer in the Pacific.

Euphausia gibboides, Ortmaun.

E. gibboides, Ortmann, 1893.
E. gibboides, Hansen, 1911.
E. gibboides, Hausen, 1912.
Discovery.

Lat. 36° 27^' S., long. 8° 20' W.— One female, 22 mm.

This specimen differs from the description and figures given by HANSEN (1912)
in the form of the lobe on the first joint of the antennular peduncle. This lobe has a
bifid extremity, the outer process quite minute, and much smaller than the inner and
main extremity. The specimen, however, agrees otherwise so well with E. gibboides
that it has seemed best to include it in that species for the present, at any rate until
male specimens are forthcoming and the copulatory organs on their pleopods can be
investigated.

Euphausia longirostris, Hansen.

E. lonyiroftris, Hansen, 19086.
E. longirostris, Hansen, 1911.
Discovery.

Lat. 37° 47' S., long. 3° 59' E.— One adult male, 19 mm.

When first I examined this specimen, I determined it as a variety of E. spinifera,
G. 0. Sars, with which it agrees very closely, except in regard to the lobe from the

(ROT. soc. EDIN. TRANS., VOL. XLIX., 877.)

188 DR WALTER M. TATTERSALL ON

first joint of the antennular peduncle. This lobe in E. spinifera, extends right across
the peduncle and has its anterior margin irregularly digitate. In the present specimen,
the lobe does not stretch right across the peduncle, and the extremity is bifid. It
thus agrees, in this respect, with E. longirostris. The copulatory organs on the first
pleopods, however, agree almost exactly with those figured by SARS for E. spinifera.
HANSEN has not, up till now, described the male of E. longirostris, so I am unable to
compare my specimen from this point of view.* HANSEN says that E. longirostris is
closely related to E. spinifera, and is only distinguished by the antennular lobe. On
that character, therefore, I refer my specimen to that species.

E. longirostris is known, at present, only from the Antarctic Ocean to the south
of the Falkland Islands and in the neighbourhood of South Georgia. The present record,
therefore, is the most northerly one yet known for the species.

Genus Thysanoessa, Brandt.
Thysanoessa macrura, G. 0. Sars.

T. macrura, G. O. Sars, 1885a.

Scotia.

Station 319, lat. 61° 05' S., long. 43° 20' W., 214 fathoms. — One female,

14 mm.
Station 414, lat. 71° 50' S., long. 23° 30' W., vertical net, 0-1000 fathoms.

— One female, 28 mm.

If the evidence of the antennular flagellum be accepted, the smaller of these two
specimens is correctly referred to this species, since it is distinctly shorter than the
two distal joints of the peduncle. The larger specimen seems clearly to belong to
T. macrura. The species has a circumpolar distribution in Antarctic waters.

Thysanoessa gregaria, G. 0. Sars.

T. gregaria, G. O. Sars, 1885a.

Scotia.

Station 98, lat. 34° 02' S., long. 49° 07' W., tow-net.— One.
Station 458, lat. 42° 57' S., long. 8° 13' W., tow-net.— Twenty-six.

Discovery.

Lat. 37° 47' S., long. 3° 59' E.— Two.

A post-larval specimen of the genus Thysanoessa taken by the Scotia at Station
137, lat. 57° 42' S., long. 46° 33' W., cannot be referred to its adult species. It may
belong to T. gregaria or to T. vicina. Hansen (1911).

* Sri- note.', j). 204. In this paper HANSEN describes and figures the copulatory organs of the male of this species, and
points out minor dillt-rtnces from those of E. spinifera.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 878.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 189

Genus Nematoscelis, G. 0. Sars.

Nematoscelis »i/rro/«, G. O. Sars.

.V. microps, G. O. Sars, 1885n.
N. microp*, Hansen, 1905/1 and c.
N. microps, Hansen, 1910.
A', microps, Hansen, 1912.
Scotia.

Station 12, lat. 22° 19' N., long. 22° 07' W., tow-net.— Ten.
Station 14, lat. 21° 28' N., long. 22° 40' W., tow-net.— Thirteen.
These specimens are all small and in rather poor condition. I cannot be quite
certain of their identity, but I believe they ought to be referred to this species.

Genus Stylocheiron, (4. (_). Sars.
Stylocheiron carinatum, G. O. Sars.

S. carinatum, G. 0. Sars, 1885a.
& carinatum, Hansen, 1910.
S. carinatum, Hansen, 1912.
Discovery.

Lat. 13° 59' S., long. 34° 35' W.— One.
Lat. 17° 15' S., long. 32° 05' W.— Thirteen.

This interesting and easily recognisable species of the genus Stylocheiron is widely
distributed in the various tropical waters of the globe.

Order STOMATOPODA.

Family SQUILLID.E.
Genus Squillci, Fabricius.

Squilla ttrmata, Milne-Edwards.
Scotia.

Station 481, N.W. off Ijzer Fontein Point, Cape Colony, 35 fathoms, sand,
trawl. — Two, 116 mm. and 98 mm.

Genus Lysiosquilla, Dana.

Stomatopod larvse, referable to this genus, were taken on two occasions, in surface
tow-nets.

Station 64, lat. 6° 30' S., long. 34° 25' W., off Brazil, tow-net.— One, 3 mm.
Station 66, lat. 7° 9' S.; long. 34° 30' W., off Brazil, tow-net.— Two, 2 mm.
and 3 '5 mm.

These specimens, belonging to the genus of larval Stomatopods known as Lysierichthus,
cannot be referred to their adult species.

(HOY. SOC. ED1N. TEAMS., VOL. XUX.,

190 DR WALTER M. TATTERSALL ON

Order ISOPODA.

Tribe FLABELLIFEKA.

Family EURYDICID/E.

Genus Cirolana, Leach.
Cirolana, Hansen, 1890.

Cirolana hirtipes, Milne-Edwards.
C. hirtipes, Hansen, 1890.

Scotia.

Station 482, Saldanha Bay, Cape Colony, May 1904, 8-10 fathoms, trawl. —
One, 25 mm.

This species is only certainly known from the Cape, from which MILNE-EDWARDS'
type was procured, and from which HANSEN has since recorded a single specimen.

Cirolana sulcata, Hansen.
C. sulcata, Hansen, 1890.

Scotia.

Station 482, Saldanha Bay, Cape Colony, May 1904, 8--10 fathoms, trawl. — One.

Recorded from the shores of Simon's Bay by HANSEN (1890), and from Somerset
West, False Bay, by STEBBING (1902). Not known from any other locality.

Family CORALLANID^E.
Genus Lanocira, Hansen.

Lanocira sp. ?
Scotia.

Station 482, Saldanha Bay, Cape Colony, May 1904, 8-10 fathoms. — One,
12 mm.

The telson of this specimen is, unfortunately, damaged, so that it is not possible to
identify the species, if known, or to describe the specimen satisfactorily. Moreover, the
sex of the specimen is uncertain. The absence of oostegites and of the external
setiferous lobe to the maxillipedes indicates that the specimen is a male. On the other
hand, I can find no appendix internet on the second pleopods.

The structure of the mouth parts clearly indicates the generic position of this
specimen, and, in the form of these appendages, it approaches very closely, among the
described species, to L. zeylanicus, Stebbing (I905o), witli which it agrees in the strong

(ROY. soc. EDIN. TRANS., VOL. xux., 880.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 191

development of the hook of the first maxilla and in the great inequality in the length
of the two seise, arming the distal joint of the second maxilla.

The shape of the body has been considerably altered, presumably by the method of
preservation. The segments are more or less separated from each other, and the whole
body considerably distended, so that it is impossible to get a fair idea of the normal
form. As it is, the specimen has not the compact, broadly oval, rotund form of the other
members of the genus. The body is nearly three times as long as broad, and the
epimeral plates seem to be more in evidence in dorsal view than in the figures of other
species of the genus. The length, moreover, is almost twice that of any other
species. The largest described form is L. Gardineri, Stabbing (1904), which is 7 mm.
long, while females of L. rotundicauda, with young in the brood-pouch, measure only
5'25 mm. The present specimen is 12 mm. long.

The body is provided with setse only on the fourth and fifth segments of the
mesosome and the whole of the telson. In this respect it is more setose than
L. Gardineri and L. rotundicauda, but less so than in L. zeylanicus.

The broadly oval inner branch of the uropods bears seven spines at its extremity,
and the outer branch, which is just slightly shorter than the inner, bears three spines.

I prefer to leave the identity of this species an open question till more specimens,
not deformed or mutilated in any way, are available. In the meantime, the genus has
not, so far as I am aware, been recorded from South Africa previously.

Family CYMOTHOID^E.
Genus Glossobius, Schiodte and Meinert.

Glossobius linearis (Dana).

G. linearis, Hansen, 1895.
Scotia.

Station 36, lat. 8° 42' N., long. 25° 28' W., tow-net— One young, in the
second stage, 3 mm.

This specimen agrees very closely with the specimens described and figured by
HANSEN (1895). I would point out, however, that both HANSEN'S specimen and my
own show four well-developed teeth on the dactylus of the first thoracic legs.
SCHIODTE and MEINERT (1879-84) show only three well- developed teeth and ;v
rudimentary one for G. linearis, but four well-developed teeth for (1. laticauda. In this
character, therefore, and in the shape of the eyes the present specimen approaches more
closely to G. laticauda (M.-Ed.), a Pacific species. On the other hand, this specimen
agrees so well with HANSEN'S specimens, and is so obviously the same species, that I
accept his decision as to the name it should bear.

(ROY. si«'. i.nix. TKA.N'S., vui.. xi.ix., 881.)

192 DR WALTER M. TATTERSALL ON

Family S
Genus Limnoria, Leach.

Limnoria lignorum (Ratlike).
Scotia.

Station 118, 51° 41' S., 57° 51' W., Port Stanley, Falkland Islands, January
1903. — One.

This specimen was found among other Isopoda from Port Stanley, with no special
note as to how it was come by. I cannot say, therefore, if it was engaged in its usual
practice of destroying wooden structures. I can find no appreciable differences from
northern specimens of the same species. The nearest recorded place of capture is
Port Elizabeth, South Africa (STEBBING, 1908).

Genus Exosphseroma, Stebbing, 1900«.
Exosphxroma, Hansen, 1905«.

Exosphteroma gigas, Leach.

E. yifjas, Stebbing, 1900a.

Scotia.

Station 118, 51° 41' S., 57° 51' W., Port Stanley, Falkland Isles, January

1903. — ca. fifty, all sizes.

Station 349, 51° 41' S., 57° 51' W., Port William, Falkland Isles, January
1904.— Nineteen.

I have nothing to add to STEBBING'S detailed account of this species. It was
infested, as seems usual with the species in this part of the world, with lais pubescens.

Exosphseroma tristense (Leach). (Plate, fig. 1.)

Splixroma tristense, Leacb, 1818.

Sphxroma tristense, Hansen, 1 905«.
nee. Sphxroma tristense, Krauss, 1843.
nee. Sphan'oma tristense, Stebbiug, 1910.

Scotia.

Station 461, lat. 40° 20' S., long. 9° 56' 30" W., Gough Island, 21st-22nd
April 1904. — One male, 10 mm., and three females, 5-6 mm., from the
shore and from floating weed.

I refer these specimens from Gough Island to LKACH'S rather obscure species,
which does not seem to have been recognised since it was described in 1818, from

(llOY. SOC. EDIN. TRANS., VOL. XLIX., 882.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. I !>:;

specimens taken at Tristan il'Acunlia. The female specimens agree with LEACH'S
brief description in its main points — body smooth, terminal segment of the abdomen
terminating in an obtuse point and having at its base two elongated and rather
indistinct tubercles. The description of the terminal segment of the abdomen and
the uropods may be amplified somewhat. The last segment of the abdomen is
triangular in shape, narrowing to a rather produced apex, the actual tip of which is
bluntly rounded. The segment is not evenly vaulted from its edges, as, for instance,
it is in E. yiyas ; but some little way in from the. margins there is a shallow impressed
groove running more or less parallel with the margins all the way round. The central
portion thus marked off is evenly vaulted, and bears anteriorly two elongated but
only slightly pronounced tubercles. The inner and outer uropods are about equal in
length, and barely reach the apex of the abdomen. The inner one is truncate at its
distal extremity, the outer one evenly rounded.

The male specimen, 10 mm. in length, which I refer to this species, differs from
the female in having the seventh segment of the thorax produced into a short, blunt
median process, which projects slightly beyond the anterior margin of the last seg-
ment of the pleon (Plate, fig. 1). Moreover, the tubercles on the latter are very
obscure and almost obsolete. But otherwise the agreement with the female speci-
mens is very close, especially in the form of the pleon and uropods, as described
above, though the latter are, perhaps, a little more fully developed. I have very
little doubt that the male specimen should be referred to the same species as the
females, and I think I am correct in regarding both as examples of LEACH'S species.

If my identification is correct, the generic position of the species requires con-
sideration. As regards the mouth organs and the structure of the pleopods, the
specimens are in complete harmony with the genus Exosphseroma. Moreover, they
show the closest agreement with the type of that genus, E. yiyas, Leach, in the
general form and the structure of the various appendages. They differ from E. gigas
in the form of the pleon and uropods in both sexes, and in the process from the
seventh thoracic segment in the male. HANSEN'S amended definition (1905«)ofthe
genus Exosphseroma, however, runs as follows : — " Last thoracic segment unarmed in
both sexes. End of the abdomen at most somewhat produced, but not acute." If
this definition be accepted, LEACH'S species would be excluded from the genus
Exosphferoma by the characters of the last thoracic segment of the male, and
would fall into one or other of the genera Zuzara and Isocladus. These latter
genera are, however, further characterised by the great development of the uropods
in the male, greater in Zuzara than in Isocladus, but much greater in both than in
the present species, in which the difference between the sexes in this respect is almost
negligible. The females of all three genera are very much alike, and UANSEN himself
has indicated the great difficulty of separating the genera in a satisfactory manner.
Moreover, as STEBBING (1910) points out, he has at least implied a modification of
the definition of Exosplui'roma, quoted above, by including in the genus Sphaeroma

(BOY. SOC. KDIN. TRANS., VOL. XLIX., 883.)

194 DK WALTER M. TATTERSALL ON

Stimpsonii, Heller (1868). HELLER says of the latter that the hind margin of the
seventh segment of the thorax is produced into a conical process, and that the telsouic
segment has an acute apex. The adjective " acute," as applied to the telson, is, I take
it, meant relatively to the shape of that organ in such a species as E. gigas, and in no
way intended absolutely. I have accepted this implied emendation to the characters
of the genus Exosphseroma, and, as modified, include & tristense, Leach, within its limits.
I may, perhaps, be allowed to suggest the probability that E. tristense, Leach, and
E. Stimpsonii, Heller, are synonymous. I have already referred to the close similarity
between E. tnstense and E. gigas as regards their appendages and general structure.
The above remarks on E. Stimpsoni apply equally well to the male of the specimen I
refer to E. tristense, and I think it highly probable the two forms are one and the same
species. I have not the necessary material to pronounce a definite opinion here, but I
make the suggestion, for any future worker with more material at his disposal to decide.
If the suggestion is upheld by future research, LEACH'S name has priority. The females
of E. tristense agree very closely with WHITE'S types of S. leiicitra, which I have
examined at the British Museum. This species was named by WHITE (1847), but never
described. It is not unlikely that it will be found to be synonymous with S. integrum,
Heller, described from specimens taken off Chile, near the same locality as that from
which WHITE'S types came. HELLER'S name would have preference, since WHITE'S name
can only be regarded as a nomen nudum. HANSEN refers S. leucura to Exosphseroma,
and S. integrum to, possibly, Isocladus or Zuzara. I believe both should be referred
to the genus Exosphseroma, and suspect that the males will be found to have the same
form as those of E. tristense. It would not surprise me if Sphseroma Stimpsonii,
S. leucura, and S. integrum were all eventually found to be synonymous with S. tristense,
though females of allied species of Exosphssroma are notoriously difficult to separate.

Exosphseroma Kraussii, sp. nov. (Plate, figs. 2 and 6.)

1 Sphseroma tristense, Krauss, 1843.
1 Sphxroma tristense, Stebbing, 1910.
Scotia.

Station 483, entrance to Saldanha Bay, Cape Colony, trawl. — Three females,
8 mm.

Specific Characters. — Sexes similar ; body microscopically granular, especially on
the pleon and uropods ; segments of the thorax with four very obscure small tubercles
equidistantly placed, the tubercles most pronounced on the last thoracic segment, and
becoming almost obsolete on the anterior segments ; a pair of larger and more definite
rounded tubercles on the centre of the combined first three segments of the pleon ; last
segment of the latter triangular in shape with a pointed apex, having a pair of closely
approximating, conspicuous elongate tubercles at the centre of the anterior part, the
tubercles separated by a shallow groove, from the distal end of which a light carina runs

(ROY. soc. KDIN. TRANS., VOL. XLIX., 884.)

SOHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 105

to the apex of the telson ; epimcra visible in dorsal view ; uropods subcqmvl in length,
sliglitly shorter than the telson. inner ranius bluntly rounded, outer ramus acute, the
extremity of both branches minutely and irregularly denticulate when seen under the
low power (§") of the microscope (Plate, fig. 6).

I believe this species to be the one recorded by KUAUSS (1843) as Sphmroma
tristense, Leach. KRAUSS' description may be quoted in full : " Die 2 langlich
Hockerchen auf dem letzten segmente, so wie die stumpfe Spitze des Abdomen bestimmen
mich, meine Examplare fiir diese von LEACH nur sehr kurz bcschriebene Art zu halten,
jedenfalls gehoren sie zu der Abtheilung der Spezies, deren 2 letzte Ringe des Thorax
wie die vorderen gebildet siud ; aber alle Riuge haben in der Mitte 4 sehr undeutliche
Hockerchen und an den Seiten eine ahnliche Anschwellung. Die Lamellen der hinteren
falschen Fiisse haben glatte Rander und sind gerade so lang als die Spitz des Abdomen.
In der Tafelbai. Lange 5 '2 linien."

The adjective " stumpfe," it is true, does not accurately describe the apex of the
pleon in the present form, but the character which I rely on mainly for the identification
of this species with the one observed by KRAUSS is : "all the segments have in the middle
four very obscure tubercles." This does not apply to S. tristense of LEACH, which has
the thorax smooth ; but it accords very well with the present species, though the
tubercles are almost obsolete on the anterior segments. The " similar intumescence " on
the sides of the segments, mentioned by KRAUSS, is present on the segments of
E. Kraassii, as a slight swelling in the region of the junction of the body segments
with their epimeral plates. If E. Kraussii is not identical with S. tristense, Krauss,
I am unable to identify it with any described form.

In the British Museum I found several specimens of this species, unnamed, from
Cape Town. Among them were two or three males, which agree in all respects with the
females, and have no processes on the thoracic segments. In the characters of the mouth
parts and pleopods, the species is in agreement with the genus Exosph&roma. 1 have
named the species in honour of the only worker on South African Crustacea (previous
to the recent researches of STEBBINC), who most probably had the species before him in
compiling his catalogue.

The species is, so far, only known from Cape Colony in the neighbourhood of
Cape Town and neighbouring bays.

Exosph&roma Coatsii, sp. nov. (Plate, figs. 3 and 4.)
Scotia.

Station 118, lat. 51° 41' S., 57° 51' W., Port Stanley, Falkland Islands,
January 1903. — Six females, 4-10 mm.

Specific Characters. — Body capable of rolling up, or at least doubling up ; epimeral
plates not visible in dorsal view, projecting down at right angles to the rest of the
segments of the body, from which they are sharply marked on' by a strong ridge which

(ROY. BOC. EDIN. TRANS., VOL. XL1X., 885.)

196 PR WALTER M. TATTERSALL ON

projects laterally and hides the plates from the dorsal aspect ; segments of the thorax
with four small, equally distant tubercles on the dorsal surface, the tubercles more
pronounced than in E. Kraussii ; a pair of larger tubercles in the centre of the combined
first three segments of the pleon ; terminal segment of the latter triangular in shape
with the apex somewhat produced but the actual tip rounded ; centre of the telsonic
segment evenly vaulted from a point some little way in from the margins, bearing in
the centre two pairs of large tubercles, the anterior pair slightly narrower and more
elongate than the posterior pair ; behind the latter in the median line are two smaller
tubercles, one behind the other; the lateral parts of the telsouic segment with a number
of minute tubercles or granulations ; a very light carina runs from the base of the median
tubercles to the apex of the telson ; inner and outer uropods shorter than the telson ;
apex of the inner one truncate, with the outer corner somewhat produced into a sharp
angle ; outer uropod almost sabre-shaped, apex sharpely acute.

My report was almost completed when I received the large adult female specimen,
on which this description is based, from the Kev. T. R. R. STEBBING, who found it and
another example among some Decapod crustaceans from the same place. I had already
figured as the type the largest specimen then in the collection, a female, 6 mm., and I re-
produce that figure here because it illustrates the differences between the young and adult
of this species, and as a contribution to our knowledge of the changes which a Sphseromid
may undergo during growth. A comparison of the two figures will show that there is
considerable difference between the young and adult stages. In the young stage, the
telsonic segment is less produced and its apex more obtuse than in adult specimens.
It bears only the two pairs of larger tubercles of the adult stage, the two median
tubercles and the lateral minute tubercles being absent. The tubercles throughout the
body are less developed in the young stage. The differences in the uropods are quite
considerable. In the young example both uropods have more or less evenly rounded
extremities, quite distinct from the form of the uropods in the adult as described above.

Having seen no male specimens, I am unable to say whether this species exhibits
any marked sexual differences. Otherwise the species seems clearly referable to the
genus Exospliseroma, as far as the characters of the mouth organs and pleopods go.

The arrangement of the epimeral plates is quite characteristic. The first plate
projects forward under the eyes, the next four are rectangular or rhomboidal in shape,
while the sixth is larger than any of the others and projects backwards so as to almost
hide the small seventh plate. When the animal is doubled in two, the epimeral plates
form a very good protection for the sides of the body.

The arrangement of the tubercles will allow the species to be recognised at once.
It differs from E. Kraussii in the more pronounced nature of the tubercles of the body,
in the possession of two pairs of large tubercles in the telsonic segment, and the shape
of the uropods, as well as in the form of the epimeral plates.

I have seen a specimen of this species, unnamed, in the British Museum, from the
island of St Paul.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 886.)

SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 197

Genus Cymodoce, Leach.

Cymodoce uncinata, Stebbing.
C. uncinata, Stebbing, 1902.

Scotia.

Station 482, Saldanha Bay, Cape Colony, May 1904. — Two males, 6 mm.
and 12 mm.; four females, 7-8 mm.

The females of this form do not seem to have been hitherto observed. They have
two submedian, blunt and rounded bosses on the telsonic segment, which correspond to
the large bosses found on the male, but are very much less developed. The bosses
are not so well developed in the largest female as in the smaller male, in which they
have reached almost adult proportions. The apex of the telsonic segment of the
female is trifid, with the median lobe well developed and bluntly rounded, and the
lateral lobes marked off by mere notches. The apex of the telson in the young male
is of the same form. It seems to me that the form of the apex of the telson character-
istic of the adult male is reached by the greater development of the lateral lobes of
the female, and consequently of the notches which separate them from the median lobe.

The most characteristic feature of the present species is the scythe-like termination
of the outer uropod, and the sharply truncate extremity of the inner uropod, which
are the same for both sexes. STEBBING mentions both of these points, but hardly
emphasises them.

Sphaeromidds oj uncertain identity.

Two specimens of eubranchiate Splueromidse, representing two distinct species, are
present in the collection. Both are female and, as such, cannot be referred to their
correct genera, though they appear to be very closely allied to Dynamenclla. I have
not attempted to refer them to their proper species. They may briefly be noted
as follows :—

(1) Station 118, 51° 41' S., 57° 51' \V., Port Stanley, Falkland Isles.— One
female, 4 mm. (Plate, figs. 7 and 8.)

The most characteristic feature of this specimen is the form of the telson, which
I have represented on Plate, figs. 7 and 8. Looked at from above, it takes the
form of a triangle narrowing rapidly to an abrupt apex ; but from the ventral surface
the lateral margins are seen to be folded in to a certain extent, though they do not
meet in the mid-ventral line to form a definite tube as in Cymodocella, nor is the half
tube thus formed as long as in the. latter genus. Looked at from the posterior end,
therefore, the end of the telson appears as a semicircular nutcli ; hut this cannot be
considered as a notch in the apex of the telson, since il is formed by the infolding of

(ROY. soc. EDIN. TRANS., VOL. xux., 887.)

198 DR WALTER M. TATTERSALL ON

the lateral margins. The body presents no processes or tubercles of any kind, though
the integument appears granular to a certain extent, due probably to extraneous
matter and not a definite character of the integument itself. The mouth parts and
pleopods agree with the characters of the sub-family Sph&rominse eulranchiata.

(2) Station 482, Saldanha Bay, Cape Colony, May 1904. — One female, 6 mm.
(Plate, figs. 9 and 10.)

The body is quite smooth, without tubercles or processes of any kind whatever.
The colour is a golden yellow minutely flecked with black pigment. The telson is
similar in form to that of the last specimen, but less abruptly narrowed, with a wider
apex, and less inwardly folded margins. The result is that, from the dorsal aspect,
the apex of the telson appears slightly emarginate.

The telsonic segment and uropods are shown on Plate, fig. 9. The most
characteristic feature of the specimen is the form of the superior antenna (Plate,
fig. 10), in which the second joint of the peduncle is very broad, with a strong ridge
throughout its whole length, so that in cross-section it would appear triangular and
not lamellar. The third joint is quite small and distally expanded.

Further specimens of both these forms are desirable before their systematic position
can be determined.

Tribe VALVIFERA.

Family IDOTEID.E.
Genus Paridotea, Stebbing, 19006.

Paridotea ungulata (Pallas).

P. ungulata, Stebbing, 19005.
Scotia.

Station 478, Table Bay, Cape Town Harbour, May 1904.— Five.

Station 482, Saldanha Bay, Cape Colony, May 1904, 5-25 fathoms. — Abundant.

Specimens were also procured from the stomach of a dogfish, caught in Saldanha
Bay.

Genus Synidotea, Harger.

Synidotea hirtipcs (Milne-Edwards).

S. hirtipes, Stebbing, 1902.

Scotia.

Station 478, Table Bay, Cape Town Harbour, May 1904.— Four.

Station 482, Saldanha Bay, Cape Colony, May 1904, 5-25 fathoms. — Abundant.

(HOY. SOC. EUIN. TRANS., VOL. XLIX., 888.)

SCHIZOPUDA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA. 199

Genus Idotea, Fabricius.

Idotea metallica, Bosc.
Scotia.

Fifteen specimens belonging to this species were found among the collec-
tions submitted to me, without any note as to the locality of their
capture. I suspect they were taken among the Gulf weed, through which
the Scotia passed between 22nd June and 30th June 1904.

Station 482, Saldanha Bay, Cape Colony, May 1904. — Two.

Family ASTACILLID/E.
Genus Antarcturus, zur Strassen.

Antarcturus ornatus, sp. nov. (Plate, fig. 5.)
Scotia.

Station 482, Saldanha Bay, Cape Colony, May 1904. — One female, 7 mm.

Plate, fig. 5, gives a general idea of the form of this species and shows its most
characteristic feature, namely, its ornamentation or armature of short, stiff bristles on
all the segments of the body.

The head and the first three segments of the thorax equal together the length of
the large middle segment. The last three segments of the thorax are widely separated.
The metasome has two segments partially marked off with transverse sutures.

The setee are found on the dorsal surface of the animal, on all the segments, as well
as on the head and metasome. The surface of the body is irregular, roughened, and
microscopically spinulose, but there are not any distinct tubercles. The first three and
the last three segments of the mesosome are elevated dorsally when seen in lateral view,
and the setse are arranged in a broad band across this elevated part, and are most
numerous in the centre. The well-marked intervals between the last three segments
of the mesosome are devoid of setae. The middle segment of the body shows two
setigerous areas, a wide and broad anterior one and a narrow posterior one, separated
by a shallow depression devoid of setse. This is shown very well in lateral view.
Both the setigerous areas are elevated and roughened ; the non-setigerous band, smooth
and depressed.

The eyes are moderately well developed and lateral. The superior antenna reaches
to the level of the distal end of the second joint of the peduncle of the inferior antenna.
The flagellum is equal in length to the last two joints of its peduncle and bears olfactory
filaments.

The inferior antenna is two-thirds of the entire length of the animal from the
front of the head to the posterior end of the metasome. The fourth joint is equal in

(ROY. soc. EUIN. TRANS., VOL. SLIX., 889.)

200 DR WALTER M. TATTERSALL ON

length to the preceding three joints, and slightly longer than the fifth joint. The
flagellum is short, about half as long as the fifth joint, composed of three joints, the
last joint terminating in a spine. I cannot see any denticulations on the inner margin
of the flagellum. The inferior antenna is armed with a few scattered stout setse
similar to those which are found on the body, but there are no teeth or spines
of any kind.

The first gnathopods are broken off on both sides. The remaining legs present no
special features.

Only one other South African Arcturid is known, Antarcturus kladophorus,
Stebbing (1908). From this species A. ornatus is readily distinguished by the quite
different character of the armature of the body, by the shorter and stouter inferior
antenna, and by the shorter flagellum to the latter. I know of no species of this group
with which A. ornatus can be confused. Spiny and tuberculous forms are common, but
no setigerous species have been described.

The generic position of this form is doubtful. STEBBING (1908) gives a table for
the discrimination of genera belonging to this family, based primarily on the number of
marsupial plates. KOEHLER (1911) has shown that all the genera of Arcturid^e possess
three oostegites, and that therefore this character is useless for generic separation.
This discovery increases the difficulty of deciding the generic position of the species of
the family. In the general form of the body, A. ornatus approaches Antarcturus and
Arcturella, and 1 provisionally refer it to the former genus. It cannot be referred to
Arcturina, Koehler, because the second and third thoracic limbs are not robust,
but conform to the type met with in Antarcturus. Male specimens are necessary to
decide whether it should be referred to the genus Arcturopsis, Koehler. The
separation of the epimeral plates would seem to exclude it from the genus Pleuroprion,
zur Strassen.

Tribe ASELLOTA.

Family JANIRID.-E.

Genus lais, Bovallius.

lais pubescens (Dana).
1. /mbescens, StebMng, 1900a.

Scotia.

Station 118, lat. 51° 41' 8., long. 57° 51' W.

A large number of specimens of this curious and interesting commensal Isopod
were found in the bottles containing Exosp/iwroina gigas from Port Stanley
and Port William, Falkland Isles. They were, presumably, living on the latter
species when captured. I have nothing to add to STEBBING'S description of the
species.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 890.)

SCHIZOPODA, STOMATOPODA, AND NON- ANTARCTIC ISOPODA. 201

Tribe EPICARWEA.

Family BOI-YRID.K.
Genus Probopyrus, Giard and Bonnier.

Probopyrus latreuticola (Gissler).
Scotia.

Station 538, lat. 32° 11' N., long. 34° 10' W., tow-net.— Eight, from
Latreutes ensiferus, captured among tbe Gulf weed.

Family DAJIID.E.
Genus Heterophryxus, G. 0. Sars.

Heterophryxus appendiculatus, G. 0. Sars. (Plate, figs. 14 and 15.)

H. appendiculatus, G. 0. Sai's, 1885«.
Scotia.

Station 39, lat. 6° 43' N., long. 25° 48' W., tow-net.— One female, with
attached male, free in a tow-netting containing many Euphausia
americana, Hansen.

Station 512, lat. 0° 22' N., long. 18° 43' W., tow-net.— One female, with
attached male, from Euphausia americana.

It is almost certain that the specimen from Station 39, found unattached, was
originally parasitic on Euphausia americana, of which there were over one hundred
specimens in the same gathering. Thus both specimens in this collection were from
the same host.

These specimens differ from those described and figured by me (1905) from
specimens taken from Euphausia Krohnii in the form of the last pair of legs.
I figure on the Plate these limbs from one of the present specimens (fig. 14) and
from a specimen taken in the North Atlantic to the west of Ireland (fig. 15).
It will be seen at once that, in the specimens from E. americana, the inner branch
of these peculiar appendages is shorter and stouter than in the specimens from
E. Krohnii. These figures illustrate incidentally the most frequent position of the
limbs in preserved specimens. I cannot decide at present whether this difference
is of specific value. In the first place, the host of the type specimen must be
considered uncertain, in the light of HANSEN'S recent work. It was called Euphausia
pellucida by SARS, but HANSEN has shown that SARS confused several distinct species
under that name. The host of the type specimen was taken in the North Atlantic,
near to Cape Verde. This is just the locality given by HANSEN for E. americana,
and though there is no improbability that the specimen is a true A'. Krohnii, it is
more probable that it is E. americana, the same species from which the present
specimens were taken. I have examined the type host and parasite in the British

(ROT. soc. EDIN. TRANS., VOL. XLIX., 891.)

202 UR WALTER M. TATTERSALL ON

Museum, but the}1 are mounted in Canada balsam, and are only to be seen in
lateral view. It was not possible to make sure either of the species of the host or
the form of the legs in the parasite. Until this matter can be settled, it is not
desirable to consider the differences noted above as specific. If the type host is
E. Kro/inii, the differences can only be regarded as varietal ; but, if E. americana,
the grounds for considering them specific are strengthened. On the principle
enunciated by GIARD and BONNIER, each species of host has a separate species of
parasite. The genus Heterophryxus would seem to be a favourable one in which
to test the truth of this axiom, for the form and shape of the last pair of legs
seem to afford more definite characters than are usually to be found in the species
of this group. It would be necessary to examine a large number of specimens taken
from definitely and accurately named hosts, to decide the point. In the meantime,
it seems to me to be best to refer the present specimens to the type species, with
a note on the differences they exhibit.

Genus Dajus, Kroyer.

Dajus siriellfB, G. 0. Sars.
D. sirieUx, G. O. Sars, 1885a.

Scotia.

Station 12, lat. 22° 19' N., long. 22° 07' W., tow-net.— Several in the

Cryptoniscan stage.
Station 14, lat. 21" 28' N., long. 22° 40' W., tow-net.— One female., with two

males attached, from the incubatory pouch of Siriella Thompsonii.
Station 56, lat. 0° 42' S., long. 31° 20' W., tow-net.— One female with male,

free, but almost certainly from one of the thirty-eight specimens of

Siriella Thompsonii in the same gathering.
Station 62, lat. 4° 15' S., long. 3D0 38' W., tow-net. — Two females with

males, from the incubatory pouch of Siriella Thompsonii.

This species was originally described by SARS from specimens taken from Siriella
Thompsonii captured during the cruise of the Challenger. It has not, so far as I
am aware, been recorded since that time, until last year, when HANSEN (1912) noted
the presence of an Epicarid, probably this species, from the same host, captured in
the Eastern Pacific.

The occurrence of two males with the same female is of interest.

SARS' type specimens, both male and female, were immature, the male being only
in the Cryptoniscan stage. The present female specimens, presumably mature, have
the incubatory pouch prolonged backwards in two bluntly rounded projections, which
meet in the mid-ventral line and extend beyond the end of the pleon. The pouch is
also prolonged in front beyond the head, on either side. The segmentation of the

(ROT. soc. EIJIN. TRANS., VOL. XLIX., 892.)

SCH1ZOPODA, STOMATOPODA, AND NON-ANTARCTIC 1SOPOHA.

body is distinct throughout, and there is no cordon for the support of the male. Tin-
pleon in the female is terminated by a pair of biramous uropods, not uniramous as
figured by SARS. In this respect the species diii'ers from the genus Dajus, to which
it was referred by SARS, and it is possible that this difference may, in the future, be
thought worthy of generic significance; but I do not feel able at present to go into
the matter fully, as 1 have not enough material for the purpose. I do not know of
any other genera of Dajiidse in which the uropods of the female are biramous.

The adult male has the pleon uusegmented, narrow, pointed, and slightly curved,
and terminated by a pair of uniramous uropods.

Both the adult male and the Cryptoniscan larvae are characterised by the excessive
development of dark pigment, so well illustrated in SARS' figure. This feature enables
the larvae to be identified in tow-net gatherings.

LIST OF REFERENCES.

COUTIERE (H.), 1906. — Expeil. Antarct. Francaise : Crust., Schizo/t. et Decap.

HANSBN (H. J.), 1890. — " (Jirolanidse et familiae nonnullee propinquiB Musei Haunieusis," Kgl. Dansk.

Vid. Selsk, Skrift. (6), nalur. og. math. Afd., I3d. v., Hft. 3, Copenhagen.
— - 1895. — Isopoden, Gumaceen u. Stomntopoden der Plankton- Expedition.

1905a. — "On the Propagation, Structure, and Classification of the Family Sphteromidaj," Quart.

Jour. Micros. Science, vol. slix., N.S., part i.

— 1905A — "Preliminary Report on the Schizopoda collected by H.S.H. Prince Albeit of Monaco during
the Cruise of the Princess Alice in the year 1904," Ball. Mus. Ocean. Monaru, No. 30.

- 1905c\ — " Further Notes 011 the Schizopoda," Bull. Mus. Ocean. Monaco, No. 42.

1908a — Danish Ingolf Expedition, vol. iii., 2, Crustacea Malacostraca, i.

. 19086.— "Schizopoda and Curnacea," Result at s du Voyage du 8.Y. " Belyica" en IS'J?, 1W8, 18'M :

Zooloyie.

— 1910. — "The Schizopoda of the Siboya Expedition," Siboga Expeditie, xxxvii.

1911. — "The Genera and Species of the Order Euphausiacea, with Account of Remarkable Variation,'

Bull. Instil. Ocean. Monaco, No. 210.

- 1912. — "Reports Scient. Results Expedition to the Tropical Pacific . . . Steamer AUmtruss: The
Schizopoda," Mem. Mus. Comp. Zonl. Harvard, vol. xxxv., No. 4.

HKI.LKR (C.), 1868. — " Crustaceen," Reise der Outer re i,:h. Fmjatte " Novara," Zoo/, fht-i/, Bd. ii.,

lift. 3, Wien.
HOLT (E. W. L. H.) and TATTERSALL (W. M.), 19U6. — " Preliminary Notice of the Schizopoda collected by

H.M.S. Discovery in the Antarctic Region," Ann. May. Xal. ///.-•/., ser. 7, vol. xvii.
ILUO (G.) 1905. — "Eine neue Art der Uattuug Thysanopoda," Znot. Air.., lid. xxviii., No. 19-20.
1906. — " Bericht iiber die neuen Schizopodengattungen und Aitrn der Dt-.utsclien Tiefsee-Expedition

1898-99," Zool. An::., Bd. xxx., Xo. 7.
IvjEULBU (R.), 1911. — " Arcturidus nuuveaiix [irovcuant des campagnes de la lji inces.-ie Alice ou appartenant

au Musee Oceauographique de Monaco," Bull. Instit. Oc^in. Mmttiru, No. 214.
RKAUSS (R.), 1843. — Die Sudafrikaniachen «.'rusi;ici;en.

LEACH (W. E.), 1818. — Diet, des Sciences Naturellex, vol. xii., Strasbourg and Paris.
OHTMANN (A. E.), 1893. — Decapodvn nn<l Schizopoden <!< r Plankton-Expedition.
1894. "The Pelagic Schizopoda. Report Dredging Operations ..IT tin' \Vest Coast of Central America

to Galapagos, . . ." Bull. Mus. Comp. Zool. Harvard, vol. xxv., No. 8.

(ROY. soc. EDIN. TRANS., VOL. XLIX., *'M.)

204 SCHIZOPODA, STOMATOPODA, AND NON-ANTARCTIC ISOPODA.

SARS (G. 0.), 1885a.— " Report on the Schizopoda collected by H.M.S. Challenger," Zool. Chall. Exped.,
pt. xxxvii., vol. xiii.

18856. — Norske Nordh.-Ex/i. : Crust., i.

ScHi6DTE(J. C.)andMEiNERT(F.), 1879-1884.— "Symbol® ad monographium Cymothoarum," Nat. Tuts.,

xii.-xiv., Copenhagen.
STBBBING (T. R. K.), 1900<i— "On some Crustaceans from the Falkland Islands collected by Mr Rupert

Vallentin," Proc. Zool. Soc. London.

19006. — "South African Crustacea," pt. i., Marine Investigations in South Africa, vol. i.

- 1902. — South African Crustacea," pt. ii., ibid., vol. ii.

1904. — " Marine Crustacea : Isopoda," Fauna and Geoyr. Maid, and Law. Arch., vol. ii., pt. 3.

1905a. — "Eeport on the Isopoda collected by Professor Herdman at Ceylon," Ceylon Pearl Oyster

Fisheries, 1905, Supplem. Rep., xxiii.

19056.— "South African Crustacea," pt. iii., Marine Investigations in South Africa, vol. iv.

1908. — "South African Crustacea," pt. iv., Ann. Soutli African Mux., vol. vi., pt. i.

1910. — "General Catalogue of South African Crustacea," Ann. South African Mus., vol. vi., pt. iv.

ZUR STRASSEN (0.), 1902. — Zool. Am., vol. xxv.

TATTERSALL (W. M.), 1905. — " The Marine Fauna of the Coast of Ireland, pt. v., Isopoda," Fisheries, Ireland,
Sci. Invest., 1904, ii.

1908. — "Schizopoda," National Antarctic Expedition: Natural History, vol. iv.

THOMSON (G. M.), 1900. — "On some New Zealand Schizopoda," Jour. Linn. Soc. London, vol. xxvii.
WHITE (A.), 1847. — List of the Specimens of Crustacea in the Collections of the British Museum.

EXPLANATION OF THE PLATE.

Fig. 1. Exoxphxroma tristense (Leach), male. x 8.

Fig. 2. Exosphxroma Krausii, sp. nov., female. x 8.

Fig. 3. Exosphxroma Goatsii, sp. nov., female, immature. x 8.

Fig. 4. Exosphxroma Coatsii, sp. nov., adult female. x 8.

Fig. 5. Antarcturus ornatus, sp. nov., female, x 10.

Fig. fi. Exosphxroma Krausii, sp. nov., extremities of the uropods.

Fig. 7. Dynamenella sp. ?, telson and uropods, dorsal view.

Fig. 8. Dynamenella sp. 1, telson, ventral view.

Fig. 9. Dynamenella sp. 1, telson and uropods, dorsal view.

Fig. 10. Dynamenella sp. 1, superior antenna.

Fig. 11. Boreomysis Brucei, sp. nov., dorsal view of anterior end.

Fig. 12. Boreomysis Brucei, sp. nov., antennal scale.

Fig. 13. Boreomysis Brucei, sp. nov., telson and uropods.

Fig. 14. Heterophryxus appendiculatus, G. O. Sars, last leg from a specimen taken from Euphausia
americana.

Fig. 15. Heterophryxus appendiculatus, G. 0. Sars, last leg of a North Atlantic specimen from Euphausia
Krohnii.

NOTE ADDED IN THE PRESS. — Wink- this paper was in the hands of the printers, I received a copy of H ANSEN'S Report
on (lie Crustacea Schizopoda of the Swedish Antarctic Expedition, but I have been unable to insert references to it
in the main body of the text or to modify the latter in any way in accordance with HANSEN'S latest observations.
Three new Antarctic Mysida- are described, and further records of known species given.

(ROY. soc. EDIN. TRANS., VOL. XLIX., MM.)

Scot. Nat. Ant.Exp. Vol. VII.

TATTERSALL : " SCOTIA" SCHI/OI-ODA AND NON-ANTARCTIC ISOPODA.

i.

£ 4
3

£•
i

•

4.

MTarlanr t Erekine. Lith.. Edin.

PART VIII.
POLYCH^ITA.

(Family NEREIDS.)

VIIL-POLYCHJ]TA (Family NEREtD*) OF THE
SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

BY L. N. G. RAMSAY, M.A., B.Sc.,

Carnegie Research Scholar, Christ's College, Cambridge.

(WITH ONE PLATE.)

VIII. — Polychseta of the family Nereidse, collected by the Scottish National
Antarctic Expedition (1902 1904). By L- N. G. Ramsay, M.A., B.Sc., Carnegie
Research Scholar, Christ's College, Cambridge. Communicated by Dr. J. H.
AsmvoRTH. (With One Plate.)

(MS. received October 7, 1!)13. Read DeuemW 1">, 1913. Issued separately March 30, 191J.)

The collection of Nereidte brought home by the ticnti't proves to be of considerable
interest. As other expeditions have indicated, the family is but poorly represented in
the antarctic or sub-antarctic regions ; and although a large number of specimens were
collected at the South Orkney Islands, these have all proved to belong to one species,
N. kerguelensis M'Int. No nereids were obtained at any of the deep-water stations
farther south — the family being decidedly littoral in its range.

The chief interest, however, lies in the material collected so assiduously throughout
the vessel's wanderings. Six other species were obtained, including one from the
Falkland Islands, hitherto undescribed.

The investigations of the more recent workers on the Nereidse continually tend to
widen the distribution of hitherto known species, and to link up forms which have been
described as distinct — in some cases apparently from the supposition that it was incon-
ceivable that one and the same species could inhabit areas so widely separated as, say,
the Indian Ocean and the North Atlantic. It is becoming more and more evident,
however, that in the determination of species, locality must not be taken into account.
Several species are already recognised as having a range that is practically cosmopolitan
(e.g. N. dumerilii, from the North Atlantic, North Pacific, and Persian Gulf; N. mirabilis,
from the West Indies, Red Sea, Persian Gulf, and Malay Archipelago). It is therefore
clear that the world-wide distribution of any particular species is a contingency that
must be reckoned with.

The means of dispersal which render this world-wide distribution possible can only,
for the present, be a matter of surmise ; the carriage of pelagic larvae by ocean currents
may be sufficient to account for it ; and in this connection, also, the investigations of the
Scotia naturalists supply a very suggestive hint in the discovery of nereids living and
breeding among floating gulf-weed in mid-Atlantic. These, it may be noted, have been
described as a variety of the species AT. dumerilii.

The excellent condition of the Scotia material has rendered its examination and the
determination of the species comparatively easy. The descriptions following are in no
case based on a single specimen ; where only a few specimens were found, these were all
examined in detail ; where examples were numerous, the individual variation has been
taken into account. This is of cnii-iilrniUr importance, as the great latitude of indi-

(KKl'RINTED FROM THE TRAXSA' TI".VS OK Tilt UOYAL SOCIETY OK EDINBURGH, VOL. I.., IT. 41-48.)

210 MR L. N. G. RAMSAY ON POLYCH^TA OF THE FAMILY NEREIDS,

vidual variation, in such characters as the numbers and grouping of paragnaths, and the
relative lengths of appendages, has apparently led astray several former investigators of
the group, and has resulted in the creation of a deplorable number of perfectly indistin-
guishable species. This is undoubtedly due in some measure to the laborious nature of
the examination necessary for the adequate determination of the characters of a nereid.

In the preparation of this report, the extensive collections of Nereidae in the
Cambridge University zoological laboratories, hailing from all regions of the globe, have
been of great assistance ; for these I am indebted to Mr F. A. POTTS. It may be
mentioned here that the writer is at present engaged on a revision of the family as a
whole, and on this account one or two points, such as the relationships of the species
allied to N. falldandica, or of the species comprised in the sub-genus or group Platy-
nereis, have not been discussed in the present paper.

I have to thank Professor J. STANLEY GAKDINER and Mr F. A. POTTS for helpful
advice. My best thanks are also due to Professor F. JEFFREY BELL for kindly allowing
me access to the collections at the Natural History Museum for purposes of comparison ;
finally, to Dr W. S. BRUCK himself for permitting me to examine the material.

TABLE OF SPECIES, WITH LOCALITIES.

Nereis kerguelensis M'Int. . . . South Orkneys.

Falkland Islands.

Nereis eugenic (Kbg.). .... Falkland Islands.
Nereis mirabilis Kbg. .... Brazilian coast.
Nereis pelagica Lin. .... Cape of Good Hope.
Nereis falldandica n. sp. . . Falkland Islands.

Nereis australis (Schmarda). . . . Falkland Islands.

Cape of Good Hope.
Nereis dumerilii Aud. et Edw., subsp.

nov. comata. ..... North Atlantic (mid-ocean).

Nereis kerguelensis M'Int.
Nereis lienjvdensits M'Intosli (10), p. 225, pi. xxxv. figs. 10-12.

Station 325. Scotia Bay, South Orkneys ; many specimens.

Station 118. One small specimen, Port Stanley, Falkland Islands, January 1903.

This species was the only representative of the Nereidse obtained at the South
Orkneys, where it was apparently fairly common in the littoral zone at Scotia Bay.
Many specimens were collected in 9-10 fathoms ; they are accompanied by tough
membranous tubes in which they lived.

Two specimens from Scotia Bay (season ?) are undergoing change to the heteronereid
form.

It appears to be a peculiarity of this species that the spinigerous setfe of the lower

(HOY. soc. EDIN. TRANS., VOL. L., 42.)

COLLECTED BY THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 211

neuropodial bundle are homogomph in form. This, at any rate, is constant in the
Falkland Islands specimens, as well as in those from the South Orkneys.

N. kerguelensis has a wide distribution in the sub-antarctic regions, having
been recorded from Kerguelen, South Georgia, the Falkland Islands, and the South
Shetlands and Graham Land, besides the Smith Orkneys. Recently VON MARENZELLER
(12, p. 15) has recorded specimens from the eastern Mediterranean, collected by the
Austrian oceanographic vessel Pola, and he mentions also some from the Azores in
the collections of the PRINCE OK MONACO.

Nereis eugenise (Kbg.) Char, emend. Ehl.
Nereis eugenise Ehlers, (3), p. 67, pi. iv. figs. 94-105.

Station 349. Two specimens, dredged in 15 fathoms, sandy bottom, off Tussock
Islands, Falkland Islands, December 2, 1903.

These are both large specimens, and agree in all respects with EHLERS' full descrip-
tion and figures.

This species has been recorded from the shores of various parts of the southern
extremity of South America, but not previously from the Falkland Islands.

Nereis (Ceratonereis) mirabilis Kbg.
Nereis (Ceratonereis) mirabilis, Ehlers, (2), p. 117, pi. xxxvii. figs. 1-6.

Station 81. One specimen, dredged in 36 fathoms, coral bottom, about 200 miles
off the coast of Brazil, 18° 24' S., 37° 58' W., December 20, 1902.

It is a small example, incomplete posteriorly, and showing the characters typical
of the species.

The type specimen of N. mirabilis was obtained off the Brazilian coast in 9° S.
latitude (9, p. 170). The species has subsequently been recorded from Florida
(2, p. 117), the Red Sea (7, p. 172), Porto Rico and Bermuda (13, p. 193), Amboina
(14, p. 33G), and the Persian Gulf (15, p. 392).

Nereis pelagica Lin. (Plate, figs. 1,2.)

Nereis pelagica M'Intosh (11), p. 267, and figs.

Station 478. One example from the shore, Table Bay, South Africa, May 14, 1904.
Station 483. Two examples, trawled in 25 fathoms, sand and kelp, entrance to

Saldauha Bay, South Africa, May 21, 1904.

The three are all small specimens, the largest being 25 mm. long ; fortunately
the proboscis is in all cases everted, and this greatly facilitates examination.

I can find no differences of importance between these and British specimens of
N. pelagica ; the general form of the body, the parapods, seta?, and the prostomiinn
and its appendages all agree, as does the armature of the proboscis, except that the

(ROY. soc. EDIN. TRANS., VOL. u, •;:?.)

212 MR L. N. G. RAMSAY ON POLYCH^TA OF THE FAMILY NEREID^E,

ventral paragnaths of the basal ring (VII. -VIII.) do not exhibit an anterior row of
markedly larger points, such as is usual in examples from Britain (o. figs. 1 and 2).

The noto-cirri are rather shorter than in typical British specimens, but nevertheless
overreach the noto-ligules all along the body.

N. lucipeta Ehl., from Great Fish Bay, some distance farther north, very
closely resembles the present species, as EHLERS himself states (6, p. 71). EHLERS'
specimens were all in the heteronereis condition, and his main grounds for separating
the species from pelagica were : the swollen outline of the noto-cirri of the first seven
pairs of parapods, the number of parapods in the anterior (unaltered) portion of the
body, i.e. 19-20 pairs, and the longer tentacular cirri (the longest reaching to the
9th-l 1th setigerous segment).

I have before me a small example of IV. pelagica, obtained at Plymouth in March
1913, 25 mm. in length, and in the heteronereis condition ; this specimen, as regards
the parapods and their modifications, exactly agrees with EHLERS' description and
figures of N. lucipeta. The change in the form of the parapods occurs at the 21st pair,
and the lst-7th pairs — of the anterior region — have the basal part of the noto-cirrus
swollen to a sausage shape, and terminate in a short, unswollen tip, exactly as figured
by EHLERS for N. lucipeta.

The length of the tentacular cirri is not of much importance, as it is subject to
considerable variation.

While the present record forms an important extension of the known range of
N. pelagica, it is not surprising that the species should be found to occur on the
coasts of South Africa, as it has previously been reported from such widely separated
regions as northern Europe, the northern Pacific (both the Japanese and 'American
shores), and the west coast of South America.

Nereis (Perineris) falklandica, n. sp. (Plate, figs. 3-10.)

Station 118. One specimen, Port Stanley, Falkland Islands.

Station 349. One specimen, dredged off Tussock Islands, Falkland Islands, in
15 fathoms, sandy bottom, December 2, 1903.

These two are in excellent preservation, but one is incomplete posteriorly. The
complete example measures 80 mm. in length by 6 mm. in width over parapods.
There are 93 setigerous segments ; the width hardly decreases till the 80th, behind
which the body rapidly tapers. The other example is smaller.

The anal cirri are very short, as are also the peristomial cirri. In the preserved
animal (in alcohol) the tentacles and peristomial cirri are livid, in marked contrast to
the deeply-pigmented prostomium ; the palps and the basal divisions of the peristomial
cirri are intermediate in colour, or slightly pigmeuted. The eyes are unusually small,
and each is surrounded by a pale ring. In the incomplete specimen the pigment is
more developed than in the larger one, and the dorsal surface of the body-segments

(UOY. SOC. IODIN. TRANS., VOL. L., 44.)

COLLECTED BY THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 213

is dull purplish-brown like the head. In the other the body is practically unpigmented,
although the head agrees with the above description (r. fig. 3).

The proboscis (tigs. 4, 5) was in both cases retracted, but owing to the good

preservation of the specimens, examination by dissection gave very satisfactory results.

The two were found to be identical as to the form and arrangement of the paragnaths.

These, numerous, black and conical, and of very diverse size, are arranged as follows : —

I. An irregular area of numerous very minute black points, with a few larger

ones situated medio-anteriorly.
II. A closely-packed triangular group of about fifteen large paragnaths.

III. A small close-set transverse group of about ten, of fairly large size.

IV. About twenty-five, of varying size, forming a transversely-elongate group.
V. One large median paragnath.

VI. One large paragnath, slightly elongate transversely.

VI I. -VIII. A wide band of very numerous cones, smallest and most numerous in
the mid- ventral region, becoming fewer and larger towards the dorsum and
also towards the anterior margin of the belt, and extending dorsalwards
nearly to VI.

The jaws are nearly black, opaque, with one big " tooth " or notch near the tip, the
remainder of the cutting edge being only slightly crenate (fig. 6).

In the anterior region of the body the lobes of the parapods are short, stout,
rounded, and of more or less even size. Proceeding towards the posterior end, a
gradual transition of form takes place ; the base of the dorsal ligule becomes enlarged
and more enlarged, assuming an oblong form and carrying the cirrus with it ; the
rounded tip of the ligule projecting below the cirrus becomes less prominent. At the
same time the lower lobes of the parapod are reduced in size, especially the ventral
ligule (figs. 8-9). Thus the outline of the parapods changes gradually from a quad-
rangular to a triangular one.

The setae, numerous, short, closely-packed, and of a horny yellow colour, are thus
distributed : —

A. Notopodial bundle : homogomph spinigers.

B. Upper neuropodial bundle : homogomph spinigers (above).

,, ,, heterogomph falcigers (below).

C. Lower ueuropodial bundle : heterogomph falcigers.

The heterogomph spinigers which appear on the upper margin of C in most species
of Nereis have apparently been eliminated. The bristles do not vary appreciably in
form from the 1st parapod to the 70th at least. The falcigers have stout shafts, and
short stout appendages (fig. 10).

In the general external form, and in the characters of the parapodia, seta?, and the
prostomium and its appendages, the present species closely resembles N. vanegata Gru.,
and others of the same group: A', ma/rionii \. et E. (11, p. 295), N. melanocephala
M'Int. (10, p. 21G), y. (I'M'iK/uiiereis) aitomala Grav. (7), N. rujiceps Ehl. (6),

(ROY. soc. KUIN. TRANS., VOL. L., 45.)

214 MR L. N. G. RAMSAY ON POLYCH^TA OF THE FAMILY NEREIDS,

X. gallapagensis Kbg. (8). From all these, however, the armature of the proboscis
exhibits considerable differences. No other species of nereid, to my knowledge, exhibits
a similar arrangement in group L, i.e. very numerous, minute, scattered conical parag-
naths. Again, there is in N. falklundica none of the markedly pectiniform arrange-
ment of the paragnaths of groups II., III., and IV. exhibited by N. variec/ata,
N. anomala, and N. gallapagensis. VII.-VIII. also are distinctive.

This group of species, in common with the rest of the family, stands in considerable
need of revision. In the case of N. variegata, two different types appear to have been
confused under the one name by different authors.

Nereis (Platynereis) australis (Schmarda).

Nereis (Platynereis) australis, Ehlers (5), p. 26, pi. iii. f. 16-20, iv. f. 1-2.
Station 349. One specimen, dredged in 15 fathoms, sandy bottom, off Tussock

Islands, Falkland Islands, December 2, 1903.

Station 118. One specimen, Port Stanley, Falkland Islands, January 1903.
Station 478. One specimen, shore, Table Bay, South Africa, May 1 903.
Station 482. Four examples, in 2-8 fathoms, shells and sand, Houtjes Bay,

Saldanha Bay, South Africa, May 21, 1903.
Station 483. Numerous examples, in 25 fathoms, sand and kelp. Entrance to

Saldanha Bay, South Africa, May 21, 1904.

The Scotia specimens are all small, none exceeding 4 mm. in width over pararodia,
by about 50 mm. in length. None is in the heteronereid state. Some are accompanied
by membranous tubes similar to those of N. dumerilii.

EHLERS (4, p. 104), and BENHAM (l, p. 238), have indicated the wide range of this
species in the southern oceans, and reference should be made to these papers for the
synonymy.

Nereis (Platynereis) dumerilii, And. et Edw., subsp. n. comata. (Plate, figs. 11-13.)
Nereis (Platynereis) dumerilii, M'Intosli (11), p. 302, rind figs.

Station 536. 27° 23' N., 33° 06' W. One example, June 28, 1904.

Station 538. 32° 11' N., 34° 10' W. Two examples, June 30, 1904.

These were taken in mid-Atlantic, with the tow-net, among the floating gulf- weed
of the Sargasso Sea. Those from Station 538 were in typical membranous tubes among
the branches of the weed. They are small specimens, about 20 to 35 mm. long. One
at least is full of ova.

The proboscis was in all cases inverted ; the two larger specimens were therefore
dissected ; the proboscis of one of these, detached, flattened out and mounted in balsam,
shows very clearly an armature of the type usual in N. dumerilii, that is, pectiniform
series of paragnaths in the areas III., IV., VI., and VII.-VIII. The last are practically
uniserial, and extend nearly half-way round the basal ring ; but some short combs, each
of a few teeth, occur in front, representing the more typical biserial arrangement.

(HOY. SOC. EDIN. TRANS., VOL. L., 46.)

COLLECTED BY THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 215

The bristles are of the forms characteristic of N. dumerilii. Two or three homo-
gompb falcigers, with free appendages, occur on the inferior margin of the notopodial
bundle in the more posterior para pods.

The parapods (figs. 12-13) are of normal form, except that the notopodial cirri are
of extraordinary length, and the neuro-setigerous lobe in the round-lobed parapods
(5th— llth pairs) is more prominent than usual.

The noto-cirri are longest about the 15th~'JOth pairs, behind which they grow
somewhat shorter. The ueuro-cirri are also unusually long, distinctly overpassing the
neuro-ligules.

The tentacular cirri are of great length, the longest reaching approximately to the
20th setigerous segment.

On the head there are peculiar markings (fig. 11): a broad ring of dark brown
granular pigment exists on the basal joint of each of the palps, and a band of similar
nature on the forehead surrounds the bases of the tentacles. Otherwise the specimens
—preserved in alcohol — have lost their colour.

The extraordinary elongation of the parapodial and tentacular cirri in these
examples seems to justify their being assigned a distinct sub-specific name. So far
as I am aware, the only other species of nereid which shows a similar development
of cirri is N. mirabilis, a member of quite a different group.

Judging from the position in mid-Atlantic where these worms were collected, one
may suppose they had drifted across from the neighbourhood of the Gulf of Mexico
with the south-west wind drift; but whether this variety of N. dumerilii characterised
by such peculiarly elongate cirri has been evolved solely under the conditions of the
drifting weed of the Sargasso Sea, or whether it is also to be found in the littoral
areas of the seas of Central America, is a matter for future investigation.

EXPLANATION OF PLATE.

Figs. 1 and '2. Nereis pelarj><-«, Lin. Smith Africa.

Fig. 1. Bead and everted proboscis, dorsal view, x 18.

,, 2. Proboscis everted, ventral view. x 18.

Figs. 3 to 10. X. falLlawU'-ii, n. sp. Falkland Islands.

Fig. 3. Head, from above. x 15.

„ 4. Maxillary ring of proboscis, dissected from the ventral side. x 15.

,, 5. liasul ring, dissected from ventral side. x 15.

„ 6. Jaw dissected out, ventral aspect. x 15.

,, 7. 10th parapodium. xi'l.

„ 8. 51st ,, x 21.

9. 70th „ x21.

,, 10. Falcigerous seta, from inferior neuropodial bundle "f loth parapndimn. Zeiss 4 D.

(ROY. soc. EDIX. TRANS., vm.. L., 47.)

216 MR L. N. G. RAMSAY ON POLYCHJETA OF THE FAMILY NEREIDS.

Figs. 11 to 13. N. dumerilii, A. et E., subsp. n. cantata. Sargasso Sea.

Fig. 11. Anterior extremity from above. x 17.
„ 12. 8th parapodium. x 32.
13. Hth x32.

REFERENCES.

(1) BBNHAM, W. B., Report on the Polijch&ta of the Sub-antarctic Mantis of New Zealand, Wellington,
1909.

(2) EHLBRS, E., " Florida- Anneliden," Mem. Mus. Comp. Zool. Harvard, vol. xv., 1887.
(3) Polychaten der Hamburger Magalhaensischen Sammelrcise, Hamburg, 1897.

(4) - —"Die Polychaten des magellanisehen und chilenischen Strandes," Festschr. K. Ges. Wins.
Gottingm, 1901.

(5) "Neuseelandische Anneliden," Abhandl. K. Ges. Wiss. Gottintjen, Matli.-phys. Kl., N.F., iii.,

1904.

(6) " Bodensa'ssigen Anneliden," Wiss. Ergelm. d. Deutschen Tiefsee-Exped., XVI. i., Jena, 1908.

(7) GRAVIER, CH., " Annelides Polychetes de la mer Rouge," Nouo. Arch. Mus. a" hist, nat, Paris,
4° serie, ii. et iii., 1900-01.

(8) - - "Annelides Polychetes recueillies a Payta (Peru)," Arch. Zool. Paris, Ser. 4, x., 1909.

(9) KINBKRG, J. G. H., "Annulata Nova," Ofvers. af K. Sv. Vetensk. Akad. Fiirh., 1865.

(10) M'lNTOSH, W. C., " Polycliata," Sei. Result?, H.M.S. Challenger: Zool, xii., London, 1885.
(11) The British Annelids, II., i. and ii., The Ray Society, London, 1910.

(12) MAHBNZELLBK, E. VON, "Polychaten der Grundes-Exp. S.M.S. ' Pola,'" DenkscJu: K. A kail. Wien,
Ixxiv., 1904.

(13) TEEADWELL, A. L., " Polychajtous Annelids of Porto Rico," Bull. U.S. Fish. Comm., xx. (ii.), 1900.

(14) MALAQUIN, A., et DEHORNE, A., "Annelides Polychetes de la Baie d'Amboine," Revue Suisse Zool.,
xv., 1907.

(15) FAUVEI>, P., "Annelides Polychetes du Golfe Persique," Arch. Zool. Expcr. (5°), vi., 1911.

(ROY. soc. EDIN. TRANS., VOL. L., 48.)

Scot. Nat. Ant. Exp.

i \" XKKKI H.K.

Vol. VII

"*»•»''• •'•'*''• .*V;""

.••.v.->-' • • .,•:•• Vv:

if .\Hy.Vt -•;••

10

'' \ Rr.kinr I HI, Rdll

PART IX.
CH^TOGNATHA.

IX.-THE CH^TOGNATHA OF THE
SCOTTISH NATIONAL ANTARCTIC EXPEDITION,

BY A.. PRINGLE JAMESON, B.Sc.

(WITH CHART IN TEXT.)

IX. — The Chaetognatha of the Scottish National Antarctic Expedition of 1902
1904. By A. Pringle Jameson, B.Sc. ('<nnin<nii<'<tt<><1 //// Dr AV. S. BRUCE.

(MS. received .Tune il, 1913. Read July -2\, 1913. Issued scparat.-h I'Vlmiary 24, 191 I.)

A considerable number of Chsetognatha were collected on the Scottish National
Antarctic Expedition of 1902-1904. Some two or three hundred tow-nettings were
examined, and about ninety yielded some of these forms. Beside the tow-netting
samples there were twelve tubes of these animals taken in the vertical net or
the trawl.

It will thus be seen that most of the material consists of tow-nettings which,
almost without exception, were taken at the surface of the water. If we accept
Dr BRUCE'S definition of the Antarctic Regions,* namely, the area to the south of the
average limit of floating ice according to the British Admiralty Ice Chart, No. 1241
(vide text-map), we may expect Antarctic forms there. Chsetognatha were taken by
the Scotia at ten stations south of this line, including only four out of the many
surface tow-nettings taken by the Scotia south of this line, whilst a fifth, although
taken in latitude 39° 48' S. (Station 468), is in close proximity to it, and actually
within the extreme limit of floating ice. The ten stations are Nos. 273, 280, 325, 34G,
398, 414, 416, 422, 448, 450, and besides the eleventh above mentioned, viz. 468.

The Antarctic species are :—

1. Sagitta. gazellie (Ritter-Zahony).

2. Sagitta maxima (Conant).

3. Sagitta planctonis (Steinhaus).

4. Eukrohnia Inimata (Mobius).

5. Heterokrohnia sp. ? (Ritter-Zahony).

The catches of the vertical net and the trawl were, with one exception — namely, the
trawling made on April 29, 1904, in 39° 48' S., 2° 33' E.— taken south of 40° S. and
south of the average limit of floating ice. Thus this one catch is hardly an exception,
so that practically we may say all these catches belong to the Antarctic series. All the
other Chsetognatha captured were taken north of this line as far north as 36° N. latitude.

The tropical and sub-tropical species are :—

1. S'i</itt(( hexaptera (d'Urbigny).

2. S'i(//t/u 1'njlntx (Grassi).

3. Sni/iltn Hi'n-ii/n<l<'iil<tta (Krohn).

4. Saijiiin bipunctata (Quoy and Gaimard).

5. SIII/I//K rnlmstn (Doncaster).

6. Pterosagitta </rn<-<> (Krohn).

7. Krohnitta snl>tilix (Grassi).

* " Ant.-iri-tii- IJinls," by AVu.MAM S. BRUCE, Kii'"/ •'"'<;•. S.-]iti-niln-r 1, IS'.ll, ].. 208.
(UKriiiNTiou I.-IIOM mi: TRANSACTIONS OK THK KOYAI. MICIKTY <>i- KIIINI<UK<;II, vm,. xi.ix., IT. 1)79-1)89.)

J"

± -*•*

222

MK A. PRINULE JAMKSON ON THE

Thus the collection is divided into two sections, one a general collection from the
colder waters of the South Atlantic and Antarctic Seas, the other a series of surface
tow-nettings made in tropical and sub-tropical Atlantic waters. These two sections

— b SOUTH(~

^mh

y^^v/rr

^mplV/ xY / /
.5 ^%W/ /H—

/Xx / /^ --/-— a

°LE |^^L_f ! | 1

VV" \JRSHAM I/AND / /

^t*AC^

^^'••^SOO/H /O-A^

4rv\ vS^/T^S» r 7/^J

^A/vP\7 /^ Ax^

- ^ / •' l^mK3

90

0

70

\^M\. \""«S-YS

ft*V- y.

^\ / r^tLz

113p ^T \X / ^yy

V7-4J

I- tJreTp ' .„ XS^GeorX

(50

JSQ-1

-P\ "VxOvf

J^JV-A / ^ /^i

50

>it

•— -^fe^j \ ^ / )

\S450 \ \ S V / \

-^\^y^/

«^/ •-, r

...-• \ "~^^\ /\^ /

/^ -

35 ^/ /

,.*. . \ J\- V /\ ,

2s ^L^

±14c^ \ ^/ \ / x

20 10 EAST

$ 10 WEST 20 . 10 *0

Chart showing average limit of Moating ice in the Atlantic Ocean and Stations to the south of which

Chfi'tognatha were obtained.
S = Station. D = Depth in fathoms.

prove on examination to be more than a merely convenient grouping of the methods
of capture — they form two distinct sets of species : the species which occur in the
Antarctic samples not occurring in those from the warmer seas, and those taken in
the warmer waters of the Atlantic never being found in the Antarctic collections.
Doubtless the fact that only a small number of Chsetognatha were obtained from

(BOY. soc. PTDTN. TRANS., VOL. xi.ix., 980.)

CHTETOGNATHA OF TIIK SCOTTISH NATIONAL ANTARCTIC KXPEDITION. 223

these colder waters is responsible for this very sharp division, and it must also be
remembered that if collections had been made from the depths of the tropical and
sub-tropical Atlantic these would naturally have yielded cold-water forms. Yet,
despite these facts, this sharp line dividing the present collection into two distinct
groups serves as a very clear example of how great a part temperature plays in the
distribution of plankton organisms.

In the Scotia collections made south of 40° S. there are found five species of
Chsetognatha — - Smjitta gazelles, Sagitta maxima, S/n/itta /i/am-foHix, fiiikrohnia
/Htina/'t, and possibly a species of J/i-h-rnkruJuiia. These are all typically cold-water
forms ; their presence in these latitudes was to be expected, and adds nothing to our
knowledge of their distribution, all these species having been found by the German
South Polar Expedition (1901-03) under very much the same conditions. Eukrolmi«
foirleri is a type which one might have expected to meet with, hut it was not present
in this collection. Sugifta lir.i'x/itfra, Sagitta, serratodentata, and Krohnitta sulti/i*
have all been reported from the seas south of 40° S., but they do not seem to be typical
Antarctic species, only occasional specimens being got. None of these was present in
the Scotia material from this area.

The remaining species, seven in number, were all taken between roughly 36° N.
lat. and 38° S. lat., and were thus living under tropical and sub-tropical conditions.
.Many of the specimens were small and evidently immature, thus supporting
GUNTHKR'S view that large, mature individuals occur at some distance below the
surface, only small and mostly immature specimens being obtained from the very
uppermost layers of the sea.

No species new to science was discovered, the only noteworthy feature of the
collection being the very large size of four specimens of Sagitta gazelhi'. These four
specimens seem to be the largest recorded for the species, and the largest of all
is probably the largest " Arrow Worm " that has been captured up to the present
time. Detailed measurements of this specimen are given under the notes of this
species.

The difficulty of identifying Chsetognatha has within recent years been considerably
lessened by the excellent papers of FOWLER and RITTEK-ZAHONY on this class. These
have been of the greatest service, and I desire to express my great obligation to these
two authors. In selecting features for aid in identification FOWLER has been followed
in his introduction to the Chsetognatha of the Biscayan Plankton (1905), where a most
helpful discussion of this difficult subject is given. As a rule, tables of measurements
and numbers of jaws and teeth have been given only when the species has not been
frequently described, or where there might be any doubt as to the identification. It
seemed to be fruitless to expand these notes unduly by including figures which had
been given lYc(|ncn! ly before, and which were <|iiite typical and did not add to our
knowledge of the species in any way. It may be mentioned that the measurement of
length dues not include the tail (in ; as this was so frequently damaged, it seemed to be

(ROY. SOC. KPIN. TRANS.. Vol.. XI.IX., 9s].)

224 MR A. PKINOLE JAMKSON ON THE

more accurate to omit this altogether, and indeed it seems to be the custom of many
investigators not to include this fin in their measurements.

As alcohol was the fluid in which the specimens were usually preserved, the identi-
fication has in many cases been attended by some considerable difficulty, for, as is
now well known, the only successful method of preserving Chsetognatha is by using
formalin, as alcohol causes these animals to shrink very greatly and become opaque.*

Sagitta he.ca.ptcra (d'Orbigny).

This very well-marked species was taken at fifteen stations in the warmer Atlantic
waters. It seems to be undoubtedly a tropical and sub- tropical species, as pointed out
by RITTER-ZAHONY (19 LI), no specimens occurring in the Antarctic collections.
Probably the forms of this species reported from the Antarctic are to be referred in
great part to the somewhat nearly related species S. gazellse.

Horizontal distribution : 22° 19' N., 22° 07' W.-150 54' S., 4° 59' W.

Vertical distribution : surface.

Temperature range : 72-2°— 80'5° F.

Stations: 12, 13, 14, 17, 19, 36, 54, 56, 59, 60, 62, 69, 498, 501.

Sagitta gazellae (Ritter-Zahony).

This was the most abundant Antarctic form found ; over a dozen easily identified
specimens were got, and a considerable number of fragmentary or badly distorted forms
are probably to be referred to this species.

It seems somewhat strange that this species, which is well marked and apparently
abundant, should have been recorded so rarely. The species was first established by
RITTER-ZAHONY in 1909 from the material of the Gazelle expedition, and it was also
found very abundantly in the collections made by the German South Polar Expedition
(1901-1903). Some of the forms described by FOWLER (1907) as S. hexaptera, from
the material of the British Antarctic Expedition (1901-1904), and others unnamed,
from New Zealand (FowLER, 1908), have been assigned to this species by EITTER-
ZAHONY (1911).

The most characteristic features of this species are undoubtedly those which suffer
most in preservation. Among the smaller forms the tail-percentage and the number of
teeth and jaws are sometimes very similar to those of young <S. hexaptera, consequently
immature specimens of the two species are very likely to be confused. With larger
forms these features become extremely characteristic for both species, and all likelihood
of confusion disappears. The diagnostic features which seem to be of most assistance
in identifying the species are the following : — The tail-percentage is characteristically
low, about 12-14 per cent, for the larger specimens; the anterior fin is comparatively

* Alcohol was usud as ;i preserving niu'liiim mi account of its suitaliility over formalin for the majority of
organisms in the plankton.- W. S. I!.

(HOY. sue. KIIIX. TKANS., VOL. XI, ix., !)82.)

CHvETOGNATHA OF THE SCOTTISH NATIONAL ANTARCTIC KX H-MTK >N. -_"J5

large, very much longer in proportion than that of S. hexaptera, and almost touching
the posterior fin ; the anterior teeth are small, somewhat conical in shape, ami overlap
one another, quite unlike the long divergent teeth of S. In :r,ij>/, ,•</ ; the teeth are few
in number in young forms and more numerous in older ones, the reverse being true with
S. hexaptera; the anus is situated well in front of the tail septum. In the Scotia
specimens the position of the seminal vesicles is a little different from that described
by KITTKK-ZAHONY ; they hardly extend as far forward as the posterior fin, and the tail
fin is only slightly removed from them.

Some of the specimens in this collection are noteworthy on account of their very
large size ; indeed, one of them seems to be the largest " Arrow Worm " that has been
taken up to the present time. Five specimens were over 70 mm. in length, the largest
of all being no less than 90 mm. long. Some measurement of this relatively enormous
animal may help to give an idea of its size. The length exclusive of the tail fin is
88 mm., the tail fin being at least ~2 mm. long, but as it is damaged rather badly it was
very probably somewhat longer than this. In any case the total length of the animal
is certainly not less than 90 mm. The tail portion, exclusive of the fin, is 11 mm.
The size of the head is of little importance, as this varies so much according to the
state of contraction, but it may be mentioned that in this specimen the head is 4 mm.
long and 6 mm. broad. The broadest part of the body is 8 mm. in width, and this
point is in the region of the ventral ganglion, 22 mm. from the tip of the head. From
this point the body tapers gradually towards the tail and somewhat more abruptly
towards the head, the body being .'! mm. in width at the tail septum and 4 '5 mm. at

Length.

Tail-
percentage.

Jaws.

Ant. Teeth.

1'ust. Te.-tli.

88

12-5

5

5

?

80

12-5

6

7

9

79

12-6

6

5

?

78

12-9

5

7

9

72

12-5

6

7

5 +

65

11 '5

6

9

Id

00

14

5

8

9

59

13

5 +

5 +

?

58

1 15-5

7711

51

13

8 5

9

50

14

8

y

9

47

12-8

7 7 9

42

13

7 4 s

in

12

10 6

33

15

7

1

5

the neck. The anterior fin commences 5 mm. behind the posterior end of the ventral
ganglion — the ganglion itself is 2 mm. long — and extends backwards for about 29 mm. ;
there is a space of about 5 mm. between the anterior and the posterior fins; the
posterior fin is about 17 mm. long, and lies almost entirely on the trunk portion of the

(ROY. SOU. BDIN. TRANS., VOI,. XI. IX., 983.)

2-26

MR A. PRINGLE JAMESON ON THE

body, there being only 3 mm. of the fin on the tail. As the fins are rather badly
damaged it is impossible to be quite certain of the accuracy of these measurements, and
no idea of the breadth of the fins can be formed. The jaws are 3 mm. long. The
reproductive organs are prominent, but they do not seem to be quite mature ; the
ovaries extend forward for a length of 16 mm.

Horizontal distribution: 39° 48' S., 2° 33' B. (depth 2772 fathoms), Station 468.
71° 22' S., 18° 15' W. (depth 2370 fathoms), Station 416.

Vertical distribution : surface (3 fathoms) — 2772 fathoms.

Temperature range: 31°-41°P8 F. (only 3 records).

Most of the specimens were obtained from deep water ranging in depth from
600-2772 fathoms. In Scotia Bay, South Orkneys, Station 325, one medium-sized
specimen was obtained in a tow-net at 3-10 fathoms; while on Burdwood Bank,
Station 346, another medium-sized specimen was obtained from 56 fathoms. These
are the only "surface-water" records; all the rest, as far as data are given, are from
deep water, and with one exception were from south of 50° S.

Sagitta maxima (Conant).

Two specimens which seem to be referable to this species were obtained, one in the
trawl, the other in a vertical net. The tail-percentage and the number of jaws are a
little lower than those usually given for this species, but otherwise these two specimens
agree very closely with the descriptions of S. maxima. The somewhat widely
separated, slightly curved anterior teeth, the prominent vestibular ridge with large
mammiform papillae, the fins without fin rays in the inner portions, and the very
considerable distance between the anus and the tail septum, all seem to be minor
features which taken together are characteristic of the species and help considerably
in its identification.

Length.

Tail-
percentage.

Jaws.

Ant. Teeth.

Tost. Teeth.

53

18

5

13 +

6

39

21

5

6

7

The vertical plankton-net haul was made at Station 414, 71° 50' S., 23° 30' W., and
the net was lowered to a depth of 1000 fathoms and raised to the surface. The
dredged specimen was obtained from Station 450, 48° 00' S., 9° 50' W. , at a depth of
1332 fathoms.

enfata (Grassi).

This very typical tropical and sub-tropical species was obtained from twenty-nine
stations distributed over the warmer parts of the Atlantic. It was rather interesting

(ROY. soc. EDIN. TRANS., VOL. XLIX., 984.)

CII^ETOGNATHA o|' Tllli SCOTTISH NATIONAL ANTARCTIC I X 1'EDITION. 2'J7

to note that, although this animal has a very delicate and rather flaccid bndv. it sccius
to suffer from the effects of preserving fluids less than most of the Cha'tognatha.

The specimens were all very typical and call for no comment.

Horizontal distribution : 2(5° 23' N.; 20° 20' W.-300 5' S., 45° 28' W.

Vertical distribution : surface.

Temperature range : 72°-8r5° F.

Stations: 7, 12, 15, 18, 19, 21, 25, 26, 27, 29, 32, 33, 35, 36, 37, 39, 44, 46, 47,
64, 66,. 73, 86, 93, 506, 515, 525, 526.

serratodentata (Krohn).

The above widely distributed and well-marked species was found in fifty-nine tow-
nettings. It was thus present in more samples than any other species, and, in
addition, it was very frequently more abundant than the other species in the haul ;
indeed, the only species which was ever as numerous as this was S. bipunctata, but
this last was only occasionally obtained in considerable numbers, so that on the whole
S. serratodentata was the form most typically found in the surface tow-nettings.
It is to be noted, however, that no specimens of this species were taken south of 40° S.,
although it has been found by other expeditions as far south as Magellan Straits and
the Falkland Islands.

Horizontal distribution : 26° 23' N., 20° 20' W.-38" 6' S., 14° 32' E.

Vertical distribution : surface.

Temperature range : 64'9°-8r5° F.

Stations: 7, 11, 12, 13, 14, 15, 17, 18, 19, 20. 21, 22, 25, 26, 27, 28, 29, 31, 32,
33, 35, 36, 37, 38, 39, 41, 44, 46, 47, 49, 54, 55, 56, 58, 59, 60, 61, 62, 63, 64, 71,
87, 88, 90, 93, 94, 95, 473, 476, 498, 501, 502, 504, 506, 515, 519, 524, 526, 529.

Sagitta bi/mnctata (Quoy and Gaimard).

This common species was fairly frequently met with, having been found in fifty-four
samples. On one occasion only was it taken in very considerable numbers ; this was
at a point eight miles from Cape Peninsula, South Africa, where a surface haul of
some ten minutes' duration captured several hundred specimens. As a rule,
however, the specimens taken in any individual haul were not numerous, and were
rather small in size. It has been found impossible to determine exactly what stations
might be called coastwise, so that this collection can throw but little light on the
question put forward by STEINHAUS (1896) as to whether this is mainly a neritic form
or not. Sagitta bi/»ni<-ti<f.a seems certainly to be found under oceanic conditions ; but
it may be noted that in the only haul mentioned as being taken very near land — the
one above — this species appeared to the exclusion of all others, and the specimens
captured were larger than most of those taken at other stations. As far as can be
judged, also, the members of this species caught on the high seas were usually rather
small in size and immature.

(HOY. soc. EDIN. TRANS., VOL. xi.ix.. iis.'i.)

228

MR A. PRINGLE JAMESON ON THE

Horizontal distribution : 36° 5' N., 30° 50' W.-340 43' S., 17° 15' E.

Vertical distribution : surface.

Temperature range : 6r3°-8 1 '5° F.

Stations: 7, 8, 13, 17, 18, 20, 21, 22, 25, 26, 27, 28, 30, 31, 35, 37, 38, 39, 41,
42, 44, 47, 49, 59, 65, 66, 69, 70, 72, 82, 85, 86, 87, 90, 91, 93, 94, 95, 476, 477,
498, 501, 502, 506, 510, 511, 515, 522, 524, 525, 526, 529, 533, 540.

Sagitta robusta (Doncaster).

Considerable difficulty was experienced in identifying this species, and even yet
the results do not seem to be quite satisfactory, for young forms of Sagitta robusta
are rather liable to be confused with Sagitta bipunctata on account of a certain
correspondence between the tail-percentages and the numbers of the jaws and teeth
of these two species. This confusion, of course, does not arise when the specimens are
in good order, and when all the features are clearly seen ; but damaged and strongly
contracted specimens of N. robusta are frequently very like S. bipunctata, and require
some care in determination. The features on which I have relied for the identifica-
tion of this species have been the small number of the jaws, never more than seven
— RITTER-ZAHONY (1911) in his monograph admits species with eight jaws, but under
the circumstances I have thought it best to fix seven as the typical number ; the
presence of a distinct collarette ; and an opaque body, all the parts of which, including
the teeth and the jaws, are of a stout build. An important feature would seem to be
the position of the seminal vesicles near or touching the posterior tins, but unfortunately
seminal vesicles were not observed in any of the specimens.

Length.

Tail-
percentage.

Jaws.

Ant. Teeth.

Post. Teeth.

? 14

?

7

8

12

13-5

26

7

7

? 10

12

25

7

6 or 7

12

11

27

1

7

12

S

28

7

6

13

Horizontal distribution : 32° 11' N., 34° 10' W.-230 8' S., 39° 40' W.

Vertical distribution : surface.

Temperature range : 74C-S1'5° F.

Stations : 15, 18, 11), 26, 29, 33, 39, 60, 73, 85, 526, 538.

It may be noted that .S. rubusta is said to have a smaller temperature range than
S. bipunctata, or at all events that it does not range so far to the north and south.
So far as my results go they bear out this statement.

(ROY. SOC. BDIN. TRANS., VOL. XLIX., 986.)

CH^-ETOGNATHA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 229

Sagitta planctonis (Steinhaus).

Tliree specimens, taken at one station in the Antarctic Ocean, are probably to be
referred to this species. The tail-percentage and the numbers of the teeth and jaws
agree with those given by RITTER-ZAHONY, and the general habit — the stiff opaque
body — seems to be the same. A very characteristic feature of these specimens is the
size and arrangement of the anterior teeth : these are long towards the inner end and
considerably shorter towards the outer end of each row ; they do not overlap, but each
tooth is distinctly seen in every part ; these features are all admirably seen in a figure
given by RITTER-ZAHONY (1911). As the specimens are small and young, the collarette
is only slightly developed.

Length.

Tail-
percentage.

Jaws.

Ant. Teeth.

Post. Teeth.

12

25

9

5

10

8

25

10

4

8

This species is given by FOWLER (1906) as a tropical surface form, but it was
repeatedly taken in the Antarctic on the German South Polar Expedition, even as far
south as 66° S. (RITTER-ZAHONY, 1911), so that its occurrence in the Scotia Antarctic
collections is not an isolated case. Its furthest south range is, however, extended by
about three degrees.

Station : 280, 68° 40' S., 30° 18' W. Between surface and 500 fathoms.

Temperature: 32'65° F.

Pterosagitta draco (Krohn).

Only a few specimens of this apparently common and typical tropical and sub-
tropical form were obtained at seven stations. This scarcity is probably due to the
fact that P. draco does not commonly occur at the very surface of the sea, but usually
at some little depth. This point would seem to be in agreement with the results of the
German South Polar Expedition (RiTTKR-ZAHONY, 1911), where this species was got
only twice in the surface tow-nets, whereas it frequently occurred in the vertical nets.

Horizontal distribution : 26° 23' N., 20° 20' W.-388 6' S., 14° 32' E.

Vertical distribution : surface.

Temperature range : 64'9°-78'6° F.

Stations: 7, 11, 12, 14, 26, 56, 473.

Eukrohnia hamata (Mobius).

It might have been expected that this form would have been taken very frequently
in the higher latitudes, but, strange to say, it has been found in only six samples. The
specimens found were in every way typical.

(ROY. soc. EDIN. TRANS., vol.. SLIX., 987.)

230

MR A. PRINGLE JAMESON ON THE

Horizontal distribution : 49° 25' S., 9° 21' W.-690 22' S., 26° 36' W.

Vertical distribution : 800 fathoms-surface.

Temperature range : 28'9°-39° F.

Stations : 273, 280, 448, 325 (2) vertical net 23/3/04, 422.

? Heterokrohnia sp. (Kitter-Zahony).

A single specimen taken in a vertical -net haul might possibly belong to this genus ;
the state of preservation, however, will not admit of any certain determination.

The body is slender and firm. Transverse muscles are developed both in the trunk
and in the tail. The lateral fins are almost totally destroyed, but there does not seem
to have been more than one pair. The tail fin is also much damaged, but it appears
to have been somewhat rounded and is very full of fin-rays. The percentage of tail
to body length is probably about 25 '6; the lowest percentage given by RITTEK-
ZAHONY (1911) is 32'3. The head is not well expanded, so that the teeth and anterior
portion are not well seen. There are certainly not fewer than 14 posterior teeth.
There are at least 1 1 jaws, rather long, slender, and knife-like, and only slightly curved,
in general somewhat like a scythe blade.

On the whole, these features are those given by PaiTER-ZAHONY in his diagnosis of
this new genus Heterokrohnia. Indeed, the presence of transverse muscles in both
trunk and tail would seem to be enough to determine the genus. It is, however,
impossible to determine the species. The only species described is H. mirabilis,
characterised by a high tail-percentage, and very numerous anterior and posterior
teeth. This specimen does not show these features clearly enough to warrant any
statement being made as to the species.

Length.

Tail-
percentage.

Jaws.

1
Ant. Teeth. ' Post. Teeth.

22-5

25-6

11

? ? 14

Taken in a vertical net on February 29, 1904, at Station 398, 68° 25' S.,
27° 10' W. Surface temperature 30° F., and temperature at 1000 fathoms probably
about 32° F.

Krohnitta subtilis (Grassi).

Specimens of this species were not abundant ; it was found in only seven samples,
and there only in small numbers. The explanation of this apparent rarity is probably
to be found in the fact, pointed out by FOWLER (1906), that this is a species which has
its habitat in cooler waters of the "lower epiplankton and upper mesoplankton,"
seeming to avoid " the warmer surface water." One would naturally expect that such
a form would be taken only rarely in surface tow-nettings. Further, the specimens
obtained were all very evidently young forms, the largest being 5'5 mm. in length.

(ROY. soc. EDIN. TRANS., VOL. XLIX., 988.)

CH^TOGNATHA OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 231

FOWLER (1906) gives the most southerly point at which K. subtilis had been taken
in the Atlantic up to that date as 25° 39' S., 3G° 51' \V., but since then RITTKR-
ZAKONY (1911) reports one specimen taken as far south as 60° 8.; the furthest south
record for the Scotia collection is 30° 25' S., 45° 45' \V.

Horizontal distril.ution : 23° 50' N., 21° 34' W.-300 25' S., 45° 45' W.

Vertical distribution : surface.

Temperature range : 73°-81° F.

Stations: 11, 17, 32, 33, 38, 94, 519.

LITERATURE.

Full bibliographies are given by FOWLER (1906) and RITTER-ZAHONY (1911). The fullowing have
mentioned specially in the preceding text :—
FOWLER, G. H., "Biscayan Plankton collected during a Cruise of H.M.S. Research, 1'JOO. l-'t. III.: The

Uhfetognatha," Trans. Linn. Sue. London, ser. 2, Zoology, vol. x., 1905.
— "The Chsetognatha of the Sibotja Expedition," " Siboga" Expe.il., xxi., Leiden, 1906.

- "Chsetognatha," National Antarrfi,- K.i-/i/>,lifit>n, 1901-1904: Natural, History, vol. iii., London, 1907.
RITTER-ZAHONV, R. v., "Die Chaetognathen der Gazelle Expedition," Zool An?.., lid. xxxiv., 1909.

- "Revision der (..'liiitognathen," Di'utsrhe Hiidpolar-Expedition, 1901-1903: Zoologie, lid. v., lierliii, 1911.

- "Chaetognathi," Das Tierreich, Lief, xxix., Berlin, 1911.

(Hoy. S"0. K.DIN. TRANS.. VOL. XI. IX., 989.)

PART X.
SEA-ANEMONES.

(tiKNl'S PoRPONIA AND RELATED GENERA.)

X.-THE ACTINIA

(GENUS PORPONIA AND RELATED GENERA)

OF THE SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

\

BY OSKAR CARLGREN,

Professor in the Zoological Institution of the University of Lunrl.

(WITH ONK PLATE AND EIGHT TEXT-FIG' HUES.)

X. — On the Genus Porponia and Related Genera, Scottish National Antarctic
Expedition. By Professor Oskar Carlgren, Universitetets Zoologiska Institution,
Lund. Communicated by Dr W. S. BRUCE. (With One Plate and Text-
figures.)

(MS. received August 12, 1913. Read January 19, 1914. Issued srpurati-ly March 30, 1914.)

In an Appendix to the Actiniae of the Challenger Expedition, Pi. HERTWIG. 1882,
described a peculiar genus, Porponia, with two species, P. elongata and P. robusta,
which he characterises in the following manner : " Actiniarien (Hexactinien ?) mit 1
Schlundrinnen ohne Ringmuskel, mit diinnwandigen Tentakeln, deren Basen auf der
iiusseren Seite durch span gen formige Verlangerungen des Mauerblatts gestiitzt werden."
Partly owing to the badly preserved material, however, he did not venture to indicate
definitely its systematic position, though he considered it conceivable that it repre-
sented a transitional form between the Zoanthidae and the true Actiniae, or Hexactiniae.
HERTWIG expresses the following opinion regarding the systematic position of the
genus : " Die doppelreihige Stellung der Tentakel, die Abwesenheit vollstandiger
Geschlechtsepten (Macrosepten) und unvollstandiger, sterilen Septen (Microsepten)
sind Merkmale, welche an die Zoantheen erinnern, die Zahlen der Tentakel und der
Septen passen ebenfalls am meisten fur diese Gruppe, da sie weder von dem Numerus 6
wie bei den Hexactinien noch von dem Numerus 4 wie bei den Paractinien bestimmt sind.
Auf der anderen Seite nahert sich die P. elongata durch den Besitz von zwei Schlund-
rinnen wieder mehr den Hexactinien, unter denen sie am meisten mit den Antheo-
morphiden iibereinstimmt. Ich halte es daher fiir sehr wahrscheinlich, class P. elongata
eine Mittelform ist, welche den Ubergang von den Hexactinien zu den Zoantheen
bildet." Porponia possibly, he thinks, belongs to the Antheomorphidte, a family
supposed to be separated from the family Antheadae chiefly by the absence of a
sphincter and by the weak development of the musculature.

Since R. HERTWIG described this genus, it has not been made the subject of any
close examination, nor for this reason has its systematic position been discussed in
detail. Yet it is only right to mention that M'MuRRlCH was inclined to refer the
genus Halcurias (Endoccelactis) to the neighbourhood of Porponia. " In fact," he
writes (p. 226, 1898), " 1 was inclined at first to associate it (Ilalcnrias) with Porponia,
and was only deterred from doing so by the simplicity of the arrangement of the
mesenteries." (That H<tl<-uri<ts possesses a peculiar arrangement of the mesenteries,
which agrees with what I have described, 1897, for the genus Endocaelactis, was not
known to M'MuRRiCH at that time.)

A closer examination of the material which I received for investigation from the
(REPRINTED KKU.M THE TRANSACTIONS OK THK ROYAL SOCIETY OK KDINBDRQH, vol.. L., PP. 49-71.)

242 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

Scotia Expedition has proved, however, that the arrangement of the mesenteries, as
also indeed a number of other characters, indicates a close relationship between
Halcurias (Endoccelactis) and Porponia, though each genus has its own distinctive
characteristics. They must of necessity be referred to the same family, though this
cannot be the Antheomorphidse set up by HERTWIG, with its genera Antheomorphe and
llyanthopsis. As the following description will show, I have no hesitation in retaining
the family set up by me, Endoccelactidse, for Halcurias and Porponia.

After describing in detail the Porponia species of the Scotia Expedition, I shall
discuss the mutual relationship of Halcurias and Porponia, and the position of the
family Endoccelactidse in the classification. At the same time I shall take the
opportunity to discuss the genera Antheomorphe and Hyanihopsis, as also the family
Antheomorphidee.

I. THE STRUCTURE OF PORPONIA ANTARCTICA, N. SP.

Place of discovery.— OS Coats Land, 71° 22' S., 16° 34' W., 1410 fathoms, 16th
March 1904. 17 specimens.

Dimensions of the largest specimens. — Length 6-8 cm., breadth of the foot 3-4
cm., breadth of the disc 8-9 cm., length of the inner tentacles 3-7 cm.

External appearance. — The fresh colour is creamy white, tinged, especially on the
tentacles, with pale lavender. The base is expanded and often attached to a round
stone, which it more or less encloses ; it is arranged in coarse, irregular folds, and
secretes a fairly extensive cuticle.

The body-wall is more or less beaker-shaped, arising from the less or greater
contraction of the individuals. The distal part of the animal is thus wider than the
proximal, and that often to a fairly considerable extent. The thick body-wall is
provided with irregular longitudinal and transverse furrows, by means of which it is
divided into irregular areas, as a rule very prominent, since the thin ectoderm has
fallen off from almost all the specimens. The distal, uneven edge of the body-wall is
not marked off by any definite groove or line, but passes irregularly into the bases of
the tentacles. In large specimens the tentacles are typically 68 * in number, yet this
may be exceeded, as the arrangement of the tentacles and even the grouping of the
mesenteries may be somewhat irregular in one quadrant (or several ?), resulting in a
somewhat greater development of tentacles here than in the other quadrants. The
tentacles have the appearance characteristic for Porponia ; on the outer side they are
greatly thickened and like cartilage ; towards the oral disc, on the inner side, they are
thin and resemble ordinary tentacles. Owing to the mesoglcea being greatly thickened
on the outer sides of the tentacles, it looks as if the tentacles here were provided with
bridge-like outshoots from the body-wall. Further, the tentacles are conical, curved

* The normal number of tentacles seems to be developed in the species at a comparatively early stage, as specimens
of only half the size of the largest in the collection have already the typical number of tentacles. A small specimen,
on the other hand, had considerably fewer tentacles ; but as it was much damaged, I have not tried to ascertain exactly
the number of tentacles or their arrangement.

(ROY. sue. EDIN. TRANS., VOL. L., 50.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 243

like a sabre towards the oral disc, thick in the basal part but gradually narrowing
towards the tip, and of moderate length. Their arrangement is irregular, and recalls

Fio. IA.

FlO. In.
Flo. 1. — Poi-ponia anlarelicn ; A from the side, H from the disc.

the condition in Halcurias (Endoccelactis). In the latter genus the true arrangement
is difficult to find, and this is even more the case in Porponia owing to the greatly
swollen bases of the tentacles. So far as I could find, the arrangement of the tentacles

(ROY. sot1. KPIX. TRANS., VOL. i,., 51.)

244 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

on each side of the sagittal plane is as follows : 1 (dt.), 2, 1, 4, 3, 5, 3, 4, 1, 2, 1, 4, 3,
5, 3, 4, I, 2, 1, 4, 3, 5, 3, 4, 1, 2, 1, 4, 3, 5, 3, 4, 1, 2 = 34 (dt. = directive tentacle).
Altogether, therefore, there should be 18 tentacles of the first order, 10 tentacles of the
second, 16 of the third, 16 of the fourth, and 8 of the fifth. But it has to be remarked
that some groups of tentacles of the first and second orders, namely, the 5 (1, 2, 1, 2,
l) in the sagittal plane on both sides of the angles of the mouth, and the 3(1,2, l) on
both sides in the transverse plane, arise somewhat further in than the other tentacles
of the first and second order. The tentacles of the first order should therefore, perhaps,
best be divided into 2 circlets (10 + 8), and similarly those of the second order into
2 (6 + 4). The underlined tentacles of the fourth order stand somewhat further out
than the others of the fourth order. The arrangement of the tentacles is thus 18
(10 + 8) +10 (6 + 4) +16 + 16 (8 + 8) + 8 = 68 (text-fig. 2). As in Halcurias, the
peculiar arrangement of the tentacles is connected with a characteristic dislocation of
the mesenteries.

The oral disc is wide, expanded, marked by distinct radial furrows corresponding to
the insertions of the mesenteries. The oesophagus is wide, oval, long, and provided
with longitudinal furrows to a number of about 18. There are 2 resophageal grooves,
lying typically in the angles of the mouth, and broader in the aboral than in the oral
part. No hyposulcus is developed.

Anatomical structure. — The ectoderm of the base is high, and consists chiefly of
supporting cells, which secrete a fairly thick cuticle. The mesogloea is extensive, as
also the entoderm, in which the nervous system seems to be well developed.

The ectoderm of the body-wall by comparison with the thick mesogloea is thin and
provided with numerous spirocysts of varying size up to 60 M long and 11 M broad.
Further, there are generally thick-walled capsules (length 26-34 M, breadth 3 M). In
cross-section there seem to be thickenings at the base which resemble transversely cut
muscles, but are probably in reality thickened basal parts of the ectodermal cells
(see below under Ilyanthopsis elegans). The mesogloea is thick, and provided with
numerous small cells with outshoots. The entodermal musculature is weak, and no
sphincter is present. The entoderm is thin like the ectoderm. The entoderm of the
body-wall and of the oesophagus is pigmented.

The ectoderm of the tentacles is somewhat high, and contains fairly numerous
spirocysts and thick-walled nematocysts (length 36-50 M, breadth 3-(5) M). The
longitudinal musculature on the outer side of the tentacles is weak, but gradually
becomes considerably stronger towards the inner side, so that the muscular wall is here
about half the height of the supporting cells. The mesogloea agrees in structure with
that of the body-wall. On the inner side, in the lower part of the tentacles, it is, as a
rule, just as thick as, or thicker than the ectoderm ; on the outer side, however, it is much
thickened (Plate, fig. 6). The thickness decreases towards the tips of the tentacles
(Plate, fig. 5), so that at the ends the mesoglrea is almost equally developed round the
tentacles. The entoderm is thin, and the ring-musculature weak. The radial muscula-

(ROY. SOO. EDIN. TRANS., VOL. L., 52.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 245

ture of the disc is in the ridges almost as well developed as the longitudinal musculature
on the inner side of the tentacles, but in the furrows it is considerably feebler.

The ectoderm of the resophagus shows the usual structure, but in addition numerous
thick-walled nematocysts (length 43-48 M, breadth 5 /«), and in fair numbers spirocysts
of the same appearance as in the body-wall. The mesoglcea, which is very thick, agrees
in structure with that of the body-wall.

The arrangement of the mesenteries is shown by the accompanying schematic
fig. 2 ; 6 pairs of mesenteries of the first order, with 2 pairs of symmetrical, direc-
tive mesenteries are present. While all the primary exocoels are reduced, so that no
mesenteries could be formed in them, eacli of the 4 lateral endocoels contains a pair
of mesenteries of the second order, in which, however, the longitudinal muscles face
outwards, as in the directive mesenteries. The mesenteries of the first and second
order, all of which are perfect, are thus 20 altogether; of these 16 are grouped
in 8 pairs with longitudinal muscles as the directive mesenteries, owing to the develop-
ment of the mesenteries of the third order in the secondary endoccels, and the last 4
mesenteries are unpaired at the sides of the directive mesenteries. The arrangement of
the 20 oldest mesenteries thus agrees completely with the corresponding condition in
Halcurias. In the endoccels of the second order we find mesenteries of the third,
fourth, and fifth order. The mesenteries of the third order (8 pairs) are perfect, and
have longitudinal muscles typical of the Actiniaria (muscles facing inwards). The
mesenteries of the fourth order are unequally developed. On one side of the
mesenteries of the third order, between these and the mesenteries of the second
order, there is an imperfect mesentery of the fourth order, but on the other side a
pair, consisting of one imperfect mesentery, lying nearest the mesenteries of the
third order, and a perfect mesentery. Between the latter and the mesentery of the
first order there is an imperfect mesentery of the fifth order. The arrangement of
the mesenteries is thus : 6 — 4 — 8 — 8 + 8 unpaired — 8 unpaired ; thus in all 26 pairs,
+ 8 unpaired of the fourth and 8 unpaired of the fifth order, or 68 mesenteries.
While this is typical, the development probably proceeds somewhat further in some
ways in very old specimens.

The mesenteries are somewhat thick, owing to the thickness of the mesoglcea. The
longitudinal musculature is fairly well developed, though not so much as in Halcurias.
In the distal part the folds are thick, but grade towards the proximal end to a weak, only
a little condensed, pennon-like region (Plate, figs. 7, 8), which approaches the body-wall
and fuses with the well-developed parietal parts of the longitudinal musculature. The
parietobasilar musculature is narrow but fairly well developed (Plate, figs. 7, 8), and
woes far out towards the distal end. There are no separate basilar muscles, though in
the foot end the muscles pass in a transverse direction, also on the longitudinal muscles'
side of the mesenteries ; but these muscles which run transversely are continuations of
the longitudinal musculature, which at the base of the column bend in a transverse
direction (cf. p. 252, Ilyanthopsis elcgans).

(BOY. soc. EDIN. TRANS., VOL. L., 53.)

246 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

The filaments are well developed and extremely broad, owing to the strong develop-
ment of the mesogloea (Plate, fig. 9). The vacuolar streak is little differentiated. Both
in the intermediate part of the ciliated tract region and in the nematocyst glandular
streak there are sparse spirocysts and fairly numerous thick-walled nematocysts,
especially in the latter (length 36-41, sometimes even 46 M, and about 3 M broad). Of
fairly common occurrence further in the glandular streak are nematocysts with distinct
basal part to the spiral thread (length 37-41 /u, breadth 7 M). Even the entoderm of
the filament is pigmented, especially on the region of the border-streak.

The sexual organs occur on all well-developed mesenteries, even on the directive
mesenteries. The animals are dioecious.

For the Porponia obtained by the Scotia Expedition I have set up a new species,
P. antarctica. Of the species of Porponia already known it comes nearest to
P. robusta, R. Hertwig, both in the form of the body and the appearance of the
tentacles. To set up good characters between these two species is, however, distinctly
difficult, as HERTWIG'S description of P. robusta is so incomplete, and both, this species
as well as the two specimens of P. elongata, do not seem to have the number of
mesenteries and tentacles typical of P. antarctica. It is probable that the mesenteries
of the fifth order have not been laid down in HERTWIG'S form, to judge from the
number of tentacles, which HERTWIG gives to be 54 in P. elongata. According to
some notes made by me in 1897 on revising the Challenger Actiniae, the distribution and
size of the spirocysts and nematocysts in P. robusta and P. elongata were as follows : —

The spirocysts in the body-wall were in P. robusta very numerous and about
40-44 M long, in the tentacles of P. elongata about 56-72 /* ; there were also spirocysts
in the ossophagus of these two species. The nematocysts in the oesophagus were 48 M
long in both P. elongata and P. robusta, in the tentacles of P. elongata 48 M long.
It is, however, noticeable that there were only fragments of the ectoderm in the
Challenger species.

II. ON THE SYSTEMATIC POSITION OF THE GENUS PORPONIA.

According to the anatomical account of the genus Porponia given above, there
should be no doubt remaining that Porponia and Halcurias (Endoccelactis} are very
nearly related to each other. Common to both is the structure of the body-wall, and
also of the oesophagus, both, among other things, being provided with numerous spiro-
cysts. Even the anatomical structure of the filaments and distribution of the sexual
organs show agreement. Most striking, however, is the characteristic and similar
arrangement of the tentacles and mesenteries, which differs from that in all other
known Actiniaria, as can be seen more clearly from the following scheme for the two
genera. In both Halcurias and Porponia the same displacement of the original
mesenteries and tentacles has clearly taken place. After the formation of the first 6 pairs
of mesenteries, 1, 1, etc., in the ordinary way, 2 mesenteries (2, 2), with the longitudinal
muscles faced outwards, have arisen in the lateral endoccels, and these mesenteries

(ROY. soc. EDIN. TRANS., VOL. L., 54.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 247

form normal pairs with neighbouring mesenteries of the first order. At this stage,
therefore, the genera have 10 pairs of mesenteries, 2 of which, lying medially
opposite to each other, are directive mesenteries, whilst the other 8 are formed in
pairs with typically arranged longitudinal muscles. In the 8 new endocoels arisen
through the growth of the mesenteries of the second order, there is a further develop-
ment of 8 normal pairs of the third order, so that at this stage we have 10 (6 + 4) + 8
= 18 pairs of mesenteries. Thereafter the development proceeds in a more normal
manner, the mesenteries of the fourth and fifth order in Porponiq not developing in
the endocoels of the third order, but on both sides of a pair of the third order ; in an
exocoel in relation to the mesenteries of the third order, but in an original endocoel of
the second order. I have endeavoured previously (1897) to show that the develop-
ment proceeds in this way in Etidoccductis, as with fairly great certainty I considered
it possible, partly from the unequal development of the mesenteries at the basis of the
column, to distinguish the first 6 pairs of mesenteries (1, 1, etc.) from the 4 pairs of the
second order, and thence from the obviously great development of these 10 pairs in
comparison with the others to distinguish them clearly from the subsequent orders of
mesenteries. In Porponia I had admittedly not been able to study the development
of the mesenteries so clearly as Endoccelactis, as the difference between the mesenteries
of the said orders were not so distinct as in this latter form, owing to the fact that
in Porponia several more complete mesenteries occur than in Endoccelactis. The
arrangement of tentacles in Porponia, as also the whole arrangement of mesenteries,
indicates, however, that in this genus the mesenteries arise in the same manner, and that
also after reaching a stage with 6 normally placed pairs of mesenteries, a development
of mesenteries in the endocoels has taken place. The groups of tentacles of the first and
second orders, which arise near the directive mesenteries 1, 2, 1 (dt), 2, 1, and those
which lie in the transverse plane 1, 2, 1, are inside the tentacles of the corresponding
order which stand at the other 4 pairs of stronger mesenteries, a condition that is to
some extent indicated on the schematic figure, but which in reality is considerably
greater than the figure shows. This indicates that we have to arrange the first G pairs
of mesenteries by this plane. With regard to the arrangement of the tentacles other-
wise, this is in the main the same in the two genera, great displacements occurring in
the cycles with the development of certain mesenteries in the endocoels, wherewith, so
to spuak, a doubling of the tentacles in the lateral endocoels of the first and second
orders arises. Above all, the arrangement of the 28 innermost tentacles is the same
in both genera, as the figure shows. It is characteristic of both genera, therefore, that
all the mesenteries of the second and third orders develop in the eudoccels, and that
in consequence great displacements in the position of the tentacles take place.

Nevertheless, there are a number of differences in the structure of the two genera.
In a number of less important characters, such as the form of the body — in Porponia
beaker-like, in Salcurtus more cylindric — in the presence of only one cesophageal
groove in Halcurias, whilst Porponia has two, there are certainly differences between

(HOY. soc. EDIN. TUANS., VOL. L., .j.~> )

248

PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

FIG. 2.

FIG. 3.

Fics. 'J and 3. — Schematic sections through the oesophagus region of Purpunia anlardim (fig. 2) and
Iliilrnriiix I'lii-li/ri'iii (lig. 3). The position of the tentacles (the dark rings) is drawn in. The rect-
angular, dark figures on the mesenteries indicate the longitudinal muscles. The shaded lines denote
the directive plane. In Po/ponia the ridges of the oesophagus are not drawn in.

(ROY. sue. KIUN. 'CHANS., VIM,. L., 56.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 249

the genera, but also others more important exist. In the two genera the tentacles are
formed in a different manner, having the usual appearance in Halcurias, whilst they
are greatly thickened in Porponia on the outer side, so that it looks as if the thick
mesoglona of the body-wall extended in the form of bridge-like outshoots into the
tentacles. An important distinction between Porponia and Halcurias is also seen in
the development of the mesenteries. Whilst Halcurias has only 10 pairs of stronger
mesenteries — the mesenteries of the third order being thus but feebly developed, and
that only in the most distal end — there is a much larger number of stronger mesenteries
in Porponia, which are connected with the oesophagus in its whole length or the
greater part of this. Very characteristic also is the fact that the development of the
mesenteries, which appear after the mesenteries of the third order, is different in the
two genera. In Halcurias the mesenteries of the fourth order arise normally, so that
the two mesenteries in the same pair appear simultaneously ; the development then
ceases, though it is possible that in some cases mesenteries of a fifth order may be laid
down. In Porponia, on the other hand, the development has taken a different line,
and comes to resemble that of the later mesenteries in Actmostola and Stompliia. The
mesenteries of the fourth order, namely, do not appear at the same time ; on the one
hand, the one mesentery is developed much earlier than its pair ; on the other, the
development of the mesenteries of the fourth order is delayed in certain regularly
arranged exoccels, so that here only one unpaired mesentery instead of a pair arises.
This displacement of the mesenterial appearance has had the result that an unpaired
mesentery of the fifth order arises on the side where the strongest and perfect mesentery
of the fourth order lies. In fact, the development of the mesenteries of the fourth and
following orders in Porponia comes under the same law as I have indicated for the
development of later mesenteries in Actinostolidfe, and which means that the develop-
ment of a stronger mesentery in a previous order leads to an earlier development
of the mesenteries of the subsequent orders in the area which lies nearest this
stronger mesentery. Finally, it has to be noted that irregularities not rarely arise in a
quadrant of Porponia which shows a somewhat larger number of mesenteries than the
normal. If we indicate the mesenteries with numbers, those of the first order with 1,
of the second with 2, and so on, the irregularity in the arrangement of the mesenteries
on each side of the directive plane can be seen from the following scheme for the two
genera (dm. = directive mesenteries) :—
Halcurias (Endoccelactis).

1 (dm.) 1, 4, 4, 3, 3, 4, 4, 2, 2, 4, 4, 3, 3, 4, 4, I, 1, 4, 4, 3, 3, 4, 4, 2, 2, 4, 4, 3,

3, 4, 4, 1, 1 (dm.) = 34.
Porponia.

1 (dm.) 1, 5, 4, 4, 3, 3, 4, 2, 2, 4, 3, 3, 4, 4, 5, 1, 1, 5, 4, 4, 3, 3, 4, 2, 2, 4, 3, 3,

4, 4, 5, 1, 1 (dm.) = 34.

The arrangement of the tentacles is also somewhat different in the two genera (see
fio-ures). This difference is due entirely to the different arrangement of the mesenteries

(ROY. S<ic. EDIN. TKANS., VOL. I.., 57.)

250 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

of the later developmental stages, yet it has to be noted that the position of the
tentacles, especially in Porponia, is difficult to determine, and it is just possible that
the agreement is somewhat greater than is represented in the schematic figures.

While Halcurias and Porponia are thus closely related to each other, there is still
the question whether they have any near relationship to the genus set up by R. HERTWIG,
Antheomorphe, as M'MuRRicn and HERTWIG maintain, the one for Halcurias, the other
for Porponia. In my opinion, such a relationship does not exist, for Antheomorphe,
according to HERTWIG'S poor description of this genus, appears to have normally arranged
mesenteries and tentacles. On the other hand, it is not impossible that Halcurias and
Porponia are related to the form which WASSILIEFF (1908) has described as Ilyanthopsis
elegans. The abnormal development of the mesenteries in Ilyanthopsis elegans, ac-
cording to the obviously imperfect data given by WASSILIEFF, indicates, namely, the
possibility of a mesenterial arrangement such as in Halcurias and Porponia, and even
the habitus of the animal resembles that of Halcurias. To settle this, I have obtained
a specimen of Ilyanthopsis elegans for investigation from the Conservator of the Bavarian
State's collection in Munich. As WASSILIEFF'S description of the animal leaves much to
be desired,* and as it is by no means so schematically constructed as this author believed,
I give here a description of this species, a description, however, that cannot be considered
complete, as the material, which could not be dissected, was insufficient for the purpose.

The body in Ilyanthopsis elegans is elongated, cylindrical, the base distinctly
flattened, but the boundary between body-wall and base not sharply marked. The
body-wall is provided with irregular transverse and longitudinal furrows, so that the
parts of the wall between the furrows have the appearance of raised areas, which, how-
ever, do not differ in their structure from the remaining part of the body-wall. The
tentacles are of moderate length, 2-2'5 cm. (length of body over 6 cm.), and distinctly
furrowed longitudinally, not thickened at the base and gradually narrowing towards
the distal end. The arrangement of the tentacles was difficult to determine and by no
means so simple as WASSILIEFF imagined. There is no arrangement into two circlets
only ; on the contrary, the arrangement shows distinctly a certain resemblance to that
in Halcurias and Porponia, as some displacement of the tentacles has taken place,
here also assuredly connected with an irregular arrangement of the mesenteries. In the
first place, we have in each angle of the mouth (in the sagittal plane) quite the same
arrangement of the tentacles as in Porponia and Halcurias. They further agree in
this, that at certain points groups of tentacles occur, where two tentacles of the first
order stand on each side of one of the second order. Lastly, the other tentacles also
show a resemblance in arrangement to the corresponding tentacles in these genera.

* The Actiniaria described by WASSILIEFF, especially some of the species, require revision. Thus, the Actinia
described by WASSILIEFF under the name of Halciimpellu minuta is quite certainly no Halcampella-, but rather a
Halodava. The occurrence of warts and the appearance and structure of the tentacles, which are bulb-like and swollen
at the tips, speak in favour of this. The arrangement of the mesenteries and the siphonoglyphe agree less definitely,
but this may be due to the specimen being a young individual, as is indicated by the small number of tentacles (15).
If it cannot be referred to Halodava or Eloactis, it may represent the type of a new genus.

(ROY. soc. EDIN. TRANS., VOL. L., 58.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 251

The arrangement, however, is difficult to determine in detail in Hya/nthopsis, as the
grouping of the mesenteries is not known, and only one specimen has been at my
disposal, which, moreover, showed an irregular grouping of the tentacles in certain parts.
Starting from the one directive tentacle (that where the directive line passes through
an 03sophageal groove), the arrangement seems to be the following (1, 2, 3, etc., tentacles
of the first, second, third, etc., order ; dt. =the directive tentacle) :—

1 (dt.)-2-l-5, 4-5-4-5-3-4-3-5-4-5-4-5-1-2-1-5-4-5-4-5-4-5-1-2-1-5-4-5-
4-5-4-5-1-2-1-5-4-5-4-5-4-5-1-2-1- (dt.)-2-l-5-4-5-4-5-4-5-l-2-l -5-4-5-4-
5-4-5-4-5-1-2-1-5-4-5-4-5-4-5-1-2-1-5-4-5-4-5-4-5-1-2.

It is to be noted here, that among the last mesenteries of the first order, nearest to
the last mesentery of the first order, there may be a few tentacles more. A small part
of the edge of the body, namely, had been torn away, and it is possible that some
tentacles, at most 4 probably, have gone with it. In this specimen, therefore, the
number of tentacles is somewhat higher, yet not so high as 100. The group 3, 4, 3, is
also somewhat uncertain, in so far as it cannot be determined whether it is a 1, 2, 1
group or not ; in any case it seems to stand somewhat inside the other groups of the
first and second order. But to clear up the arrangement of the tentacles definitely
requires an investigation of more material with the developmental forms.

The oral disc is somewhat broader than the base, with deep, radial furrows corre-
sponding to the insertions of the mesenteries. The mouth was partly projecting, wide.
The oesophagus is wide, long, almost half the length of the body, with a siphonoglyphe.
WASSILIEFF states that two grooves are present, but I have not been able to find more
than one ; the other supposed to be present by WASSILIEFF does not differ — at least in
the specimen examined by me, which has also been investigated by WASSIIJEFF — from
the deep longitudinal furrows with which the oesophagus is also provided. Including
the siphonoglyphe, the number of these furrows amounts to about 16. No well-marked
prolongation of the groove downwards below the oesophagus is present.

The anatomical structure resembles that of the Endocoelactidse. The ectoderm of
the body-wall is not extensive, especially by comparison with the thick mesoglcsa. It
contains very numerous spirocysts of varying size, and fairly numerous thick-walled
nematocysts (length 34-43 n, breadth 5 M). In transverse sections at the base of the
ectoderm we find structures packed close together, which greatly resemble cross-sections
of ectodermal, longitudinal muscles. In the beginning also I was inclined to take
them as such, but a closer examination showed that this was hardly the case. In
longitudinal sections (Plate, fig. 2) I found similar structures, which should not have
been there if it was a question of longitudinal muscles. The muscle-like parts are,
indeed, so far as I could determine, no other than the greatly thickened bases of
ectoderm cells, which can also be clearly seen when the section cuts through a wall of
the ectoderm and passes through the ectoderm cells a little way from the base. In
these it can be seen that the ectoderm cells are fixed in the mesoglcea by greatly
thickened bases. I lay stress on this, as M'MuRRiCH was of the opinion that in

(ROY. 80C. KDIN. TRANS., VOL. L., 50.)

252 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPOMIA

Halcurias pilatus he had found longitudinal muscles in the ectoderm of the body-wall.
Though I have not examined this species, I consider it possible that the same condition
prevails there as in Ilyanthopsis — in other words, that the longitudinal muscles are
no other than thickened, basal parts of the ectoderm cells. The mesoglcea of the
body-wall is very thick, and provided with extremely numerous cells with bipolar,
irregularly prolonged ends. In the inmost part of the mesoglcea there are closely
packed fibrillar folds arranged as Plate, fig. 1, shows. The sphincter I have not
closely examined, but I think it probable, from an external investigation, that this is
completely wanting, as WASSILIEFF states. The entoderm is somewhat broader than
the ectoderm, and thicker in the middle between the insertions of the mesenteries than
at the sides. WASSILIEFF has described the structure of the tentacles and emphasised
the longitudinal muscles of the ectoderm. The ectoderm contains very numerous
spirocysts of varying size, and also numerous thick-walled nematocysts (length about
36 n). The longitudinal musculature of the ectoderm is comparatively weak. How
far fine outshoots run out from the mesogloea into the ectoderm, as is stated by
WASSILIEFF, I am unable to determine ; it seems to me that these outshoots are
nothing else but the thread-like basal parts of the ectoderm cells. The ectodermal
nematocysts and spirocysts of the oral disc agree with those in the tentacles, but are
not so abundant. The ganglion and nerve layer is also well developed here, as in the
tentacles. The ectodermal radial musculature is strong, especially in the ridges, and
the muscle folds just as high as, or higher than, the epithelial parts of the ectoderm.
The inner parts of the muscle folds show a tendency to be mesogloeal in the ridges
(Plate, fig. 3). The structure of the oesophagus agrees with that of the body-wall.
The mesenteries are all complete, corresponding in number to the tentacles.

As WASSILIEFF states, the musculature is very weak, and only somewhat more
strongly developed where the mesenteries join the body-wall (Plate, fig. 1). Here we
can distinctly distinguish the muscles, both the longitudinal and the parieto-basilar,
which are weak. On the very thin mesenteries there is no trace of protuberance of
the longitudinal muscles. In longitudinal sections through the mesenteries and trans-
verse sections through the base it looks as if the basilar muscle occurred as a wreak
fold of the musculature of the mesenteries. But specially differentiated basilar muscles
do not seem to be developed, for the transverse layer of muscles near the base on
the side of the mesenteries where the longitudinal muscles are, is formed by
the latter muscles, bending almost at a right angle a little way from the base,
thus forming what may be called pseudo-basilar muscles (Plate, fig. 7), the same
condition as in Porpoma, as I was able to determine from preparations of the
mesentery. With regard to the arrangement of the mesenteries, it is impossible to
determine whether all of them are equally developed. So much can be said, however,
that in the specimen I have examined, in addition to the directive mesenteries with
longitudinal muscles on the outer side, the other mesenteries seem to be arranged in
pairs with the longitudinal muscles on the inner side. The Halcurias stage, or rather

(HOY. SOC. EDIN. TKANS., VOL. L., 60.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 253

the Porponia stage, which ll//<tn//iopsis probably passes through during development,
is thus not apparent in the older individuals. The filaments seem to agree with those
in Halcurias and Porponia, but were so badly preserved that I could not obtain any
clear picture of them. In addition to spirocysts there are thick-walled nematocysts
like those in the body-wall (length 36 M), as also numerous thick-walled capsules with
distinct base to the spiral thread (length 31-34 M, greatest breadth 5 /u). Sexual
organs are present on all the mesenteries. The animal is hermaphrodite. Well-
developed testes occurred distally inside the filament region in each mesentery, while
a few grape-like eggs were found in the proximal part.

The investigation of Ilyanthopsis elegans has thus led to the result I imagined it
would, namely, that this is nearly related to Halcurias and Porponia. Though the
material is too small to permit of a detailed statement of the grouping of the tentacles
and mesenteries, there can be no doubt that they should be placed together. The
arrangement of the tentacles shows the same characteristics as in these genera, and
certain features of the mesenterial arrangement are the same apparently. Even the
external appearance agrees well with that of Halcurias : spirocysts occur in the body-
wall and oesophagus, as in the latter genus and Porponia. Ilyanthopsis shows most
resemblance to Halcurias, and it might be a question whether these two genera should
not be joined as one. For the time being, however, such a grouping would not be so
fortunate, as Ilyanthopsis has a much greater number of mesenteries than Halcurias .
further, in the former all the mesenteries are perfect, while in the latter about half are
perfect. Add to this that the longitudinal musculature of the mesenteries is strongly
developed in Halcurias, very weak in Ilyanthopsis, and it is evident that Ilyanthopsis
has its own developmental characteristics. It seems, moreover, more probable that
Ilyanthopsis has passed through a Porponia stage than a Halcurias stage, if the
mesenteries are taken into consideration. If we imagine all the mesenteries in Porponia
to be perfect, it is quite easy from them to derive the arrangement of the mesenteries
in Ilyanthopsis. In Halcurias, on the other hand, the stronger, not-directive mesen-
teries occur as unpaired mesenteries. How the development has proceeded we can
only learn from the younger stages. I consider it advisable, therefore, to set up a
new genus, Synhalcurias, for the species Ilyanthopsis elegans. The genus Ili/nn/finpyis
must be abolished, as the type species of this genus, Ilyanthopsis longifilis, R. Hertwig,
is no other than Condylactis passiflora, as stated by PAX (1910); I had also come to
this view in 1897 on examination of type specimens of the Challenger Actinise in
London.

We know of one more genus that might possibly be allied to Porponia, namely,
the genus Actinemus, founded by VERRILL. From R. HERT WIG'S description of
Polysiphonttt tuln-rosa (= Actinemus tuberosus M'Murrich) and from .M'MuRRicn's
description of A. plebeius, however, we can hardly conclude that a close relationship
exists between Porponia and these forms. According to my observations on a
specimen of Polysiphonia tuberosa from the Challenger Expedition, the arrangement

(ROY. soc. EDIN. TRANS., VOL. L., 61.)

254 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

of the tentacles does not seem to show the displacement seen in Porponia ; in fact,
as far as I can see, the tentacles are at no places grouped in such a way that two
tentacles of the first order border on a tentacle of the second order, even though
certain changes in the size of the tentacles have been observed, so that accord-
ing to R. HERTWIG the exocoel-tentacles are not the smallest in size. Though
Actinernus plebeius and A. tuberosus do not suggest any close relationship to
Porponia, it is yet not impossible that the type specimen of the Actinernus genus
may show greater similarities, a question I may leave unsettled at present, as I have
riot had the opportunity to examine this specimen.*

For the present, therefore, we must be content with a comparison between
Halcurias, Porponia, and Synhalcurias. The question is now, where we are to place
these genera, and would it be of advantage to separate them from other forms ?
M'MuRRiCH in 1901 dealt with this question with regard to the genus Halcurias.
"There are, apparently, three courses open for the disposal of the genus. It may be
referred to a family already existent, the definition of the family being changed, if
necessary, to accommodate it, or it may be taken as the type of a distinct family, as
CARLGREN has done, or, finally, it may be separated altogether from the Hexactinise
and regarded as the type of a separate tribe. It seems to me that this last procedure
is quite unnecessary and would probably be entirely out of harmony with the phylo-
genetic relationship of the genus. We have learned within recent years how extensively
nearly allied forms may differ, and how great all the modification which the hexactinian
type may undergo. The entire facies of Halcurias is that of an hexactinian." I am
entirely in agreement with the above citation from M'MuRRiCH, and, like this author,
I am of opinion that it is unnecessary to set up a separate tribe for this genus and
Porponia, as the whole development indicates that the initial stage is a typical
hexactinian with six pairs of mesenteries. M'MuRRiCH comes further to the conclusion
that Halcurias need not be placed either in a separate family, as 1 had done in 1897,
but considers it preferable to refer the genus to the family Actinidse (Antheadae).
" The peculiar mode of development of the secondary and tertiary mesenteries is of
minor importance, and I see no more reason for separating Halcurias as the type of
a new family than I do for separating an octamerous sagartian or one with a multi-
plicity of mouths and many siphonoglyphs from the rest of the members of that family."
He supports this view because " occasional endocoelous development of mesenteries
have been already recorded, as in Bunodes thallia, in Actinioides dixoniana and
papuensis" — a condition already pointed out by me in 1897.

But is this view of M'MURRICH justifiable ? So far as I can understand, this is not
the case, as variations irregularly arisen through asexual propagation, or through
regeneration and regulation in the symmetry of certain species — in the case of the
phylogeny — cannot directly be compared with similar variations from the normal type
arising during the ontogeny — a condition not hitherto taken into consideration, but

* Compare Appendix !
(ROY. sou. BDIN. TRANS., VOL. L., 62.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 255

which I must strongly emphasise. A species, for example, that normally, through
unequal development of mesenteries, through stopping of the growth of certain parts
and more rapid growth of others during the ontogenic development, e.y. from being
a 6-rayed becomes an 8- or 10-rayed type, which is constant or nearly so for the
species, cannot in phylogenetic respects be compared with another species where the
same stages are obtained through accidental, asexual propagation or by regeneration.
In the first case, the 8- or 10-rayed type is constant for the species, and occurs onto-
geuetically and phylogenetically ; in the latter case, on the other hand, it is a mode of
adaptation in a less or greater part of the individual, and is dependent on the course of
the asexual propagation, and the greater or less reduction of the old mesenteries in the
separated or damaged fragments, a condition which has been further dealt with in my
studies on the regeneration and regulation stages in the Actiniae, 1904, 1909. In so
far as it has arisen ontogenetically, an 8- or 10-mesentery stage is thus of direct use for
the phylogeuy, but not in other cases. What applies to the occurrence of an 8- or 10-
rayed type of Actiniaria also applies to the varieties that arise through the develop-
ment of the mesenteries in the endocoals. In such cases the conditions are in full
agreement with those found in 8- or 10-rayed forms. Porponia and Halcurias leave
no doubt that the regular development of mesenteries in the endocrels has taken place
ontogenetically, whereas the irregular and chance development of mesenteries in the
endocoels in Bunodes, Actinioides, and others stands in intimate connection with the
regeneration or possible early dislocations of tissues during development. In Porponia
and Halcurias the development of mesenteries in the endocoels is of importance for the
classification, whereas the abnormal occurrence of mesenteries in the endocosls in
Bunodes, etc., is of no use for this purpose.

The peculiarity that mesenteries occur regularly in the endocosls during the course
of development is thus quite specific for Porponia and Halcurias, and probably also
for SynJialcurias, and has not been observed in other Actiniaria. The question is still
left open, if this peculiarity is of such great importance that it necessitates the setting
up of a separate family. As mentioned above, M'MuRRicn connects the development
of mesenteries in the endoccels with the occurrence of an 8-rayed type, with the develop-
ment of several mouths and several siphonoglyphs. Just as little as we separate the
forms showing such variations from the normal Actiniaria type ought we, in his opinion,
to separate Halcurias from allied forms on account of the development of mesenteries
in the endocoels. That the multiplication of mouths in a genus of Actiniaria does not
involve a separation of the genus in question from other closely related species is
evident from the above, as this multiplication has not arisen ontogenetically, but by
asexual propagation. The same is certainly also the case with the multiplication of the
siphonoglyphs. It is now left to take into consideration the abnormal development of
the mesenteries. An 8- or 10-rayed type derived ontogenetically from a 6-rayed one is,
as already known, by no means a seldom occurrence within the Actiniaria group, and
may obviously arise within different families and genera that are in no genetic connec-

(ROY. SOC. EDIN. TRANS., VOL. L., 63".)

256 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

tion with each other. M'MuRRicn is therefore quite right in saying that a genus or
species need not be separated from other genera or species because it has been trans-
formed into an 8- or 10-rayed type. It must be pointed out, however, that such a type
may in certain cases be of great importance for the classification, namely, in cases where
8 or 10 rays are observed in all species of a certain genus, as the variation in the
symmetry can naturally be used as a good generic character. We know of no case
where a number of the pairs of mesenteries differing from 6 has led to the setting up of
a separate family.

As shown above, both Porponia and Halcurias, from an assumed typical 6-paired
mesentery stage, are transformed into one having 10 pairs of mesenteries. Where the
transformation takes place in the ordinary manner by the belated appearance of certain
mesenteries in certain areas and through the arising of other mesenteries in the
exocosls,* it seems unnecessary to separate these genera, but as the 10-rayed condition
arises in such a specific way by development of mesenteries in the endocoals, a develop-
ment that is continued during the following cycle, I consider it absolutely necessary to
set up a separate family for these genera, the more so as such an ontogenetic develop-
ment of mesenteries in the endoccels has not been observed in any other Actiniaria of a
higher type. As far as we know, no such displacement of the tentacles has been
observed in other forms of Actiniaria than the above mentioned. I place Porponia and
Halcurias together in one family, therefore, to which already in 1897 I gave the
appropriate name of Endocoelactidse.

III. RELATIONSHIP OF THE FAMILY ENDOCCELACTID^E TO OTHER
ACTINIARIA — ORIGIN or THE RUGOSA TYPE.

As already mentioned in the introduction, R. HERTWIG stated the possibility that
Porponia, owing to the arrangement of the macro- and micro-mesenteries, might form a
transitional stage between the Hexactiniaria (Actiniaria) and Zoanthidse (Zoantharia).
This explanation of the position of Porponia and the family Endoccelactidse cannot,
of course, be maintained, after we have ascertained the facts on which the relation-
ship between stronger and weaker mesenteries depends. There is nothing in the
organisation of the family Endocoelactidse that might indicate a close relation to the
Zoanthidse, as the development of the mesenteries in this family takes place in quite a
different way from that in the latter characteristic group of Anthozoa.

In my paper on Endoccelactis ( = Halcurias} I pointed out that in Minyas there is
a strong tendency to widen the endoccels at the expense of the exoccels, causing an
alteration in the grouping of the mesenteries, which had some resemblance to the
alteration in the grouping of the 10 stronger mesenteries in Endoccelactis. How this
grouping of the mesenteries has taken place in Minyas is still unknown, but it may
possibly have arisen in connection with the development of mesenteries in the endocosls,

* It is also to be noted that not all 8- or 10-rayed types are homologous with each other, for the 8- or 10-rayed
condition is not always obtained in the same way ontogenetically.

(ROY. SOC. EDIN. TRANS., VOL. L., 64.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 257

though this is not absolutely necessary. The arrangement of mesenteries in Minyas
may be explained quite simply through an enlargement of the endocoels. In any case
I consider the similarity in the Endoccelactis and Minyas arrangement as due to con-
vergence, a view which I am now able to further confirm, as Muiyas, i.e. the species
described by me in 1895, and a closely related species, probably M. olivacea, later
examined by me, are stichodactyline Actiniaria, which are nearly allied to the family
Aurelianidse (the genera Aureliana and Actinoporus}.

With regard to the position of the family Endoccelactidse, I pointed out in 1897
that it must be placed fairly low in the system of Actiniaria, a view that has also been
taken up by M'MuRRiCH. This is indicated not only by the absence of the sphincter
and the presence of spirocysts (thin-walled nematocysts) in the ectoderm of the body-
wall and oesophagus, but also by the absence of true differentiated basilar muscles.
Thus, the Endocrelactids must be Actiniaria, though they are not developed in the same
way as the elongated genera provided with physa (e.g. Edwardsidse, Halcampidse).
From a theoretical point of view we must assume the occurrence of forms which con-
stitute a link between the Protactininse and the Athenaria among the Actinim«, i.e.
we must take for granted the occurrence of original Actiniaria, which by the retention
of the original body-shape with flat base (thus without development of a physa) have
lost the longitudinal muscles in the body-wall, but, on the other hand, have not yet
developed true basilar muscles ; in the same way as I pointed out (1900, p. 57) that
the family Discosomidje forms a link between the Protostichodactylinse and Stichodo-
dactylinse. Among the Actininse type similar conditions would then prevail with
regard to the family Endocoalactis, if my supposition that this family has no longi-
tudinal muscles in the body-wall proves to be correct. Should it be the case, on the
other hand, that M'MuRRiCH is right in saying that such longitudinal muscles occur
in Halcurias pilotus, the family Enclocoelactidse must be referred to the Protactininse.
In this case the thickenings of the basal parts of the ectodermal cells in the body-wall
may be considered as rudimentary epithelial muscles, a view, however, I do not hold,
and a question that can only be answered by means of good material of maceration.
For practical reasons it would possibly be advisable in future to combine the family
Endoccelactidse with the Protactininae, and the Discosomidfe with the Protostichodac-
tylinse, a grouping which I already, in 1900, p. 57 (77), pointed out as possible with
regard to the Discosomidse.

In my opinion, the family Endocrelactidse must thus belong to the lowest Actininse,
or possibly to the more differentiated Protactininse. Probably an intimate relation to
any other ActininsB family does not exist.

Before concluding the account of the relations of the family Endocoelactidaj to
other Actiniaria, we might just point out that these variations are of importance for
the study of the other Anthozoa. As already known, the skeleton-forming Madreporaiia
show similar variations from the usual symmetry, as the Actiniaria, as 8- and 10-rayed,
radial or more bilateral forms are found even there. As we have seen that such a

(ROY. soc. EUIN. TRANS., VOL. L., 65.)

258 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

peculiar symmetry as that in Endoccelactidfe may also be found in the free Actiniaria,
it seems reasonable to conclude that among the variously attached Madreporaria, with
their varying adaptation to the under-layer, still more complicated and varying
arrangement of the mesenteries may be found. In my opinion, the arrangement of the
mesenteries in Endoccelactidse opens up the possibility of a more intimate connection
between Rugosa and Madreporaria, and more readily leads to an explanation of the
conditions of symmetry in Rugosa, like the one proposed by me in BRONN'S Klassen
und Ordnungen, p. 1 50. Whilst the development of the mesenteries in the Endocoe-
lactidfe gives a greater possibility of interpreting Rugosa, it makes the question still
more complicated, as in Rugosa there might be a development of mesenteries in the
endoccels. Though we shall naturally never be able to reach finality with regard to
the position of Rugosa as compared with the typical Madreporaria, but have to be satis-
fied with a hypothetical explanation, so long as we do not know how the mesenterial
musculature is arranged, I shall nevertheless give a picture of the way in which we
might imagine the origin of the Rugosa type, if the mesenteries after the 6-pair stage
have developed in the endoccels. I presuppose that the hypothetical, separating walls,
sarcosepta, are taken as mesenteries, the skeletal dissepiments, sclerosepta, as septa.

We start, therefore, from a stage with 6 pairs of mesenteries arranged typically, but
with the lateral endocoels larger or at least as large as the exocoels. In each of the 6
endocoels a septum has been formed (text-fig. 4A). In the next stage the development
of mesenteries of the second order takes place in the same way as in the Endoccelactidae,
i.e. in the lateral endocoels, 4 pairs of the second order with the longitudinal muscles
turned outwards. Each of these mesenteries of the second order forms a new pair with
neighbouring mesenteries of the first order. In these new endocoels 4 septa are formed
(text-fig. 4B).

Owing to this arrangement of the mesenteries and their occurrence only in the
lateral endocoels, 4 zones of development have arisen instead of the 6 found in the
exoccels of a normal Madreporaria. These zones of development lie one in each
quadrant of the animal. This results in an asymmetrical development of the
mesenteries, together with an irregular growth of the walls, due to the fact that the
animal is generally attached along the one side of the goblet, or at least has been so
once. The consequence is now that in each quadrant of the dorsal side of the animal,
i.e. the side turned away from the siphonoglyphe, a complete suppression of the
mesenteries of the next order takes place, while the development in the ventral part is
continued. In the ventral endoccels 4 pairs of mesenteries arise with the same arrange-
ment of the musculature as those of the second order. These mesenteries form new
pairs with adjacent mesenteries of the first and second order. In the 4 new endoccels
4 septa are formed (text-fig. 4c). The development is continued in this way with the
next order, with suppression of the mesenteries and septa in the dorsal endoccels of the
third order in each quadrant. At the end of the development, or at least at a late
stage, septa develop in the exoccels (text-fig. 4n).

(ROY. SOC. EDIN. TKANS., VOL. L., 66.)

AND RELATED GEN'ERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 259

As we see, the development of the Rugosa type may be explained by a similar
development of the mesenteries as in Halcurias or Porponia, though with the

Fiu. 4A.

FIG. 4B.

FIG. 4c.

FIG. 4D.

Flos 4A-D. -Scheme to illustrate the origin of Rugosa, assuming that the development of the mesenteries to begin with
has proceeded as in tl.,- Eudocuilactid*. The black tongues are endocoelic septa, the shaded tongues exoccehc septa.

difference that the development of new mesenteries is continued in the entloco3ls, and
that in each quadrant a suppression of mesenteries on the dorsal side takes place in each

(ROY. soc. EDIN. TRANS., VOL. i,., <J7.)

260 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

order from and with the third, a suppression which probably stands in connection with
the animal's mode of life.

In this hypothetical explanation of the arrangement of the Rugosa septa, I have
mainly intended to direct attention of palaeontologists to the fact that the Rugosa type
may be explained in various ways ; moreover, a closer examination of Rugosa has
shown that the development is different in different genera, that some of them retain a
bilateral and others a more radial arrangement ; the development of septa in this
group, however, requires further examination.

The hypothesis put forward by me with regard to the origin of Rugosa seems to me
to speak for itself. The presence of the 4 growth-zones after the development of the
first 12 mesenteries in Rugosa may be fully explained by supposing a development of
certain mesenteries like those in the Endocoelactidse. The enlargement of the 4
primary lateral endoccels has led to developmental zones for the new mesenteries being
removed to these areas instead of to the 6 primary exoccels. The origin of a 4 (8)-
rayed type in certain Rugosa may be explained in this way. In any case, I consider
the above explanation as good as, if not better than, that put forward by DUERDEN, to
the effect that Rugosa must stand in a certain relation to the Zoantharia (Zoanthidse).
In consequence of this view, he also considers the latter group as very old, a view,
however, I have some difficulty in accepting, as the Zoanthidas are obviously rich in
species, and presumably form a group which is still in process of differentiation. See
also my work in BRONN.

Finally, it seems convenient further to characterise the family Endoccelactidse with
the genera Halcurias. Porponia, and Syrihalcurias.

Family Endocoelactidiv.

Atheuaria (Protactininae ?) with thick, sometimes cartilaginous body-wall, without
sphincter and fossa, with spirocysts in the ectoderm of the body-wall and oesophagus.
Arrangement of the mesenteries quite different from that of the normal Actiniaria type,
owing to the development of the second and third order of mesenteries in the endocoels.
In consequence, the arrangement of the tentacles very different from the normal type
(among others, 1 0 tentacles of the first order immediately border on those of the second
order). Sexual organs present on all the stronger mesenteries from and with those of
the first order, including directive mesenteries.

Genus Halcurias M'~M.m-ric}\ = Endoccelactis Carlgren.

Endocoelactidaa with ca. 68 mesenteries, 36 of which are perfect. Four cycles of
mesenteries. The mesenteries of the fourth order regularly arranged on each side of
those of the third order, the mesenteries of the same pair equally developed. The
perfect mesenteries arranged as follows : 20 (6 + 4 pairs) + 16 (8 pairs), of which the
first 20 str6ng, extending over the whole length of the body ; the others only developed
in the distal part, and weak. The body cylindrical. The tentacles not bridge-like

(ROY. soc. EDIN. TRANS., VOL. L., 68.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 261

thickened on the outer side. On the first 20 mesenteries the pennons of the longi-
tudinal muscles well developed. One oesophageal groove.

H. pilatus M'Murrich.

H. carlgreni* M'Murrirh (Endoccelactis sp. Carlg.).

Genus Porponia R. Hertwig.

Endoctelactidse with (54 ? to) ca. 68 mesenteries, of which 44 perfect. Five cycles
of mesenteries. The mesenteries of the fourth and fifth order are regularly arranged,
but show unequal development, so that the mesenteries of the fourth order on the one
side of the mesenteries of the third order consist of a perfect and an imperfect
mesentery, on the other side only of an imperfect mesentery ; but the mesenteries
of the fifth order are not paired, and only developed between the mesenteries of the
first order and the perfect mesenteries of the fourth order (as in Actinostola). The
arrangement of the perfect mesenteries is 20 (6 + 4 pairs) +16 (8 pairs) +8 (these
form pairs with imperfect mesenteries). The body goblet-like, sometimes cylindrical.
The tentacles on the outer side bridge-like and greatly thickened. The pennons of the
longitudinal musculature on the mesenteries hardly indicated. Two oesophageal grooves.

P. elongata R. Hertwig.
P. robusta R. Hertwig.
P. antarctica Carlgren.

Genus Synhalcurias Carlgren.

Endoccelactidae with considerably more than 68 mesenteries (ca. 100), all of which
are perfect, arranged in pairs, and frequently agreeing in the size and distribution of
the sexual organs. The irregular arrangement of the mesenteries probably due to the
development of the mesenteries of the second and third order in the endocoels. Origin
of the mesenteries of the fourth order and the following (?). The body cylindrical. The
tentacles are not thickened on the outer side. The longitudinal muscles of the mesen-

o

teries weak, not forming pennons, and almost equally developed on all mesenteries.
One cesophageal groove (2 ?).

S. elegans (Wassilieff).

In a coming work I intend to give a description of the other Actiniaria, ca. 20 in
number, which have been collected by the Scotia Expedition.

* As further characterisation of this species, I may give the following information about the nematocysts : —
Spirocysts occur in quantities, especially in the tentacles, but are also common in the body-wall, the ectoderm of the
oesophagus and in the lil.-iments. They are of greatly varying sizes, generally as large as the corresponding thick-
walled nematocyst capsules ; hut smaller as well as still larger ones occur, the latter especially in the tentacles, where
they reach a length of up to 43 ^, breadth 7 /*. In the body-wall the thick-walled nematocysts reach a length of
22-26 /*, in the tentacles 20-34 ^, and in the filament and oesophagus ca. 26-29 ^. In the latter places are also found
nematocysts with distinct basal part to the spiral thread, of almost the same length as the preceding, but broader at
the basal end. Tin- thick-walled nematocysts are most numerous in the tentacles.

(ROY. soc. KUIN. THANS., VOL. L., 69.)

262 PROFESSOR OSKAR CARLGREN ON THE GENUS PORPONIA

APPENDIX.

Now that I have had occasion to examine two highly retracted and badly preserved
specimens of the type of Actinernus, A. nobilis Verr. (place of discovery 43° 18' N.,
60° 24' W., Gloucester Fisheries, 1879, U.S. Fish. Com.), as far as I can see from the
bad material, the tentacles of the first and second order are arranged as in Porponia.
There are also spirocysts in the ectoderm of the body-wall. Therefore I think that
Actinernus nobilis (but not Polysiphonia tuberosa, and probably not A. plebeius — 1
have not seen this latter species) must be placed in the Endocoelactidse. Whether
Porponia and Actinernus are synonyms I cannot say for the present, but it is not
impossible.

. February 4, 1914.

BIBLIOGRAPHY.

CARLGREN, O., " Studien liber nordische Actinien," K. tivenska Vet. -Alt ad. HandL, Bd. xxv., No. 10, 1893.

"Zur Keimtnis der Minyaden," 6/vers. K. Vet.-Ahad. Forli., Stockholm, 1894.

"Zur Mesenterienentwicklung der Aktinien," Ofvers. K. Vet.-Akad. Fork., Stockholm, 1897.

" Ostafrikanische Actinien," Mitt. Nat. Mus. Hamburg, xvii., 1900.

- " Anthozoa," Bronn's Klassen und Ordnunyen, L. 1-6, 1903-1908.

" Studien iiber Regenerations- und Regulationserscheinungen, 1, 2," K. Svenska Vtt.-Akad. HandL,

Bd. xxxvii., 1904, Bd. xliii. 1909.
DUBRDBN, J. E., "Relationships of the Rugosa (Tetracoralla) to the living Zoan these, " John Hopkins Univers.

CircuL, Jan. 1902.

"The Antiquity of the Zoanthid Actinians," Sixth Ann. Report Mich. Acad. Sc., 1904.

HERTWIG, R., " Die Actinien der Challengerexpedition," Jena, 1892.

M'MuRRiCH, J. P., " Report on the Actinias collected by the United States Fish. Com. Steamer Albatross

1887-1888," Proc. U. Stat. Nat. Mus., xvi., 1893.

" Report on the Actiniaria collected by the Bahama Expedition of the State University of Iowa, 1893,"

Ball. Lab. Nat. Hist. Univ. Iowa, 1898.
— " Contributions on the Morphology of the Actinozoa, 6. Halcurias pilatus and Endocculactis," Biol. Bull.,

ii., No. 4, 1901.

PAX, F., "Studien an westindischen Actinien," Zool. Jalirb., Suppl. 2, H. 2, 1910.
WASSILIEFF, A., " Japanische Actinien," Abh. Bayr. Acad. Wiss. Mat. Nat., Classe, 1908, 1 Supplb.

EXPLANATION OF PLATE.

bac., thickenings of the basal part of the ectoderm cells?

cm., circular muscles.

ct., ciliated tract of the filament.

dp., directive plane.

ec., ectoderm.

en., entoderm.

lm., longitudinal muscles.

m. mesoglcea.

inf., fibrillous folds of the mesoglrea.

(HOY. soc. EDIN. TRANS., VOL. L., 70.)

AND RELATED GENERA, SCOTTISH NATIONAL ANTARCTIC EXPEDITION. 263

mp., muscles' pennon.

nd., enido-glandulai tract of the filament.

p/n/1., parieto-basilar muscles.

pm., parietal part of the longitudinal muscles.

psb., paeudo-basilar muscles.

rm., radiated muscles of the disc.

fp., spirocysts.

t., tentacles.

Figs. 1-4, Synhalcurias elegans ; 5-9, Porponia antarctica ', 10. flalcurias carlijreni.

Fig. 1. Transverse section through the body-wall with a part of two mesenteries. ^.*

Fig. 2. Vertical section through the body-wall. Only a part of the mesoglcoa is designed. $- with out-
drawn tube.

Fig. 3. Transverse section of ihe mesoghca and muscles of the disc. ?..

Fig. 4. The basal part of a inesenterium with the longitudinal muscles and the pseudo-basilar muscles.
Schematic.

Fig. 5. Cross-section through one tentacle above the middle. ^.

Fig. 6. Cross-section through the same on the basis §.

Fig. 7. Cross-section through a not-directive mesenterium. The same section as in fig. 8. The whole
breadth of the mesoglcea is not drawn, f •

Fig. 8. Cross-section through a part of a directive mesenterium and the body-wall in the region of the
aboral end of the stomatodaeum. ^.

Fig. 9. Transverse section of the ciliated tract region of the filament, -jj.

Fig. 10. Section through the upper part of the body to show the arrangement of the mesenteries. Twice
magnified.

* Magnifications refer to REICHERT'S system, "Austria." Figures drawn in the level of the micro-
scope's foot.

(ROY. soc. EDIX. TRANS., VOL. u, 71.)

Scot. Nat. Ant. Exp.

CAKLOUEX : "SCOTIA" <«KXUS I'OKI-OXIA.

...

Vol. VII.

,

mf

'

•

•' • •

•

•
:

,
,

.tm

? / ' ', '''.'
• ' ' " '•

2

_, • .

kec

oskar Carlgrcn. del.

Kn-kine. Lith . Edin

PART XL
M'DIBEANCH MOLLUSCA.

(BATHYDORIS.)

XL-ANATOMY OF MTHYDORIS BROWNE.

AND AFFINITIES OF THE GENUS:
SCOTTISH NATIONAL ANTARCTIC) EXPEDITION

BY T. J. EVANS, M.A. (Uxoii.),

Lecturer in Zoology in the University of Sheffield.

(WITH TWO PLATES.}

XI. —The Anatomy of a New Species of Bathydoris, and the Affinities of the
Genus : Scottish National Antarctic Expedition. By T. J. Evans, M.A. (Oxon.),
Lecturer in Zoology in the University of Sheffield. Communicated hi/ L)r J. H.
ASHWORTH. (With Two Plates.)

(MS. received October 3, 1913. Read January 19, 1914. Issued separately April 1, 1914.)

INTRODUCTORY.

The genus Bitf/ii/tloris was created by BERGH in 1884 in his Report on the
Nudibranch Mollusca collected by the Challenger. In his account of the anatomy
of the new genus BERGH draws attention to the anomalous combination of characters
possessed by the animal, and gives it an annectent position between the Dorids and
the Tritonids, but places it among the Dorids on account of the predominance of Dorid
features. The single specimen of Batlu/doris abyssorum was dredged off New South
Wales in 2425 fathoms. A second specimen of this peculiar genus was obtained by
the Danish Ingot f Expedition and described by BERGH in 1900. This specimen came
from 1870 fathoms in Davis Strait, and resembled B. abyssorum, with specific
variations. Thus Bathydoris came to be regarded as an isolated genus with the
characters of a connecting link, and appropriately a denizen of deep water. Our
anatomical knowledge of the animal is derived almost entirely from BERGH'S accounts
of the two species mentioned, and is moderately extensive, considering the rather
imperfect state of preservation of the material and the fact that he was dependent on
single specimens in each case.

Two more specimens were brought from the Antarctic by the Discovery and
described by Sir CHAKLES ELIOT in the Report of the National Antarctic Expedition
published by the Natural History Museum in 1907. These constituted separate
species resembling BKRGH'S, but their state of preservation was such that ELIOT was
able to add little to our knowledge of the anatomy of Bathydoris. The discovery of
one of ELIOT'S species in 100 fathoms dispels the idea that the genus is confined to
the great depths.

A specifically distinct specimen was also taken by the German Antarctic Expedi-
tion, but THIELE, in the Report on the .Mollusca recently published, confines himself
to a superficial description in order to retain the rare animal intact as a museum
specimen.

It would seem to be the fate of polar expeditions to bring back single specific
specimens of this genus; for the species which forms the subject of the present
memoir is based on one specimen dredged in 1410 fathoms by the Scotia in March
1904. When the zoological material of the Scottish National Antarctic Expedition

(REPRINTED FROM THE TRANSACTIONS OF THE ROYAL SOCIETY OK EDINBURGH, VOL. L., PP. 191-209.)

270 MR T. J. EVANS ON

was distributed for identification, the animal went astray, and Sir CHARLES ELTOT, who
received the Nudibranchs, comments with surprise on the complete absence of Dorids.

Dr BRUCE has provided the following notes on the haul in which the animal was
included: " Locality, 71° 22' S., 16° 34' W. Bottom temperature, 31° "9 F. Surface
temperature, 29° '9 F. Depth, 1410 fathoms. Bottom, blue mud." The haul included
specimens of nearly every group of animals from sponges to fishes.

It is proposed to name the new species Bathydoris brownii, in honour of Dr R. N.
RUDMOSE BROWN, naturalist to the expedition. To him and to Dr BRUCE the author's
thanks are due for permission to undertake the investigation of its anatomy.

The six known species of Bathydoris may be tabulated as follows :—

1. B. abyssorum, Bergh. 2425 fathoms, off New South Wales. " Challenger."

2. B. ingolfiana, Bergh. 1870 fathoms, in Davis Strait. " Ingolf."

3. B. hodgsoni, Eliot. 100 fathoms, off Coulman Island. "Discovery."

4. B. inflata, Eliot. Depth not stated, off Coulman Island. "Discover}/."

5. B. clavigera, Thiele. Depth not stated, Gauss Station. " Gauss."

6. B. broivnii, sp. nov. 1410 fathoms, off Coats Land. "Scotia."

The specimen was preserved in about 5 per cent, formaldehyde, and suffered very
little distortion or contraction. The viscera were also in excellent condition for
dissection, and even the histological preservation was found to be remarkably good
when certain tissues were cut and stained for identification.

EXTERNAL FEATURES.

The animal, as preserved, was 75 mm. long, 40 mm. broad, and 35 mm. in height.
A thin, flabby foot margin extended about 8 mm. beyond the body all round and
was rather bluntly pointed behind, while anteriorly its thickened edge ran transversely
across the body behind the head, and had a deep glandular furrow extending into the
lateral margin for some distance. The contraction of the foot when the animal was
killed probably accounts for the distension of the dorsal integument and the extruded
condition of the genital organs, which were forced out to the extent usual among
Dorids during copulation. The mouth lies entirely on the ventral aspect, and is
surrounded by crinkled lips surmounted by a bulging forehead, which extends laterally
into rather long cylindrical oral tentacles. Nearly a third of the length of the ventral
surface is occupied by the buccal region, the enormous size of which is a striking
feature of the genus. The arched dorsal surface is covered by an integument which is
delimited all round the body by a slight fringe at a distance of 4 or 5 mm. from the
foot. Laterally this fringe or reduced mantle edge is represented by little more than
a linear thickening of the skin. The dorsal integument is thin, transparent, and
destitute of spicules, but, as in hodgsoni, inflata, and clavigera, the whole surface seems
to have been originally studded with papillae. Those that remain on exposed parts

(ROY. sou. EDIN. THANS., VOL. L., 11)2.)

THE ANATOMY OF A NEW SPECIES OF BATHYDORIS. 271

of the back are smallest ; while round the dorsal tentacles and along the nottual
margin they are much larger. They seem to be more numerous in brownii than in
any of the other three species possessing them. They are undoubtedly caducous, and,
as recorded by ELIOT, the points of attachment of detached papilla? are marked on bare
places by small circles with a spot in the centre. The only other conspicuous features
of the dorsal surface are the dorsal tentacles and the anal complex, consisting of the
anus, the renal pore, and the branchise. The rhinophores are club-shaped and perfoliated
all round. There is no trace of a pocket. The branchiui are two in number, and are
placed symmetrically in front of the prominent anal papilla. Both numerically
and structurally the branchue appear to differ very considerably from those of all the
other species of Bathydoris. Thus, B. abyssorum has live and /ni/n/finna ten arranged
in a circle in front of the anus, while ELIOT'S species have eight and five or six
respectively similarly arranged. The two tufts in 11. l>/-<>/n/ii are united by a ridge
which, on dissection, is found to contain the afferent and efferent vessels of the
branchiae. As the structure of the gills will be described later, it is sufficient here to
mention that they are not fine dendritic structures like the gill-plumes represented by
BERGH. The anal papilla is much shorter than in the other species, but this is probably
due to the relatively greater local distension of the hinder end of the specimen. The
renal papilla is inconspicuous, and lies in the median line between the anus and the
base of the gills.

The extruded genitalia stand out conspicuously on the right side between the
notgeal margin and the foot, and are thrown further back than in the Dorids by the
great size of the buccal region.

ANATOMICAL DESCRIPTION.

The importance of Bathydoris in any discussion of the interrelationships and
classification of the Nudibranchs was recognised by both BKRGH and ELIOT, and, during
the examination of the anatomical structure of the present species, this importance
became increasingly impressive as system after system was considered. This was
deemed to justify a more complete and detailed account of the anatomy than has
hitherto been given, especially as facilities for such an account were amply provided in
virtue of the excellent preservation of the specimen. In the course of this memoir it
will be noted that the description disagrees with those of BERGH and ELIOT on matters
the bearings of which are of great theoretical interest. These contradictions are some-
times so striking that the inclusion of the species in the genus Bathydons seemed
jeopardised. The points of agreement, however, form such convincing evidence of
o-eneric identity that the serious divergencies here given must be regarded as corrections,
based on examination of a more favourable specimen, of observations partially frustrated
by the poor condition of the material observed by previous workers.

The various systems will now be considered in order.

(ROY. sou. KDIN. TRANS., VOL. L., 193.)

272 MR T. J. EVANS ON

THE ALIMENTARY SYSTEM.

It will be seen in fig. 2 that the general plan of the alimentary system of Bathydoris
resembles that of the typical Doric!, but a detailed examination shows that there are
very important differences. The great size of the buccal mass (b.m.) was emphasised
by BERGH as a characteristic feature of the genus. Indeed, BERGH claimed its supposed
relationship with the Tritonids solely on the character of the buccal mass and the radula.
The mouth is surrounded by two sets of lips, an outer and an inner. The outer lips are
crinkled and fleshy, and leave a wide gape into which the inner lips project. These are
merely the thickened rim of the outer integument limiting the buccal opening. The
inner lips (p.m., figs. 2 and 3) are a pair of lateral pads enclosing a relatively small open-
ing leading into the buccal cavity. The pads have the consistency and appearance
of hyaline cartilage Histological examination shows that the hyaline substance is a
cuticular secretion which is continued in varying thickness as a lining of the alimentary
tube from the buccal rim to the stomach. The underlying epithelium consists through-
out of tall slender columnar cells, and corresponding columns of the secreted matter are
faintly visible in the cuticle.

The buccal cavity contains a pair of dark-brown horny jaws (J., fig. 3) supported on
muscular pads (P.J., fig. 3) which separate them from the globular odontophore (o.)
occupying the middle of the floor. The free edges of the jaws are quite blunt, and each
is produced into a slight beak opposite the mouth ; so that they are probably used as
a prehensile organ. The mouth must be capable of far more extension than might be
supposed from its preserved state in order to bring the jaws into action. The radula
(R.) is narrow in front and broad behind where it enters the radula sac, and the sac is
entirely contained in the substance of the odontophoral mass. The first row of teeth has
three teeth on either side of the rhachidian tooth and the radula broadens to 90. 1. 90,
the formula of the youngest row. The total number of rows is about 50. The rhachidian
tooth (Rh., fig. 4) has a broad base on which stands a backwardly directed cusp. The
laterals (L.1 to L.v) are formed on the same plan, but the cusp bends from the base
towards the middle line. The first four laterals differ from the rest in having shorter
and blunter cusps. Unlike those of the other species, the extreme laterals show little
sign of reduction. The teeth are firmly fixed in the specially thick cuticular covering
of the radula mass, which is itself bilobed with a deep median depression. The direction
of the cusps and of the underlying muscles suggests that the radula is used for
gripping the food during trituration. The approximation of the cusps of the first
laterals thus entailed would also explain the broken and irregular appearance of the
rhachidian cusps as well as the reduction in length of the cusps of the first few laterals.
Comparison with the figures published by BERGH and ELIOT shows that the present
species can be identified by its radula alone.

The (esophagus (ce. and cr., fig. 2) seems to differ from that of the other species in
that it turns to the left, even at its origin from the buccal mass. It is a broad, sigmoid

(ROY. soc. KDIN. TRA.NS., VOL. L., 194.)

THE ANATOMY OF A NEW SPECIES oK BATHYDORIS.

tube with muscular walls lined, as already mentioned, by a thick cuticle. At its lower
end it ends in a thin-walled sac, the stomach (fit., fig. 5), lying below the liver, and
partially imbedded in its substance. The lining of the, oesophagus is, throughout its
length, thrown into twelve raised longitudinal bands (!>.<>., tig. .r>) which are covered with
minute, blunt, brown cones of various sizes. As shown in fig. 6, the cones are partly
imbedded in the cuticle of which they are specially resistent local modifications. The
dark tint of the cuticle of the first half of the oesophagus and the extensive crinkling
of its walls obscured the presence in it of the longitudinal bands which were only seen
after clearing. ELIOT, who describes the denticulate bands in the second half of the
oesophagus, may have overlooked them in the first half for the same reason. The same
writer, in his account of />'. hodyxoiu, names the two regions of the (esophagus the first
and second stomach— names which seem inappropriate for a thickly outicularised tube
which serves merely for the delivery of the food to the sac in which it is actually
digested. The denticulate cuticle would, moreover, be quite ineffectual for the purpose
of mastication, and probably serves as a protective layer against the coarse diet of mud,
sponge, and small shells The thin-walled stomach stands at the junction of oesophagus,
intestine (int., fig. 5), and main liver ducts (I.t.) into the expanded ends of which the
food enters for some distance, as in Dorids generally. At its junction with the oeso-
phagus the stomach has a small pocket (.v.r.) like that of the Dorids, which was not seen
in any of the other species. Its function is not known, though such inapplicable names
as "pancreas" and "gall-bladder" have been applied to it by different authors.

It may be mentioned that the similar stomach recess in Doris tuberculata secretes
a glassy, refringent substance which is also found as a granular deposit on the mucosa
of the intestine and on the massed Halichondria . spicules passing down that tube. This
suggests a protective function for the organ, its secretion acting as a lubricant for the
passage of spiculose excrement down the intestine. The liver (/., figs. 2 and 5) is a
bulky organ which is not invaded by gonad or kidney, and is unlobed except in so far
as the intestine and the lower end of the oesophagus lie in furrows on its surface. The
intestine is a rather broad, smooth tube making an arc round the pericardium and end-
inCT by a sphinctered opening on the anal papilla.

The alimentary tract contained one large piece of undigested sponge, and sponge
spicules were present in all parts of the stomach and intestine. There were also found
much mud, bits of old shells, small pebbles, and the spines of Kchinids. The animal is
therefore probably an omnivorous feeder, though the prevalence of sponge suggests that
it has predilections for that group, like the Dorids.

The salivary glands (s.fj., fig. 2) are tiocculent and voluminous, forming a mass on
each side pressed against the wall of the oesophagus, and opening by stout ducts on
the hinder wall of the buccal cavity. BKRCJH and ELIOT mistook them for the blood
o-lands, but their histological structure puts their salivary nature beyond doubt.
Moreover, the true blood glands were found elsewhere.

(ROY. soc. EDIN. TRANS., VOL. L., 19f>.)

274 MR T. J. EVANS ON

THE NERVOUS SYSTEM. (Fig. 7.)

The brain of Bathydoris broadly resembles that of the Doricls, but with much less
concentration of the ganglia. It lies on the top of the buccal bulb, the cerebral,
pleural, and pedal pairs being quite separate, but lying close together. As fig. 7 shows,
the ganglia are asymmetrical in shape and disposition. Each cerebral ganglion (e.g.)
gives off four nerves from its anterior edge which go to the lips and oral tentacles,
dividing as thev go into a number of smaller nerves. The last bifurcations have small

o J O

ganglionic swellings at the point of division, as in some Tectibranchs. On the posterior
edge of the dorsal surface of the cerebral ganglion stands a small, almost sessile,
proximal rhinophorial ganglion (p.rh.g.) which sends a stout nerve (rh.n.) to the dorsal
tentacle. A distal rhinophorial ganglion swelling (not shown) marks the point of sub-
division of the rhinophorial nerve as it enters the tentacle. No sub-cerebral commissure
was found, so that the cerebrals are connected together below the alimentary tube by
the stomato-gastric loop (s.g.l.) only. The pedal ganglia (ped.g.} are connected by a
stout pedal (ped.com.) and a slender parapedal commissure (p.ped.com.). In the notch
between the pleural and pedal ganglia of the right side is placed a small genital ganglion
(gen.g.) which is broadly united to the pedal and connected with the pleural by a baud
of fibres from its lower aspect. From it four nerves go to the genitalia, both male
and female, over the surface of which they distribute themselves with a number of local
ganglionations on their courses.

The pleural ganglia (pl.g.) give off two main lateral nerves on each side which
supply the whole of the dorsal integument, with the exception of the anterior region
which receives a number of very fine nerves from the pleurals not shown in the figure.
The longer pair of pleural nerves pass back to the anal region, where they anastomose
with each other and with a visceral nerve (br.n.) from the under surface of the right
pleural ganglion. From the ganglia on this plexus the gills are supplied as in Dorm
tuberculata. The visceral ganglion, so obvious on the under side of the right pleural
ganglion of Doris, is not represented as a discrete mass in BatJiydoris. The visceral
ganglion of Doris would appear to include the visceral centres as well as the penial
centre usually associated with the pedal. In Bathydoris, however, the genital centres
of the visceral seem to be segregated from the rest and to be associated with the penial
centre on the pedal to form a special genital ganglion. The long, finely ganglionated
visceral loop (v.l., ?>./.') about the middle of its course sends backwards the chief visceral
nerve (v.n.), which, after giving a branch to the gastro-cesophageal anastomosis on the
stomach and liver, continues its course as the reno-cardiac. Two delicate nerves from
the pleuro-pedal angle (see diagram) supply the branches of the aorta. The stomato-
gastric ganglia (at. (/.;/.) are relatively very large, but their size is not surprising when
we remember the dimensions and muscular complexity of the buccal mechanism which
they innervate. The inequality in length of the two cerebro-buccal connectives is pro-
bably the result of the sharp sinistral bend of the oasophagus. There are no separate

(ROY. soc. EDIN. TRANS., VOL. L., 196.)

THE ANATOMY OF A NEW SPECIES OF BATHYDORIS. 275

gastro-cesophageal ganglia, ami the paired gastro-cesophageal nerves (g.o.n.) arise
directly from the stomato-gastric ganglia. The paired nerves pass back in the con-
nective-tissue investment of the oesophagus to the stomach, where they form a joint
ganglionated anastomosis with the gastric branch of the visceral, as already described.
From this plexus the stomach and liver receive their innervation, while the salivary
glands are supplied from the gastro-cesophageals on their way down the (esophagus.

This compound system of gastro-hepatic ganglia is paralleled in the Dorids, where
it was described by ALDER and HANCOCK, while DREYER has lately shown that an
analogous arrangement is present in Aeolids and Tritonids.

Like the other species examined anatomically, It. hroirnii showed no trace of organs
of sight, and otocysts could not be found by careful surface examination. They were,
however, found on staining and clearing and also in sections of the brain. They are
two small sacs placed close to the pleuro-pedal connectives on their lower aspect
and partially imbedded in the connective-tissue capsule that surrounds the brain. The
author has found small otocysts similarly placed in some of the Polyceratidse. Otocysts
were not found in the species examined by BERGH and by ELIOT.

THE EXCRETORY SYSTEM. (Fig. 8.)

The kidney of Bathydoris is unusually well developed. This was also noted by ELIOT,
who described two fern-like organs lying over the liver as well as the renal syrinx.
ELIOT, however, misconceived the nature of the renal organ, since he took the paired,
fern-like structures to represent the main portion of the kidney, whereas they are
merely outgrowths of its floor or ventral wall in the posterior half. The whole renal
organ is a huge sac extending from almost the extreme posterior end of the body to
within a short distance of the head, but narrowing in front on account of pressure
between the alimentary canal and the genital mass. Posteriorly, its delicate dorsal
wall is overlain by the pericardium, to which it is connected by fibres. Elsewhere it
bulges free except where it is pinched by the intestinal loop (int.). This dorsal wall
is throughout non-glandular, except at two points — namely, at its extreme anterior
corner (b.y.'} and at a place in front of the pericardium (!>.[).") where a diverticulum of
it lies as a flap across the intestine. These two points will be further mentioned in
connection with the vascular system, because the two phagocytic or blood glands lie
here adherent to the wall of the kidney. The glandular part of the kidney is therefore
almost entirely confined to its ventral wall and to those parts of it which are folded
inwards into the renal cavity as the two fern-like structures seen by KLIOT. These,
however, are not two but six in number, the posterior pair being more fern-like than
the other two which lie on the surface of the gonad (//.;/.). These glandular regions
coincide with the areas of distribution of great branches of the aorta, and the narrow
strip-like folds forming the two anterior pairs may easily he mistaken i'<>r the arterial
trunks themselves, which actually lie within them below the renal wall. The vascular

(ROY. soc. EDIN. TRANS., VOL. L., iy7.)

276 MR T. J. EVANS ON

supply of the kidney is therefore purely arterial, and all the renal arteries arise from
an aortic bulb (a.c.) opening into the ventricle at the point o.v. in fig. 8, and continu-
ing forwards as the cephalic artery (c.art.). The extensions of the renal arteries into
the gonad are not shown in fig. 8. The blood delivered by the renal arteries — which,
as we have seen, occupy the crests of the glandular folds — passes on into venous lacunae
which lie deeper in the substance of the folds, and open into a great median venous
space, lying between the kidney and liver behind, and between the gonad and liver in
front. Into this median venous space also passes the blood that has traversed the
gonad and liver.

The main collecting reservoir of the kidney lies behind the gonad, and in its hinder
wall is seen the opening into the renal duct which leads to the exterior. In it
originates the reno-pericardial duct (r.p.d.), consisting of a median tube opening in
front by a funnel into the renal chamber, and a renal syrinx (r.s.) opening on the floor
of the pericardium. The syrinx is a bulbous structure with a wide lumen which is
almost filled with delicate laminate ingrowths of the epithelial lining. Sections of the
floor of the kidney show that the gland cells lying in connection with the renal arteries
contain concretions, often of large size, which stain faintly with basic dyes. The con-
cretions collect in big vacuoles, which finally burst and liberate the excreted contents.
The foliations of the wall of the syrinx are covered with cells of two kinds. The distal
part of a lamina — namely, the free edge towards the middle of the lumen — is covered
by ciliated cells only, the cilia being extremely long. The proximal part — namely, that
nearer the wall of the syrinx — is glandular, and the cells contain fine granules of a
substance which takes acid dyes. These cells are continued on to the wall of the
pericardium. The renal organ of Bathydoris and its vascular supply are thus Dorid
in type, the reuo-pericardial duct, especially, being almost identical in structure with
its homologue in Doris. The association with the blood glands, non-functional though
that may be, the absence of ramifications into underlying organs, and its forward
extension into the head region are features not paralleled among true Dorids.

THE VASCULAR SYSTEM. (Figs. 9 and 10.)

In general, the vascular system of Bathydoris resembles that of the Dorids, but in
several respects distinct affinities with the blood system of the Pleurobranchids are
exhibited. It may be conveniently described under the following heads :—

(1) The Heart.

The most obvious feature of the heart and pericardium is their asymmetrical
disposition, since the antero-posterior axis, unlike that of the true Dorids, lies at an
angle to the long axis of the body. The pericardium is a spacious cavity lying
posteriorly on the surface of the kidney, with the reno-pericardial opening in its

extreme right-hand corner.

(ROY. soc. EDIN. TRANS., VOL. L., 198.)

THE ANATOMY OF A NEW SPECIES OF BATHYDORIS. 277

The asymmetry of position mentioned above is an insignificant matter compared
with the structural asymmetry shown by the heart itself. The typical Doric! heart is
roughly an isosceles triangle with three efferent ducts opening into its base, the efferent
branchial in the middle and the two lateral integumental sinuses at the corners. The
auricle of Bathydoris, on the contrary, receives but one efferent vessel, which enters it
at the right-hand side, the efferent branchial vessel and the lateral sinuses being con-
fluent outside the pericardium altogether, as in the Pleurobranchids. The left side of
the auricle is fused for some distance with the pericardia! wall, along which it sends a
muscular wing. This asymmetry, as we shall see later, is only one of many pre-l)orid
and ancestral opisthobranch features exhibited by Bathydoris.

(2) The Arterial

Although the arterial system possesses no striking feature, it is proposed to describe
it somewhat fully, because no comprehensive account exists of the arterial system of
any Dorid except HANCOCK and EMBLETON'S account of Doris tuberculata in their
famous article in the Philosophical Transactions of the Royal Society. The ventricle
(v.) is immediately followed by a large aortic bulb (a.c.) from which arise the renal
arteries (see kidney), as well as vessels to the intestinal loop, the gonad and the
periphery of the liArer mass lying below. The aortic bulb is continued forwards as the
main cephalic artery (c.nrt.) This gives off on the left the visceral artery (v.art.),
running below the intestine and supplying the liver, stomach, and oesophagus. In
fig. 9 the arterial trunks lying below the outlined viscera are dot-shaded. After
giving off the genital arteries ((/.art.) on the right, the cephalic artery bifurcates, one
branch passing over the (esophagus to the left and the other below the buccal mass to
the right. The left branch provides both salivary glands (sal.g.), the brain (rcr.art.),
and the buccal muscles of both sides, while the right branch goes direct into a spacious
iufra-pharyngeal lacuna (lac.), in which the left also ends. This lacuna was also found
in Dorix tuiicrcidata. It should be noted that the cephalic artery forms a complete
collar round the oesophagus and buccal bulb. From the central lacuna under the bulb
arise a number of vessels. A median vessel passes straight up into the bulb (buc.art.) ;
three run forward into the lips (lab. art.) and floor of the mouth, while a broad median
vessel dips into the foot and bends backward in its substance, to continue throughout
its length as a median pedal artery (ped.art.)

(3) The General H&mocoele.

The irregular lacunar blood-space in which lie all the viscera is in /)»t//t/>loris
nowhere spacious. It receives the blood that has passed through the tissues from the
arteries, except the renal, gonadial, and hepatic blood, which is collected in another
way already indicated in the description of the kidney. The ha-moca-lic blood passes
partly into the gills and partly into the dorsal integument, l>ut the proportion of blood

(ROY. soc. EDIN. TRANS., VOL. L., 199.)

278 MR T. J. EVANS ON

that passes into the integument is much in excess of that, in the Uorids. The thin
skin, studded with papillae, is conducive to this amplification of the tegumentary
respiration in Bathydoris, while the thick, glandular, and spiculose skin of the Uorids
has vitiated this system and necessitated the extension of the special gills. The dorsal
wall of the hsemoccele is a membrane more or less bound to the underlying organs by
conjunctive-tissue fibres. This membrane is separated from the dorsal integument by
an empty space, but runs into it at the side of the body all round. Thus, when an
incision is made through the dorsal body- wall, the space entered is not the hremocoele,
but this closed cavity between the body-wall and the dorsal wall of the haemoccele.
The same arrangement is present in the Dorids alone among Nudibranchs, though the shell-
cavity of Pleurobranchus closely resembles the problematic dorsal cavity of the Dorids.
Whatever be the nature of the cavity, HANCOCK and EMBLETON'S name — peritoneum
—for its lining should not be perpetuated. The passage of blood from the underlying
hsemoccele into the skin and its papillary outgrowths takes place below the level of the
edge of the dorsal hsemoccelic wall all round. The blood that runs from the hsemocoele
to the gills passes along two narrow conduits on the posterior aspect of the liver (h.v.,
fig. 10). This must be regarded as of secondary importance in the afferent branchial
system.

(4) The Afferent Branchial System. (Fig. 10.)

Blood enters the branchiae from two sources : (a) from the hsemocoele by the small
paired ducts (h.v.) already mentioned, and (b) from a great median venous space (m.s.)
lying above the liver, which receives the blood from the kidney, liver, and gonad. Just
before narrowing in order to enter the gills (a.b.v.), it receives the paired ducts from
the haemoccele (a). (See also the description of the kidney.) The afferent space at the
base of the gills is not a circle, as in the Dorids, but a transverse expansion of the
afferent vein from which ramifying tubes run up the branchiae.

(5) The Efferent Branchials.

The afferent and efferent venules in the gill-leaflets form continuous loops from the
afferent to the efferent side of a gill lobe, and the efferent veins from the two gills join
together to form a transverse space at the base of the gills similar to the contiguous
afferent space. This space is connected with the auricle by a tube (c.b.v.) running to
the right and entering the auricle at its right-hand corner.

(6) Tlie Efferent Tegumentary System.

The blood that enters the skin and its papillae from the hsemoccele returns from all
sides into a circular sinus (c.s.) running round the edge of the pericardium. The
efferent tubules returning blood into the sinus were described by BERGH as renal tubules
in B. abyxsorum. The sinus opens behind into the efferent branchial vein just before

(HOY. soc. EDIN. TRANS., VOL. L., 200.)

THE ANATOMY OF A NEW SPECIES OF BATHYDORIS. 279

it reaches the right corner of the pericardium to empty itself into the auricle. The
circular sinus is also represented in the Pleurobrunchids, where it also opens into the
efferent vessel, but takes a wider sweep round the body. In the Dorids it is repre-
sented by the two lateral sinuses, which there, however, enter the auricle separately
and directly.

(7) The Blood Glands. (\>.<j! and b.y." Figs. 8 and 9.)

The structures commonly called " blood glands " are characteristic of some Tecti-
branch families (Bullids and Pleurobranchids) and of the Dorids. They are lymphatic or
phagocytic glands situated on the course of the cephalic artery and supplied by it. In
the Dorids they lie in the head region near the brain, but in the Tcctibranchs they are
placed further back. In Bathydoris they form two separate masses united to the wall
of the kidney. It is of some interest to note that in the Prosobranchs possessing them
they are also associated with the kidney, so that in this, as in many other respects,
Bathydoris presents features more primitive than the true Dorids, the equivalents of
which are found among Tectibranchs rather than Nudibranchs. As already mentioned,
BERGH and ELIOT confused the unusual salivary glands with the blood glands, the
identity of which they did not recognise.

THE RESPIRATORY SYSTEM.

As already indicated in connection with the vascular system, the general pallial
respiration is rendered more effective by the papillary outgrowths, and the blood so
oxygenated returns into the circular sinus. The special respiratory organs or branchiae
are two tufts placed symmetrically on a transverse ridge in front of the anus. Each
tuft stands on a broad base or stalk in such a way as to give the appearance of a roughly
pinnate condition. The lobate units of the tuft resemble the pinnae of the gill of
Pleurobranchaea, being laminate on opposite faces of a wide rhachis, while the afferent
and efferent vessels occupy its edges. The laminae vary in size from mere ridges across
the face of the rhachis to longish leaves which may themselves be provided with laminae.
In this way an irregular bipinnate condition is simulated. It will be seen that, by
narrowing the rhachis so as to bring the ascending; and descending vessels nearer

o o o o

together and regularising the pinnation, the Dorid plume would be produced. On the
other hand, if the lamina; were equal in size and the tufts stretched along the ridge in
a regular row. the sessile portion of a Pleurobranchid ctenidium would result. The gill
of Bathydoris In-mritii would therefore appear to be in a condition intermediate between
a typical Dorid rosette of plumes and a Tectibranch gill. There is, however, no indica-
tion of a circumanal circlet either in the gill itself or in the underlying vessels, and though
the tufts are provided with muscles capable of reducing their height, they cannot be
retracted below the general level of the integument.

(ROY. soc. EDIN. TRANS., VOL. L., 201.)

280 MR T. J. EVANS ON

THE REPRODUCTIVE SYSTEM. (Fig. 12.)

No adequate description or figure of this system in a Batkydoris has hitherto been
given, but both ELIOT and BERGH give a somewhat hesitating impression that it is
constructed on the Dorid plan. Since the universal triauly of known Dorids is one of
their most striking characteristics, it is essential that on such a critical point our
knowledge should be clear and definite.

The hermaphrodite gonad (h.g.) lies posteriorly below the kidney and above the liver
within the arc made by the intestine. It is a yellowish, bi-convex lenticular body,
truncated in front and with a minutely lobulated surface. The specimen was captured
at the stage of male activity in the protandric cycle, since the male acini and ducts
are full of sperms, while the eggs are small and lightly yolked. Its blood supply is
an extension of the renal arterial system, branches from which pass through the lower
wall of the kidney into its substance.

The common hermaphrodite duct (c.h.d.) leaves the gonad as a single slender
tube. It is ampullated (amp.}, as usual in Nudibranchs, but its extreme length escaped
previous notice. After a short, slender portion beyond the ampulla, it divides into
two tubes, the vas deferens (v.d.) and the oviduct (o.d.). The vas defereus is a
comparatively short, coiled tube, expanded by the glands in its walls into a prostate
for nearly the whole of its length. It enters the penis sac (p.s. ) some distance
from the end and runs a straight course to the tip of the everted penis imbedded
in loose connective tissue and muscle fibres. The mode of extroversion of the
penis, deducible from dissection of the everted organ, is represented in the section-
diagrams (fig. 11, « and b), the dotted area representing loose fibrous tissue the
perfusion of which with blood from the hsemoccele causes the extroversion. The penis
is seen to be a partial introvert, since the end is retracted into the sac unchanged.
This terminal portion (p.) presents a remarkable appearance on account of the sucker-
like pits covering one side of it. It is possible that the pits, under control of the
blood-pressure in the penis, really act as suckers on the smooth surface of the female
atrial wall.

The oviduct soon enters the massive mucus-albumen gland complex (m.g. and a.g.),
the structure of which could not be investigated on account of its stony hardness. The
albumen gland could be recognised on the upper surface by its yellowish-brown colour
and its granular consistency. The coils of the mucus gland end distally as the broad
tube which opens into the female atrium. The atrium is turned out as in copulation :
the first part of it has a highly crinkled surface, but inside this arise two leaf-like lips
(a.L), or folds of its surface, which between them enclose the entrance (f.o.) into the
female channel. These valvular lips appear to be a characteristic feature of Bathydoris,
because they are also partially shown in surface view in B. davigera by THIELE.
Within the valve on the posterior wall of the channel opens the vagina (vg.), which
consists of a stout tube ending blindly in a globular recurved bursa copulatrix (b.c.).

(HOY. SOC. BDIN. TRANS., VOL. L., 202.)

THE ANATOMY OF A NEW SPECIES OF I'.ATII VDORIS. 281

There is no second sac on the course of the vagina, which is also the only connection
between the buisa and the female channel. From this it follows that the herma-
phrodite duct of Bathydoris divides but once — namely, into a male and a female duct.
The Dorid duct bearing the so-called spermatoryst, and connecting the bursa with the
region of fertilisation in the course of the ovidue.1, is totally unrepresented. In short,
Bathydoris In-mniii is typically diuulic, like Ti-it'mi« and the Pleurobranchids.

Externally, the everted organs are surrounded by a rim representing the limit of
the common genital vestibule of the male and female system.

SYNOPSIS OF SPECIFIC AND GENERIC CHARACTERS.

When we come to analyse the differences between the foregoing account and those
of the two previous investigators of the genus, we encounter considerable difficulties in
deciding which are specific differences and which may be put forward as corrections.
To the latter category we may presume to relegate all differences in regard to which
previous statements have been qualified or made with reservation.

The specific distinctness of Bathydoris liroirnii is undoubtedly more striking than
that of any of the other recorded species, and, since the animal was mature, there can
be no question of its being the young of any one of them. Among characters pre-
sumably not of generic rank which distinguish the species may he mentioned the
following :—

1 I ) The uniformity in size and shape of the lateral teeth of the radula.

(2) The pitted pad on the glans penis.

(3) The immediate sinistral bend of the oesophagus at its origin from the buccal

mass and the unequal lengths of the cerebro-buccal connectives. The
causal connection between these two features stamps them as true anatomical
constants.

(4) The two tufted gills placed on a traverse ridge symmetrically in front of the

anus.

On the basis adopted above, it is likely that the following features in which the present
account differs from those of BERGH and ELIOT may be regarded as of generic value :—

(1) The diauly of the reproductive system.

(2) The circular canal embracing the pericardium and collecting blood from the

dorsal integument.

(3) The foilicular nature and great size of the salivary glands.

(4) The presence of two blood glands on the wall of the kidney (the structures

described as such by BKUCH and ELIOT turned out, on histological examina-
tion, to be the follicles of the salivarv glands ; the true blood glands were
not seen by these authors).

(5) The asymmetrical opening of the efferent vein into the auricle.
((>) The great saccular kidney with its ventral wall thrown into folds.

(ROY. soc. EDIN. TRANS., VOL. L., 203.)

MR T. J. EVANS ON

(7) The segregation of the genital elements of the visceral into a distinct ganglion

on the surface of the brain.

(8) The possession of a proximal and a distal rhinophorial ganglion.

(9) The cuticularisation of the wall of the alimentary canal as far as the stomach

(the horny cones imbedded in the cuticle were described by ELIOT for B.
hodgsoni, but no armature was found by BERGH in B. abyssorum and B.
ingolfiana; its systematic value cannot therefore be assigned).

(10) The presence of a ga,stric caecum.

(11) The presence of small otocysts below the pleuro-pedal connective.

The genus may now be defined in the following terms, of which some are supple-
mentary to BERGH'S original definition :—

Body highly arched and elliptical in outline. The edge of the notseum slight or
wanting. Dorsal papillae present or absent, Rhinophores placed rather far back, non-
retractile, perfoliated. Gills in front of the anal papilla, variable in number, non-
retractile. Buccal mass very bulky. Radula sac not an appendage. Dental formula
n. 1. n. Buccal cavity with a thick cuticle extending down the oesophagus. Powerful
jaws present. (Esophagus may have horny cones. Liver massive and unlobed, not
invaded by any other viscus. Salivary glands follicular, flattened, with a stout
duct. Cerebral and pleural ganglia distinct. Stomatogastric loop very long. No
gastro-ctisophageal ganglia, but the long gastro-cesophageal loop arises from the
buccals. Eyes absent. Kidney saccular with laminate ingrowths of its ventral wall.
Branchial and pallial efferents join before entering the right side of the auricle. Penis
unarmed and massive. Hermaphrodite gland a compact mass. Reproductive system
diaulic.

THE AFFINITIES OF BATHYDORIS.

BERGH and ELIOT have invested Batliydoris with a certain importance as a type
combining the features of the Dorids with certain Tritonid characters, with prepon-
derating affinities to the Dorids. The Tritonid features accentuated by BERGH were
the buccal apparatus and the unarmed penis, while ELIOT rightly passes over the latter
resemblance unnoticed, since an armature of the penis may be present or absent among
the species of some genera, of Dorids. As to the buccal apparatus, even a superficial
examination shows that, when reference has been made to the great size of the buccal
muscles and the jaws, the sole resemblance has been stated in full. In Tritonia the
odontophoral mass arises from the dorsal wall of the buccal cavity and bulges down-
wards, while in Batliydoris that organ arises from the floor of the buccal cavity and
bulges upwards. The mandibles are also quite differently placed and used in the two
animals, those of Tritonia having their long, finely serrulated cutting edges facing the
floor of the mouth cavity below the radula mass, while the blunt beaks of the mandibles
of Bathydoris jut into the mouth above the radula. Moreover, the oesophagus takes
its origin on the hinder aspect of the globular buccal bulb of Bathydoris, while the

(ROY. SOC. EDIN. TRANS., VOL. L., 204.)

THE ANATOMY <>K A XKW SPECIES OF BATHYDOKIS. 283

oesophagus of Tritoitix rises out of the dorsal surface of the bulb and well forward, the
main bulk of muscles being behinil it. These and corresponding differences in the
muscular mechanism suffice to make good the statement made above, that the supposed
resemblances are confined to size ami the presence of powerful jaws. It should be
mentioned that BERGH refers specially to B<>rncll<( in assigning Tritonid features to
Batlnjilorix; without discussing the problematic relationship of Born<'ll« to the
Tritonids, suffice it to state that the large buccal apparatus of that form differs from
that of both Triton/'/ and /{ntltydorix.

It is indeed likely that these three cases of powerful and mandibulate mouth parts
are examples of convergence in unrelated types. The only other reference to a non-
Doricl affinity of Batlujdoris is made by ELIOT when he compares the armature of the
"stomach" with that found in Honn-l/K. This comparison is strange, coming from an
author who has since, in the Ray Society's monograph, separated the two genera in
his first cleavage of the Nudibranchs. In any case the comparison is untenable, since
the two armatures are totally unlike in structure and position, that of Bornella being
situated in a region of the alimentary tube /tostcrior to the point of entrance of the
liver ducts. On the foregoing grounds we must regard the proposed Tritonid and
Bornellid affinities of fiatli//</<>rix as inadmissible. It is, however, obvious that the
investigation of this last species has brought out certain features of the genus which
render necessary the reopening of the discussion of its affinities and, as we shall see
later, those of the Dorids generally. It is no less certain that the genus presents a
combination of characters far more significant than that considered by BERGH when he
assigned its affinities — namely, a Doric! gill of a primitive form, an asymmetrical heart
and efferent branchial system, blood glands placed far back on the course of the aorta,
a thin integument with scattered branchiate outgrowths, a diaulic reproductive system,
a liver distinct from the gonad and kidney, a brain with separate ganglia, a nerve
collar embracing the buccal bulb and not the oesophagus, and finally, but perhaps least
significant, a powerful buccal apparatus.

That Jtathi/doris must be definitely placed among doridiform animals follows from
its possession of the following striking Dorid characteristics :—

(a) The collocation of the anus, renal pore, and gills in the median line posteriorly.

The gill is, however, more primitive than the typical rosette form common
among Dorids, though primitive gills are also found in such types as
Trcrrl i/u mi and Nembrotlm.

(b) With the exception of the buccal mass and the protected oesophagus, the

alimentary canal is Dorid, even to the possession of a gastric caecum, ami
those divergent features are adaptations to a coarser and more omnivorous
diet. The enlargement of the salivary glands is probably due to the same
cause.

(c) The kidney is a Dorid structure, the reno-pericanlial tube and syrinx being

practically identical with those of Doris as described by HANCOCK
(ROY. soc. EDIX. TRANS., VOL. L., 205.)

284 MR T. J. EVANS ON

and by HECHT. The absence of ramifications is, doubtless, a primitive
character.

(d) The blood system is built essentially on Dorid lines, but presents a greater

number of primitive features than any other. Chief among them are the
possession of but one auricular efferent opening, the union of the circular
collecting canal of the integumental system with the efferent branchial, and
the position of the blood glands. It is noteworthy that these primitive
features are points of agreement with the Tectibrauchs, especially the
Pleurobranchids.

(e) As to the reproductive system, its diaulic condition makes it more primitive

than that of any other known Dorid ; but, apart from that very important
divergence, it closely resembles that of Doris, since the separation of the
gonad from the liver is found in a typical Dorid like Alloiodoris.

(f) The nervous system, in spite of a close similarity to the Dorid type in most

respects, differs from it in several important points. Of these, the length
of the nerve collar and the position of the brain on the top of the buccal
mass are paralleled in Tritonia and the Pleurobranchids, and should probably
be regarded as primitive, while the distinctness of the ganglia of the brain
and the absence of separate gastro-ossophageals, if primitive features, take
us back to a condition earlier than that found in the Pleurobranchids and
Tritonia. The fusion of the ganglia of the visceral loop with the pleurals
is, on the other hand, a modern feature, as is the loss of eyes consequent
on the adoption of a deep-water habitat.

We conclude, therefore, that Bathydoris is a highly primitive Dorid possessing
some characters that adapt it to a specialised habitat and mode of life, while those that
are primitive connect it with the Tectibranchs, particularly the Pleurobranchids among
existing forms. The derivation of the Dorids from Pleurobranchid ancestors is,
however, no new proposition. GUIART, for example, has recently advocated their union
into one group, and PELSENEER has derived all Nudibranchs from the Pleurobranchids
with Tntonia as an intermediate link.

BERGH'S advocacy of a special relationship between Bathydoris and Tritonia on the
evidence of the buccal apparatus has already been criticised. PELSENEER'S position,
however, takes a wider outlook, but takes no cognisance of Bathydoris at all. He
bases his contention of the Tritonid origin of all Nudibranchs on the possession by
Tritonia, in common with the Pleurobranchids, of a large number of primitive Nudibranch
characters which are not found together in any other Nudibranchs. These are : — a
frontal veil, formed by the fusion of the oral tentacles of the Pleurobranchid, a wide
foot, a ventricle turned to the right, a broad radula, a nervous system placed on the
buccal bulb, an cesophageal crop, extensive salivary glands, a saccular, unramified
kidney, a long reno-pericardial tube, pericardial glands on the auricle, male and female
openings in a common vestibule, and a lateral anus. Of these, it will be noticed that

(ROY. soc. EDIN. TRANS., VOL. L., 206.)

THE ANATOMY OF A NEW SPECIES OF BATHYDOEIS. 285

Bathydons possesses all except the lateral anus. Further, it retains oral tentacles in
a condition more strongly reminiscent of those of Pleurobranchxa than that of the oral
veil of Tritonia, while the separate ganglia of the brain and the separate gonad of
Baihydoris can certainly not be regarded as new and derived features in that genus.
Thus, excluding the case of the lateral anus, which will be considered separately, Batliy-
doris, which is essentially Dorid in construction, exhibits all the primitive features of
Tritonia, some indeed being more primitive than the corresponding ones in Tritonia,
the supposed ancestor. At this reductio ad dbsurdwn we arrive by considering only
those primitive features selected by PKLSENEEH, without calling in the evidence of the
blood and respiratory systems wherein Tritonia, with its symmetrical auricle receiving
blood from symmetrical lateral sinuses, appears very modern indeed. It is in the complete
avoidance of any comparison between vascular and respiratory systems in Tritonia and
Pleurobranchids that the weakness of the Tritonid theory of Nudibranch descent lies,
and it is significant that on the characters of these very systems is primarily based any
discussion of gastropod and even molluscan affinities. Previous application of this
criterion in the Opisthobranchs has resulted in their cleavage into Tectibranchs with
a ctenidial gill and Nudibranchs with pallial outgrowths of varied form and distribution
replacing the lost ctenidium. Of these neomorphic gills the lateral tufts of Tritonia
have been regarded as an early type, but it is not clear whether the Dorid circlet was
derived from them by concentration or by local specialisation round a posterior anus or
was evolved independently. Nor is it clear why modern writers on the Opisthobranchs
have always accepted the neomorphic nature of the Dorid circlet. It is true that a
comparison of the highly specialised, multipimiate plumes placed in a pit in the
tuberculate dorsum of some Dorids provides no suggestion of homology with the
ctenidium of a Tectibranch ; but it is not such a comparison of extremes that evinces
homologies. In Batliydoris, however, the gill is in two portions only, joined by a
crinkled ridge, it shows but the beginnings of pinnation, its lobes have the broad laminae
running from the afferent to the efferent side seen in the ctenidium of the Tectibranch,
and there is no suggestion of the circumanal ring in either the -gill or the underlying
vessels. From this point of view the extreme similarity of the condition of the
auricle, the efferent branchial vessel, and the circular sinus in Batliydoris and the
Pleurobranchids acquires a special significance. Evidence derived from the nature of
the innervation is perhaps of doubtful value ; but, so far as it goes, it is favourable
to the present contention, since the Dorid gill is jointly innervated from pleural and
visceral centres, while other Nudibranch gills receive no visceral nerves unless invaded
by ramifications of the liver. The dorsal position of the Dorid gill should present little
difficulty, since the pallial edge of the Dorids is undoubtedly a new formation of
mechanical value which progressively increases in width within the group and is absent
in many genera. In any case, the same difficulty would apply to the anus and renal
pore, and there is no proposal to class them as new formations in the Dorids. The separa-
tion into two or more parts also forms no objection to the ctenidial nature of the Dorid

(ROY. soc. EDIX. TRANS., VOL. L., 207.)

286 MR T. J. EVANS ON

gill, since it is a progressive process in the group and is incipient in many Tectibranchs,
including Aplysia. Moreover, at least one Dorid, Trevelyana crocea, has a single un-
divided laminate gill indistinguishable from a ctenidium. Furthermore, the three residual
units of the pallial complex — namely, the kidney and its pore, the intestine with the anus,
and the gill — occupy in the Dorids just those relative positions which they would occupy
had they been turned over from the Tectibranch position into the median dorsal line.
Here one is inclined to ask what is the nature of the great cavity, cut off from the
underlying hsemoccele, which lies under the dorsal integument of the Dorid, but is
absent in all other Nudibranchs. An exactly similar cavity in Pleurobranchus or
Oscanius contains a shell-remnant, and is the shell-cavity. In the absence of any
information regarding the metamorphosis of the veliger of either Dorids or Pleuro-
branchids, it is difficult to find any satisfactory reason for contradicting the homology
of these two spaces.

It is on the above grounds proposed to define the Dorids as ctenidiate Opisthobranchs
that have retained the shell-cavity and in which the elements of the pallial complex
have moved dorsally into the median line. In this position the ctenidium has under-
gone progressive modification within the group, the retractile circlet being its highest
development.

In Tritonia, on the other hand, the residual members of the pallial complex have
remained in a more anterior position than they occupy in many Tectibranchs, and in
that position the old molluscan gill has been lost. Whereas in the Dorids and Pleuro-
branchids the connection of the auricle with lateral integumentary sinuses is supple-
mentary to the ctenidial connection, in Tritonia it is the sole remaining connection of
the auricle with respiratory sinuses. As a primitive actenidiate animal, however,
Tritonia retains many common features with the Dorids and Pleurobranchids. its
nearest ctenidiate relatives.

BIBLIOGRAPHY.

(1) BERGH, R., "Report on the Nudibranclriata," " Challenger" Expedition, vol. x., 1884.

(2) " Nudibranchiate Gasteropoda," Danish " Ingolf" Expedition, vols. ii.-iii., 1900.

(3) DRETER, T. H., " Uber das Blutgefass- und Nervensystem der Aeolididaaund Tritoniadae," Zeitschr.

f. wiss. Zool, Bd. xcvi., 1910.

(4) ELIOT, Sir CHARLES, A Monograph of the British. Nudibranchiate Mollusea, part viii., Kay

Society, 1910.

(5) " Report on the Mollusea Nudibranchiata collected by the Discover//," National Antarctic

Expedition Reports, 1907.

(6) GUIART, J., "Les Mollusques Tectibranches," Causeries scientifiques de la Soc. Zool. de France, 1900.

(7) - - Contribution a I'etude des Gasteropodes Opisthobranches, Lille, 1901.

(8) HANCOCK, "On the Structure and Homologies of the Renal Organ in the Nudibranchiate Mollusea,"

Trans. Linn. Soc. Land., xxv., 1865.

(9) HANCOCK and EMBLETON, "On the Anatomy of Doris," Phil. Trans., London, 1852.

(10) HECIIT, K., "Contribution a I'etude des Nudibranches," Mem. Soc. Zool. de France, vol. viii., 1895.

(ROY. soc. EDIN. TRANS., VOL. L., 208.)

THE ANATOMY OF A NEW SPECIES OF BATHYDORIS.

(11) HE LACAZB-DUTHIERS, H., "Histoire anatomique et physiologique du Plcurobranclie orange," Aim.

Sci, Nat. (4), xi., 1859.

(12) MOQUIN-TANDON, G., Recherches anatomiques stir I'Ombrelle de la Mt'dilerranee, These de Paris, 1870.

(13) PELSENEER, P., Recherches sur divers Dpi stob ranches, Gand, 1893.

(14) THIELE, J., "Die autarktisohon Schnecken und Muscheln," Deutsche Siidpolar Expei/>

Berlin, 1912.

EXPLANATION OF I'LATES I. AND II.
Bathydoris bronmii.

Fig. 1. The animal seen from the dorsal side, natural size.

Fig. 2. General view of the alimentary system from above, b.m., buccal mass ; «•., oesophageal crop ;
int., intestine ; /., liver ; ce., oesophagus ; p.m., lateral pads of the inner lips ; s.g., salivary glands.

Fig. 3. Buccal cavity laid open from above. J., jaws; o., odontophore ; OB., oesophagus ; P.J., pads of
the jaw ; p.m., pads of inner lips ; R., radula.

Fit;. 4. Portion of a half-row of the radula. L.'-L.T, lateral teeth 1 to 5 ; liJi., rhnchidian tooth.

Fig. 5. Stomach and adjacent parts of the alimentary canal laid open ; the cut is continued into the
posterior lobe of the liver, b.o., bands on the wall of the crop ; 1., liver ; int., intestine ; l.t., liver ducts ; s.r.,
gastric caecum ; st., stomach.

Fig. 6. Section across an oesophageal band, b.c., brown cones; rf., cuticle; ep., epithelial layer; m.,
muscle layers.

Fig. 7. Nervous system, b.co., buccal commissure; c.ij., cerebral ganglion; yen.y., genital ganglinii ;
g.o.n., gastro-cesophageal nerves; ped.g., pedal ganglion; ped.com., pedal commissure; p./n'i/.i-om., paraped:d
commissure ; pl.ij., pleural ganglion ; s.g.L, buccal loop; st.g.g., stomato-gastric or buccal ganglion; v.l., i:l.1,
visceral loop; o.n., visceral nerve.

Fig. 8. Kidney with thin dorsal wall removed, (i.e., aortic bulb ; b.g.\ b.g.li, lobes of the kidney to which
the blood glands are attached; c.art., cephalic artery; h.g., hermaphrodite gonad ; int., intestine; o.v.,
opening of aortic swelling into the ventricle ; /•.</., renal tube to exterior ; r.p.d., reno-pericardial duct ; r.s.,
renal syrinx.

Fig. 9. Arterial system, (i.e., aortic bulb; b.i/.\ h.g.", blood glands; c.art., cephalic artery; cer.art..
cerebral artery; buc.art., buccal artery; cj.art., genital artery; lab. mi., labial arteries; lac., lacuna- under
buccal mass; ped.art., pedal artery ; sal./j., salivary gland ; o.art., visceral artery.

Fig. 10. Diagram of the afferent and efferent vessels, a.b.v., afferent branchial vein; aw., auricle;
c.s., circular sinus ; e.b.v., efferent branchial vein ; h.v., hsemocoalic vessels; m.s., median sinus.

Fig. 11, a and b. Diagram showing the relation of the penis to its sheath in the retracted and protruded
condition.

Fig. 12. General view of the reproductive system. a.L, atrial lips ; a.y., albumen gland ; amji., ampulla ;
b.c., bursa copulatrix ; c.h.d., common hermaphrodite duct ;/.«., female opening; h.g., gonad : m.g., mucus
gland; o.d., oviduct; p., pitted pad on penis; p.s., penis sac; v.cL, vas deferens ; vg., vagina.

(ROY. SOC. KPIX. TIIAXS., VOL. I . JO! I.)

Scot Nat. Ant.Exp.

Vol. VII. Plate I

EVANS: NEW SPECIES OF BATHYDORIS.

..-p.m.

.h.m.

; rli.ri

, L'.ii.ii.

..Ct.
...rp.

' • ' M :; Aut.Exp

EVANS: NEW SPECIES OF BATHYDORIS.

Vol. VII.

r.|nl

s^

J. E.. DEL.

PART XII.
PULYCH^TA.

(SCLEROCHEILUS.)

XIL-A NEW SPECIES OF 8CLEROCHEILUS,

WITH A REVISION OF THE GENUS:
SCOTTISH NATIONAL ANTARCTIC EXPEDITION.

BY J. H. ASHVVORTH, D.Sc.,

Lecturer in Invertebrate Zoology in the University of Edinburgh.

(WITH ONE PLATE AND FOUR TEXT-FIGURES.)

On a New Species of Sclerocheilus, with a Revision of the Genus. By J. H.
Ashworth, P.Sc., Lecturer in Invertebrate /oology in the University of
Edinburgh.* (With One Plate.)

(MS. received M;iy 4, 1!)]-). I lead May 2f>, 1914. Issued separately July 2, 1915.)

CONTENTS.

Observations on a Specimen of " EUHU uia ncnlata" 303

i-H'-lirilim Grube, emend. .... 306

Sclerocheilus minutiis Grube . . . :jn7

Selerorheilus antarctic us n. sp 309

! iption of Plate 310

Description of the Specimen collected in Scoi.ia Ray,

South Orkneys ....... 293

Systematic Position ot'thu Specimen from Scotia Bay 297

Observations on Scli'roclieilus minittus ( Jrube . . 298

Observations on Sclerocheiltts pacificus J. P. Moore . 302

Sclerocheilus euros Saint-Joseph .... 303

The Polydin'te family Scalibregmidae comprises seven genera, the limits and inter-
relationships of which are, however, still imperfectly known. The present paper
results from a detailed study of one of these genera, a new species of which is de-
scribed, chiefly from a specimen collected in Scotia Bay, South Orkneys, and entrusted
to me for examination by Dr W. S. BRUCE.

DESCRIPTION OF THE SPECIMEN COLLECTED IN SCOTIA BAY, SOUTH ORKNEYS.

This specimen, the only Scalibregmid found by the Scottish National Antarctic
Expedition, was dredged on a stony bottom in ten fathoms at Station .->>25, in Scotia
Bay, South Orkneys (hit, (it)0 4:j' 42" S. ; long. 44° :}«' 33" \V.), in August 1903.

The worm, which is yellowish brown in colour (in alcohol), is ] '.) mm. long. The
anterior portion is broad ; the maximum breadth, 3 mm., is reached about the level
of the 10th segment; from this region the worm tapers gradually to the anal seg-
ment. The dorsal surface of the worm is strongly convex ; the ventral surface is
flattened, and there is a well-marked depression which extends alon^ the mid-ventral
line from the 2nd eha't ii'erous segment almost to the anus.

The prostomium is drawn out at cadi side into a stout, bluntly conical process
(Plate, fig. 1 ). On the middle region of the dorsal surface of the prostomium
there is a A-shaped area of dark-brown pigment representing the eyes. The point of
the A is median, and is situated near the centre of the prostoniiuin ; each limb of this
pigmented area passes obliquely backwards and down the. side of the prostoniiuin.

The perist om i urn consists of a single, aeha-tmis ring, which is incomplete yen
trallv, where it forms the antero-lateral borders of t lie mouth (tigs. 1, •_>).

* A grant in .ii'l Hi ill.' i-xp.-inr* inniriv.d dimn;,' lliis iv-caivli hi- lin-n rt-i-i-ivi-d from tin- Karl of Moray Endow-
ment of the University of Kflinlmrnli. Tin- cost, of the woodcuts and of ihe rrprodnrt.ion of the liguras on the jilate
h.i- liri'ii defrayed l>v tin1 Carnegie Trust for tlie Uiiiveraitii'" of So.tUnd.

, ••• (KEI'KINTKI) KKOM TIIF. Tit ANSACTIONS OF THK RdY.M. SOCIKTV OF Kill X IITHi ;H, VOL. L, PP. 405-4'2i.)

294 DR J. H. ASHWORTH

There are 43 chsetiferous segments, the last of which is small and evidently
recently formed ; this is succeeded by the anal segment or pygidium (fig. 5).

The first chretiferous segment is narrow mid-dorsally, wider at the sides, i.e.
where the parapodia are borne, and enlarged mid-ventrally, just behind the mouth,
to form a well-marked " under-lip," the front margin of which is lobate — about six
lobes being indicated (fig. 2). This segment consists of two annuli, the anterior of
which is much the larger and bears the parapodia.

The second chsetiferous segment (figs. 1,2) is tri-aunulate, there being a
small ring in front of and another behind the chtetiferous annulus. These smaller
rings are visible on the dorsal and ventral aspects, but are not seen in the lateral
view of the worm.

The third chsetiferous segment has a similar constitution, but the fourth is
composed of four rings — the chsetiferous annulus together with two rings in front
and one behind. All the succeeding segments up to and including the 41st are
also four-ringed ; the auuulation is much less clear in the next segment, and the
43rd segment is a single narrow ring bearing very small parapodia.

In the fifth and succeeding chsetiferous segments the annuli are subdivided on
the dorsal surface * by antero-posterior furrows, so that the skin is marked out into
more or less quadrangular areas, as is the case in other Scalibregmidse. This
sculpturing ©f the skin is well marked up to about the 23rd segment, but in the
following segments is exhibited only by the chsetiferous annuli.

The anal segment or pygidium is a short ring on which ventrally there are two
slight enlargements (one on each side of the middle line), of which the left bears a
slender cirrus about '5 mm. in length (fig. 5). Other cirri have been lost ; it is not
possible to say with certainty how many, but probably the original number was four.

There is no trace of gills in this specimen.

Parapodia.

The notopodia and neuropodia of the first segment are little elevated above
the general surface, but in the following segments there is right and left a prominent
elevation extending over the whole lateral region of the segment from each of which
the notopodium and neuropodium arise. From about the seventh segment onwards
the raised area presents papilliform elevations the epidermis of which is glandular.
There is usually one of these papillse anterior and another posterior to the lip of each
di;etal sac, that just behind the lip of the chsetal sac being especially well marked,
forming in most of the segments a prominent, rounded knob.

A cirrus is present on the posterior face of the 18th left neuropodium, and on all
the succeeding neuropodia up to and including the 40th. Cirri were probably
originally present also on the, 41st and 42nd segments, the parapodia of which are

' Tins subdivision of Lliu annuli is IVelily marked on the ventral surface.
(ROY. soc. EDIN. TUANS., VOL. L, 40(i.)

ON A NKW Sl'KCIKS OK SCLI'-IH" 'II I-.I I.I >'.

295

slightly damaged ami no longer bear cirri. In tin1 L8th ami
1'Jth cliR'tif'erous segment s t IK- neuropodia] cirrus is a short conical
process, '05-'08 inin. in length; lint those of the succeeding seg-
ments (lig. i;) rapidly increase in length, so that the cirri of the
3ord to -Kith segments arc linger-shaped ami ':2f>-':! nun. long
(rig. 5). There are no notopodial cirri.

A lateral si- use-organ is present in each parapoiliuni imnie-
diatelv ventral to the base of the notopodium ; but as it is small,
ami usually hidden in a depression, it can be seen satisfactorily
only in sections. The surface of the organ which bears the sense-
hairs is oval in outline, and its longer diameter is not more than
about 40//. This area is sunk below the level of the surrounding
epidermis of the papilla on which the organ is situated. Ventral
to the papilla bearing the sense-organ is a larger eleva-
tion the epidermis of which is glandular (fig. <>)•

ChsetSB.

The first notopodium bears ehfetse of three or
four different kinds, the relative positions and detailed
structure of which have been studied as far as has been
possible on the single intact notopodium available.

(i) There is an anterior series of about ten almost
straight chretre (text-fig. 1, A), approximately '5 mm.
long and 9-HV in maximum diameter. Each of these
chretee tapers rapidly in its distal fourth to a fine
point, and the preparations indicate that the tips of
these chfetse project little beyond the lips of the
chietal sac.

(ii) Close behind the cli;rt;e just described is a
series of about fifteen stronger chaetse, each bent in a
characteristic manner (te.xt-fig. 1. 15). These chretce
are '(>-'G5 mm. long, and their maximum diameter is
\n-\~2fjL. Kadi ch.-i'ta tapers somewhat abruptly at its
free end. and, when unworn, has along both sides,
tor a distance of ' I ."> mm. behind the fine-pointed tip,
a narrow and very delicate lamina which readilv
breaks up into a (dose-set scries of minute, pointed
processes, so that this region of the cli:t.-ta appears
to be finely spinctus. It is possible that these chaetse
arc really in two rmvs, an anterior row of about

TKXT-KII:. 1.— Cli.-i'tse fnnii tin
|»iiliuni uf tlio specimen fnuu
(x200.)

tirM. nolo-
S :otia l'.;iy.

(ROY. 80C. EU1N. TRANS., VOL. 1., 10,.)

I)K J. H. ASH WORTH

teli and a posterior row of about five ; but if so, the two rows stand very
close together.

(iii) Posterior to the foregoing are tapering, capillary chsetse, which appear to be
of two types, shown in text-fig. 1, C and D. There are about
forty chsetse of the type shown in fig. C ; each of these is
'9-1 '2 mm. in length and 7—!)^ in maximum diameter, and
tapers very gradually to a long fine point ; consequently the
distal portion of the chseta is very slender. There are fewer
chretse — about five — of the type shown in fig. D. These are
about '5-'G mm. long, and taper more rapidly than the pre-
ceding. They are originally laminate near their tips, at any
rate along one side, but the lamina is exceedingly fine.

The first neuropodium bears capillary chsetse like those
in the notopodium, and one chseta similar to that shown in
text-fig. 1, A, was observed, but there are no stouter, bent
chtetse like those occurring in the notopodium.

The chsetse of the second notopodium include repre-
sentatives of all the four types present in the first noto-
podium ; there is a single row of eight or nine of the bent
chsetse.

The second neuropodium contains capillary chsetse
similar to those of the first neuropodium, but there are also
present two of the peculiar furcate chsetse (see below), so
characteristic of the family Scalibregmidse. There are no
stouter, bent chsetse like those in the notopodium.

The armature of the third notopodium consists of (l)
an anterior row of eight or nine curved chsetse, similar to
those of the first two neuropodia ; (2) several furcate chsetse
standing near the former ; (3) a posterior series of capillary
chfetfe of two types, similar to those shown in text-fig.
1, C, D.

The third neuropodium bears capillary chsetse of the
usual two types and several furcate chsetre.

In the following segments the capillary ehretre become
TEXT-FIG. 2.-A. Furcate ch«tu considerably stouter and longer, attaining a length of 17-P8
x,ui;j mm. and a diameter of 12-1 3 /n, and the differences between
end of same chreta. t]ie ^wn types gradually disappear, so that in most of the

segments the two types are not distinguishable.
The furcate chsetse (text-fig. 2) form in both rami of the parapodium, from the
third onwards, a fan-shaped series situated at the base and in front of the capillary
chsetse. The. shaft of each furcate chseta tapers gradually in its distal portion to the

(ROY. soc. EDIN. TRANS., VOL. L, 408.)

('•'<

ON A NKW SPECIES OK SCLEROOHKJLUS. 297

base of (lie lurk. The prongs, \vliicli arc unci|iial in length, arc usually curved, their
fine tips pointing away from each other. In the anterior ami middle segments the
prunes are about 50/u ami :'>."i,« in length respectively, ami iu the posterior segments
:il >out 4(V and 25^. Kadi prong hears along the inner edge of its proximal tlirce-
fourtlis a series of regularly-placed pointed processes. Only tin- fork and a short
portion of the shaft proximal to it project beyond the lips of the chsetal sac. In the
middle region of the body there arc aliout eight to twelve furcate cha't;e in cadi of
the rami of the parapodia.

SYSTEMATIC POSITION' OF THK SPECIMEN FROM SCOTIA BAY.

The characters of the prostomium, parapodia, and di;rta', and especially the
presence of the distinctive furcate di.-ct:o, show dearly that the specimen described
above belongs to the family Scalibregmidse. It is referable to the Scalibregma-
section * of the family, for the body is sub-fusiform and the prostomium T-shaped,
its lateral angles being drawn out to form short tentacular processes. To this section
of the family belong the following genera : — Scalibregma Rathke, Pseudoscalibregma
Ashworth, Sclerocheilus Grube, Asclerocheilus Ashworth, and Oncoscolex Schmarda.

The Scotia Bay specimen differs from Scalibregma in several striking respects,
e.g. the absence of gills and dorsal cirri, the ventral cirri are much more slender,
and there are stronger choetre in the first and second notopodia, whereas such chaetse
are not present in Scalibregina. Further, Scalibregma rarely possesses eyes.t

The genus Pseudoscalibregma was suggested by the writer J to contain certain
little-known, abranchiate worms resembling Scalibregma in general appearance, but
their real relationship to Scalibregma remains to be ascertained. These worms
appear to agree in their external features with Scnfibrci/ina', except in regard to the
absence of gills, and so differ from the Scotia Bay specimen in the parapodial and
chsetal characters mentioned above.

The characters of Axderwlu'iliix are little known, but the single species —
Asclerocheilus imtervnedius (==Lip6bra,nchius § intermedius Saint-Joseph)— referred

* See the olassitication suggested liy the writer in ',"""''• -fonrn. Micr. Sri., vol. xlv (1901), p]i. 296, 297.

t I have examined about sixty specimen* of Nr,j//7,,v(//m/ varying in length from I nun. to .">*! mm., and including
five epitokons examples. Eyes are present in only t.wu specimens, both onlin.-irv non-cpilokoiis forms. For one of
llicsc spccimi'iis, which lias been in my possession twelve years, I am indebted to Dr E. J. AI.LKN, F.R.S., who
collected it near Plymouth ; for tin- othrr, colh-i'ted in 1!»11 .it (';i\< \,i-v\, near Cherbourg, I have to thank Professor
F.vrvEL, who, on finding that this specimen possessed eyes, kindly si-nt. it to me for examination. Moth specimens
are similar in size, lint only one — the Plymouth specimen — is complete; it, is 3d mm. in length. In both, the eye
are on the right and left sides of the prostominm, and are wide apart, i.e. do not approach each other like those of
Kderoclieilits. 1 have stained and cleared the Plymouth specimen, which exhibits on each side two eyes adjacent to
each other, composed of a series of closely associated simple eyes, which in section are found to lie similar in structure
to those of Sclerncheilit*.

I See Quart. J»nrn. Min: Xn., vol. xlv (1901), pp. 291, 292.

§ The genus Lipobranchiti* founded by Messr.- CCNXINCIIAM ami I;\MV;I to contain the species L. jrffnysii
( = Eumenia jeffrujfii M'Intosh) cannot well contain also /,. int, nii>-ilin< Saint-Joseph, wliich differs from the former
in several respects, but especially in the nature of the ch.i t.i of tin- first, and sei-emd notopodia. A thorough revision

(ROY. soc. KIIIN. TUANS., \oi. i., 409.)

298 DR J. H. A8HWORTH

to this genus lacks eyes and aeuropodial cirri, and in these respects, as well as in the
form of the anal cirri, and the chfetfe of the first and second notopodia, it differs
from the Scotia Bay specimen.

The structure of Oncoscolex is also imperfectly known, but, as represented
by the type-species 0. dicranoch&tus Schmarda, it differs from the Scotia Bay
specimen, for in the former the segments are composed of not more than three rings,
the parapodia do not bear cirri, and stronger chsetse are not present in the first and
second chsetiferous segments.

The specimen under consideration cannot, therefore, be referred to any of the
four genera just discussed, nor do its characters entirely agree writh those generally
accepted for the genus Sclerocheilus. The differential characters of this genus, as
represented by the type-species, S. minutus Grube, have been emended and stated
thus by SAINT -JOSEPH * : — Prostomium small, with thick lateral processes and with
eyes ; notopodium and neuropodium of first ehsetiferous segment bear capillary
chsetse and, a little above the neuropodium, five or six large, retractile, curved
acicula, together with an e<jual number of slender and shorter acicula ; the para-
podial rami of the other segments bear capillary and furcate chsetse ; small neuro-
podial cirri are borne on the last segments only ; anus surrounded by five or six
cirri ; gills absent.

I have recently re-examined the series of specimens of S. minutus at my dis-
posal, and have also, by the courtesy of Professor ANT. COLLIN, had the privilege of
examining the entire series of the original specimens collected by GRUBE, and
preserved in the Konigliches Zoologisches Museum, Berlin, and am able to modify
and supplement the characters given above.

OBSERVATIONS ON SCLEROCHEILUS MINUTUS GRUBE.

The prostomium (fig. 7) is drawn out antero-laterally into two stout processes,
and bears eyes dorsally. The eyes t vary in development in different specimens ; in
some they are comparatively small, and the right and left eyes are separated by a
distinct interval, but in other specimens they are larger, and in several cases are in
contact, or are actually fused in front, forming a A-shaped, brown pigmented area.
Near the posterior margin of these large eyes there is, on one or both sides, a simple
eye, about 9/x in diameter, consisting of a cup-shaped mass of pigment-spherules and
a spherical lens. This eye is situated below the epidermis, on or in the brain, and
is therefore not seen except in cleared specimens.

Between the prostomium and peristemium there is on each side a protrusible

of the genera Lipobranchius and Eumenia is required, and the relationship of L.jeffreysii to E. crassa should be re-
investigated. I have collected considerable material in view of undertaking such revision, but until I have the
opportunity of examining certain types, which have been imperfectly described, the work is necessarily at a standstill.

* Ann. Sci. Nut. Zoul., vii Stir., tome xvii (1894), p. 104 et seq.

t For an account of the structure of the eyes, see A. et L. DEHORNE, Arch. Zoul. Exper., tome liii (1913), pp. 85-90.

(HOY. soc. EDIN. TRANS., VOL. L, 410.)

ON A NE\V SPKC1KS OF SOLEROCHEILUS.

299

nuchal organ, which, when everted, forms a sausage-shaped mass of considerable
size (see fig. 7).

The peristomium is ;i single ach;etous ring, incomplete ventrally.
The first two or three ch;etiferous segments are usually lii-anmilnte, but the re-
maining segments, except a few of those last formed, are subdivided into four rings.
The stronger, curved dia'tie ("acicula" of SAINT-JOSEPH) on cadi side of the first,

B

TEXT-FIO. 3. — A, Pi, 0, D. Ch:i-tic from the first notopodium of a specimen of Sclerorhfilus

niiitii/iis, l'S'5 mm. long. (x200.)

B'. A clurta which hud mil i;ome into use, from tlif lirst notopodium of a
s|irnimcn 11 mm. long. ( x300.)

duutiferous segment are undoubtedly part of the armature of the notopodium, and
form a portion of the anterior series of ch;rta>. The following is a brief account of
the chief eha-tal characters resulting from a stud}' of specimens I I to 18'5 mm. long.

The first notopodium contains three kinds of chietse :—

(i) Strong and almost straight eha-ta', which taper distally to a fine point
(text-fig. 3, A). Kaidi of these cha'ta- is about '."> mm. in length and I f>-2(V in maximum
diameter, ami in all the specimens examined only the tips * of these chsetse pi'oject

* Nut nidri' th.'Ui alunil f|i| nun.
(ROY. SOC. KOIN. TRAMS., VOL. I., 411.)

300 DR J. H. ASH WORTH

beyond the lips of the chfetal sac. There are generally four to six fully formed
chsetse of this type, and two to five in course of formation, constituting the most
anterior row of choatse in the notopodium.

(ii) Stronger chsetse (the "acicula" of SAINT-JOSEPH), which lie in a row close
behind the dinette just described. These cliEetse (text-fig. 3, B), which are about
'5-'55 mm. long and attain a diameter of 24-28^, are curved distally, and project
considerably beyond the lips of the chsetal sac (Plate, fig. 8). Each chseta usually
presents a blunt tip, but if one of these chsetse be examined before it has come into
use it is found to be pointed at the tip and to be there enveloped by a flat sheath
(visible as a delicate lamina on each side of the chseta), which also runs out into a fine,
flexible tip (text-fig. 3, B'). When the chseta conies into use, the tip of the sheath is
worn away ; then the laminae break up into minute pointed processes and soon
disappear, and the tip of the chseta itself becomes blunted, assuming the form shown
in text-fig. 3, B. There are usually three to five of these chsetse in use, and two or
three in course of formation.

(iii) Capillary chsetse, the maximum diameter of which does not exceed 4/x
(text-fig. 3, C). These taper gradually to fine points. From eight to fifteen fully
formed chaetse of this type, each about '5— '55 mm. long, are present, together with a
few in course of formation. In one specimen the first notopodium of each side bears,
among the capillary chsetEe just described, one or two slightly stouter and shorter
bristles (text-fig. 3, D) which taper more rapidly at their free ends. These chsetse
are about '3 mm. long and 5/x in diameter.

The series of stronger chsetse (text-fig. 3, A, B) arise in two closely apposed rows
near the bottom of the large chsetal sac, while the slender chsetse (C, D) arise more
dorsally and on the posterior wall of the chaetal sac. The two series of stouter
chsetee stand well in front of the slender ones. The latter usually point backwards,
while the strong, curved chsetse (the straight chtetse (A) in the anterior row are
usually scarcely seen on external examination of the specimen) point in a quite
different direction, either antero-dorsally or latero-dorsally. As the stout and slender
chsetfe are so different in appearance and direction, and are comparatively widely
separated from each other at their exits from the chsetal sac, it is scarcely surprising
that the curved chsette have not been considered to be part of the notopodial
armature, but have been referred to a region between notopodium and neuropodium,
the notopodium being regarded as having capillary clisetae only. There can, how-
ever, be no doubt that the strong chsetse belong to the notopodium. The lip of the
rli;i't;il sue immediately to the outer side of the curved chsetse is very well developed,
forming a prominent feature of the first chsetiferous segment (fig. 8).

The first aeuropodium is a simple conical elevation bearing capillary chsetse
only. These appear to be arranged in two rows — an anterior row of about eight or ten
cli;eUi> similar to that shown in text-fig. 3, D, and a posterior row of thirty or forty
longer and more, slender c.liaetse like that represented in text-fig. .'», (J.

(ROY. SOC. KIIIN. TRANS., VOL. L, 412.)

ON A NEW SI'KCIES OF SCLEROCHEILVS. 301

The armature of the second notopodium and neuropodium is composed of

two series of capillary eh;i>t;r. of about ten ami forty re-peel i\ ely. similar to lint
rather longer than those in the first neuropodinm. The ch;et,-e of tin- anterior
ro\v are proportionately longer, and hence there is less difference in length between
them and the ch;rta- of tin- posterior ro\v. In several of the specimens examined
there are in the second notopodiuni or neuropodium, or in both, two, three, or four
furcate chaetae, but these are neither so large nor so well formed in regard to their
prongs as those found in the following segments. They are very fragile, and
perhaps this explains \vliv they are not seen in preparations of the second parapodinm
in all cases.

In the notopodiuni and neuropodium of the third and following segments the
capillary chteta? are similar to those of the second parapodium, lint in addition there
is an anterior series of furcate chsetse the prongs* of which agree almost exactly
in size and proportions with those shown in text-tig. 2 (p. 29G), but the shafts are
rather shorter.

The parapodia of N. mi until* are very similar to those of the Scotia Bay
specimen; the figure of one of the latter (fig. <>) would serve equally well for a
parapodium from about the 36th segment of N. minutus, except that the latter is
smaller (about three-fourths the size shown).

SAINT-JOSEPH states that behind the 22nd segment the parapodia. bear nenro-
podial cirri, and u similar statement is made by Professor M'lxxosH and by
.M. and Mme. DEHORNE. I have, examined fifteen well-preserved specimens from
different localities — the Adriatic-, Saint-Vaast, Plymouth, and the west coast of
Ireland — and find that the most anterior neuropodial cirrus is situated on the 25th,
•_'7th. 2Uth, or 31st chsetiferous segment in the various .specimens. Of five specimen-
from the Adriatic (collected by GRUBE) in which the neurocirri are preserved, three
bear minute cirri on the 2.~>th segment ; in the other two the first neurocirri are on
the 27th segment. Of the four specimen- from Saint-Yaast, two have the first
neuroeirrus on the 27th segment, the other two on the :» I st segment. The specimen
from Plymouth presents its first neuroeirrus on the 27th segment. Of the five
specimens from the west coast of Ireland, two have the first neuroeirrus on the 2!)th
segment, and three on the :!|st. In <!I;VJBE'S figure t of N. minutus the first neuro-
eirrus is shown on the :'.()th ehsetiferons segment. As (!IU:I:K^ remarked, the cirri are
not respiratory structures, for blood does not enter them. They are purely sensory,
as the presence upon them of sense-hairs indicate.-, and are composed of epidermal
cells surrounding a thin axial strand, formed apparently of nerve fibrils.

• >ne of the neuropodia examined, from a specimen IV 1'dack-od Bay. Co. -Mayo,

* In a neuropodium from one nf the specimens from Blacksod Hay, Co. Mayo, Ireland, there occurs a fun-ate
rh;i'ta with an additional (third) prong, arising from between the bases of the I wo normal ones. This prong is shortrr
than either of the normal ones, the lengths being — normal prongs, 45^ and 27^ respectively ; additional prong, lsu.

t Arcli.f. Xuturg., Jahrg. xxix, Bd. i (1863), Taf. v, fig. 3.

J 46 Jahresber. Schles. Ges. (1868), 1869, p. 67.

(ROY. soc. F.DIN. TRANS., VOL. i., ii:< i

302 DR J. H. ASHWOETH

Ireland, possessed, in the usual position, two cirri practically equal in length and
arising close together.

A small lateral sense-organ is present just ventral to the base of each
notopodium, i.e. in a position corresponding to the " Seitenorgan " of Scalibregma,
and its retractor muscle is usually related to the notopodial musculature. Each
sense-organ is (in the larger segments) a convex elevation of the epidermis, about
5<V long and 20,u wide, traversed dorso-ventrally along its middle by the " hair-field "
bearing long, delicate sense-hairs.* Between the sense-organ and the neuropodium
there is invariably an epidermal papilla of considerable size, the cells of which are
glandular, some of them (" bacilliparous follicles," CLAPAREDE) producing curious rod-
like secretions, usually sinuous in form.

The number of cha?tiferous segments in the complete specimens examined
varies from 44 to 52.

The anal segment or pygidium bears long cirri, slightly thickened at their
distal ends (fig. 9). These cirri are solid, like the neurocirri, being composed of
epidermal cells with an axial strand of connective tissue and nerve-fibrils. There
is evidently some variation in the number of the anal cirri. Of the six specimens
with complete anal cirri which I have examined, one has five and each of the others
four cirri. GRTJBE described and figured four cirri in his original account of the
species ; SAINT-JOSEPH states that there are five or six ; and Professor M'!NTOSH
and M. and Mine. DEHORNE found five in their specimens.

The Scotia Bay specimen agrees with Sclerocheilus minutus in the form of the
prostoinium and the presence thereon of pigmented eyes, in the division of the
segments into four annuli, in the general characters of the parapodia and in the
presence thereon of lateral sense-organs, in the possession by the posterior segments
of neuropodial cirri, in the presence of anal cirri, and in the absence of gills. The
stronger, bent cha?t;e present in the first three notopodia of the Scotia Bay specimen
are evidently homologous with the much stronger chfetce of the first segment of
N. minutus, and the three other types of cheetfe present in the first notopodium of
the Scotia Bay specimen have their homologues in the corresponding notopodium
of S. mini//!/*. These agreements in structure show that the Scotia Bay specimen
is closely related to N. minutus, and should certainly be placed in the genus
Si-li'rodii'il a* as a new species, for which I propose the name »S. antarcticus.

OBSERVATIONS ON SCLEROCHEILUS PACIFICVS J. P. MOORE.

Only two species of Sclerocheilus have hitherto been described — namely,
S. m'nnitiix, the chief external characters of which have been already stated, and
N. /><«•/ fii-ii* J. P. Moore, 1909, from Monterey Bay, California. By the courtesy of

* Since this was written the sense-organs have been briefly described by M. and Mme. DEHORNE in Arch. Zool.
Exper., tome liii (1913), p. 72.

(ROY. soc. EDIN. TRANS., VOL. L, 414.)

ON A NEW SPECIES OF SCLEROCHEILUS. 303

Dr J. PERCY MOORE I have been able ti> examine the co-type of the latter species.
The specimen is not in good condition. and I cannot add much t<i the account which
Dr MOORE * lias given of the species, lint my examination of the specimen leads me
to the conclusion that it should not lie placed in the genus Sc/i'i-ix-lif/'/n*. Dr MOOUK'S
specimen differs from N<7ov><7/<'/7"x in the following character: — (l) the absence of
Stronger chsetse in the first notopodium ; (l!) the alisence of neuropodial cirri; (•">)
the segments are not four-ringed lint three-ringed. It is, of course, possible
that neuropodial cirri were originally proent and liave lieen lost, liut had they
been destroyed there would, I think, have lieen more evidence of damage to the
parapodia. Further, Dr .Moo UK states that in the larger and better-preserved type-
specimen there are no neuropodial cirri, and it may therefore lie concluded that
these organs were not present in life. Dr MOOKF.'S species seems to be much more
nearly related to the v,enus Oncoscolex than to Sclerocheilus, lor it agrees with
the former genus in : — (l) the segments from the 5th to about the 30th are tri-
annulate, those further back (as far as about the 50th t) are bi-annulate : (L!)
the absence of parapodial cirri; and (3) the absence of stronger chsetee in the tir>t
notopodium. So far as I can see, N. /^r//?r//.s differs from Oncoscolex dicranochaetus
only in the shape of the eyes,t a difference which is probably of little account.
Until I have had the opportunity of examining the better-preserved type-specimen.
I am not prepared to give a final opinion on the systematic position of 8. /»/<•* //ms.
but the information at present available indicates that the species should not be
referred to the genus Sclerocheilus, and I believe it will prove to belong to the
genus Oncoscolex.

SOLEEOCHEILUS CJECUS SAINT-JOSEPH.

SAINT- JOSEPH § recorded the capture of examples of N. csecus, but this species
never described, the onlv information given about it being contained in the si
phrase, " Sfli'>-ni-li<'!liis C.VC/'N, ditferaiit seiisiblemeut du Sclerocheilus minutus (!r.
N. r,rr//.v is therefore a nomen //"'/"///, but there can be little doubt that the
specimen- referred to were those snl)M'ijueiitlv described by SAINT-JOSEPH under the
name Lipobranchius inter

( IUSKUVATIOXS ox A STKCIMKX OK L' A'r.i/A'.v/.i OGULATA."

The worm recorded by Dr ('if. ( JI;AVIK.I: *' as En-menia or"/"/" Mhlers appeared to
me to be closely related to the Scotia Hay specimen. Dr < ii:.\viK.n has kindly lent

* I'm: Aanl. Xat. Set. liu'/'i,Ulj,lt;<i, 1909, p. iJsj.
t The specimen has 62 cluutiferous segments.

I Which are approximately round or oval m 0. dicranochsetus, m-i inv^iil ,rh triangular in 5. poe(/?cu».
§ C. R. Acad. Sci. Paris, tnim- ci (1885), p. 1511.
|| See above, p. ln!i.
IT Deuxieme iVy.W. .\ntnrct. f'nnir. : Am^'lnl. s I'uiijdi.l'S, lllll, p. 111'.

(ROY. SOC. EDIN. TKANS., VOI. I., II . I

304 DR J. H. ASHWORTH

me his specimen, and I am alile to add some details to those given in his short
account of it. This worm was taken at low water on the east coast of Petermann
Inland (lat, 65° ll' S., long. C>4° 10 ' W.), off Graham Land, on 31st October, 1909.
The specimen is in two portions ; the, anterior region is moderately well preserved,
but the posterior portion is very fragile and is no longer intact behind. The worm
was originally 11 mm. long. Its maximum width is 1'4 mm.

The prostomium is only partially seen, for both nuchal organs are everted, form-
ing two lobulated masses which conceal a considerable portion of the prostomium
(fig. 3). The antero-lateral processes of the prostomium are stout and rounded at
their tips. The eyes, which are present on the dorsal surface of the prostomium,
diverge as they pass backwards ; they are not in contact at their anterior ends.

The peristomium consists of a single achaetous ring which is broader ventrally,
where it is incomplete (tig. 4).

The first and second chaetiferous segments are each bi-annulate, the third
.•I nd fourth show indications of subdivision into four rings, and the fifth and
following segments, as far as they are still present (i.e. up to the 28th ; there
were originally 34 chffitiferous segments), are clearly four-ringed. The rings are
subdivided, on the dorsal side at any rate, into quadrangular areas.

The posterior end is now wanting, but Dr GRAVIER states that there were four
short, slender, anal cirri.

The contours of the parapodia are not well preserved, but they were evidently
of similar form to those of the Scotia Bay specimen (cf. fig. 6). Finger-shaped
neuropodial cirri are present in the 21st* and following chsetiferous segments.

There are no gills.

The specimen is not sufficiently well preserved to have retained its lateral sense-
organs, and these have, therefore, not been looked for.

Each of the first three notopodia bears, in front of the slender, tapering chsetee, a
number of stronger, shorter, bent chsetae, which taper much more abruptly at their
free ends. In the second notopodium, which I excised for examination, there are
four kinds of chsetae practically identical in form with those described from the Scotia
Bay specimen (see pp. 299, 300). The chsetse from Dr GRAVIER'S specimen are,
however, smaller, being about two-thirds the length and thickness of those of the
Scotia Bay specimen. There are several chsetse like that shown in text-fig. 1, A;
these arc about '35 mm. in length and 6,u in greatest diameter. Nine stronger, bent
chtetfe (text-tig. 4) are present ; these are about '4f> mm. long and 9^ in maximum
diameter. The longer capillary chfetse (cf. text-fig. 1, C), of which there are about
thirty, arc approximately '8 mm. in length and G/* in greatest diameter ; among these
are several shorter capillary chietfe (cf. text-fig. 1, D) about '5 mm. long. I have not
thought it necessary to figure all these types of cha?tte, but a figure of one of the

h Assuming, as is apparently the case, that there are no segments missing between the anterior and posterior
1 mi-til ins into whir] i the specimen' is now divided.

(ROY. soc. EDIN. TRANS., VOL. L, 416.)

iN A NEW SPECIES OF SCLEftOCHKILUS.

305

bent chastse is given to show that it is practically identical with tho-c of the Scotia

P>ay specimen (<;/! text-fig. 1, 11, p. •_".>">, and text-tig. 4).

The first neuropodium bears capillarv ch;et,-i- of l he t \\ o t vpcs shown in text-
fig. I. ('. D. The third notopodium and neuropodium were not excised lor examina-
tion, but the former contains an anterior row of eighl stronger, bent
chaetre, like those in the first two notopodia.

The fourth and following pa ra pod ia. w hidi do not bear -Wronger,
bent ehaetae, contain in each ram us capillarv ch;et;e and about half a
do/en furcate cheetse identical in form and in the dimensions of their
prongs with those described on p. 'Jilii. but their shafts are shorter
(about '4 mm. long).

I have recentlv had the opportunity of examining the tvpe-
specinieii of /''n ii/i'iiiu m'/i/'t/a Ehlers. : which was sent to me from
the Koiiigliches /oologisches Museum. Berlin, through the kindness
of ProfesMir A XT. COLLIN, and I find that Dr ( JKAVIKR'S specimen does
not agree with the tvpe. for in the latter most of i he chsetiferous
segments are three-ringed, stronger chaetse are not present in the
anterior notopodia, and there are no neuropodial cirri. /:'. »i-n/nt<t
lias been placed by Professor EHLERS i as a synonym of Oncoscolex
<!i<-r»n<«-/i;i'f/i!< Sehmarda, after he had examined the type-specimen
of the latter.

Direct comparison of Dr (JitAViEii's " Knnn'mu oculata" with the
specimen collected in Scotia Bay, allowance being made for thedis-
paritv in si/.c and age. shows that they belong to the same species.
The di.-i't.-r of the two specimens exhibit an agreement in structure
which is particularly striking, and prove that the specimens are
specifically identical. There are only two slight difference- between
the specimens: — (1) tin- eyes of the Scotia Bay example are in

contact, while those of Dr ( i l: A VIKl;'s specimen are separate, but a

similar variation in the grade of development of the eyes is met
with ill ditl'ereiit individuals of SclerOcheiluS nl'muliix (see p. •_".)«) ;
(•_') the lir-i neuropodial cirrus is borne on the I Dili duet ifcrotis TF.XT-FIG. 4.— Bent

I Cl j.1 T» • 1 • \ \ ch:i la flOlll tin1

segment in the Scotia Bay specimen, and on the L'lsl in Dr second notopodium

GRAVIER'S, but a greater variation is met with in N. minutus, ^ ''"*"

where the tir>t iteiirocirrus may be borne by the L'.'ith. 'J7tli. L".»th, (xyfta^fig.**0]^

or :Jlst chu'tiferou- segment (-ec p. •'!(»!). A minute comparison P- 295.)

of the two specimen- shows that I hey arc both examples of the same SpecieSj

N. antarcticus, n. sp.

•* /mil. .I'lLfii., Sii]i[>l. v (11)1)1), ji. ^i;.",. A I'lillrr 'li-.-i-i-i]itinti ni' llic .-] .i-i'i iii.-ii (ft <] li'i-t i'i I at Tun 1 1 '(_•-, ( '1 1 i lr) \va -

III.KKS iii l-'e.<t*i-lu: A". . 1 '.>< '1, ]pp. is], ls-2.

t Al'lniiiill. h'. ' Math.-Phyi. A'/., X.R, IM. iii, No. 1 (['.mi), p. r,\ • X,,ii,,,,,,i Antarcti, I

Nat. //..-•'., vol. vi (1912), p. L'<; ; //. „ . 1:1111-1903, Kd. xiii, /,<><•]., v(1913), pp. 537, 538.

I KoY. SOC. KMN. TKAXS., VOL. L, -417.)

306 DR J. H. ASHWORTH

The only other Scalibregmid with stronger chsetse in the anterior notopoclia
which calls for mention here is that described by Dr ATJGENER * under the name
Oncoscolex (Eumenia) heterochastus, in which the first three notopodia bear stronger,
bent ehsetse. The description is not as complete as could have been desired, but it
affords sufficient evidence — the segments are three-ringed, and neuropodial cirri and
eyes are absent (or at any rate not mentioned) — to indicate that this worm is not, in
spite of its stronger chastse, a Sderocheiln*.

The foregoing considerations show that the genus Sclerocheilus is represented by
only two valid species — S. minutus Grube and the new species, S. antarcticus,
described in the present communication. The diagnosis of the genus Sclerocheilus,
as given by GRUBE, and emended by SAINT-JOSEPH, requires some further emenda-
tion on the addition of the new species here described. The diagnosis t may be given
as follows : —

Sclerocheilus j Grube, 1863, emend. Ashworth.

GRUBE, Arch. f. Natunj., Jahrg. xxix, Ed. i (1863), p. 50.
SAINT-JOSEPH, Ann. Sci. Nut. Zoo!., sur. 7, tome xvii (1894), p. 103.

Abranchiate sub-fusiform Scalibregmidae ; the prostomium is drawn out into two
blunt antero-lateral tentacular processes, and bears, dorsally, eyes which are either
separate or are united to form a A-shaped pigmented area ; the peristomium is a
single acheetous ring ; the chsetiferous segments (except the first two or three and a
few of the most posterior) are subdivided into four annuli ; the two rami of the
parapodia are similar in form, and are simple oval elevations arising, in most of the
segments, from a glandular raised area of the integument ; stronger, bent cheetse are
always present in the first notopodium (in S. fuifarcticus they are present also in the
second and third notopodia), but not in the neuropodium ; in the 3rd and all
succeeding chsetiferous segments each ramus of the parapodium bears slender
capillary chsetae and, anterior to these, furcate chsetee ; the segments of the posterior
region possess conical or digitiform neuropodial cirri, and the anal segment or
pygidiuni carries four or five (or six) digitiform anal cirri ; a small lateral sense-
organ (often recognisable only in thin sections) is present on each parapodium
between the notopodium and neuropodium, but nearer the base of the former.

Type species, S. minutus Grube.

* Hull. Mux. Uij-niji. Zuol. Harvard, vol. xliii (1906), p. 159.

t The external features only are considered ; tliere is not sufficient material of 8. antarcticus to enable me to
furnish a description nf llie internal organs, or to state if any "f these may be employed as generic characters.

J inc\-t\p&s, hard ; xeiXos, lip. A name derived from the same roots, but spelt slightly differently — Sclerochilus —
was employ, >d soon afterwards by G. 0. SARS to designate a new genus of Ostracods of the family Cytheridae. See
Forhandl. Videiisl,:-M.ik. I'hrMania (1860), 1866, p. 89.

(ROY. soc. KDIN. TRANS., VOL. L, 418.)

ON A NK\V SPKCIKS OF SCLEROCHE1LUS. 307

The two known species may l>e separated thus: —

1. Stronger chaetse prc-ent in iirst notopodium only ; tin1 -cement which bears the
first iicuropodial cirrus varies from the 2.5th to the .".1st . N. minutus, p. 307.

2. Stronger cha-tsi- are present in the first, second, and third notopodia. Init are
not so proportionately stout as in minutus; the iir-t aeuropodial cirrus is present on
the hstli to the 21st segmenl . >'. antarcticus, p. 309.

Sclerocheilus minutus (irube. L863.

GRUBE, Arch. f. Naturg., Jahrg. xxix. IM. i (1S63). !'• SO; 'I'"1'- v, »\'- 3 (Lussin piccolo, Crivi//.,i,

Xeresine, all on Lus-in Island, Ailriatir).

GRUBE, /'/• Insel /.//x.</« and i'//re J/ee/w/V<«B<i, Lireslau, 1801. p. s.~i (Ncn-sim., Ciivi^a, Civile).
GRUBE, Abltandl. S'-h/>-*. Ges., Natuna. M't. (1868), isil'.i, pp. 105, 127 ( Saint- Vaa&t).
(iiiUBE. 46 Jahresfar. M,l.-t. (,V.--. (1868), 1869, p. G7 (note Oil rim).
ASIIWURTH, Uitfirf. Journ. Micr. >'<;/., vol. xlv (1901), pp. 293, 2'.i7.

lUciniKNK, A. et L., Arrh. Zool. A'.'7»V., t.iuie liii (1913), p. 61 (Le Portel, near K'-u^yne).
M'lNTOSH, Ann. May. X<it. ///.<, ser. 8, vol. i (1908), pp. 3sn, :!8l (Cuernsey, Menu).
Marine Biol. Assoc., Journ. Mar. ISinl. Awe., N.S., vol. vii (1904), ]>. i':'.l (I'lymouth).
MARION et BOBRETZKY, Ann. Sei. A"'</., s^r. 6, tome ii (187"i), p sd ((iulf of Marseilles).
MARION, Ann. Sri. Nat. Zool., ser. G, tome viii (1879), Art. X<>. 7, p ~> (off Marseilles).
PRUVOT, Arch. Zool. A>;><r., S/T. 3, tome iii (1895), p. 043 i lianynls) ; «/'• cit., tome v (1897), p. 16,

"Catal. ties Invertebn's, etc. " ( I'.rittaiiy ; Gulf nf Lyons).
SAINT-. IOSKPU, Ann. Sr.i. A'nf. ZuoL, s.:i'. 7. tnme xvii (1S94), ]>. 103 (l)iuaril ami Saint-Malo) ; uji. cit.,

ser. 8, tome v (1898), p. 213 (Saint- Vaast) : op. cit., bum: x (1899). p. 1G4 (Paimpol); op. cit.,

s«-r. 9, tome iii (1906), pp. 147, '2:!0 (Cannes).

A Nc/c/v»(7/f//(/.s- hearing in the tir.-t notopodium — (l) stout lint stfai^lit. tapering
chreta? ; (2) very stronu eh;i-t;r eiirveil at their tips; (.".) slemler eapillarv eha'tie.
Stout eh.-i'tje are not present in anv of the Mieeeeilin^- notopoilia or in the neuropodia,
the armature of which consists only of slemler. capillary clm't;e and furcate chsetse.
Neiiro|poilial eiri'i appear alioui i he -_'7tli segmenl (rarelv anterior to this, and in
some specimens not until the 31st segment), and arc pre-ent in all the succeeding
chsetife rous >(-^ments (except iii one or two of the last formed).

HISTORICAL Accorxr.- >'. minutus was described h\- (iuniK trom specimens
collected liv him on the shores of the island of Liissin. in the . \driatii-. He, howexer.
mistook the dorsal for the ventral surface. The generic- name is liased on another
misconception, for the two dark plates dc~oril>ed l>v <!r. n:i-: ("-ulitus ad os laminis 2
ni'jri- eorneis tricuspidibus armatus") as liornv and mi tlie ventral side of the head-
lolie are the eyes which arc situated on the dorsal side of the prostomium. The rest
of the description and the li'jiiie- are admirably clear, and the species was so well
defined livtiurr.K that -uli-diiieni workers have had no difficulty in recop-nisino it
when it came into their hands. (Ji:i r.i. also, with his usual insight. iecOLrnised the
close afKnity of his new ^eiius with S,-<ili1>n'<jin<t.

(ROY. soc. tuix. TRANS., VOL. L, 41'.).)

308 DR J. H. ASHWORTH

The late Baron DE SAINT-JOSEPH revised and extended the description given by
GRUBE, and added observations on some of the internal organs, e.g. the alimentary
canal, nephridia, and reproductive organs. Quite recently M. and Mine, DEHORXF,
have given an account of the morphology of this species, dealing especially with the
nervous system, eyes, and nephridia, with a discussion of the nature of the funnels of
the latter. The other authors cited have either simply recorded the capture of
specimens or have briefly referred to their external features. (For an account of
the external features see pp. 298—302.)

BIONOMICS. — S. minutus lives in shallow water, and has been obtained either by
shore-collecting between tide-marks or by dredging. The greatest depth from which
it is recorded is 33-35 fathoms (at Cigale ; GRUBE, 1864).

This worm has been frequently found between the lamellse of old oyster-shells,
and SAINT-JOSEPH suggested that the strong chsetae of the first segment served to
excavate a shelter for the worm between the lamellfe. On emerging from its retreat
S. minutus swims in the water by means of rapid wriggling movements, but soon falls
to the bottom and lies on its dorsal surface with the anterior and posterior ends
raised and the intervening portion curved. The worm probably feeds on mud and
fine debris.

COLOUR. — S. minutus is usually reddish brown in colour, but may be a uniform dull
brick-red (M'lNTOsn). A female specimen full of eggs was greyish white (DEHORNE).

SIZE. — This species has not been found to exceed 20 mm., and most specimens
are from 7 mm. to 15 mm. in length. The largest specimens seen by the writer are
18'5 rnrn. long ; the smallest is one 2'2 mm. long, taken on the surface of the sea oft'
Sark, and kindly lent to me by Professor M'lNTOSH.

DISTRIBUTION. — I have examined specimens from most of the localities from
which >S'. minutus has already been recorded — Lussin Island (Adriatic), Plymouth,
Guernsey, Herm, Dinard, and Saint- Vaast.

Two additional localities may be given here. There are, in the Konigliches
Zoologisches Museum, Berlin, several specimens collected by GRUBE on the shores of
Lesina, an island about one hundred and thirty miles south of Lussin. Through the
kindness of Dr R. F. SCHARFF and Mr R. SOUTHERN, I have examined a collection of
Scalibregmidse from the west coast of Ireland, in which there are specimens of
N. minutus from Blacksod Bay and Clew Bay, Co. Mayo. These are the first recorded
Irish specimens, and they extend the area of distribution of the species north and
west of its previously known range.

The range of distribution may be stated thus : — S. minutus occurs on the west
coast of Ireland, in the English ( 'hannel,* along the south coast of France, and on the
eastern coast of the Adriatic,

Type-specimens in Kiinigliches Zoologisches Museum, Berlin.

It will be observed that there is no record of the species from any point in the North Sea.
(ROY. soo. EDIN. TRANS., VOL. L, 420.)

ON A NEW SPECIES OF SCLEROCHK1LUS. 309

Sclerocheilus antarcticus, n. sp.

(Synonym : Euinenia oculafa Gravier nee Ehlers.)
Kumenia oculata. Gravier, Deusieme Expi'd. Antard. Frani;.: Anw'Mes I'nlychHes, 1911, p. 112.

A Sclerocheilus bearing stronger, bent dia't;r in tin1 first, second, and third noto-
podia, lint these arc not so proportionately stout as in >S. num/tux. Stouter, brut
eluvt.-i1 an- not present in the succeeding notopodia or in the neuropodia, the arma-
ture of which consists only of slender capillarv rhieta1 and furcate ch;et;e. Neuro-
podial cirri are present on all the segments (except one or t\vo of the last formed)
from about the I Nth or -Jlst.

Only two specimens are known. One. the type-specimen, 11) mm. long, dredged
in 10 fathoms in Sent ia Hay, South Orkneys (see p. "J93) ; the other, the para-type,
11 mm. long. collected at low water on the east : roast of Petermaiin Island, off Graham
Land (see p. 3(i:i).

Type-specimen in Scottish Oceanographical Laboratory, Edinburgh.

1'ara-tvpe in Museum d'Histoire Naturelle, Paris.

Sclerocheilus rn'c»* Saint-Joseph is a uuntm /i>nl>/i/> (see p. 303), and N. pacificus
Moore is apparently not a Solerocheilus, but should be referred to the genus Oncoscolea
(see p. 304).

ni (Jam* -Jl, I'Jlo). — While this paper has been lying in type Mr
SOTTHKKX has recorded the specimens of Srt<>r<>r/i<'il>ix mhmtux from Blacksod Bay
and (.'lew Bav, mentioned on the opposite page, in /'roc. f\. Ii-ixli Acrid., vol. xxxi.
('/ni-i' Ixlniul Surri'i/. pt. xlvii (I'.)I4). p. 137. Prof. M'lxTosn has given a general
account of S. minutus in his Monograph of British Annelids, vol. iii, pt. i (1915),
p. 4:2. Dr E. -I. ALI.KX, in a list of the Polvelueta of Plymouth (Journ. Mm-inc
Iliu/. J.N-.s-(*t\, N.S., vol. x (11)15). p. 640), states that N. itiiiiutiix was formerly
frequently taken near Plymouth, but it has not been found during the last two
or three years in spite of .special search.

\Jnin1 -~2. 11)15. X(it<> !>>/ (ii'iKT'tl Si't-ri'turi/. — The proofs of this paper and of
the accompanying plate were passed bv the author on June 1, l'J14; but owing to
the war and the unavoidable delay in obtaining delivery of the copies of the plate
from abroad, the paper could not lie scut to press until to-day.]

(ROY. sou. EDIN. TRANS., VOL. L, 421.)

310 ON A NEW SPECIES OF SCLKROGHE1LUS

DESCRIPTION OF PLATE.
List of Reference Letters.

AN. C. Anal cirrus.

E. Eye.

Mo. Mouth.

NEUR.' First neuropodiuni.

NKUR.S|J Tliirty-nintli neuropodiuiu.

NEUR. C. Neuropodial cirrus.

NOT. Notopodium.

NoT.1 First notopodiuni.

NoT.40 Fortieth notopodiuin.

Nuc. O. Nuchal organ.

PER. Peristomiuni.

PR. Prostomium.

S.O. Lateral sense-organ.

TENT. Tentacular process of prostoniium.

Fig. 1. Sclerocheilus antarcticus, n. sp. Type-specimen from Scotia Bay, South Orkneys. Dorsal aspect
of prostomium, peristomium, and Hrst four chpetiferous segments. Tlie prostoniium is retracted into and
partially hidden by the peristomium. For further description see pp. 293, 294. (x 25.)

Fig. 2. Ventral aspect of same specimen. ( x 25.)

Fig. 3. Sclerocheilus (intarctii-us, n. sp. Para-type ( = " Eumenia oculata " of Dr GBAVIBK) from Petermann
Island. Dorsal aspect of prostoniium, peristomium, and first four chsetiferous segments. Note the two
everted nuchal organs. For further description see p. 304. ( x 35.)

Fig. 4. Ventral aspect of same specimen. ( x 35.)

Fig. 5. Sclerocheilus antarcticus, n. sp. Type-specimen. Ventral aspect of the 39th-43rd chsetiferous
segments and the anal segment. Neuropodial cirri were probably originally present on the 41st and 42nd
chfetiferous segments, but have been lost therefrom. The anal cirri are also reduced in number, only one
remaining. For further description see p. 294. ( x 30.)

Fig. 6. Sclerocheilus antarcticus, n. sp. The 24th parapodium of the type-specimen, posterior aspect.
Immediately ventral to the notopodiuni is the papilla bearing the lateral sense organ ; ventral to this is a
much larger papilla the epidermis of which is glandular. For further description see pp. 294, 295. ( x 30.)

Fig. 7. fjcleroclieilua minutus Grube. Specimen 13'5 mm. long, from Plymouth. Dorsal aspect of
prostomium, peristomium, and first two chfetiferous segments. The right and left nuchal organs are everted.
For further description see pp. 298, 299. ( x 35.)

Fig. 8. Sclerocheilus minutus Grube. Specimen 18'5 mm. long, from 1'lymouth. Lateral (somewhat
ventro-lateral) aspect of peristomium and first chfetiferous segment. ( x 40.)

Fig. 9. St'li'i'tH'lteilus minntug Grube. Specimen 10 mm. long, from Saint- Vaast. Ventral aspect of
posterior end, showing two chsetiferous segments, another segment recently formed and as jet achcetous, and
the anal segment with four cirri. ( x 80.)

(HOY. SOC. KDIN. TKANK., VOL. if, 422.)

Scot. aClt. .Int. /:.*•/>.

V OIj. VII.

I

Vat!

ffaif.'.

\ruri'.

In I

Not"'

ArU

A K K.--.

PART XIII.
S P ( ) N (1 K S (SUPPLEMENT).

XIII -SPONGES COLLECTED BY THE "SCOTIA" IN
THE ANTARCTIC. SUPPLEMENT.

BY EM1LE TOPSENT,

Professor of Zoology in the University of Dijon.

(WITH FIVE TEXT-FIGURES.)

Spongiaires recueillis par la " Scotia" dans 1'Antarctique (1903-1904). Supplement.
Par Emile Topsent, I'rofesseiir ii l;i Faculte des Sciences de Dijon. J
l>«r /c Dr. \V. S. BRUIT..

(MS. received June 1, 19ir>. llt-ad June -JS, lillij. Issued separately September 22, li)K>.)

|)c])iiis l;i publication de nu>n memoire snr les SjHHif/inircx <fc I' Expedition
anta/rctigue u<iii<nn<l<' t:<-i>x*nix<' (9), j'ai recu de M. le Dr. W. S. liitucK quelques
hi] Mince's antarctiques faisant aussi partie dcs collections de la Scot in niais dout le
triage s'etait trouvr retarde.

fndependamment de fragments en doulile. d'especes mentioniiees dans inon
memoire, ce second lot contenait plusieurs formes qu'il me parait utile de signaler
on de decrire.

Dendrilla arctica, Topsent.

Station 325, avril-aout 1903; Scotia Bay, Orcades du Sad, 60° 43' 42" lat. S.,
44° 38' 33" long. W. ; 9-10 brasses. Un rameau.

Cette Dendroceratide a ete recueillie plus a 1'ouest, aux Shetland du Sud et
au-dela, dans les campagnes du Frnii.oti* (8, p. ll) et du Pourquoi Pas? Elle
parait etre repandue dans la region americaine de 1'Antarctique, au voisinage des
terres et par des profondeurs mediocres.

Eumastia attenuata, n. sp. (Figs. 1 et 2.)

.lanvicr 1903, Port Stanley, lies Falkland, greve.

I'n seul specimen, incomplet, sans support, long dc 7 centimetres, large de 3
centimetres en son milieu. Mince sur Fun de ses cotes qui, de contours doucement
arrondis, represente son bord naturel intact, il est, sans compter les papilles, epais
de 10 millimetres du cote oppose, qu'un instrument coutondant a tranche uettement.
L'outil a certaiuement laisse en place une partie plus ou moins etendue du corps,
mais il a, pour ce qu'il en a detache, suivi de tres pres le support. Le specimen est
done un morceau d'une Eponge en pla<|ue, a bords libres, et de quelque epaisseur en
son milieu. II revet un aspect particulier parce qu'il souleve toute sa surface en
processus digitiformes. Hautcs de 5 a 9 mm., epaisses dc 1'2 mm., en movenne, ces
papilles sont sensiblenient cylindriqncs, droites ou un pcu tordues ; generalement
simples et independantes les unes des autres, elles devieunent souvent plus ou moins
concresceutes par deux ou trois, tout en rcstant distinctes sur toute leur longueur ;
ijuelques unes se divisent en deux branches. Leur nombre est tel que les intervalles
qui les separent restent etroits. Elles sont lisses, ainsi, d'ailleurs, que la surface
generate a nu dans leurs intervalles. Aucune d'elles, de celles dont 1'integrite est

(KEl'RINTED FROM THE TRANSACTIONS OK THE ROYAL SOCIETY OF EDINBURGH, VOL. LI, PP. .35-43.)

316 EMILE TOPSENT: SPONGIAIRES RECUEILLIS PAR LA "SCOTIA '

certaine, ne parait percee d'un orifice a son extremite ; toutes, au contraire, se
immtrent en re point opaques et plus fortement teintees que le reste. L'absence <le
pnpillcs ouvertes au sommet au moment oil elle a ete recueillie, n'empeche pas de
emisiilerer 1'rtponge comme representant une espece du genre Eumastia. Outre que
certaines des papilles peuvent avoir normalement joue le role d'oscules et se montrer
contractees, il est evident que toutes ont fonctionne comme organes aquiferes. Sous

FIG. 1. — Eumastia atlcnuaia, n. sp. x£.

leur ectosome spiculeux, mince, transparent, rampe un systeme de cauaux qui occupe
la majeure partie de leur intrrieur, le reste contenaut un axe ramih'e, irregulier, de
place en place relie aux parois par des brides tenues.

A part cela, comme chez Eumastia sitiens, la structure est celle des Ilalichondria,
taut dans le choanosome que dans 1'ectosome. Le choanosome est assez dense,
quoique fragile ; 1'ectosome se laisse detacher en lambeaux translucides. Le tout,

FIG. 2. — Kxtrt'iuites des oxes de Eumastia atlcnuata. x B4U.

apres un long sejour clans des alcools renouveles, dont le dernier est incolore, a
conserve une teinte orangee generalement pale, assez vive pourtant, comme il a ete
dit, au bout des papilles. La coloration etait due u un pigment ditfus, car je note,
en fait de cellules spheruleuses, des elements spheriques de O'OOS mm. a 0'012 mm.
do diame.tre, incolores, tres brillants, composes de spherules iudistinctes avec un
noyau central sombre.

Eumastia attenuata a pour spicules uniquemeiit des oxes, courbes, fusiformes,
inegaux ; les plus faihles mesun-nt O'o.'i mm. de longueur sur O'OO:? mm. d'epaisseur
ct les j)lus forts 0'48 mm. sur O'Ol mm. Les spicules greles ont les bouts pro-

(ROY. soc. EDIN. TRANS., VOL. LI, 36.)

DANS L'ANTARCTIQUE (1903-1904). 317

•ji cssi\ ement aminds en deux pointcs fort longues et acerees (fig. 2, a) ; mais ton.-
ceux, ct e'est 1 immense majorite, (\\\\ depasscnt inin i mm. d'epaisseur, se terminent,
an contraire, d'unc faeon singulicrc : 1111 amincissement soudain leur forme une sort.-
ilc miicroii (lig. -. /') nil l'on vnit peiictrcr h- canal axial. C'cst. comme unc atrophic
limit tons sc troiivcnt frappcs au.x ilcux bouts. Kn presence d'nn specimen unii|ue,
il cst ditlicilc d'attinncr ipie ccttc ciiricusc disposition ne resulte pas d'une aptitude
toute individiiclli'. Totitefois. sj ,-e ca.ractcre niain|iiait de eonstancc, la spiculation
dc E. iitti'u IUI/K sc distinguerail encore dc cdlc dc /-,'. xi/irnx Schmidt par Ics
dimensions plus faiMcs dc scs elements c| dc cdlc dc I'-. Sc/nniil/i Dcmly (1, p. -4H).
par 1'inc^alitc dc leur taille.

Homceodictya microchela, n. sp.

Station :14C.. I" decembre 1903; I'.auc de ?.urd\v 1, f>4° -J.V lat. S., .r>7° :',-j'

lonu'. \V. : profondeur, .")') brasses.

Uu specimen blane, dresse, simple, haut dc 105 mm., cpais dc 10 mm. an plus,
progressivement clar^i dcpuis la hasc. ipii mcsure 7 mm. a. peine, jusqu'au voisinage
du sommet, ou il atteint 30 mm. de largour. Depourvu d'axe compact, il est tics
flexible. Sa eharpente interne consiste en un rcscau ladie de fibres trcs riches en
spongine, fort resistantes, cpaisses ct mal^i't' tout iucolores. Les fibres secondaires
qui s'elevent de ce systtmc sont plus tcnac.es ({lie dans la plupart des autres
Homceodictya, la spongine les .iccompa^nant jusiju'a leur terminaison. II n'cxiste
pas de membrane ectosomique distincte ii la pcriphcrie du corps. La surface, assez
c.ii'alc, a unc hispidation nctte, courte et serrcc ; ses orifices inhalants s'ouvrent au
fond dc depressions siniiciiscs, plus ou moins profondcs et plus ou moins rapprochecs.
ijiii la dccoiipcnl en dcs sortes dc villosites irrcii'idicrcs. Les deux faces portent
i|iidi|ucs osculcs noil surelcvcs, de 1 mm. a '2 mm. de diamctrc ; mais c.'est surtout
Ic lon^ dcs bords du corps lin^uiforme ,|ue les orifices exhalants sc localisent, disposes
en serie longitudinale, au nomlirc dc (i a 8 de chaipic cute.

Les oxes, longs de 0'4 mm., cpais de 0'013 mm., sont courbes. aceres, a pointes
assez longues. Us partageut ce dernier caraetcre avec les megasdcrcs dc I/,>ni<eo-
iln-li/it. kfrt/itt'/i'iK'iixix Ridley et Dendy (5, p. I 10) ct se distiiiguent p;l'' (M'':I mcmc de
ceux des Sponges de la Discovery appelccs par K niKr.vTRicK Desmacidon (II.) kcr-
var. tintni-c/irn (2, p. 37) ct ipii, pour moi, reprcsentent une espece a part.

i' cpaisscur est moimlre ipie cdle notec jusiju'ici des ox.es de H. kerguelenensis,
mais, comme on sail dcja ijue lc^ dimensions de ceiix-ci sont variables, mi nc pout
attacher d'importance a cette Ic^cre diff6rence.

Ce ipii, dans sa spiculation. distingue le miciix I/(ninr<></i<-l i/>i microchela, ce SOUt
36S isochdes, de mcmc tvpc ipic cciix de //. Jeerguelenensis, mais bicn plus faibles.
Tres abondants, ils n'ont (pic 0'OI7 mm. ;'i iro|s mm. de longueur, raremcnt. (CO'J mm.
Us sont done notablemenl plus courts ipic ceux memc dcs specimens dc la collection

(ROY. 80C. KDIN. TKAXS., Vn|.. LI, 37.)

S18 EMILE TOPSENT: SPONGIAIRES RECUEILLIS PAR LA "SCOTIA"

<lu Francois. Us sont, en outre, beaucoup plus minces que tous ceux que j'ai vus
chez ties II. kerguelenensis de provenances diverges. Leur tige, sans tubercules aux
extremites, dessinc une courbe assez forte et assez reguliere ; leurs ailes ne s'apergoivent
presque pas de profil ; enfin, leurs dents, aux replis generalement etroits, ont une
courbure assez simple.

Esperiopsis Scotise, n. sp. (Figs. 3 et 4.)

Station 417, 18 mars 1904; 71° 22' lat. S., 16° 34' long. W. ; profondeur, 1410
brasses.

Un specimen. C'est, fixee sur un petit galet, une Sponge ficiforme, grise surtout
sans doute a cause de la vase qui la souille. D'une base d'insertion etroite, dont elle
est actuellenient presque detachee, elle s'eleve en un pied qui bientot se renrle en un
corps massif, comprime. Vers le haut, un peu lateralement, s'ouvre un oscule unique,
entoure d'un cone membrano-spiculeux trauslucide, assez haut. La partie renflee du
corps est lisse, marquee de nombreuses depressions circulaires ou ovales, petites et
tres peu profondes, pareilles a celles de beaucoup de Grayella et autres Myxillinse et

FIG. 3. — Eapcriopsis Scotia;, 11. sp. x f;.

representant comme elles des aires inhalantes. Le pied s'est trouve tres endommage,
depouille de son revetement et en partie effiloche. La hauteur totale de 1'animal est
de 37 mm., dont 10 mm. pour la longueur du pied. Celui-ci n'a guere que 5 a G mm.
de diametre. Le corps atteint 20 mm. de largeur, mais n'a pas plus de 5 a 8 mm.
d'epaisseur. Sa consistance est assez elastique.

La portion denudee du pied montre la structure du squelette interne. C'est, dans
cette region tout au moins, un paquet de fibres ascendantes, polyspiculees, blanches,
usscx fortes mais sans spongine et, par suite, mal delimitees. Dans le corps, ces
fibres se diviscnt en faisceaux spienlcux de plus en plus greles a mesure qu'ils gagnent
la peripherie, ou ils se terminent juste au-dessous de 1'ectosome. Comme d'habitude,
Irs elements de ces faisceaux centrifuges tournent leur pointe vers 1'exterieur.
L'ectosome est une membrane assez resistante quoiijue tres mince, percee d'orifices
microscopiques nonibrcux, et pourvue seulement de microscleres, mais en quantite
considerable.

(ROY. soc. BDIN. TBANB., VOL. LI, 38.)

HANS L'ANTAKCTIQUE (1903-1904). 319

Spiculation. — 1. Megascleres: I. StafeiyZosfa/Zes presque droits, UD pen fusiformes;

base pen rentiee, allnngee, rarement marquee il'ini boiirrelct annnlaire situt'1 assex
loin de son extreinite, suivie d'un con pen retreci inais assex. long ; tige doucement
polytylote, attoignant son maxiiiiuin d'epaisseur on son milieu ; pointe breve, mueronee.
Us out, pour la plupart, O'D nun. dc longueur sur <)'<>I2 mm. de largcur de base et
O'Olf) nun. d'epaisseur do tige, niais lour longueur pent oscillei1 entre 0'75 mm. et
1'25 mm. et leur epaissenr entre O'OIH mm. et <>'()2 mm.

II. AI icroscleres : 2. Isochele* arques, extremement nombreux ; tigo fortomont

FIG. 4.— Spiculcs de EsiKriopsis Smliir. fi, extn'miU's ri'un sul)ty]ostyle ; i, isochelcs. x 540.

courb6e, surtout dans sa partie modiane, a peine moins large (ju'rpaisse (0'0029 mm.
de face pour 0'003 mm. de proh'l on 0'0035 mm. pour O'OlKi.H mm.) et d'ogales
dimensions sur toute sa longueur ; dent courte, crochue, triangulaire de face avec
tubercule large et pointe aceroe ; ailes courtes, repliees en deliors, dontiformes. La
longueur totale est de 0'04 mm. a 0'043 mm. ; I'ecartement des ailes n'est guore que
do (HM mm., leur longueur de O'OOS mm.

Suberites montiniger, Carter.

Station :54(i, l"r decembiv I !)().", ; J'.anc de Burdwood, 54° 2f/ lat, S., 57° 32' long.

\V. ; ;>r, brasses.

Un soul specimen, complet, sans la moindre trace d'attache a un snp])ort. par
nmsei|uoiit libre ; globuleux, avec un enfoncement, cependant, correspondant peut-
etre ,-iu ente ]iar loijuol il reposait sur le fond. Diametre moyen, Id mm. Coloration
brunatre a la surface, jannatro interieuroment. Consistance assez ferine. Snrfare
lisse portant de tros iiombreuses papilles surbaissees, inegales et irrogtiliorement
distributes. Les plus grandes, qui peuvent atteiudre 1 '2 mm. do diamotro, out la
forme d'un bourrolot oin-ulairo on saillio autour d'nno depression contrale pen profonde ;
les plus petitos apparaissent comme de simples verrues minuscules sans perforation
clistincte. 11 y on a partout, sauf sur la partie la plus bombee dii corps, opposee a son
enfoneement ; olios sont geiioralomont dispersees ; los plus grandes, au nombre d'unc
\ringtaine, sont cependanl gi-oiipees d'un cod'1 et separoes par des intorvallos ot ruits ;
la face concave on pnrte u |uinxaine do petites.

Par ees jiapilles, I'Kjjongc doit avoir lino oei'taine rossemblanee avec Suhrritt-s
ci-t'/lnitfi's Marenzeller (4, p. 4). Sa sti-ucture est hi meme : les spiciiles. assex. laehe-
meiil entrecroises dans rinteriour, Inrmoiit a la surfaco line c.oiio.lie plus dense ou

(ROY. SOC. BDIN. TRANS., VOL. I.I, 39.)

320 EMILE TOPSENT: SPONGIAIRES RECUEILLIS PAR LA "SCOTIA"

beaucoup se tiennent tangentiels tandis que d'autros se placent clebout, la pointe vers
le delmrs. Le rapprochement est d'autant plus indique que les spicules out, de part
et d'autre, une forme rare chez les Suberites : ce sont des tylostyles a tete allongee.
Ceux de S. crelloides atteiguent, d'apres les mesures des dessins qui en out et£ publics,
0'275 mm.; ceux de Sponge du Bane de Burdwood varient entre 0'32 mm. et 0'4 mm.
de longueur sur O'OOS mm. d'epaisseur. Independamment de leur taille, il y aurait
une difference de forme entre ces spicules, la tete de ceux du S. crelloides des parages
ilc Jan Mayen n'atteignant que loin en arriere son maximum dY'paisseur. Pour le
reste, il y a similitude remarquable, la tige etant souvent un peu onduleuse, se renflant
doucement pour devenir aussi epaisse que la tete et se terminant en une pointe breve,
mucronee. Le canal axial est visible jusque dans la tete des spicules de Sponge
de la Scotia et s'y terniine sans rennement vesiculaire.

Si je ne rapporte pas 1'Eponge en question a 1'espece S. crelloides, c'est surtout a
cause de la tete de ses spicules que je crois plus semblable a celle des tylostyles de
Siiberites montiniger Carter. Mais les dessins de MARENZELLER et de VOSMAER con-
cernant ces Sponges sont-ils d'une exactitude rigoureuse ? Je ne suis pas absolu-
ment convaincu que S. crelloides diftere specifiquement de S. montiniger. MARENZELLER
les a separes en tenant compte avant tout de 1'etat de leur surface ; mais il existait
quand meme des papilles chez S. montiniger, puisqu'il en est dit : " Vents, one large
at the summit, fringed, the rest small, on the sides" (10, p. 31). Le nombre de ces
eminences pourrait etre plus eleve chez certains individus sans distinction specifique
possible.

J'ai vu, de 1'Atlantique Nord, une Eponge que j'ai determinee aussi S. montiniger
(7, p. 130), qui n'a pas de papilles du tout et dont les spicules, par leur tete plus
elliptique qu'ovale, ressemblent plus quo. tous les dessins a ceux du Suberites de la
Si-nfia, ; ils sont de meme grosseur qu'eux, mais plus courts et droits. Leur longueur
(0'24 mm. a 0'295 mm.) est a peu pres celle des tylostyles de S. crelloides.

Quant au Suberites de la cote occidentale de TAmeritjue du Nord, appele par
LAMBE (3, p. 128) $. montiniger Carter, il produit des tylostyles bien plus epais
(O'OIG mm.) quo toutes les Eponges precitees et appartieut peut-etre a une espece
differente.

Latrunculia brevis, Eidley et Dendy. (Fig. 5.)

Station 34G, ler decembre 1903 ; Bane de Burdwood, 54° 25' lat. 8., 57° 32' long.
W. ; 56 1 trasses.

Un specimen assez grand, a papilles tout-a-fait semblables a celles du specimen de
Latrunculia 'ipicalis Ridley et Dendy qui provient des Kerguelen (5, pi. xliv, fig. 4),
ma is a discasters constamment depourvues de prolongement apical. Longues de
()'(((> mm., ces discast.crs rcsscmlilcnt beaucoup a celles de L. lin-rii* llidley et Dendy
(5, p|. xliv, fig. LO) : elles ont une base elargie, disciforme, a marge armee de fortes
cpincs dirigi'-cs obliquement vc.rs le bas ; un vei'ticillc d'epines ]tlus etroit la surmonte,

(ROY. soc. KPIN. TRANS., VOL. LI, 40.)

DANS L'ANTAKCTIQUE (1903 1904).

321

ordinairement regulier. assc/ ecarte d'elle et netteinent distinct, quelquel'ois ditt'us <•)
plus on inoins GOnfondu avee die; puis vient UN vcrtieille de 0'0:>:i nun. de dlametre,
!(.• plus large de tuns, liieii perpcndiculaire an grand axe du spicule, lanielleiix, decoupe
par des ineisuiTS en trois lobes marques en leur bord de plusieurs indentations; puis
encore tin verticille semblable nmis nioins large et a lobes retrousses ; enfin, une
eouronne terminale composec de deux vertieilles d'epines de plus en plus etroits, tres
rapproches, emboites, rinf'erieur I'ortemcnt olili(|iie vers ]e haut, le superieur dresse.
Sur les spicules iinparfaits, ectte extreinite se presente connne une dilatation ovoide,
simple mi dejii subdivisee par un le^'iT bourn-let annul/lire ; je ne I'ai vue qu'tme seule
fois, a eet etat. sunnoiitee d'une poiute apieale, loiigue et, naturellement, fort grele.

VP \? X^

S S Z

FIG. it. — Discasters de Lnlritnculia ln-crin. x 540. 1-5, discasters imparfaites ; d, discaster
]iai-faite ; b, sa base vue par dessous ; v, un grand verticille, de face.

Que cette exception devienne hi regie chez certains individus et les discasters
pai faites ressembleraient ii cclles de Latrunculia apicalis. C'est une possibility que
1'existence de L. apicalis var biformis Kirkpatrick (2, p. 14) rend admissible.

Les megascleres sont des styles, Icgerement polytylotes, souvent courbes ou
tlexueux, a pointe breve, 11 base simple, plus mince que la tige ; ils mesurent 0'5 mm.
a 0'53 mm. de longueur sur 0'012 mm. environ d'epaisseur. Ils ressemblent, par
coiisci|iieiit, a la fois a ceux de L. brevis et a ceux de L. apicalis.

11 y a. I'll soinine, (juelques raisons d'admettre que ces deux especes n'en forment
(|u'uiie seule, dont, par comparaison aux autres Latrunculia, L. apicalis ne serait
pas la forme typique.

rtis Sc<>tin', Topsent.
Station 417, is mars l'J()4; 71° 2~2' lat. S., 1(5° :!4' long. \V. ; I4K) brasses. Un

-|n''ri|nen et 1111 pedolicllle detadie.

]jc specimen a le corps en tier et le pedoncule brise. Le corjis, d'une tres grande
mollesse, est disciforrne a contour sinueux et me.snre pres de I 1 centimetres de
dia metre; epais de 4 a 5 mm. en son centre, il devient tres mince sur les bords,
ijiii se I'eplient plus ou inoins par dessous. La face superieure (gastrique ou cloacale)
n'est pas plane niais soulevee en une infinite de petites bosselures ; elle a son revete-
ment presque complet, fonnant sur ses orifices, inegaiix, un tamis it mailles regulieres.
Le re\ etcinent de la face inferieure (ou dermique) est nioins bien conserve; ses

(ROY. 30C. KU1N. TRANS., VOL. LI, 41.)

322 EMTLE TOPSENT : SPONGIAIRES RECUEILLIS PAR LA "SCOTIA"

deehirures laissent a nu deux sortes d'orifices, les tins grands mais de diametre
d'autant phis large (ju'ils sont voisins du centre, les autres petits, distributes entre
les premiers. Le pedoncule, cylindrique, epais de 9 mm., s'attache au milieu de la
face iuferieure ; il est creux, mais sa cavite n'aboutit pas a un orifice externe ; elle
communique avec les lacunes creusees dans 1'epaisseur du corps. II n'en existe plus
qu'un troncon, long de G centimetres seulement, de calibre uniforme, assez souple et,
au moms par places, finement hispide.

Par sa forme comme par ses dimensions, ce specimen ressemble beaucoup plus au
type de Cauloph acnt* Intus F. E. Schulze (6, p. 124, pi. xxiv) qu'a celui de C. Scotia
Topsent (9, pi. ii, figs. 1-3). Pourtant, certains traits de sa spiculation sont caracte-
ristiques de C. Scotiw. Ses hexactines dermiques, quoique de dimensions un peu plus
faibles que dans le specimen geant primitivement decrit, out toujours 1'actine distale
renflee en pinule notablement plus courte que les autres actines, meme que la proxi-
male, qui est toujours bien plus breve que les tangentielles. Ses autogastralia sont,
au contraire des autodermalia, de taille fort inegale ; ce caractere, deja note d'apres
le specimen type de C. Scotiie, est en opposition avec ce que montre la reconstitution,
d'apres SCHULZE, d'une coupe verticale de C. latus. Les discohexasters, a rayons
secondaires de meme longueur que les primaires, portent des boutons terminaux
petits et sans denticules distincts.

D'autre part, certains traits de la spiculation sont communs a celle du type de
C. Idtn-is. Aiusi, il existe, parsemees dans le revetement dermique, des hexactines a
actine distale barbelee longue, que je n'ai point trouvees dans les portions examinees
du type de C. Scotia). Puis, les autogastralia sont ici un melange d'hexactines et de
pentactiues, avec predominance de ces dernieres. Enfiu, les microscleres, de deux
sortes seulement, sont d'une grande uniformite.

Les parties hispides du pedoncule doivent leur aspect a un revetement compose
de spicules pareils aux autogastralia, encore plus inegaux qu'eux de taille et presque
toujours a 1'etat de pentactines, leur actine proximale se reduisant a un tubercule.
Sa cavite longitudinale a une paroi lisse, sans spicules de revetement et limitee
simplement par un feutrage, sans synapticules, de diactines de longueur et de grosseur
variables.

Le pedoncule detache, provenant de la meme station, est cylindrique, long de
13 centimetres, un peu moins gros que celui de 1'Eponge precedente et solide. II
porte par places encore un peu de son revetement sous forme de pentactines a actine
distale longue et barbelee, de dimensions tres inegales.

(ROY. soc. KDIN. TRANS., VOL. LI, 42.)

DANS L'ANTARCTIQUE (1903-1904). 323

1NHKX BIBLtOGRAPHIQUE.

(1) DENDY (A.), (..'a/nfni/nf <>f Non-calcareous Sponyes collected by J. Bvacebridffe Wilson in the

ii- i-jltboiu-liii'-'t <>J 1'ii/i 2'/iilliji lli-mlt, I'ail I (Proc. IJov. Soc. of Victoria, vn], \i, m-l. 24.
Melbourne, 1894).

(2) KlRKPATRIOK (1!.). I'lirifnni, 'l''truj'"iii'l,i (National Antarctic Expedition, Natural History, vol. iv,

1908).

(3) LAMRE (L. M.), $/>nntjes fn/>n the Western Count <>f Xnrtli America (Tr.m>. Hoy. Soc. of Cunailu,

section iv, art, s, Ottawa, 1894).

(4) MAUENZELLEB (E. VON), I'ori/cren, Ant/iozoen, Ctenophoren uml Win-mi-i- von .Inn MHI/,I/ (Die

intcrnatioiiale I'olavforsrlnmg, 1882-1883, Wien, 1886).

(5) RIDLEY (S. O.) and DKN'DY (A.), Report on the Monaxonida (The Voyage ul H-M-^. "Challengi •

Zoology, vol. xx, Edinburgh, 1887).

(6) SCHULZE(F. E.), It f port on the Hexactinellida (The Voyagr of 11. M.S. "Challenger," /Dol,,-^, \^l.

xxi, Edinburgh, 1887).

(7) TOPSENT (E.), Contribution fi Vt-tude ilm H/i<m<jiftir':< Ji- /'At/antiijue NurJ (Kn-ult. camp, scient. du

Prince de Monaco, fasc. ii, Monaco, 1892).

(8) TOPSENT (E.), Spitngiaires (Expedition aiitarcticjue franyaise, 1903-1905, commandc'e par le Dr.

Jean Charcot, Paris, 1908).

(9) TOPSENT (E.), Spongiaires de V expedition antarctique i/ationaL ecoseitisi (Trans. Roy. Soc., vol. .xlix,

part iii, no. 9, Edinburgh, 1913).

(10) VOSMAEU (G. C. J.), Report on the Sponyes dredged nji in the Arctic Sf/t by the " Will: ,,i Hurfiit.* /;.•
the years 1878 and 1879 (Nederla'nd. Arch, fur Zool., Ld. i, Suppl., 1882).

(ROT. soc. EDIK. TRANS., vor.. 1 1, \:>,.)

.*

*

.*