« <• - w S
N Jf ,*i#u

-^ •>

** ^i
^.

y ,

lr-
t

R-; 5

•^ **Sr

»>>, r^'^

L.-TI — n — TV

REESE LIBRARY

UNIVERSITY OE CALIFORNIA.

,1906.

.

ON

FROM THE GREAT DEEPS OFF THE NORWEGIAN COAST.

PARTLY FROM POSTHUMOUS MANUSCRIPTS OF THE LATE

PROFESSOR Dr. MICHAEL SARS

BY

GEOEGE OSSIAN SAES,

With 6 Copper plates.

university-Program for the 1st half-year 1869.

CKRISTIANIA.

PRINTED BY BR0GO3H fc CHRISTIE .

1872.

Contents,

Preface V

Introduction VII

Polyzoa.

1. Rhabdopleura mirabilis 1

The Polyparium S

The Polypide 4

The tentacular arms 6

The bucal shield '. 8

The contractile cord 10

The axial cord 11

Vital phenomena 12

Occurrence 12

Characteristics of genus and species '. 13

Development 14

Concluding remarks 15

2. Flustra abyssicola 19

Conchifera.

3. Yoldia obtusa 23

4. Pecchiolia abyssicola , 25

Cephalophora.

5. Dentaliuin agile 31

6. Triopa incisa 35

7. Gonieolis typica 37

Annelida.

8. Umbellisyllis fasciata 41

9. Paramphinome pulchella 45 .

Anthozoa.

10. Mopsea borealis 50

11. Fungiacyathus fragilis 58

I*

IV

Spongise.

12. Trichostemma hemisphfericum 62

13. Cladorhiza abyssicola 65

14. Hyalonema longissimum 70

Explanation of the plates , 74

P r ef ac e.

Under the above title I purpose, if I can obtain the necessary assistance, to make
known little by little the most important discoveries which I have had opportunity to make
during my deep-sea investigations in recent years; and I shall consider it my duty first to
treat those animal forms which my Father has already partially examined and designated.
The present university-program comprises some of these belonging to 6 different classes of
animals, namely 2 Polyzoa, 2 Conchifera, 3 Cephalophores, 2 Annelids, 2 Corals and 3 Sponges.

These animal forms had already been partially destined by my Father for more detailed
notice in a university-program similar to that composed on the subject of the Rhizocrinus ;
and application for the requisite assistance was already sent in to the academical college,
when his unexpected death put a sudden end to the important investigations which he had
commenced.

It is therefore with deep gratitude that I here acknowledge the kindness of the col-
lege, in confiding to me the elaboration and publishing of the intended program; and although
I am well aware of my want of that thorough acquaintance with the invertebrate animals, and
of that clear and precise method of treatment which always distinguished my Father's works,
it is still a pleasure for me to have opportunity given for prosecuting to the best of my power
those investigations, to which my Father devoted himself with so much zeal and perseverance,
until the day of his death.

Although, as above stated, the present program was intended to be published by my
Father, still the necessary detailed investigations were scarcely more than commenced, for
which reason also no manuscript had been compiled. I find however, among my late Father's
papers, notes on most of the animal forms mentioned in the following lines, and with respect
to some of them also more complete descriptions, which have been of great help to me in
the elaboration of the present program. I have also always quoted as conscienciously as
possible my Fathers's more coherent remarks, which certainly will have their important value.
Where this has taken place, I have always expressly remarked it in the text. Finally I am
bound to express my warmest thanks for the valuable information relative to the Mollusca,
which has been communicated to me by the celebrated English Conchologist I. Gwyn Jeffreys
Esq.; and for the remarkable kindness with which the equally highly celebrated Swedish

VIII

to appear a rich and in many respects peculiar deep sea Fauna, of which only a very incom-
plete notion had previously existed. The number of new forms of animal life was therefore
also very considerable; and some of these were of peculiar interest, as more or less evidently
carrying us back to former telluric periods, especially the Fauna of the Cretaceous period. This
was specially evident in the case of an animal which my Father has therefore made the
subject of a separate and detailed treatise. I mean the little remarkable sea-lilly Rhizocrinus
lofotensis belonging to the family of the Apiocrinidse, which has been considered as long ago
extinct, but which flourished in the Oolitic period. The discovery excited great interest in the
scientific world, and may be said to have given the impulse to the comprehensive investi-
gations of the Atlantic depths, which have been instituted with great liberality by the English
government in later years. By these grand English deep-sea expeditions, in connexion with
the deep-soundings executed during the Swedish Spitzbergen expeditions in the Arctic Ocean,
which have both extended even to the enormous depth of between 2000 and 3000 fathoms,
it has now been further ascertained that the great depths of the Ocean do not form, as was
formerly supposed barren tracts or deserts, but are peopled with a richly developed and pe-
culiar animal world, which oh the whole shews a very clearly marked affinity to the Fauna of
the Cretaceous period, and in a great measure may really be considered as consisting of descen-
dents in direct line from the forms existing in that telluric period. This cannot indeed be
always so evidently proved, as in the case of the above mentioned sea-lilly, but it is also in
many other forms very distinctly expressed. We may therefore with Carpenter 1 presume that the
Cretaceous formation is continued undisturbed at this present day in the depths of the Ocean;
and we may therefore assume that just here, there has been little change during all the pe-
riods of the earth ; and we may also expect to find just here, the descendents from the chalk
period most unchanged; while the Fauna at smaller depths, and especially the shore-Fauna,
would in a relatively short time, by reason of telluric and physical revolutions, be forced
entirely to change its character. It is also very probable that many of the remarkable forms
of animal life, from the great depths off our coast, more particularly described in the follow-
ing lines, are just such ancient slightly altered forms, which thus — -, apart from purely zoological
considerations — excite a very peculiar interest, as giving us important indications with respect
to the historical, or rather the palseontological development of animal life.

See Carpenter on this subject. Preliminary report of dredging operations in the seas to the North of the British
islands, p. 192.

o 1 y z o a.

I, Rkabdopleura mirabilis, (M, Sars),

n. sp.
(PI. 1. & 2. fig. 1—24).

Halilophus vii>-aliV*. M. Sars. ,,Fortsatte Bemserkninger om det clyriske Livs Udbredning i
Havets Dybder" p. 12.

In the year 1866 I drew up in the dredge, together with other deep-sea animals, from
a depth of 120 fathoms at Skraaven in Lofoten, a small ,,Phytozoon" to which at first I paid
no great attention, as I took it to be a colony ofHydroids; I took therefore only a few exem-
plars, and put them in spirit with some other uncertain forms of animal life from the same
depth. After my return 'home however, on examining more closely this supposed Hydroid
colony, my Father found immediately that we had before us a very peculiar animal, which
quite certainly could not be a Hydroid, but seemed rather to be related to the Polyzoa;
although the shape and appearance of the single cells had undeniably a great resemblance to
the former, especially to some Campanularides. As a satisfactory examination of the animal
could not be made with the specimens brought home in spirit, my Father urged me, the next
time I visited Lofoten to examine the animal in a living state as minutely as possible ; as
hereby we should without doubt obtain interesting and instructive results. This I had an op-
portunity of doing in the following year 1867; and I then examined, as minutely and consci-
enciously as possible with the instruments at my disposal, the structure of this little and
fragile animal, which I also found peculiar in the highest degree, and different from all that
I had previously known. My Father was also greatly surprised on learning the result of my
examination, and looking over the numerous drawings which I had made from the living ani-
mal, It was clear enough that we had before us not only an entirely new genus and family,
hut even the type for a still higher division; and we found no small difficulty in referring it
to any known animal type. But as it appeared most nearly related to the Polyzoa, my Father
classed it with these, and noted it in his catalogue of deep-sea animals, compiled in 1868,

under the denomination of .,Halilophus mirabilis" (the generic name taken from a certain

1

resemblance of the tentacular arms to the so-called Lophophore of the freshwater Polyzoa).
rFhe next year appeared Allman's treatise .,on Rhabdopleura, a new form of Polyzoa from deep
sea dredging in Shetland"1; and my Father, as well as myself, at once recognised in the species
therein described and delineated, Rh. Normanni, a form closely agreeing with the Halilophus
mirabilis, but which evidently, from the form and attachment of the Polyzoarium, must belong-
to a different species from ours. Allman's communications concerning the organisation of the
animal shew indeed, as will appear from the sequel, many essential differences from what I
have had occasion to observe in our northern form; so that if these communications were
in reality* correct, there could scarcely be any doubt that both forms were also generically
distinct. But the fact is that Allman has only had the opportunity of examining specimens
preserved in spirit; and both my Father and myself know from experience how extremely
difficult it is to obtain results with specimens in this state, and what imperfect and false
notions may thus be formed of the animal's real structure. Taking this into consideration
we may really be astonished that Allman has been able to see so much as he actually has
seen, and that h-e has not misunderstood the animal's organisation in a greater degree than
appears from his description and delineations. Allman has indeed seen in the Ehabdopleura
a very aberrant form of Polyzoa, but is far from having apprehended that the form is ab-
normal in so high a degree as it is proved to be, according to investigations which I have
executed with the utmost care and minuteness.

With respect to the method of examination, there is little or no use in dissecting so
small and fragile an object as the animal of the Rhabdopleura, even if the finest imaginable
instruments are employed. It is therefore necessary to study the animal entire as it is, or
at most after separating the individual animals from their cells or tubes, which, as will be
seen, may be done with the greatest ease in operating on the living colony. In order to get
a better and sharper view of certain parts, I have found it very useful to effect a gentle com-
pression of the animal between 2 glass plates, but so that the pressure can be moderated at
will, increased or diminished. To this end I cemented a thin glass plate to the upper arm
of my compressorium, by which means the desired result was obtained far better than by the
use of the so-called „ aquatic boxes". Moreover most of the parts of the living animal may
be easily examined without pressure, and without even taking the animal out of its cell. A
colony, or part of one, can be placed under the microscope, and the most important parts will
plainly be seen through the transparent walls of the cells, even if the animal has not stretched
itself out of the aperture.

Besides the outer chitine-like tube with its off-shoots or cells (Polyzoarium) there may
be distinguished in the animal under consideration the following principal parts. 1. the Po-
lypidc itself, which again shews 3 principal parts a) the body; b) the tentacular arms and
<•) the bucal shield; 2. the contractile cord: and 3 the n.rial cord. We shall treat each of these
parts separately.

Quarterly Journal of microscopical science. Vol. IX. 1869. p. 57-63. pi. 8.

The Polyzoarium.-

The Polyzoarium (co?rio3cium) in the Rhabdopleura mirabilis (see tig. 1. -2. & 3.) has
the form of a thin, elastic flexible, chitine-like, transparent, most frequently quite colorless
cylindrical hollow tube, consisting of a stem which creeps along the bottom of the sea, now
and then attached to other bodies, irregularly winding, and only seldom, here aud there, forked,
which at short intervals sends up perpendicular, free, undivided, more or less winding branches
or continuations, of the same form calibre and nature as the stem, and all terminating with
a circular aperture. A difference between the stem and the branches, strikes the eye imme-
diately: the stem is always more or less thickly covered with extraneous particles (sand, mud,
fragments of shells, Rhizopod-shells &c.) while the branches are always, with exception of the very
lowest piece, quite free from such particles, and consequently quite transparent, appearing in
their whole extent very distinctly and ornamentally ringed. The rings, which are only exterior,
form close, equidistant, sharp, circular transverse folds, strongly prominent over the surface
of the tube, causing the edges everywhere to appear crenulated (see fig. 3). If one can sepa-
rate from the stem the very closely adhering extraneous particles, (which is connected with
no small difficulty) it will be seen (fig. 9) that this outer formation of folds is also continued
on the stem itself, although far less sharply marked aud also more irregular; the transverse
folds being often divided fork-wise, or in other words, not forming completely separate rings.
It will also be remarked (fig. 10) that the branches or free tubes are not at all sharply dis-
tinguished from the stem, but that their interior cavity is prolonged immediately into, or
continuously with, the cavity of the stem. This latter, the interior walls of which, like those
of the tubes, are quite smooth, is divided at certain intervals, by tolerably thick transverse
lamellae, or septa, into several successive cylindrical chambers, which do not communicate
with each other, but each one of which is continued immediately in one of the perpendicular
tubes proceeding upwards from the stem. Through the whole of the creeping stem, exten-
ding along through all its chambers, runs a thin cylindrical chitinous cord (the axial cord) very
remarkable from its dark nearly black color, of which more hereafter. This cord is never
continued up through the free tubes, but may now & then, rarely, divide itself fork-wise,
namely when the creeping stem divides itself in this manner.

Every one of the perpendicular branches or fubes contains an animal, which is connected
by a long cylindrical fleshy cord (the contractile .cord) near the bottom of the corresponding
chamber in the stem, with the axial cord, which thus unites all the individuals of the colony
with each other.

The creeping stem in the Kh. .mirabilis may indeed attain a very considerable length;
but it is very difficult to get the whole separated from the substances which adhere to it.
One can usually therefore only get up small pieces of the colony, and seldom more connected
portions. The largest connected piece of stem (fig. 1) which I succeeded in getting loose from

its attachment, was about 40 millimetres long, and was at irregular distances 4 times bifur-.

1*

cated; another piece was of about the same length, but only 3 times divided. These bran-
ches, which proceed from the stem at a more or less acute angle, are also creeping and
irregularly bent, and, like the main stem, produce cells, the number of which in all on these
pieces of stem amounts to about 60. The stem is thus only seldom branched, and usually
runs to great length without producing any branch.

The cells or tubes, which proceed from the stem by lateral budding, are of very dif-
ferent length, while their thickness is everywhere, and in all, about the same. The largest
are 6—7 millimetres long and V4— Y5 mill, thick. They are, as stated, erect, yet seldom,
or never perfectly straight, but always more or less bent in some part, or in the whole of their
length : sometimes like an S, sometimes in several bends or turns like a drawn out screw, but
most frequently irregularly bent, in rare instances so strongly curved that the curve is nearly
circular.

From the above description it will be seen that the Polyzoarium in the Rhabdopleura
mirabilis is decidedly different from that of the Shetland Rh. Normanni. Firstly, the cree-
ping stem of the latter, of which one surface everywhere adheres to old shells or other solid
substances, is much more strongly branched (,,subalternately") and, like the cells, quite naked,
without trace of attached extraneous particles. Then the cells themselves are not, as in our
species, free in their whole length, but at the base for some distance, like the stem itself,
fixed and creeping, for which reason also the free perpendicularly rising ringed part of the
same is much shorter than in the Rh. mirabilis. Finally there appears some difference be-
tween the two species with regard to the manner of the division of the stem into chambers.

The Polyzoarium in the Rhabdopleura does not coincide with any other species of
Polyzoa. The cells in their tubular form resemble most those of the Cyclostomata, but are
horny or chitine-like (in the Cyclostomata they are chalky) and are distinguished by their
surface, covered with prominent transverse folds or rings which are also foreign to the Po-
lyzoa, but are found in many Hydrozoa such as certain Tubulariadae and Campanulariadse.

The Polypide.

The individual animals or Polypides seem at first glance to resemble the ordinary Po-
lyzoa (see fig. 3). The body, which is only a -little over 1 millimetre long, is oblong, and
appears to be occupied almost entirely by the digestive system; on closer examination, we
find however (see pi. 2. fig. 15, 17, 18) that a thin glassy skin surrounds the digestive appa-
ratus, which therefore is not, as in all other Polyzoa, freely suspended in the ,,Perigastric
fluid", which latter, as will appear in the sequel, is entirely wanting in the Rhabdopleura. In
all other Polyzoa without exception there is besides the so-called Ectocyst, corresponding
with the Polyzoarium, also a so-called Endocyxt, which always represents a thin membrane
lining the interior of the Ectocyst or cavity of the cell, and, from the aperture, recurved and
attached round about to the Polypide under the base of the Lophophore. The interior cavity
• of the cells in the other Polyzoa is thus actually completely closed by the endocyst and the

body of the Polypide; and the so-called perigastric fluid therein- contained, wherein, the intes-
tinal canal of the animal is freely suspended, does not stand in any direct connexion with
the surrounding medium. The retraction of the Polypide into the cell is effected only, by a
folding (invagination) of the anterior elastic part of the endocyst, by which the so-called ten-
tacular sheath is produced. The case is quite different with the Rhabdopleura. Here is no
cnrfocyst at all (unless we should consider the glassy skin which closely surrounds the diges-
tive apparatus, to be an endocyst), consequently also no perigastric fluid; and further the cavity of
the cells stands in direct communication with the surrounding medium, without being closed
at the aperture by a skin connecting it with the animal's body. The retraction and protrusion
of the Polypide is therefore not effected as usual by in- and evagination, but in a totally
different manner, of which more hereafter. Allman has indeed (1. c.) imagined that he has
perceived a trace of a real endocyst of the same nature as that observed in the other Poly-
zoa, but he has quite certainly deceived himself. The Polypide of the Rhabdopleura lies
quite free in the cell, and is only attached to the colony by means of the contractile cord,
neither by any endocyst nor special muscles, as appears clearly enough from the fact that
when the contractile cord is severed, the Polypide can be taken entire and uninjured out of
its tube with the greatest ease.

From the anterior part of the body where the mouth is situated, yet, as will appear
in the sequel, not as usual in a terminal position, but rather in a ventral situation behind the
peculiar oval prominence (the bucal shield), the gullet or ossophagus, (fig. 5, 14 etc. e) rather
short, but wide, and furnished with thick walls, proceeds right downward or backward to the
stomach, from which it is separated by an also outwardly apparent constriction, and by a
sort of internal valve (the cardiac valve).

The Stomach (f.) which is simple without any armament of teeth or hard parts in
its interior, and furnished with tolerably thin walls, is elongated, rounded cylindrically, slightly
and somewhat irregularly curved, with ventral concavity, and in its anterior part, where it
has its greatest breadth, only a little wider than the gullet and occupying about 2/a of the
cell's caliber (see fig. 14). In the anterior half it is of about uniform thickness, but diminishes
towards the posterior end (Pylorus) very rapidly, and goes imperceptibly over, after turning
a little to one side, into the intestine, suddenly curving itself upward and forward. The In-
tcxti-nc (g) which is not by any constriction, nor yet by any interior valve (Pylorus-valve) dis-
tinguished from the stomach, of which it forms the immediate continuation, is narrow, cylin-
drical (its thickness scarcely V3 of that of the stomach in the widest part) only slightly
tapering towards the end (see fig. 17) and has a tolerably straight course forward, lying close
to the dorsal side of the stomach and gullet, and terminates with a circular anal aperture,
situated just behind the spot from which the tentacular arms proceed. Immediately behind
the anal aperture, between the terminal part of the intestine and the dorsal wall of the gullet,
which here forms a little concavity, there appeared a clear cellular body (fig. 15. r) in which
several evident nuclei were visible. I cannot however pronounce any decided opinion as to

the signification of this object; it can scarcely be a nervous ganglion; as it does not lie in
the substance of the body itself, but only in the thin external skin which encloses the body.
The stomach and the intestine usually shew in the living animal a bright opaque yellowish
brown color, which color seems mainly to be derived from the contents. When the stomach
and intestine are more empty, they shew a far paler yellowish white color. The walls of both
shew plainly a fine cellular structure.

It is evident that the digestive system in the Rhabdopleura differs, in many points
which have not been remarked by Allman, from the normal system of the Polyzoa. In the
latter the stomach usually consists, asjs well known, of 2 distinctly separate parts, one shorter
cylindrical cardiac part, and a longer and wider pyloric part, which ends in a large rounded
bottle-shaped Caecum (cul de sac). Therefore the intestine usually takes its origin, in these,
high up, or about on a level with the transition of the cardiac part to the pyloric part; while
in the Rhabdopleura, where no such division of the stomach occurs, the intestine proceeds
from the posterior end of the stomach or fundus, as the immediate continuation of the same.

The Tentacular Arms.

The tentacular Corona, or Lophophore, situated in the Rhabdopleura on the anterior
end of the body, is of a totally different appearance from that of the other marine Polyzoa,
while on the other hand it appears at first sight to shew an unmistakable resemblance to
that of most fresh-water Polyzoa (P. hippocrepia) and is also thus represented by Allman.
However when it is more closely examined, it displays many essential differences; although
by its strongly marked bilateral symmetry, it appears to be most nearly connected with the
same, in respect of the semilunar or horse-shoe form peculiar to them.

As in the said fresh-water Polyzoa, the lophophore or tentacular frame does not form
a circular ring, but is drawn out into 2 lobes or arms, each of which bears a double row of
tentacles. These lobes or arms (fig. 3. 4. 5. &c. d.) which also here proceed from the dorsal
side, are however considerably longer and narrower, or more cylindrical than in any other of
the known Polyzoa; and while in those freshwater Polyzoa they only form a part of the ten-
tacular corona, they represent here the whole lophophore, as it is only on these arms that
the tentacles have their place. The tentacles in the Rhabdopleura do not form, as in the
other Polyzoa, a continuous series, but are interrupted, as well dorsally as ventrally, by an
evident interval; in other words, trc /tare not one single tentacular crown, but 2 symmetrical
tentacular arm*, which take their beginning on each side of the anterior part of the body, and
extend out from the same dorsally and diverging to each side. If we examine these tenta-
cular arms in the living animal we find that they are also in many respects different from
the lophophore of the Hippocrepia. While the latter always retain unchanged their form
and somewhat inclined direction, the tentacular arms in the Rhabdopleura are in a high degree
flexible and variable in their direction relatively to the body of the polypide and to each
other. As long as the polypide i:s withdrawn into the cell, they are always (see rig. 14),

extended straight forward, and nearly parallel with each other, forming, together with the
tentacles attached to them, a close fascicle extending in the same line with the body. As
soon as the polypide reaches the aperture of the cell, they spread out from each other; but
this takes place in various manners. Sometimes they are bent with the ends only a little
out from each other, while otherwise they are nearly parallel (see fig. 3.); sometimes they
spread themselves out so widely on each side, that they stand almost diametrically opposite
(fig. 4); sometimes they bend themselves with the ends downwards (see fig. 3); sometimes
— and this is most usual, and always occurs when the animal is taken out of its tube (fig. 5),
they are bent upwards and backwards, and that often so strongly that they describe a nearly
semicircular curve, so that the extremities even touch the dorsal side of the polypide's body
(fig. 5 & 15). This great mobility of the tentacular arms or lophophore (so different from
what is observed in the other Polyzoa) which appears to be produced at will by the animal,
sometimes in one manner and sometimes in another, must certainly, although it always takes
place very slowly and with little energy, be brought about by means of auxiliary muscles or
muscular tissue. I have however only succeeded in observing very faint traces of anything
of the kind. When the animal is gently compressed between 2 glass plates one may observe
on each side some very fine fibres (Fig. 15, 18, p) passing obliquely over the gullet, proceeding from
the ventral side, where the body of the polypide forms on each side a small conical prominence
(ibid, o.o.) which may perhaps be considered as the ventral corners of the lophophore, and
losing themselves at the root of the tentacular arms. They appear to represent the retractile
muscles of the tentacular arms, which produce their flexion upwards and backwards. It has
not been possible for me to discover any distinct trace of any such thin membrane connecting
the basis of the tentacles (the so-called calyx) as is found in the fresh-water Polyzoa. As
regards the tentacles themselves, they are indeed (fig. 8) of the usual cylindrical form for Po-
lyzoa, and are as usual furnished with cilia; but in the living animal they have a very different
appearance from that of the tentacles in the ordinary Polyzoa. While in the latter they are
always in the greater part of their length extended straight, forming a regular corona (see
tab. 2. fig. 28. & 30. d.) which only slightly changes its form, the end of some one or other
of the tentacles being only sometimes bent a little in one direction or another, the tentacles
in the Rhabdopleura are always bent and curved in the most irregular manner in all directions
(fig. 4. 5. 15. 18) so that there can be no question of any regular tentacular corona. The
number of these tentacles, which as above mentioned, are attached in a double row along the
anterior side of the tentacular arms, is somewhat various (about 40 on each arm); they are
longest about in the middle, and are in this part about Y3 of the length of the arm, and di-
minish somewhat towards the base, but more towards the extremity, where they are often
quite rudimentary. The tentacular arms themselves, each of which at the base on the dorsal
side, is furnished with a little fascicle of unusually long cilia attached to a small tubercular
prominence (fig. 15. 17. n) are of very considerable length, quite as long as the whole of the
rest of the body; of narrow cylindrical form, thickest at the base and tapering regularly to

the end, which is obtusely pointed. Anteriorly they are separated (see fig. 16) by a naked,
somewhat concave part, (extending a little downwards) here contiguous to the basis of the
remarkable oval shield (c) which, both by its enormous development, and peculiar function,
below described, forms one of the most peculiar features of the Rhabdopleura.

The Bucal Shield.

Between the bases of the tentacular arms, and from a somewhat ventral point, pro-
ceeds in an anterior direction, a large and very remarkable prominence (fig. 4. 5. &c. c) situ-
ated longitudinally, which has the form of an oblong thick disk or shield, one surface of which
(the dorsal surface) is in the middle grown together with the anterior end of the body, while
the ventral surface is free and bordered by a rather thicker raised ridge, distinct from the
adjacent parts. The form of this disk is, as above mentioned oval, or rather rounded pen-
tagonal (see fig. 18. c) nearly half as long again as wide; the width about equal to that of
the body at the beginning of the stomach. Its posterior border, which at once shews itself
(see fig. 5) separated by a deep constriction from the part of the body lying behind it, is in
the middle slightly incurved (fig. 18). The side borders are, a little in front of the middle,
strongly, almost angularly bent, and then converge strongly towards the anterior freely projec-
ting extremity, which is narrowed obtusely. The whole disk is everywhere, and especially
distinctly on the edges, thickly covered with small vibratory cilia. Allmann, who also mentions
this prominence, but without having gained a correct notion of its form and connexion with
the other parts, says of it that one might take it for a largex and peculiarly developed Epi-
stome, if its position on the ventral side of the mouth, and not, as in the fresh-water Polyzoa,
between the mouth and the anus, did not oppose such a supposition. Allman supposes thus
that the mouth in the Rhabdopleura, in analogy with the other Polyzoa, is terminal, and situ-
ated above, or on the dorsal side of this prominence, between it and the anus. Such is how-
ever not the case. The anterior extremity of the body above the prominence described is
completely closed without any trace of aperture (see fig. 16). On the other hand I have by
gently compressing the animal, been able distinctly to see (see fig. 15) that the bucal aper-
ture (q) is situated just on the ventral or haemal side behind that prominence, and seems to
have the form of a cross slit, which is bordered behind by an oval lobe (m.) furnished with
vibratory cilia like a sort of underlip. By increased pressure the bucal shield could be moved
more out from the base of the tentacular arms; and it appeared then everywhere very dis-
tinctly constricted from the rest of the body, and in the middle of the dorsal surface, fixed
to the body by a sort of short stalk, while the upper and the lower part were free. Further
it was observed that on each side of the bucal shield there extended, from the base of the
tentacular arms downwards, a strongly projecting nearly semilunar .border of thin skin (1)
ciliated on its edges, so that between this and the bucal shield there is formed, on each side,
a narrow half-tube or channel leading to the bucal aperture, and through which the nourish-
ment is probably conveyed to the mouth by the abundantly ciliated tentacles. Since, as above

stated, the said bucal shield is really situated between the mouth and the anus, I think we.
may consider it as morphologically answering to the so-called Epistome in the fresh-water
Polyzoa. Its enormous development in the present instance seems however to indicate that
it must have a very peculiar and important function in the economy of this animal. My ob-
servations on the living animal have also guided me to a decided opinion, to which, however
strange it may appear, I have been forced again and again to return, and which I therefore
must retain: namely that the animal uses this bucal shield (according to my observations), as
a sort of erc<'i>i)i<i-or(/«n. !>>/ means of ich'«-li it can. drntc -itxrjf up to the aperture of its tube.
Since, as above stated, both Endocyst and all muscles of protrusion are wanting, it is in rea-
lity quite inexplicable how the Polypide, which is often found drawn back, not only to the
bottom of the free cell, but even partially into the corresponding chamber of the creeping
stem, (see fig. 10) should be able in any other manner to get forward again so far as to the
aperture of the cell. It might perhaps be supposed that this could be effected by means of
the elasticity of the contractile cord; but I have convinced myself that such is not the case,
by cutting through the contractile cord at its base; the Polypide has continued undisturbed
its slow protrusion, and has also at length really reached the opening of the cell without any
remarkable change. The direct observations made on the uninjured animal have also con-
firmed me in the view expressed above. It will be seen (see fig. 14) that during the slow
protrusion of the Polypide (which often lasts for hours) the bucal shield is always in imme-
diate contact with the wall of the tube, the whole of its ventral side being closely pressed
up against the same; it retains this position unchanged as long as the protrusion lasts;
and the protrusion does not stop until the whole length of the bucal shield is extended out-
side of the aperture of the cell; then the Polypide is completely expanded. On examining
more closely this bucal shield (see fig. 4) we observe in the middle of it an opaque part
which seems to contain an interior glandular organ. Continuing the investigation, and slightly
pressing the animal, we notice however (fig. 18) that this opaque appearance is not produced
by any such internal organ, but by a peculiar and seemingly muscular structure of the sub-
stance of the shield itself. It exhibits, seen from below, in the middle numerous small bub-
bles situated rather far from each other, or somewhat irregularly formed small cells, which
however when more closely examined (and this is particularly evident in those which lie nearer
to the periphery of the disc) shew themselves to be .the external rounded extremities of small
inwardly prolonged cylinders, which together appear to form a thick fascicle of incompletely
differenced muscular fibres penetrating into the stalk of the bucal shield.

The animal rs. with the aforesaid exception of the stomach and the intestine, which
are opaque yellow, colorless and transparent. The tentacular arms, and the tentacles, as also
the anterior part of the body before the stomach, are covered with numerous very small irre-
gularly shaped intensely dark violet spots of coloring matter which also occur on the bucal
shield, and expecially on its anterior freely projecting extremity, where they are very close
together, forming a large roundish dark spot. In specimens in spirits all these parts are

10

dark reddish brown, which probably arises from the diffusion of the dark coloring matter
produced by the spirit.

The Contractile Cord.

As above stated, the Polypide is without any sort of attachment to the cell, in which
it lies quite free. But it is attached by means of a long and thin fleshy cord to the axial
cord which runs through the creeping part of the Polyzoarium or stem. The attachment does
not take place immediately at the bottom of the free cell, but at the bottom of the corre-
sponding chamber of the creeping stem, close to the transversal septum which divides the
chamber from the next preceding (fig. 10 & 11). This cord, issuing from the body of the
Polypide (tig. 3. 5. 10 &c. h) is of very considerable length; as, when the Polypide is expan-
ded, it extends not only through the whole length of the cell, but also through the corre-
sponding chamber of the stem. The cord is however very thin, filiform, when fully extended 5
or 6 times less than the caliber of the cell, and 4 times thinner than the stomach. It is of
cylindrical form, and lies quite free without being attached at any point to the wall of the
cell; but it is nearer to the one side (the ventral) than to the other. Along all one (the
dorsal) side it is covered with the same sort of small dark violet spots of coloring matter as
the anterior part of the body, the tentacular arms &c., but is otherwise quite colorless and
transparent, and is of a soft fieshy consistency. It shews on closer inspection (see fig. 7)
in a part of its substance, an extremely fine fibrous structure of fine parallel longitudinal
lines and less sharply marked transverse lines; but in the dorsal part these fibres are
entirely wanting; and the structure of this part seems to be cellular, and its edge appears
somewhat irregularly wavy. With regard to its attachment to the Polypide, this does not
take place at the bottom of the stomach, but rather high up on the ventral side where it
seems to go over into the thin skin which encloses the digestive apparatus. Its ventral fibrous
part may still be traced (see fig. 15) to a considerable distance forward, in the form of a
rather wide clear skin-border which gradually disappears in front of the Cardia. In this
skin-border, the fine longitudinal fibres may still be distinctly observed diverging like radii;
but I was not able to trace their course further. The posterior end, which as beforesaid is
attached to the Axial cord at the bottom of the chamber in the stem which corresponds to
the cell, is (fig. 6) somwhat enlarged; .and all through of a very distinctly marked cellular
structure, without any evident fibre. In spirit specimens, the contractile cord shews itself
often irregularly thickened in particular places, and is also thus represented in Allman's figures;
but this appears to be only a result of the action of the spirit. In living exemplars I have
always found it, whether fully extended or contracted, of a cylindrical form. When the Po-
lypide, as is frequently the case, is very strongly retracted, not only to the bottom of the
free cell, but also partly in the corresponding chamber of the stem, the contractile cord is
always spirally convolved, so that the coils are closer or looser, accordingly as the retrac-
tion is stronger or weaker (see fig. 10). Also when, after severing the contractile cord

11

at its base, we take the animal out of its cell (fig. 5), the cord always convolves itself in
spiral coils.

Allman has considered this contractile cord as corresponding with the socalled funi-
culus in the ordinary Polyzoa ; although it is not as in these, attached to the end of the stomach
(the terminal caecum), but on the ventral side of the Polypide's body (Allman has represented
it erroneously as attached near the end, on the dorsal side). Moreover Allman indicates that
this funiculus is accompanied by a long fascicle of muscular fibres attached to the chitinous
cord (axial cord) at the point where the funiculus is joined to the same; and that at the
point where the funiculus is joined to the animal's body, this muscular fascicle divides itself in
2 bands, of which one goes along the right side, and the other along the left side of the
body, finally attaching themselves, each on its own side, to the Pharynx below the Lopho-
phorus. These fibres form, according to Allman, the great retractor muscles of the Polypide.
This representation, which in fact only depends on spirit specimens, does not, as may be
seen, agree with what I have had occasion to observe in our northern species, in which the
fascicle of muscles (if one really may venture to use this appellation here) is everywhere, as
an integral part, intimately connected with the contractile cord, and is produced only by a
peculiar modification of its tissue on the ventral side. Special retractor muscles cannot there-
fore, any more than other muscles, be said to be distinguished in the Rhabdopleura.

The Axial Cord (,.chitinous rod" Allman).

Through all the creeping stem there extends, as already mentioned, a filiform cord
very remarkable by its dark, nearly black color, and, unlike the contractile cord, to which
it is about equal in thickness, only slightly flexible, and of a very hard chitine-like consis-
tency. This cord (fig. 3. 9. 10. 11 &c. i) which we will call the axial cord, is freely extended
in the hollow of the individual chambers into which the stem is divided, and only attached
to the septa, which it perforates enlarging itself a little (fig. 11). In conformity with the
rarely branched form of the stem, it is only now and then forked; and when this takes place,
it is always at one of the septa. Otherwise it forms everywhere a cylindrical tube with very
strong, almost horny walls, but always enclosing in its interior (see fig. 12) a soft cellular
cord (s) of similar appearance to the contractile cord, and like it colorless and transparent
with small dark violet spots of coloring matter, but scarcely "half so thick. This fine cellular
marrow which extends through the whole length of the axial cord, seems entitled to be con-
sidered as a sort of incompletely defined nervous trunk connecting all the individuals of the
colony; as at each partition in the stem, it sends forth a branch which enters into the con-
tractile cord of each respective individual animal; and the latter cord does also probably
contain in its dorsal part the imperfectly developed elements of nerves. We may therefore
herein observe the analogon of the so-called Colonial Nervous system (so strongly developed
in the other marine Polyzoa) and specially in the marrow of the axial cord, the common
main trunk of the whole colony.

12

Allman, wlio has drawn special attention to this peculiar chitinous axial cord, (which
does not exist in any of the known Polyzoa), and has precisely derived from it his generic
appellation, calls it a ,,Blastophore" being of opinion that it is destined to bear the so-called
Statoblasts, which he represents as projecting from the posterior part of the contractile cord.
I regret that I cannot, from my own experience, give any decided opinion as to the axial cord
having likewise this destination, because I have not been able to observe the formation of
these so-called Statoblasts.

Vital Phenomena.

The animal, unlike the other Polyzoa, is very slightly sensitive, and is not much af-
fected by -having its tentacles or body touched. If the irritation is strong, it draws itself, but
only very slowly, and usually only a little way, back into its tube. This very slow and slug-
gish retraction, which may last a very long time before it ceases, contrasts strongly with the
extraordinary, almost lightning-like, rapidity with which the retraction takes place in the other
Polyzoa, and is evidently accounted for by the want of special retractor-muscles, and by the
slightly developed contractile elements, not distinguishable as evident muscular fibres, in the
contractile cord, the only instrument by which the retraction of the animal in the Ilhabdo-
pleura is effected.

The extension (protrusion) of the animal is yet far more tardy than the retraction ;
the process is extremely slow and almost imperceptible; several hours may often elapse be-
fore the animal progresses from the stem or bottom of the cell to the aperture of the latter.
Neither do- we, as before remarked, find in the animal under consideration the slightest trace
of any special muscles for such progression; since the Endocyst, and also the parietal and
parieto-vaginal muscles connected with it are entirely wanting. The protrusion seems on the
other hand, as already mentioned, to take place in a very peculiar, and in the highest degree
remarkable manner, that is, solely by means of the enormously developed Epistome (Bucal
shield) which the animal uses — strange at this may sound, — as a sort of creeping organ,
like the foot, or creeping disc of the Gasteropods, to draw itself upwards little by little along
the wall of the cell to the aperture.

• Occurrence.

The Rhabdopleura mirabilis seems to be a genuine deep-sea product, which I have
never found at a less depth than 100 fathoms; but it is probably to be found extensively at
greater depths, where it appears to be more and more plentiful. I have hitherto only found
it in Lofoten, where it is not uncommon, in soft clay bottom, at depths of 100 to 300 fathoms.
As the Polyzoarium is both very small and entirely colorless, it is rather difficult to discover.
Its presence is however easily detected by stirring the washed mud in a fine sieve with a
feather or another instrument, when irregular fibres will be noticed therein. These fibres,
covered with particles of mud, Rhizopod-shells and fragments of mussel-shells, will prove to

13

be the creeping stem, whereon by closer investigation there will be discovered the small
transparent perpendicularly projecting cells. One seldom, however, succeeds in raising these
colonies entire; they are most frequently broken into several pieces by the dredging operation
itself, or in washing out the mud.

Characteristics of Genus and Species.

We may, according to the preceding ^description, characterise the Genus Rhabdopleura
in the following manner.

Gen. Rhabdopleura. Allman.

Polyzoarium tubum form an s tenuem, flexibilem, cylindricum, chitinosum hyalinum ex
stirpe compositum repente intus septis transversis in cameras plures discretas divisa quarum
utraque in cellulam cylindricam plus minusve liberam et erectam stirpe vix angustiorem sub-
tiliter annulatam vel plicis acutis circularibus dense ornatam, oriticio simplice circular! termi-
natam exit. Stirps in tota longitudine chorda chitinosa, obscura, tenui, cylindrica, rigida, pulpa
vero molli cellulosa impleta trajecta.

Polypidcs nullo endocysto vel pallio parietibus cellularum connexi, sed modo funiculo
en nt rartili, tenui et carnoso chordae stirpis chitinosaa affixi, corpore forma elongato-ovata, ex-
tremitate anteriore paulo dilatata et in ramos divisa duos cylindricos et attenuates supra
vergentes et a se divergentes, quibus series duplex tentaculorum flexuosorum affixa est. Se-
ries tentaculorum minime continua sed et supra et infra intervallo distincto interrupta, quare
nulla adest corona tentaculorum vel lophophorus proprie dictu sed modo duo rami tcntacut/fcr/
raldc flcxibiles. Inter bases horum ramorum iuferne adest prominentia magua carnosa scuti-
formis ovalis vel subpentagonalis pedicello brevi et crasso affixa facie inferiore subplana ex-
tremitate antica libere prominente attenuato-truncata, inferiore corpori incumbente medio
leviter emarginato superficie ubique dense ciliata.

()riflcinm oris subventrale, transversum pone prominentiam scutiformem situm, postice
lobo rotundato ciliato limitatum. (Esophagus breyis et spatiosus constrictione distincta a
ventriculo sejunctus; ventriciilus simplex subteres, postice attenuatus et sine tine, ansam sub-
itam formans in intestinum transiens; infest in urn antice porrectum tenue cylindricum lateri
dorsali ventriculi et oesophagi incumbens, orificio anali circular! ad basin ramorum tentaculi-
ferorum supine sito. Tractus intestinalis, vel corpus proprium polypidis, minime nudus, sed
cuticula distincta, tenui, hyalina circumcirca arete circumdatus.

Musculi adsunt nulli distincti, neque retractores, neque protractores. Retractio Poly-
pidis solummodo funiculo contractili etfecta; protnisio singular! modo prominentia effici videtur
scutiformi praeorali, at modum solese gasteropodum. Et retractio et protrusio polypidis seg-
nissima.

Spec. Rhabdopleura mirabilis. (M. Sars).
Polyzoarium irregulariter flexuosum sed raro modo et parce ramosum, stirpe corporibus

14

alienis modo ex parte adhserente ubique particulis alienis vel quisquiliis dense obducta, cel-
lulis vero nudis in tota longitudine liberis perpendiculariter ascendentibus, vario modo flexu-
osis, valde elongatis (15ies — iGics circiter longioribus quam latioribus) stirpe vix angustioribus.
Habitat ad insulas Lofotenses in profunditate 100—300 orgyarum fundo limoso, non
infrequens.

Development.

I have unfortunately not been able to make any observations on the development of
the living animal, in which I have also in vain sought for the organs of generation. It was
only by critically passing in review the specimens which I had brought home one by one,
that my Father at last succeeded in discovering in the creeping stem a couple of polypides
in course of formation. Allman has however been more successful, and has even found in
the specimen of the Rh. normanni examined by him, a whole series of developments, which
is of great interest. I can therefore only add very little to Allman's communications on this
subject. Both of the buds observed by me (fig. 20 x.y.) had their place in separate, every-
where closed chambers of the creeping stem, without these chambers having as yet prolonged
themselves into any cell, and like the developed polypides, appeared here attached to the
axial cord near the bottom of the chambers at a short distance from the transversal septum.
The youngest of the buds (fig. 20 x. fig. 21) corresponds approximately with the youngest
stadium observed by Allman; as only 2 parts were to be distinguished, a short stalk and an
enlarged terminal part, which had not however the form indicated by Allman of 2 compressed
valvules, but of a wide scutiform slightly curved plate (see fig. 21). The stalk (h) which is
strongly, almost globularly enlarged, is continued for some distance along the concave side
of the scutiform plate mentioned; but this continuation, which is not visible from one (the
ventral) side, is by an evident instriction separated from the proper stalk, and represents the
groundwork of the Polypide's real body, whence as well the tentacular arms as the digestive
system are afterwards developed. The other bud (fig. 20 y. fig. 22, 23 & 24) will about an-
swer to the stadium delineated by Allman 1. c. fig. 6. The peculiar scutiform part (c) has
also here the form of a wide evenly curved plate, which already has assumed a somewhat
pentagonal form, and completely covers on one side the still only slightly developed real body,
from which however there project in front 2 tentacular processes (d. d.) extending beyond
the border of the shield, and slightly crenulated at the edges, representing the tentacular
arms; the stalk (h), which represents the contractile cord, has lengthened itself considerably
(see fig. 23.), and its anterior part (fig. 24. h) forms a strong enlargement, marked with evi-
dent traces of the spots of dark violet coloring-matter peculiar to the adult animal. The
real body was, as stated, still very slightly developed, and appeared only as a small rounded
part projecting dorsally between that enlarged part of the stalk and the basis of the tenta-
cular arms; on the ventral side, or that which turned towards the concave surface of the
scutiform plate, it was in the middle, and to a small extent, united to the same by growth;

15

and behind this union there appeared already an evident incurvation or incision in the body
of the Polypide as the first indication of the bucal orifice. That the large scutiform plate is
the homologon of the bucal shield in the fully developed Polypide is sufficiently evident, both
from its position relatively to the animal's body and from its shape. Allman has also recog-
nised this; however when he assimilates this plate in the Rhabdopleura-buds with the mantle
lobes of the Lamellibranchiata, the notion seems to me very hazardous and difficult to esta-
blish. In any case, the early appearance and enormous development of this part in the buds
of the Rhabdopleura, are extremely remarkable.

Allman has — guided especially by the mode of development in the Rhabdopleura — come
to the conclusion that the Polyzoa are not, as was formerly imagined, most nearly related to
the Brachiopods, but to the Lamellibranchiata, and gives 1. c. some schematic figures, in order
to represent more evidently the agreement of the Rhabdopleura-buds with a Lamellibranch.
Allman however presupposes, as taken for granted, that the Rhabdopleura is furnished with
an endocyst of the same nature as the other Polyzoa, which, as above stated, is not the
case; as also his conception of the bucal shield seems to be inaccurate.

As will be seen in the sequel, my Father has acquired a very different notion with regard
to the relationship of the Polyzoa; for — guided by the organisation of their lowest represen-
tative, the Rhabdopleura, — he has arrived at the surprising conclusion that the Polyzoa in all
probability have taken their origin from the Coelenterates, namely the Hydrozoa.

Like Allman, I sometimes found in the middle of the stem of the polyzoarium indi-
vidual chambers which, without containing any bud in process of formation, were quite closed
and not continued into any cell (see fig. 13). The interior wall of these chambers was always
of very dark horn-brown color, and so little transparent that the axial cord, also here running
in the middle, was but dimly discernible through the wall. This dark color was particularly
intense at one of the ends, and appeared to proceed from the axial cord fixed in the middle
of the septa, and here somewhat enlarged. Its exterior horny substance seemed to be directly
continued into the adjacent chamber of the stem. Allman considers these closed chambers,
wherein he has thought to perceive a stratum of large polygonal cells, — of which however I
have not been able to observe the slightest trace in any that I have examined, — as Statoblasts,
and thinks that they are formed by the posterior enlarged part of the contractile cord. This
seems to me however to be far from probable, at least in reference to the closed chambers
examined by me; for they looked far more like remains of old cells decayed, owing to the
destruction of the Polypide, as is found to be the case with other Polyzoa in the oldest part
of the polyzoarium.

Concluding Remarks.

It will be seen from the foregoing description that the Genus Rhabdopleura differs
in nearly all essential points from the ordinary Polyzoa, far more than Allman seems to have
conceived. If we compare what is here communicated with the chief points which have been

16

briefly enumerated by various authors, for instance Allman, Hyatt &c. others, as characteri-
stics of the Polyzoa in general, and which therefore are considered as essential marks most
intimately connected with the idea Polyzoa, the anomaly of this form becomes so striking, as
finally even to justify a doubt as to whether it really can be referred to the class of Polyzoa.
First and foremost stands the want of a so-called Endocyst or Mantle, which sharply distin-
guishes this form from all other known Polyzoa, all of which possess such an appendage. This
Mantle is so essential a component part of a Polyzoon that it is difficult to imagine one
without it. One would rather imagine the Ectocyst wanting, as this plays a far less important
part in the economy of the animal, generally remaining passive, and properly only serving as
a protection for the soft animal. The Mantle is likewise a characteristic for the Tunicates
and the Brachiopods, which two classes have also been united by M. Edwards with the class
Polyzoa, under the common appellation of Molluscoidse (Hackels Himatega or Mantle-animals).

Next, and as a consequence of the absence of a real Endocyst, the retraction and
protrusion of the animal in the Rhabdopleura are effected in a manner totally different from
that of the genuine Polyzoa: it moves up and down in its cell without being attached to the
opening, not by imagination and ecntjination of the anterior part of the cell, and not l>y wend
sets of special separate muscles.

The following remarks on the affinity of Rhabdopleura are from my Father's manu-
script notes:

,,The Rhabdopleura shews in many respects an unmistakable resemblance to certain
Hydrozoa. Just as in these, the individual animals are not attached to the anterior part of
the cells (in the Polyzoa the anterior involved part of the Endocyst is attached all round to
the basis of the tentacular corona); the cells are therefore open, filled (not with the so-called
perigastric liquid, but) with the sea-water entering from without; and the aperture of the cell
is of a defined and invariable shape (while the cells of the Polyzoa are always closed by the
attachment of the Endocyst to the basis of the tentacular Corona, and have therefore no
proper opening; for what is called aperture is nothing more than the part of the cell through
which the animal passes in and out)."

,,Moreover, the retraction of the animal, effected in the Rhabdopleura by means of tie
roitfraelilc cord, at the end of which the animal is suspended, coincides essentially with that
of theHydrozoa in which the part corresponding to that cord (,,the fieshy stalk or axis':, .,the
intestinal canal" (Loven), ,,the branched or unbranched Coenenchym, on which the individual
animals are situated, and which is perforated by a canal-like continuation of the abdominal
cavity of the individual animals") is indeed usually less free (often in many places attached
to the wall of the cell) and possesses a less degree of contractility than in the Rhabdopleura.
but yet in some genera, f. ex. Grammaria, also nearly approaches the Rhabdopleura in these
respects. On the contrary, the Protrusion in the Rhabdopleura is effected in a peculiar man-
ner, and different from that of either Polyzoa or Hydrozoa, namely by a sort of crecpinf/,
executed by the preoral prominence (bucal shield) which appears to answer to the Epistome
in the other Polyzoa, although in these it must have an entirely different function.''

17

,,0n the other hand, the highly developed digestive system, the presence of an anus, the
juxtaposition of the mouth and anus, and finally the bilateral Lophophore, are all characters pe-
culiar to the Polyzoa, and entirely foreign to the Hydrozoa,

,.It is clear that we have under observation in the Rhabdopleura, a form of animal life
which stands as it were in the middle between the Hydrozoa and Polyzoa, or forms a tran-
sition from one to the other; one of those ,,perplexing forms" which will not fit rightly any-
where in the systems of zoologists.

,,When finally, and as the object of the whole investigation, we will give account of
our conception of the Rhabdopleura, and decide on the class to which we will refer it, our
opinion is that these questions, like so many others, can only be properly answered through
the medium of the Darwinian theory.

,,The Rhabdopleura is undoubtedly, like many other animals which at present inhabit
the greater depths of the sea, and with some of which we have in the latter times become
acquainted, a ccnj old form, which in its organisation has still retained several features from
the time when the animal type that we call Polyzoa first developed itself from a lower type.

,,The Polyzoa, which most authors agree in referring to the main type or trunk (phylon)
of the Molluscs are usually supposed, among the other animal types, to shew the greatest
affinity with the Vcrmes; and they are even considered by many zoologists as not being mol-
luscs at all, but as genuine worms. Their affinity to worms has however not been demon-
strated by any evidence, and distinct transition-forms between the two are not known. The
Rhabdopleura shews now evidently that the Polyzoa are not most closely related to the worm-
type, but to the type of the coelent crates, and especially to the class of Hydrosoa. The Po-
lyzoa have already in the earliest primordial times (for fossil remains of them are found in
the lowest Silurian formations) developed themselves from the Hydrozoa by transmutation.
We have in the Rhabdopleura manifestly such a form of Polyzoa in course of development out
of a form of Hydrozoa. The changes which must take place in order that a Hydrozoon can
be transmuted into a Polyzoon consist in the following points. In stead of the simple abdo-
minal cavity of the Hydrozoa, with a single aperture which functions as both mouth and anus,
there is formed an intestinal canal with special walls, dividing itself into 3 sections: gullet,
stomach and intestine, which last ascends alongside of the stomach and gullet, terminating
with an exterior aperture or anus in the vicinity of the mouth. This formation is completed
in the Rhabdopleura, but no more. The following phases of this development which consist
in the formation of a wide sack-like contractile Endocyst or mantle, which in its anterior part
is detached from the Ectocyst or cell, involved in itself (invagination) and attached round
about the basis of the tentacular Corona, (formation of the tentacular sheath, vagina) whereby
the animal that formerly was free in its cell now becomes attached to the anterior part of
the cell, and the <•< Duplicated system of special retractor and protractor muscles, all these muta-
tions have not yet taken place in the Rhabdopleura. This animal has thus remained statio-
nary in the first stage of development from a Hydrozoon to a Polyzoon, but must nevertheless

3

18

be considered as belonging to the type (trunk) of the Polyzoa, since the accomplished de-
velopment of its completely organised digestive system, which is so entirely foreign to the
Hydrozoa, sufficiently stamps it as a Polyzoon.

,,Finally we remark that it may appear strange that the Rhabdopleura, which in all
probability is of so ancient origin, should possess a similar, although modified, form of ten-
tacular Corona (bilateral) to that which belongs to most fresh water Polyzoa (P. hippocrepia,
Gervais, Phylactolaemata Allman) and which is usually considered as a more perfect formation
than the circular (in P. infundibulata G). It is however possible, that the first is properly
the original form, from which the latter has subsequently arisen. The fresh waters appear,
as Hseckel lately has remarked, to contain the direct descendents of some of the eldest animal
forms which, by reason of the less complicated accidents of the fresh waters, have often in
the ,,struggle for life", only slightly altered their original more simple structure; as for
instance: among the coelenterates, the Hydra; among the Rhizopods, the Actinrtylirya, Gromia
an'd the shell-less Radiolari lately discovered by Focke; among the fish, the Ganoida &c.

2, Flustra abyssicola, M. Sars,

nov. sp.
(PL 2. fig. 25—30).

Of this remarkable deep-water species my Father had already elaborated a full de-
scription, which I here quste as exactly as possible from his manuscript.

The colony is similar to that of F. foliacea, forming an upright leaf-like stem, com-
pressed from two sides (see fig. 26) (but less compressed or rather thicker than our other
Norwegian species) sometimes undivided, but most frequently scantily branched (fig. 27) na-
mely only once or twice bifurcate, with diverging branches all extended in the same, or
nearly in the same plane, and gradually becoming much wider towards their extremity,
which is truncated, straight, or but slightly curved. The undivided specimens are more linear,
or of nearly the s'ame width everywhere, excepting in the top which gradually becomes wider.
It is only loosely attached to extraneous bodies which lie at the bottom of the sea (small
shells of conchiferae, Rhizopods, grains of sand &c.) by means of numerous very fine soft or
membranous cylindrical hyaline root-fibres proceeding from the inferior extremity, and having
their extreme points so attached. As it is rigid, not flexible, it is also brittle and easily
broken, for which reason perfect specimens are difficult to obtain. The largest specimen found
is 32 millimetres high; at the base 1V2 millimetres wide; higher up 2— 2V2, and at the end
of the branches 5 millimetres wide. The thickness is everywhere the same, a little over Y2
millimetre. Imperfect specimens were also found, the branches of which were 8 millimetres
wide at the extremity.

The cells, which form two parallel layers, attached back to back to each other (one
on each of the two broad sides of the colony) are situated in straight longitudinal and oblique
transverse rows, or in somewhat irregular quincunx (see fig. 25) as the regular arrangement
is here and there interrupted by the intervening (interposite) avicularia, of which more here-
after. At the inferior end of the stem there are a few (2—3) ; at the superior end, or at the
extremity of the branches, there may be 8 — 16 cells in a transverse row. The narrow side-
borders of the colony are formed (fig. 26) by two longitudinal rows of cells meeting at an
obtuse angle, one row from each of the two broad sides or surfaces.

The cells are of a membranous or parchment-like quality, of greyish brown color, and
tolerably transparent, so that the glassy white animals are clearly seen through them (see
fig. 28) about 1 millimetre long and l/2 millimetre broad, that is larger than in our other

Norwegian species. Their shape is elongated tongue-like, or similar to that of a bathing

3*

20

vessel; the superior or distal edge evenly rounded or curved; the proximal edge truncated,
and both the side edges or partitions only slightly convex, often nearly straight, erect or
prominent over the surface of the anterior side, which is flattened and smooth. These par-
titions are calcareous, strongly marked, and shew themselves as narrow hands of nearly opaque
white color.

All formation of thorns or bristles is wanting on the cells. Close below their distal
border lies the opening, in the form of a transversal half moon- or crescent-like fissure, the
membranous lid of which sometimes appears raised, and in other cases closed.

What is most remarkable in the present species is its avicularia: they are both in
size and shape similar to the cells. On closer examination they appear (see fig. 25 a) to be
about of the same width as the cells (sometimes, especially in the distal part, even a little
wider) but always a good deal namely V3 — 1/2 longer (or iy3 — iy2 millimetre, long and V2
millimetre wide) as their proximal part forms an acute angle, while they are otherwise exactly
of the same shape as the cells.

They are sparingly and pretty regularly or uniformly distributed over the whole colony;
their number is therefore considerably (about 7 — 9 times) less than that of the cells. They
appear always as a beginning to one of the two longitudinal rows, into which an elder row
of cells resolves itself in its accrescence. The r>i<tiutil>k> occupies the whole of the anterior
side of the avicularium, excepting the lately named lower part terminating in an acute angle,
which is occupied by the reccpfacidum, or that part of the avicularium, or the special cell (if
one prefers to call it so) which lies behind or below the articulation of the mandible, and
encloses the muscular apparatus for the movement of the mandible. This articulation is
formed by the distal extremity of the receptacle, a transversal straight line or border (one
side of the triangle) whereon the basis of the mandible, which also forms a transversal straight
line or border, is movably inserted. In dead colonies where the mandible has fallen off, there
is found in the place where it had been attached, namely close above the articulation, an
eliptical cavity in a longitudinal direction occupying nearly V3 of the length of the avicularium,
as also instead of the receptaculum there is a large triangular cavity corresponding to its
outline. The anterior side of the receptacle is flattened, but the anterior side of the proper
avicularium is evenly convex or vaulted, and projects therefore considerably over the surface
of the colony formed by the cells. This convexity is caused only by the mandible, which when
lowered, exactly and closely covers the proper avicularium or cell ; but when the mandible is
raised, it shews itself to be entirely coincident with the cell, having the same size and shape,
namely the anterior side flattened, the distal edge curved, the proximal truncated, and both
the side-borders (partitions) projecting and calcareous like small white stripes.

From what is already stated it will appear that the mandible in the F. abyssicola,
is, like the cells, of an elongated tongue-shape, but with convex or vaulted exterior surface
(the anterior side). Its opposite or interior surface is on the contrary strongly concave ; and
consequently the mandible is somewhat thin, and resembles the bmvl of a spoon.

21

On closer examination the mandible appears (fig. 3D) to consist of a middle-part occu-
pying the whole of its length, the proper mandible, and two side parts. The middle part is
in the whole of its length slightly and evenly curved inwards; in its lower or basal part, thicker;
and upwards, or towards the free end, gradually thinner being a strong or hard chitinous
plate of narrow lancet-like shape pointed at the extremity, truncated at the basis and forming
a transversal straight line, or properly a thickened stripe with which the mandible is movably
inserted in the distal, likewise straight edge of the receptacle, and of which both the side-
borders are slightly concave. These borders are of a horny-yellow color, and thickened or
like fillets, particularly the basal border, while the side-borders become gradually thinner
towards their superior part. Both the side-parts of the mandible are on the contrary extre-
mely thin, membranous, somewhat soft and hyaline; they surround the sides of the middle
part, and also its point like a broad brim, and their border is nearly straight or slightly
convex, and at the extremity of the mandible where they join, evenly curved. After dessica-
tioii the mandible retains its shape completely, its chitinous stripes retaining their horny-
yellow color, and is distinguished from the nearly dull cells by its shining surface.

The animal (the ,,Polypide" Allman) (fig. 28 b. fig. 30) is entirely colorless and trans-
parent with slightly yellowish intestinal region; and the mouth is surrounded with a circle
of 24 tentacles. On the slightest irritation, the animal withdraws itself as usual with the
Polyzoa, rapidly, one might almost say quick as lightning, and with a jerk into its cell.

The mandible of the avicularia is found extended in various positions: sometimes more
or less obliquely upwards and outwards from the colony (see fig. 27. a. a.) sometimes straight
out from it (fig. 26) and finally sometimes curved downwards or backwards (fig. 26. fig. 28 a.)
The last position is the most usual-, when the animals of the colony are in a state of full
undisturbed tranquility. In case of irritation the mandible is curved somewhat rapidly upward,
and is laid close up to the cell of the avicularium which it then exactly covers (fig. 25. a.a.)
The opening or extension of the mandible seems to take place still more rapidly and ener-
getically than its closing. A loose colony, or one separated from its attachment, lying on the
bottom of a vessel filled with sea water, was even observed to move a little way, or as it
were to make a leap, by the combined effort of the mandibles of all the avicularia, as they,
by rapidly and energetically opening themselves, and curving outward and downward, pressed
against the bottom of the vessel.

The present species, which occurs somewhat rarely in Lofoten at the depth of 120
fathoms (at Guldbrands0er) to 300 fathoms (at Skraaven) is distinguished, as has been shewn,
from the hitherto known species ofFlustra, principally by its avicularia and especially by the
great size and peculiar form of the mandible.

The largest avicularia (and those most resembling the animals' cells) which are hitherto
known, are found in the Flustra Barlei Busk (F. membranaceo-truncata Smitt). Where they
are of Y3 — Va the size of the animals' cells, and in F. foliacea Linne, where they are nearly
equal in size to the cells. Both these species have however a quadrangular receptacle, and

22

the mandible, which is of about the. same size as the receptacle, or about half the length of
the avicularium is in the form of a crescent.

The avicularia in the F. abyssicola are still larger than in any of these species, being
Vs — Va longer than the cells, to which they are also similar in form, excepting that their
proximal end is pointed, and the receptacle thus triangular. Lastly the mandible is much larger
than in those species, namely of the length and breadth of the cells and of an entirely pe-
culiar form, not like a half moon but like a spoon. The only form resembling that here
described which I have been able to find noticed by the authors with whom I am acquainted
as having written on the subject of Polyzoa, is one described and delineated by Busk under
the name of Eschara Platalea (Catalogue of the Polyzoa in the British Museum p. 90, tab. 105
fig. 1 — 3). It resembles a spoon with a short narrower handle, but is without any such thick-
ened chitinous lancet-shaped middle-plate as is found in our Flustra abyssicola, representing
the body of the proper mandible which has here, — while the avicularium is still an organ
co-ordinate with the cells of the colony — already assumed a strongly elongated triangular form,
and also in its whole length is somewhat curved.

The question as to the proper significance of the avicularia seems, in the preseiit
species more than in any other to find its solution. They shew themselves here most evi-
dently homologous with the individual animals themselves (the cell and the Polypide therein
contained) they are only modifications or reconstructions of the same, destined chiefly for
the defence of the colony, and perhaps also in some degree for its support, by retaining the
animals of which the more or less decomposed parts may serve as food for the individual
members (the Polypides).

The Flustra abyssicola is evidently one of the oldest now living forms of Polyzoa,
which seems like the Rhabdopleura to be much spread in the great depths of the Ocean.

The new species may be recognised by the following diagnosis.

Flustra abyssicola. M. Sars.

Colonia in stirpis formam frondosa fusco-cinerea, subpellucida, rigida, radicata, parce,
ramosa, ramis truncatis. Zoooecia utramque paginam stirpis tegentia, elongato-linguiforma
setis carentia. Avicularia longitudinem zoooeciorum superantia, eorumque formam referentia
sed inferne acuminata, inde receptaculum triangulare; mandibula magnitudinem zoooeciorum
aequans, cochleariformis, e parte media longitudinal! dura crassiore lanceolata (mandibula
propria) marginibus fusco-luteis et utrinque parte lateral! membranacea tenuissima molli con-
stans. Oooecia desunt. Animalia hyalina ore tentaculis 24 cincto.

Habitat rarior ad insulas Lofoten profunditate 120 — 300 orgy arum.

C o n c h. i f e i* a.

3. Yoldia obtusa, M, Sars,

(PI. 3 fig. 16—20).

Yoldia abyssicola. M. Sars. ,,Christiania Vid.-Selskabs Forhandlinger" for 1858 p. 86.

This very distinct species, first discovered by Mr Conservator Keren in the Herl0fjord
near Bergen at a depth of 300 fathoms, is also one of those peculiar animal forms which we
certainly only know from the superior limit for their occurrence; while in all probability it
exists very extensively in the greater depths of the Ocean.

My father thought it expedient to change the previously used appellation abyssicola
and to substitute the above, in order to avoid confusion with the very different arctic variety
of the Yoldia pygmsea Minister, to which the same name abyssicola had been given by Torell. 1

The Shell (fig. 16, 17) which in the largest specimens has a length of 13 millimetres,
is somewhat strongly compressed from the sides, very thin and fragile, and of a broad elip-
tical form; its length is nearly* twice the height; and its thickness is considerably (nearly V2)
less than the height. The sides are very inequal ; as the umbones, which are small and slightly
prominent, are situated far in front, at the limit of the anterior third part of the length of
the shell. Anteriorly the shell is narrowly rounded, but posteriorly widened little by little, and
at last almost abruptly truncated. The ventral border forms in its anterior part, together
with the anterior border, a perfectly even curve; while further back it is more horizontal,
going over, with a very strong and sudden curve, into the posterior border, which is nearly
straight and perpendicular. The dorsal border is, in front of the umbones, somewhat in-
clined downward, going evenly over into the anterior edges ; but behind the umbones it
is quite straight and horizontal, forming with the posterior border a very strong nearly
angular curve. The shell is externally quite smooth, with extremely fine concentric lines
of growth, and covered with a strongly shining yellow green epidermis; internally it is of a
dull white color.

The cartilage-groove is small and indistinctly defined, longitudinal, narrow, and not,

1 "Chr. Vid.-Selskabs Forhandlinger" for 1868. p. 256.

24

as in the other species, spoon-shaped. But there^is a very strongly developed external ligament
of horn-brown color, projecting from the umbones, and extending half way between them and
the posterior extremity.

The cardinal teeth are numerous, but somewhat various in number. They are however
always most numerous in the posterior part, where their number may even amount to 27.
In the anterior part there are usually about 14 teeth. In 6 specimens, the following pro-
portions were observed.

Dentes anteriores 14 posteriores 24
14 22

11 18

13 19

14 27

15 27

Of the muscular impressions (see fig. 18) the posterior is largest, and of an oval-
triangular form, and, like the anterior, situated high up under the border of the hinge. The
ptillinl impression has towards its posterior part a deep rounded sinus, which extends nearly
to the middle of the length of the shell.

Very small specimens (fig. 20) of 5 millimetres in length, have the ventral border
somewhat less curved, and the posterior border more straightly truncated. In such small
specimens, the shell is quite transparent, so that the animal may be distinctly seen through
the shell.

The. animal is highly remarkable by the proportions of the siphons, of which the anal
(a) is quite unusually long, namely more than the whole length of the shell, and usually more
or less curved or crooked downward at the extremity; while the branchial (b) is particularly
short, and scarcely projecting beyond the posterior border of the shell.

Between the two there is observed a long filiform appendage (c) nearly of the same
length as the anal siphon, doubtless a pallial tentacle.

The foot (d) is of the usual shape in this Genus, forming at the extremity a widened
creeping-disc, wliich consists as it were of 2 lobes and is crenulated at the edges. Above
this in the middle of the body, the crescent-formed transversally placed gills (e) may be
seen, and above these again a semi-circular dark line (f), curved towards the umbones, which
is the intestinal canal filled with excrement.

The present species seems to be widely diffused in the great depths off our coast.
It was, as has been already stated, first discovered in the Herl0fjord near Bergen by Mr. Con-
servator Koren. L have myself found it pretty often in the great deeps at Lofoten, in 200—
400 fathoms ; also in the Hardangerfjord, where it goes down to 500 fathoms, and lastly during
my recent stay at Aalesund, in the so-called Brei-sunds deep in 200 fathoms. It occurs every
where, like the other species of this Genus, only on a bottom of soft clay or mud.

The species may be known by the following diagnosis.

25

Yoldia oMusa. M. Bars.

Testa teuuissima, alba, laevis, perparum convexa, elliptica. maxinie inequilatera, antic e
anguste rotundata, postice altior et rotundato-truncata, epidermide luteo-virente, nitida, con-
centrice teuui-striata. Margo dorsalis posticus fere rectus, ventralis paruin arcuatus. Um-
bones parvi. Fovea ligament! minuta longitudinalis ; ligamentura vero externum magnum et
prominens. Denticuli cardinales parvi, antice 11 — 15 postice 18 — 27.

Animal siphonibus disjunctis, anali perlongo et angusto, branchiali brevissimo prse-
ditum; teutaculis pallialibus singulis valde elongatis filiformibus in vicinia siphonum.

Longit. testae majorum 13 Mm., altit. 8 Mm., crassit. (valvarum ambarum) 5 Mm.

Habitat ad oras Norvegise pluribus locis in prof. 200 — 500 orgyar.

Of the species hitherto known it resembles most the Lcda obesa Stimpson (Shells
of New England, p. 10. tab. 2. fig. 1.), but is more than twice as large (L. obesa is only 6
millimetres long) and has the posterior side both longer and higher; and lastly the cardinal
teeth (which in the L. obesa are according to Stimpson, dent. ant. 10, post. 12,) are more
numerous, viz. dent. ant. 11 — 15, post. 18 — 27.

4, Pecchiolia abyssicola, (M, Sars),

(PI. 3. fig. 21—43.)

Lyonsiella abyxsicola. M. Sars. Vid.-Selskabs Forhandliuger for 1868 p. 257.

This remarkable deep-sea form was first discovered by me at Lofoten, at the consi-
derable depth of 300 fathoms, and was noted by my father 1. c. as a new genus and species
under the denomination Lyonsidla al(i*»i<-<>la. It has been subsequently rediscovered during
the expeditions of the Porcupine, at still greater depths in the Atlantic Ocean, and referred
by Jeffreys, the first and most thorough conchologist of our time, to the genus Pecchiolia,
Meneghini (Verticordia, Searles Wood) of which we .have hitherto known only one single living
species from the tropical oceans, and some few fossils from the chalk formations. The animal
of the Pecchiolia has hitherto been entirely unknown; so that we have only had the exterior
characters (the shell) to guide us in determining its systematic place, of which also different
authors have had very different conceptions. The discovery of a living form of this genus in
our northern ocean is therefore of great interest to us; as its proper place in the system
can thereby be ascertained with greater accuracy.

In preparing the following description, I have had rather detailed notes of my Father

to consult.

4

///. xhtll has (see fig. 21) by its very inequilateral form, at first glance some resem-
blance to the shell of the genus Crenella. Brown, but differs from it, as will hereafter appear,
in many essential points. It is in the largest specimens 5 millimetres long. 4 millimetres
high in the posterior part and 3 mill, in the anterior part, the thickness across both valves
together 3 mill. It is however .often smaller 3—4 mill. long. The xhripr is rhomboidal or
oval-trapezoidal, inflated, the posterior part somewhat compressed, the middle part extending
from the umbones to the posterior ventral corner (or diagonally) strongly convex.

The anterior extremity is narrowly rounded, much lower (by nearly half) than the
posterior extremity which is vertically truncated and nearly straight, sometimes even slightly
incurved in the middle and with rounded corners. The dorsal and ventral borders are like-
wise nearly straight and parallel, or only slightly convex, and form a right angle with the
posterior border, but with rounded corners. The umbones lie quite in front, so that the anterior
extremity of the shell only advances a very little beyond them, and near to each other (sub-
contigui); they are (see fig. 24) 'strongly projecting forwards and inwards and somewhat spi-
rally involved.- There are no evident lunula. The right valve (fig. 25) is a very little larger
than the left (tig. 26) especially in the posterior part, so that both its dorsal and ventral
borders project a little beyond those of the left valve, and are curved a little outwards round
the latter (see fig. 21 & 23).

In other respects the valves close accurately without gaping at any point and tin-
edges of the shell are everywhere smooth, not dentated.

The shell is of tolerably thin and fragile consistency; whitish and somewhat trans-
parent, the exterior surface having only a slight or dull polish; the interior surface brighter,
with a weak mother-of-pearl-like iridescent lustre.

Its outer surface, where there appears to be a very thin hyaline epidermis (covered
with strongly agglutinated fine grains of sand) which howrever cannot be detached from the
shell, is, like that of the genus Lyonsia, radiately striped with fine raised smooth lines distant
from each other and in number 1G— 20 running from the umbones to the edge. The intervals
between these radiating lines are extremely finely punctuated; and the points appear under
the microscope (fig. 29. 30) to be somewhat projecting small cylindrical or truncated conical
processes in radial rows, 6—8 rows in each interval. The lines of growth are extremely fine
and close, parallel with the edge and difficult to perceive.

The interior surface of the shell is smooth, and shews (see fig. 25 A: 20) 2 roundish
or somewhat oval slight adductor-impressions, the anterior one very distinct and situated near
the middle of the anterior edge; the posterior one, less evident and situated close under
the posterior superior corner. The impression of the mantle-edge (the mantle-line) is simple
without any sinus.

The khnjc is without teeth; the ligament internal, situated in a somewhat oblique
longitudinal groove near the dorsal edge of each valve, and containing, as in the Lyonsia,
a small opaque-white free movable ossicle (fig. 26. 27. 28 a) which lies longitudinally under

2?

the hinge, and is oblong (the length being about double the width) convex below, flat-concave
above, truncated at both extremities, and here somewhat hollowed out.

The animal (fig. 31) has both lobes of the mantle grown together, with exception of
an aperture for the foot, and 2 siphonal apertures. The aperture for the foot is (see fig.
35) situated a little before the middle of the animal's length, and is elliptical (in the direction
of the animal's axis) with continuous edge, occupying about V3 of the animal's length.

The siphonal apertures are circular, sub-sessile; the branchial (fig. 31 b) not at all
prominent, the anal (a) only a little prominent like a short conical tube. The branchial
siphonal aperture (fig. 36. 39 b) is large, and its edge surrounded by a circle of 12 propor-
tionally lurge, couically tapering, or awl-shaped tentacles (ibid. t. t.). These tentacles are
(see fig. 40) pinnate, or show on each side 4—5 short blunt lateral branches, which like the.,
whole outermost narrower or tapered part of the tentacles are covered with small prolonged
awl-shaped soft papillae (fig. 41).

The anal siphonal aperture (tig. 36, 39 a) is small, scarcely half as hirge as the bran-
chial, and without tentacles; but the siphon itself is (see fig. 32) everywhere thickly covered
with fine taper-pointed papillge, (fig. 34). Its aperture, as could be clearly seen in living
specimens, is not perfectly round, but rather heart-shaped (fig. 33) the lower wall of the
siphon being somewhat incurved in the middle.

A row of 16 small papillge (Y2 — V3 of the size of the branchial tentacles) which are
simply conical, forms round both siphons or along the posterior edge of the mantle, an oval
border elevated in the direction of the animal's height (fig. 36. 39 xx).

The body (fig. 37. 39. 42) is roundish or trapezoidal, strongly inflated and of white
color. The foot which is also white, proceeds (fig. 42 d) from, or a little before, the middle
of the ventral side of the body; it is shaped, as it were, like a knee at the base, forming
nearly a right angle with the body, and is otherwise extended forwards ; it is slender, cylindro-
conical, slightly compressed at the sides and rounded at the extremity. It is, when contrac-
ted, scarcely half as long as the animal, but has, when extended in the living state (fig.
31. d) a length nearly equal to that of the animal itself. On its ventral side near the
basis, it has a slight medial longitudinal groove, whence proceeds a long, but very thin lyssus,
(fig. 35 i.) consisting of extremely fine parallel tendinous, or strongly transparent and colorless
threads. About the anterior half of the dorsal side of the animal is occupied by a glandular
mass which fills the hollow of the umbones of the shell, (in the individual examined, the
ovary) (fig. 37 o.o.) and consists of tolerably large globular hyaline cseca or capsules (fig. 38)
attached to a very short and slender cylindrical stem, and each of which has a globular or
oval yellowish white egg enclosed in, but far from filling its cavity. In the smaller eggs the
globular nucleus was very evident: in the larger, it had disappeared. Under the ovary lies the
reddish brown liver.

The gills, of which there is apparently only one on each side of the body (fig. 42 e)

are light-brown, elongated, and curved nearly in a cresent-like form: in the posterior free

4*

28

extremity, shortly pointed; and this extremity does not extend nearly so far as to the siphons.
They are, along their dorsal edge, connected with the body, and by a longitudinal groove
divided in 2 parts (fig. 43) of which the upper is only about half as wide as the lower; both
consist of numerous transverse lamellae. The palpa were not to be recognised: they appear
to be obsolete or rudimentary.

This remarkable shell occurs, somewhat rarely, in Lofoten at the depth of 120—300
fathoms on clay bottom; likewise in the Hardangerfjord, where it has also been found by
Danielsen at a depth of even 500 fathoms. Moreover it has been found at still greater depths
in many places in the Atlantic ocean during the British expeditions (with the Porcupine).

The Genus may accordingly be characterised as follows.

Gen. Pecchiolia. Meneghini.

Testa inaequivalvis, valva dextra majore, valde iimequilatera, ovata, rhombica, seu tra-
pezoidea, tenuis inflata undique clausa intus submargaritacea. Umbones prorsus antice siti,
tumidi prominentes, incurvati, antrorsum involuti subspirales. Cardo edentulus; ligamentum
internum in fissura linear! cujusque valvai sub umbones situm et ossiculum continens oblon-
gum planulatum testaceum mobile. Impressiones musculorum adductorum parvse rotundatse
seu subovales, anterior bene distincta prope marginem anteriorum sita; posterior subobsoleta
margini dorsali approximata; linea pallialis integra sinu nullo. Epidermis tenuissima parum
distincta.

Animal crassum, pallio occluso apertura oblonga ventral! subanteriore pro pede emit-
tendo exc.epta. Siphones separati subsessiles; branchialis non prominens, amplus, margine
tentaculis cinctus sat magnis utrinque pinnulatis papillisque numerosis subulatis obsitis; analis
prominens angustior subconicus, margine integro (absque tentaculis). Margo posterior pallii
siphones ambos cingens serie papillarum conicarum ornatus. Pes subcylindricus seu subulatus
longus ad basin bysso longo tenuissimo ornatus. Branchiae elongato-subsemilunares utrinque
apparent singulse, sed sulco longitudinal! in duas partes transverse plicatas inferiorem supe-
riore fere duplo latiorem divisse.

Spec. Pecchiolia abyssicola, M. Sars.

Testa parva, alba, translucida, parum nitida, arenula tecta, striis radiantibus tenuissimis
distantibus; intersticiis radiatim punctatis, antice rotundata, postice truncata, angulis posteri-
oribus rotundatis marginibus Isevibus, dorsali et ventrali sub-parallelis. Longit. 5 mill. ; altit
4 mill.; crassit. (valvarum ambarum) 3 mill.

Habitat ad insulas Lofotenses in prof. 120—300 orgyar. fundo argillaceo arenoso, nee
non in sinu Hardangerfjord usque ad 500 orgyas. Ceterum in magno abysso Ocean! Atlan-
tic! a clariss. Jeffreys inventa.

According to obliging communication from Jeffreys, the genus Pccdiiolia was established
by Prof. Meneghini in ,,Revue et Magazin de Zoologie (Guerin M^neville) 2e. serie. T. TV.

29

1852. p. 578 (a work which I have not had occasion to .see). The type of this genus is Cham a
argentea Mariti (1797) = Chanta? arietina Brocchi (1814) a Miocene fossil. The Synonyms of
this genus are: Hippagus Phil, (non Lea) and Verticordia Searles Wood (Trigonulina cl'Orh).

If we compare the characteristics noted by the different authors (e. gr. the Brothers
Adams, Chenu, Woodward) for the above-named genera, we shall find some apparently very
essential discrepancies, which justify the assumption that not all the forms thereto referred
really belong together. This applies especially to the hinge, and also to the proportion be-
tween the two valves, which the anthorities mentioned represent as equal, while in our nor-
thern form they are evidently unequal. Both these points are, as is well known, usually re-
garded as having generic significance. Jeffreys has however communicated to me that he
looks on the typical form P. Argentea Mariti, and also P. acuticostata (Hippagus) Philippi l
as undoubtedly belonging to the same genus as the northern form here described. With
respect to the latter of these two species, Philippi states indeed like the other authors,
that the shell is sequivalve; and further he does not mention the peculiar movable ossicle
lying between the hinge-borders of both valves; but both these points may easily have escaped
his attention, as he only had occasion to examine 2 single (left) valves. In the outline of
the shell, our species has no great resemblance with the P. acuticostata Phil., which is roun-
dish or heart-shaped with strongly projecting ribs and dentated edges. But it is more like
the Verticordia parisiensis Deshayes2 from the chalk at Grignon, and V. leana Desh. from
the chalk 'at Monchy8, which are both oval (longitudinally); but the first has thin, yet pro-
jecting ribs, and therefore dentated edges; the latter on the contrary has no ribs, and there-
fore smooth edges like our northern species, which is distinguished from both by its rhom-
boidal shape, and by the posterior nearly straight side or edge being much higher than the
anterior rounded side.

The animal of the Pecchiolia is, as above stated, hitherto unknown; for which reason
it has been referred to various families, only according to the presumed similarity of the
shells. The Brothers Adams, like Chenu, refer it to the family of the Trigonidaj; Philippi, to
the family Carcliidae; Woodward4, to the family Cyprinidae. With regard to the shell of our
species, it presents two characteristics — namely that its valves are unequal (the right larger
than the left) and that its ligament contains a loose ossicle, — which are foreign to the
above-named families, but exist in the family Anatinidee (incl. Pandoridse) to which it therefore
appears to be referable; although its shell does* not, as in these, gape at both ends, nor has
any external ligament. As to the animal, it agrees with the Anatinidaa in having both lobes
of the mantle united, with only one opening below and in front for the foot, and the foot —
as in some of the Anatiniclse (Lyonsia) — furnished at its basis with abyssus; but its Siphons

Fauna Molluscorum regni utriusque Sicilian 2. |>ag. 42. tab. XIV. fig. 19.

2 Chenu. Manuel de Conchyliologie 1862. Vol. 2. fig. 340.

3 Ibidem, fig. 842.

4 Manual of the Mollusca I860, p 304.

30

are sub-sessile, while in most of the Anatinidse they are long; and only the branchial siphon
is surrounded by tentacles, but not the anal ; while the openings for both siphons in the Ana-
tinidffi are fringed, or furnished with tentacles. These differences appear however to be of
secondary importance; and on that account the animal may be referred to this family. On
the contrary, it cannot be placed among the Trigonidse, Cyprinidse or Cardiidse, to which the
several authors have referred this genus; as the animals of the said three families have the
lobes of their mantles separate, and equal shells.

Of our native genera the genus Pecchiolia seems to stand nearest to the genus
Lyonsia, Turt., from which it differs in the following points : the shell is much shorter, not
gaping, the right shell is larger than the left (the contrary is the case in the Lyonsia) the
umbones lie quite in front near the anterior extremity of the shell, and are somewhat spirally
involved. The animal has separate siphons (in the Lyonsia they are united nearly in their
whole length) of which the anal is prominent, and the branchial sessile having its opening
surrounded by pinnate tentacles covered with papillae (the tentacles of the Lyonsia are simple
smooth threads) and the anal is without tentacles (which exist in the Lyonsia). The foot is
slender cylindro-conical with blunt rounded end, (in the Lyonsia it is tongue-shaped and flat)
and the aperture in the mantle through which it is extended, has an entire border (in the
Lyonsia it is according to Forbes &Hanley fringed); the lip tentacles appear to be rudimen-
tary or wanting (in the Lyonsia they are said to exist and to be triangular, slender and
folded on the interior side). The circle of simple tentacles, on the border of the mantle,
surrounding the siphons, seems also to be wanting in the Lyonsia.

It may further be remarked in conclusion that the conchifer now under examination
has with respect to its shell considerable resemblance with the genus Mytilimeria Conrad,
which (for the animal seems to be unknown) is referred by Woodward to the Mytilidse, but
byChenu (I. c. p. 39) certainly more correctly to the Anatinidse, with the following characters:
,,The shell roundish (subarrondie) more or less inflated, equivalve, fragile, covered with a thin
,,caduke epidermis. The umbones subspiral; the hinge without teeth, but formed only of
,,small linear excavations under the umbones to receive the ligament, which contains a small
,,bone (osselet). The muscular impressions small ; the mantle-impression with an obtuse sinus."
M. Nuttalii Conrad, fig. 175.

31

Cephalophor'a.

/

5, Dentalium agile. M. Sars.

n. sp.
(PI. III. %. 4—15.)

Dentalium incertum. Philippi fauna Molluscorum regni utriusque Sicilian Vol. 2. p. 207
(non Deshayes).

This species, living in the great depths of our seas, for the identification of which with
the fossile Dentalium incertum Philippi (non Deshayes) found in the tertiary formation in
Calabria, I am indebted to Jeffreys, is certainly very closely allied to I), abyssorum M. Sars;
and is perlaps also, as Jeffreys thinks, properly to be considered as only a peculiar deep-sea
variety of the same. However I note it here preliminarily as a distinct species, under the
specific denomination assigned to it by my Father, since as will appear in the sequel, it shews
not a few differences from the ordinary D. abyssorum. If it is really a variety of this latter,
the variation is in any case so considerable as to affect the idea of species. To draw a sharp
boundary line between what is to be understood by a strongly marked constant variety, and
by a species, is moreover --at least from the stand-point of the Darwinian theory, no easy
matter; for, according to this theory, the species have had their origin precisely in such va-
rieties. It seems to me also that we might with nearly as good reason include in the same
category the typical D. entalis L. which in nearly all respects, excepting the striation of the
shell, coincides with the D. abyssorum, and consider all these 3 forms as originally belonging
together, or in a comparatively new time developed into independent forms (species), in which
case the deep-sea species here described would probably be best entitled to rank as the
proper parent form.

The shell (Fig. 4 — 6. tig. 10) which in the largest specimens attains the considerable
length of 58 mill, is immediately distinguished from that of the I), abyssorum by its much
more slender, and less curved shape. Sometimes -- and this is, contrarily to the rule, espe-
cially the case in larger exemplars (see fig. 4), it is nearly quite straight with only an exces-
sively slightly marked ventral curve. Where the curve is greater (see fig. 6), it is however
quite uniform; while in the D. abyssorum the proper curve of the shell begins only towards
the posterior extremity. Neither does the shell taper so strongly and suddenly towards the
posterior end as in D. abyssorum, but always quite evenly from the anterior to the posterior
extremity. The whole shell has therefore also a more regular cylindro-coriical shape than
that of the other Dentalia. To facilitate comparison, the measurements of several specimens
of all our 3 species of Dentalia are here given.

Dattalium agile. M.
Sars.

Dentalium abyssorum.
M. Sars.

M

length of shell.

width at aperture.

width in middle.

width at apex.

I 1

58 Mill.

4 Mill.

3 Mill.

1 Mill.

2

45

3

21 —

±,

I 3

40 —

3 —

2

i -

4

51

31

a*

2.

f 5-

57

4 —

3 —

1

, 6

34{ -

3 —

o

1

1

49

r> —

4

H -

i 2
I

45

4

3-

1

1 3

46

4|

3* -

1

•

48

4J

3 —

1

/ 5

24

3 —

21

3 -

6

53 —

5 _

4

1

\

1

42

5 —

4

11

2

35

H

3* -

f

3

21

31

21

-?, —

4

36 —

5 —

4

i|

Dentaliiim entail*.
Linne.

The Shell is untransparent, or opaque, of white color, and usually quite smooth, with-
out visible longitudinal stripes or ribs. It resembles therefore in this respect more the D.
entalis, from which it is however at once distinguished, as well by its slender and elongated
shape, as by its duller and less lustrous surface. We observe on it, as on that of the D.
entalis, evident lines of growth, which are circular or somewhat oblique, parallel with the
anterior edge, and which at certain- intervals are more distinctly marked, and sometimes even
form thickened places on the shell, by which its outlines may often be rendered a little irre-
gular. In the posterior third part, the shell assumes nearly always (see fig. 4. 6/10) a darker
dirty-grey color; and on examining this part with the microscope, we perceive that the appear-
ance is produced chiefly by peculiar ramified superficial marks, composed of numerous raised
lines, anastomosing with each other in various manners, and forming, especially towards the
anterior part, (fig. 9.), an extremely fine net. Nearer the apex (see fig. 7. & 8.) the lines
become less fine and less close, taking more the form of branched tubes. It appears however
that this is not, as I was at first inclined to presume, something originally belonging to the
shell, but an extraneous body; because we find shells on which this peculiar reticulation is
entirely wanting. I have in vain endeavored to discover the origin of this extraordinary
ramified marking, which, as I have subsequently ascertained, is not only not restricted to this
species, but has also been formerly mentioned by Jeffreys, as observed in very different mol-
luscs (Littorina). Probably it is one of those (as yet little known) lower organisms, concerning

33

the relationship of which either to the sponges or to the Khizopodes, we have not yet arrived
at any clear notion. The posterior end of the shell is extremely thin, and usually almost
perfectly straight; as in our two other species of Dentalium, the apex is obliquely truncated
(fig. 8) the posterior opening of the shell below forming a tolerably deep incision (see fig 7).
Inside this opening one often sees the supplementary tube, which terminates in the same
manner, but only projects a little beyond the extremity of the original shell.

Sometimes -- although apparently more rarely — the shell exhibits a more or less
distinct longitudinal striation (see fig. 5); but this striation is always only slightly marked, and
never forms such salient longitudinal ribs as in the D. abyssorum; moreover it is always
confined tc the posterior part of the shell. '

In the smallest of the specimens found by me, the embryonic shell remained (fig. 14).
It is nearly straight, smooth and tolerably lustrous, also finely cross-ringed, whereby the out-
lines, especially towards the posterior part, appear finely crenulated. The apex is (see fig.
15) somewhat enlarged, and evenly convex, without stripes or rings, and terminates with 2
small collateral points, between which the small circular aperture is situated.

The Annual is of the usual form for this genus. The foot (see fig. 10 &11) is, when
fully extended, about 1/i the length of the shell, slender, cylindrical and at the extremity as
usual three-lobed. The middle lobe, which represents the proper extremity of the foot, is
conically pointed, terminating in a fine point, and considerably longer than the side lobes.
The latter are thin, membranous, closely folded longitudinally, and nearly surround the basis
of the middle lobe like a frill. In front of the protrusible part of the mantle, which reaches
beyond the aperture of the shell, there are to be observed numerous so-called captacula,
curved and bent in all directions, with which the animal brings food to its mouth. The pos-
terior spoon-shaped appendage to the body (fig. 12 & 13) is considerably longer and nar-
rower than in the D. Abyssorum, somewhat inequilateral, and has on the ventral side a long
and narrow incision.

The animal is remarkably active in its movements, much more so than 1 have found
to be the case with our 2 other species of Dentalium; and it is also precisely this which
has given rise to the specific name. The manner in which it uses its foot, partly to bury
itself in the mud, partly for locomotion in one direction or in another is really so energetical,
that one would at first glance be inclined to believe the animal a Siphonodentalium rather
than a Dentalium. As is the case with the former, a considerable time may pass, before one
can see these movements; for the animal will frequently remain obstinately retracted in its
shell, during which time however the anterior part of the mantle together, with a greater or
smaller number of the filiform captacula, will usually be seen projecting beyond the aperture
of the shell. By these latter, which certainly do not only serve as instruments of capture,
but also act as real feeling organs, the animal appears, as it were, to survey the ground before
extending its foot. This operation takes place very rapidly, just as with the Siphonodentalium.
the foot being advanced suddenly beyond the aperture of the shell, with the side lobes lying

34

close against the middle lobe (see fig. 10); when the foot is fully extended the side lobes are
turned energetically outwards and backwards, at the same time that the whole foot twists
itself to one side or to the other, alter which it is again withdrawn into the shell with a
rapid jerk. This is repeated at greater or smaller intervals; and each time the shell is moved
a little forwards in one direction or in another. The protraction and the retraction take place
so suddenly and unespectedly, that I had at first the greatest trouble in getting a correct idea
of the form of the foot; and it was only by many repeated observations that I succeeded
completely at last. This is very different from what we observe in our other 2 Bentalia, in
which the movements of the foot are far slower, and the whole animal is much less sensitive.
When first the foot of these species is extended, we can quietly and leisurely observe it from
all sides ; nay I have often taken such a Dentalium up out of the water without the foot
being retracted.

I have found this peculiar deep-sea form not very rarely at the fishing-place Skraaven
in Lofoten, but always only at the greatest depths, 200 — 300 fathoms, while the D. abyssorum
is found up to 40 fathoms depth. Lately I have met with the same form again far more
to the south, namely in the neighborhood of Aalesund, and also here only at the greatest
depths which occur, namely in the so-called Breisunds-deep in 200 — 220 fathoms in muddy
bottom. Our species may be characterised in the following manner:

Dentalium agile. M. Sars.

Testa teres, valde angusta, parum arcuata interdum fere recta; posteriora versus in
tota longitudine sequaliter attenuata, albida sub-opaca parum nitida; posteriore parte ssepius
obscurior, apice tenuissimo oblique truncato rimato et tubulo ornato supplementario parum
prominente. Superficies testae plerumque laevis striis modo incrnmenti circularibus obducta,
rarior in posteriore parte longitudinaliter striata, striis vero parum distinctis et vix costellarum
instar prominentibus.

Animal quam solito agilius. Pes cylindricus apice ut vulgo trilobate, lobo mediano
conico-acuminato, lateralibus membranaceis et plicatis; et protractio et retractio ejusdem
celerrima. Appendix scaphscformis corporis postice sat elongata insequilateralis incisura ventrali
longa et angusta fissa. Longit. testse majorum 58 Mm. diametr. apertura 4 Mm. apicis 1 Mm.

Habitat non infrequens ad insulas Lofotenses in profunditate 200—300 orgyarum fundo
argillaceo nee non prope urbem Aalesund prof. 200 org. Ceterum fossilis in tellure tertiaria
Calabria3 inventa (D. incertum Philippi (non Deshayes).

Philippi's short description of his Dentalium unrrtntH, which however according to
Jeffreys is different from the form described by Deshayes under the same name, suits our
northern Form precisely; and Jeffreys has also, according to his obliging communication, by
comparing specimens of our form with the fossile form from Calabria, ascertained that the
two agree perfectly. We have thus here again an interesting instance of the animals of
earlier telluric periods having been able to continue their existence undisturbed in the great
deeps of the Ocean up to our own time.

35

Triopa incisa, M, Sars,

n. sp.
(PI. III. fig. 1-3).

Of this remarkable Xudibranch I have only observed some few specimens all taken
from great depths at Lofoten. It varies in some respects so considerably from the 2 other
known species of the genus, that it ought perhaps more properly to be considered as the
type of a particular genus. However I note it here preliminarily under the denomination
assigned to it by my Father; because, owing to the scarcity of specimens, I have not been
able to institute an investigation in minute detail.

The animal is only 7 Mm. in length, that is considerably less than our other two
species, but yet appears to be full grown. The shape is rather short and plump, with a high
convex dorsal side; seen from above (fig. 1.) oblong oval, in front obtusely rounded, both the
side edges only a little curved, or nearly parallel, and the posterior extremity suddenly much
narrower and conically pointed.

The mantle is particularly large and wide, and in the upper part strongly convex,
advancing far beyond the foot in front, as also it extends beyond the foot on each side to
about l/4 of its own breadth, in the form of a sharp and thin border (see fig. 3). At the
commencement of the posterior third part of the length of the body the mantle terminates,
shewing at this part a crescent-formed incision, or going out into 2 triangularly pointed lobes
that project on each side of the conically tapered posterior part of the body. From the
anterior extremity of the mantle there proceeds upwards in the middle (see fig. 1 & 2) a
tolerably high longitudinal keel, which however is quite short, and soon becomes forked into
2 somewhat diverging branches which surround an oblong triangular area in the middle of
the dorsal side. These diverging keels, which like the single keel are thin and vertical, are
somewhat irregularly lobed, and were in one specimen resolved into several separate pro-
cesses; somewhat behind the middle of the animal, each of them terminates in an obtusely
conical lobe or process directed backwards and somewhat outwards. On each side of these
diverging longitudinal keels, between them and the edge of the mantle, there is a longitudinal
row of 4— G short cylindrical obtusely terminating processes. 4 similar but larger processes
are situated further in front, namely 2 quite close together near to the anterior edge (one
on each side of the single medial keel) and 2 further from each other somewhat larger near
to the side edges of the mantle and by the exterior side of the dorsal tentacles. Finally
there is, at the posterior extremity of the area in the middle of the back above-mentioned,
a transverse row of 5 processes, of which the middle and the two extreme ones are very
strongly elongated, smaller at the base, and somewhat enlarged at the extremity, or elongated-
claviform, while the 2 others are very small and of the same shape as those in the lateral
rows. All these 5 processes are directed obliquely backwards and upwards, and the mantle

36

is from here rather abruptly inclined downwards towards the posterior extremity (see fig. 2).
Behind these follows, still along the middle of the conically tapered posterior part of the
body, a longitudinal row of 5 — 6 irregular processes, of which the 2 anterior are longest, and
which together form an irregularly incised longitudinal keel along that part. The foot (see
fig. 3.)' is very slender, linear, tapering to a point at the posterior extremity, and scarcely
more than half as wide as the mantle, but tolerably high, (see fig. 2). It terminates in front
abruptly truncated at a considerable distance from the anterior edge of the mantle. There
is no trace of any labial tentacles, nor of any evident velum; but the anterior part of the
foot is as it were separated from the other part by a slight ventral sinus, and curves itself
strongly upward towards the bucal aperture, which is concealed in the angle between this
part and the mantle. The amis (cloaque) (fig. 1. b) is situated on the dorsal side immedia-
tely behind the incavated part bordered by. the longitudinal keels, and is partially hidden by
the long processes which project from the posterior edge of the same. The gills which
closely surround the anus are particularly small and rudimentary; so that in the specimens
examined it was not possible to determine their number or form. The dorsal tentacles (a. a)
which are far distant from each other, and attached near the edge of the mantle, are short
conical and transversely lamellated, and can be partially retracted in short sheaths. On the
right side of the body in front, under the edge of the mantle, there was remarked a small
porus (fig. 2. c) which is the genital aperture. The whole animal, together with its various
processes, is of a uniform whitish color with slightly transparent yellowish-red intestines. The
mantle is everywhere strengthened with chalk spiculse, which are especially conspicuous at
the edge and in the dorsal longitudinal keels, and here usually exhibit a radiating arrangement.

I have only met with this little peculiar deep-water nudibranch at Lofoten, namely
at the fishing-place Guldbrands0erne in 120 fathoms, and at Skraaven in 200 fathoms; in
both places on soft clay bottom. It was very slow in its movements, which only consist in
a gentle scarcely perceptible gliding forwards, during which the body is scarcely curved, nor
notably altered in shape.

This species may be recognised by the following diagnosis:

Triopa incisa. M. Sars.

Corpus breve et obesum, supine valde arcuatum, supra visum subovatum, extremitate
antica late rotundata, postica subito multo angustiore et conico-acuminata. Pallium magnum
et convexum pede multo (duplo circiter) latius et eum antice longe superans marginibus ex-
pansis et laminaribus, postice incisum vel in lobos duos triangulares utrinque prominentes
desinens, parte postica 4ta corporis nuda et anguste conica. Crista adest in anteriore parte
pallii longitudinalis, sat alta sed brevis, in ramos vero duos leviter divergentes se dividens
spatium includentes medianum depressum forma elongato-triangulari postici serie transversa
processuum elongato-claviformium 5 marginatum quorum medianus et lateralis longissimus.
Series longitudinalis processuum similium sed breviorum utrinque inter cristas divergentes et

37

marginem pallii denique 4 anteriores, 2 ad marginem anticum sibi approximati et 2 alii ma-
jores magis a se remoti raargini exteriori propiores ad basin tentaculorum adsunt. Series
longitudinalis processuum irregularium pr&terea adest in dorso partis posticse corporis. Pes
valde angustus sublinearis postice acuminatus. Tentacula dorsalia longe a se remota brevia
et crassa laraellosa ex parte retractilia. Velum obsoletuni. Tentacula labialia nulla, Anus
dorsalis pone medium situs, branchiis minimis et rudimentariis circumdatus. Color albidus;
lougit. 7. Mm.

Habitat rara ad insulas Lofotenses in prof. 120—200 orgyarum, fundo argillaceo.
This species is immediately distinguished from the 2 other species T. clavigera and T. lacer,
by the peculiar and strong development of the mantle, as also by the few (4) processes on
the anterior extremity, and the peculiar forked longitudinal carena, along the back.

Gonieolis typica, M, Sars,

(PL IV. fig. 1-11).

Gonieolis typica M. Sars. nBeretning om en i Sommeren 1859 foretagen zoologisk
Reise ved Kysten af Romsdals Amt," pag. 4.

Also this very peculiar nudibranch, which already in 1860 was shortly noticed by my
Father, seems to be a genuine deep-sea form which, as is well known, is something very
unusual in this group of molluscs which comprises scarcely any other than littoral or sub-
littoral forms.

The following description is for the most part after the manuscript of my Father.

The body resembles (see fig. 1. & 2.) a snail; it is oblong, about 12 Mm. long; in
the anterior part 5 Mm. wide; towards the posterior part gradually narrower and quadrangular,
as the back, which forms a distinct mantle with projecting tolerably thick side-borders, is
nearly quite flat or only slightly convex, and the sides of the body are high (nearly as high
as the body is wide) and somewhat hollowed. The mantle, which every where is narrower
than the foot, is (see fig. 1) widest in the anterior part, where it takes its beginning im-
mediately behind the dorsal tentacles, and becomes gradually narrower towards the pos-
terior part which terminates in a point. The foot is tolerably thick, furrowed in front and
with the anterior corners rounded; it is wider than usual in the Eolis, only a little narrower
at the posterior end, and obtusely rounded, and both longer and wider than the mantle. The
head is wide (as wide as the mantle) nearly quadrilateral, and bears on the posterior part,
immediately in front of the anterior end of the mantle, the dorsal tentacles (a. a) in front
of which it gradually (see fig. 2) inclines downward, and has here a wide straightly truncated
velum (b) (which is wanting in the Eolis) from .the upper side-corners of which the labial
tentacles (c. c) arise.

38

The dorsal tentacles (fig. 1. & 2. a. a) are nearly one third of the length of the body,
and unusually thick (much thicker than in any Eolis known to me) at the base, and gradually
tapered towards the obtusely rounded extremity, smooth, only a little wrinkled transversely
by contraction, and without sheath. The labial tentacles (ibid. c. c) are 2 — 3 times shorter,
and 3 — 4 times thinner than the former, and pointed. As in the Eolis all the tentacles are
not retractile. The eye-spots were not perceived. Both the observed specimens had the
head elongated in front in a long projecting conical proboscis (d) consisting of 2 sharply
defined cylindrical divisions (the anterior thinner than the posterior), which it does not seem
that the animal can draw in; for in the living animal it was always held extended during the
creeping, which is a very slow and regular gliding progress. The falsely so-called gills (want-
ing all the conditions for breathing) are short, elongated conical, pointed smooth papillae
(fig. 3) of unequal size, which are situated somewhat irregularly along the projecting side-
borders of the mantle, and also within the same in greater or less number scattered over
the dorsal surface. A great number of these papillae had however fallen off in the specimens
examined. The genital aperture is situated on the right side, about in the anterior third
part of the length of the body, in the shape of a prominent roundish tubercle (e) from which
a tolerably large conical somewhat twisted and pointed penis protrudes. Some distance be-
hind the genital aperture, and in a line with it, or about in the middle of the length of the
body and on the same side, the anus is situated at the top of a circular smaller tubercle (f).
The 2 vertical lateral jaws, which extend in a longitudinal direction, are well deve-
loped, and occupy the fleshy sides of the bucal mass. They are (fig. 5. & 6) thin chitinous
plates of faint light horny-yellow (particularly in the anterior thicker part) transparent color
and irregular crescent-like shape (or rather remarkably like the shell of a Lima) externally
(fig. 5) somewhat convex particularly in the anterior part, and less so behind; internally (fig.
6) concave, and behind double as high as in front. Their upper and posterior very thin edge
is curved; the lower and in front somewhat thicker edge is nearly straight, and the foremost
edge truncated and strongly thickened, especially in the middle, where it terminates in a
% broad, projecting forward three-cornered process, pointed at the end, of horny or yellow brown
color, and resembling the ear-like process (auricula) of the cardinal edge of a Lima shell.
Just under this three cornered process, and consequently at the anterior ventral extremity, is
situated the apex of the jaw which is the most convex part of the jaw. To the apex is
attached the yellow-brown ligament (,,fulchrum" Alder & Hancock) which is a peg-shaped pro-
cess projecting downwards, and connecting both jaws. This peg-shaped process (ligament)
has at its basis a roundish enlargement, and when cut through, shews short transversal close
threads, that is, a ligamentary structure. When greatly magnified the jaw exhibits innumer-
able fine concentric growth-lines running parallel with the curved edge, and crossed by still
finer and closer lines running radially from the apex to the edge, forming a sort of threaded
structure. The lowest straight edge of the jaw is without doubt the cutting edge, although
it has here no teeth; for there is here no trace of ,,the strongly curved process which ex-

w

<&.

39

tends downward and forms the cutting edge" in the Eolis. The posterior and lower half of
the cavity enclosed by the jaw (in the Eolis according to Alder & Hancock, tab. 7. fig. 6
more than two thirds) is filled up by the tonyuc (Fig. 4. a. fig. 7. & 8), which is a fleshy
mass of nearly triangular or wedge-like form, being wide behind, and in front gradually nar-
rower or compressed from the sides; so that it is here nearly double as high as wide, with
obtusely rounded anterior extremity. Along the middle of the tongue there extends the
narrow ribbon-like ravp-plate, (radula) consisting of about 20 transverse rows of teeth, which
lie hid in the so-called sheath of the tongue. It takes its beginning above, near to the an-
terior end of the tongue, where the largest teeth are found, and extends thence downward
along the same and on the ventral side \ its teeth becoming gradually smaller, and backward
to more than half the length of the tongue, where it goes over into the sheath of the tongue.
There are (see fig. 9.) in each transversal row 3 teeth, of which the middle one is thick strong
and broad not curved, but only a little hollowed at the base and conical, bent backward,
with obtuse rounded extremity (see fig. 10. & 11) and of vitreous transparent extremely faint
light yellowish color; at a short distance from, and in a straight line with this, stand (see
fig. 9) 2 others, one on each side, nearly equally long but much thinner, straight triangular,
pointed glassy and colorless. As well the middle, as the side teeth are smooth; only at the
base of the former there may be observed on each side (fig. 11) a very small roundish node.
In front of the tongue there is the secondary bucal cavity (fig. 4. b.) called ,,Bimundhule" by
R. Bergh, which according to him is peculiar to the Eolides. The rasp-plate in the Gonieolis,
corresponds by its narrowness- with the usual form in the Eolidinse, which family is by Alder
& Hancock characterised by ,,a narrow tongue, usually only with a central plate". However
it does not quite agree with any of the delineations of these authors. By the presence of
side teeth, and by their form, it resembles the Eolis pellucida (Alder & Hancock Tab. 47.
fig. 12) the central tooth of which yet differs by being denticulated on the edge; with respect
to the smooth central tooth it resembles the Eolis alba (ibid. Tab. 47. fig. 14) which species
has however no side teeth.

The color of the animal is every where yellowish white, with yellow brown liver shi-
ning through the skin along both sides of the back, and a more deeply lying oblong light
minium red intestine .(probably the ovary) which is especially visible from the ventral side.

Of this remarkable nudibranch 2 specimens were taken in the summer of 1859 at
Christiansund in 50—70 fathoms depth. A third specimen was found in the summer of 1869,
in the Hardangerfjord at Mosterhavn at the depth of 100 fathoms.

The present genus may be characterised in the following manner:

Gen. Gonieolis. M. Sars.

Corpus oblongum, quadrilaterum, subpalliatum, pallio pede angustiore absque spiculis.
Caput latiusculum, tentaculis 4 simplicibus, conico-acuminatis, non retractilibus, quorum 2

The figure 7 is represented in an inverted position. /

40

maxima dorsalia et 2 multo minora labialia ad margines laterales veil lati antice truncati
posita. Maxillae corneae in proboscide conica exserta valvulas Limae referentes, antice supeme
auricula triangulari ornatae, processu infero uullo. Lingua subcuneiformis radula seriebus trans-
versis dentium circiter 20 Isevium dente medio conico retroflexo utrinque lateral! recto acu-
minato. Papillae dorsales simplices. conicae, breves, utrinque ad latera dorsi et ad marginem
pallialem disperse. Pes ovato-elongatus, latiusculus, angulis anterioribus rotundatis, pallio
major. Orificia generationis et ani ad latus dextrum.

Goiiieolis-typica. M. Sars.

Corpus triple circiter longius quam latius, altitudine maxima in parte antica sita lati-
tudinem aequante, capite antice valde declivi. Tentacula dorsalia latitudinem corporis aequantia
postice cornuum instar curvata minime lamellata; labialia tenuissima fere filiformia lateribus
corporis appressa. Papillae dorsales minimae, inaequales, sparsa. Pes amplus postice conico-
rotundatus. Color albido flavescens subpellucidus. Longit. 12 Mm. latit. max 5 Mm. Ha-
bitat rarissima ad Christiansund prof. 50—70 orgyarum, nee non in sinu Hardangerfjord prof.
100 orgyarum fundo argillaceo.

The Genus Gonieolis belongs to the family Eolididae Alder & Hancock, but is not
properly referable to any of the sub-families formed under this head. It cannot be classed
with the Dendronotinae the tentacles of which are lamellated and retractile into sheaths; nor
yet to the Melibaeinae, which have tentacle-sheaths and claviform dorsal papillae; neither can
it be referred to the Hermaeinae and Proctonotinae, which have a dorsal anus. The sub-family
with which it best agrees is that of the Eolidinae, where however there is not to be found
any trace of the mantle, so distinctly marked in the Gonieolis, although not projecting much
beyond the sides of the body, and without spiculae, which are found in the mantle of the
family Dorididae. Among the Eolididae it has most resemblance with the proper genus Eolis
Cuv. but differs by the presence of a mantle-border, which gives the body a square form,
(while the Eolis has a round back), by its unusually large dorsal tentacles, as well as by the
labial tentacles springing from the sides of a broad velum, points that are not to be observed
in any other species of Eolis with which I am acquainted. It ought therefore certainly to be
placed nearest to the Eolis, and the characters for the sub-family Eolidinae should be some-
what modified in order to admit it.

.Annelida.

8, Umbellisyllis fasciata, M, Sars,

nov. gen. et sp.
(Plate 4. fig. 12 — 18).

I have seen some specimens of this beautiful Annelide from different parts of our
coast, and from very different depths. Usually however it appears to have its proper dwel-
ling place in the greater depths about 100 fathoms and downwards; and this is probably the
reason why it has formerly quite escaped the attention of Naturalists notwithstanding its very
remarkable and unusual coloring, which preserves itself tolerably well, even in specimens that
have been kept a long time in spirit.

In preparing the following description, and particularly the concluding remarks, I have
been greatly aided by my Father's manuscript notes.

The body attains a length of about 24 Mm., and a breadth over the middle (from foot-
point to foot-point) of 3 Mm.; towards the anterior end it is somewhat tapered (see fig. 12)
and towards the posterior end considerably so. Thus the body is not very long in proportion
to its width. The back is convex (see fig. 16) with rather abruptly declining sides; but
the ventral side is flat or rather concave along the middle, shewing a wide and shallow longi-
tudinal ventral furrow. The number of the bristle-bearing segments varies according to the
size: in the largest specimens it is as great as 65 while in smaller it is not more than 46.
The segments are (see fig. 15) very short -- 5 — 6 times broader than long; the last (the
anal segment) is (see fig. 17) very small, button-like and furnished with 2 conically pointed
anal filaments, which are about as long as the 2 nearest segments taken together.

The head (see fig. 13) which is somewhat sunk in the 1s.1 bristle-bearing segment, and
slightly smaller than the same, is nearly square, with convex side-borders, broader than it is
long, its upper surface convex; the anterior border strongly hollowed forming 2 very short
triangular frontal lobes, standing out far from each other, and somewhat curved downwards,
separated by a transverse furrow from the rest of the head; the posterior edge, which is
straight, is hidden by a membrane (a), which will be mentioned subsequently, and which fre-
quently covers also the posterior part of the head.

The head bears 3 tenlac'lc* of about equal size, which are cylindrical, tolerably thick,
very short (Y4— Vs °f the width of the head) and not articulated, with obtusely rounded ex-
tremity. The 2 paired tentacles are situated on, or properly speaking quite close behind, the

42

transversal furrow which bounds the frontal lobes, at a considerable distance from each other;
the third or single tentacle stands far behind them in the middle of the crown, not far from
the posterior border of the head, so that its basis is not unfrequently covered by the above
mentioned membrane. There are 2 pairs of eyes of round shape and bright brown red color,
which are situated somewhat nearer to the side-borders of the head than to its medial line:
the anterior eyes some distance behind the paired tentacles, the posterior eyes in a line with
the single tentacle ; the latter are a little nearer to each other than the former, and are also
nearly twice as large as the anterior, close in front of which may further be observed a pair
of extremely small black points, that may perhaps be considered as a third pair of eyes as
in several other Syllides.

At the posterior extremity of the head, or just in the transversal furrow which sepa-
rates it from the first segment, is situated a large soft thin leaf-like semilunar membrane (fig.
13. 14. a) which is placed horizontally or across the axis of the body, and with the central
part of its concave border attached to the body, but otherwise freely projecting and more or
less erect; its convex or free border lying usually beyond, and covering the posterior part of
the head; both its rounded side-corners, or ends of the crescent, being likewise free. It is
about twice as wide as high, about as wide as the head; its free border (and, as it appeared
to me, also its surface) is thickly covered with short and fine, in a living state vibratory, cilia.

The bucal segment, .which projects only a little beyond the head, shews on its anterior
extremity (see fig. 14) the rather large round mouth, on the thick border of which there are
observed 5 small raised round nodules or papillae; one ventral and 2 lateral on each side;
close behind the upper pair of these papillae, there is situated on each side of the segment
a tentacle, which in form and size completely resembles the tentacles of the head.

The feet (the pedal nodes) which enclose one single thin needle pointed at the end,
and a fascicle of very long and thin bristles, are situated far down on the body, so as nearly
to go Hush with the flat ventral side (see fig. 14. & 16). They are nearly cylindrical, yet
somewhat compressed from the front backwards, proportionally rather long (about V4 of the
width of the body), and have at the extremity, on the side of the rounded bristle-bearing node,
a conical process projecting beyond the same; and below, close to 'the extremity of the foot,
a similar one but somewhat longer, which properly seems to represent the ventral filament.
The dorsal filament is not attached to the foot, but some distance above it on the side of
the body. It is cylindrical, filiform, not articulated, longer than the feet, and even reaching
to the end of their bristles, and its extremity is obtusely rounded. At the posterior end of
the body (see fig. 17) the pedal nodes, as well as the dorsal filaments, diminish very rapidly
and suddenly in length, so that in the extreme segments they are quite rudimentary. The
pedal bristles (fig. 18) in each foot about 20 in number, project very far, and are composed,
or consist of a very long straight shaft, on the obliquely truncated extremity of which is fixed
a rather short linear, or knife-like terminal appendage. They belong therefore to the so-
called thorny bristles (setae spinosae).

43

The color of the animal preserved in spirit is white, with greyish olive-green (some-
times darker, nearly black) belts or bands across the back, usually (see fig. 15) one on every
other segment, sometimes even (see fig. 12) on 2 consecutive segments, very seldom on 3.
These colored belts are however interrupted in the medial line of the back, and do not more-
over reach down to the feet; so that each of them properly consists of 2 transversal, elon-
gated quadrangular or rectangular spots situated in one line.

I found first a few small specimens of this peculiar Annelide in the summer of 1865
at the entrance to Flekkefjord, at a depth of 80 — 100 fathoms. Afterwards I found it again
on several other points of our coast, although always rarely; namely at Lofoten in 150 — 300
fathoms; in the Hardangerfjord, where it goes up even to 40 fathoms; and finally not long
ago at Christiansund and Aalesund in 100 — 150 fathoms.

The new Genus and species may be characterised in the following manner:

Gen. Umbellisyllis. M. Sars.

Corpus vermiforme, hand longum, segmentis brevibus, cirris 2 analibus. Lobi capitis
frontales minuti, late distincti; oculorum paria 2, anteriora minora, et ante hos puncti ocu-
lares 2 minimi; tentacula 3 sequalia subcylindrica brevissima, non moniliformia: duo anteriora
ad sulcum lobos frontales a capite separantem posita, tertium postice in medio vertice. Ten-
tacula 2 oralia iisdem capitis simillima, in segmento buccali ad latera oris posita. Cirri
tentaculares nulli. Pedes uniremes, setis compositis spinosis minuti; cirro dorsali elongate
non moniliformi, a pede remoto, lateri corporis affixo ; cirro ventrali brevissimo ad apicem
pedis sito. Margo occipitalis lamina munitus cutacea sub-semilunari, transversa, ciliata, mar-
gine inferiore medio adnata ceteroquin libera et supra basin lobi cephalici prominente.

Umbellisyllis fasciata. M. Sars, unica species.

Corpus subcrassum, antice parum, postice magis angustatum, semicylindricum, dorso
fornicato, ventre applanato, segmentis setigeris circiter 65 brevibus (5 — 6'?s latioribus quam
longioribus). Caput subtetragonum, latius quam longius, lobis frontalibus triangularibus ; ten-
taculis quartam ad quintain capitis latitudinis partem longitudine sequantibus; occulis fusco-
rubris, anterioribus paulo distantioribus quam posterioribus ; lobo occipitali latitudinem capitis
sequante. Pedes subcylindrici, quartam fere latitudinis corporis partem longitudine fequantes,
cirro dorsali pede multo longiore. Cirri anales conico-accuminati, breves (segmentis 2 prox-
imis junctis longitudine tequales). Color albidus fasciis dorsalibus transversis, medio inter-
ruptis, cinereo-olivaceis in segmentis alternis aut in 2 (rarissime 3) subsequentibus. Longit.
corporis 24 Mm. latit. maxim. 3 Mm.

Habitat at oras Norvegise meridionales, occidentales et boreales in profunditate 40
— 300 orgyarum.

Although the present Annelide in respect of its proportionally short body consisting

of few segments, and the colored belts with which it is adorned, has a great resemblance to

6*

44

the Hesionides, it seems nevertheless to be properly referable to the family ef the Syllides,
among the Genera of which it appears to stand nearest to the Exogone 0rsted, especially
by reason of its wanting proper tentacular threads (cirri tentaculares) and by its simple not
beaded dorsal filaments. It presents however sufficient variations both from this and from
the other hitljerto known forms of this family, for the establishment of a new Genus which
has received the name Umbellisyllis on account of the remarkable large ciliated membrane
or screen situated at the posterior extremity of the head.

The Umbellisyllis differs from the Exogone by the following peculiarities :

1) The frontal lobes are not (as in the Exogone and Sylline Grube, Spermosyllis Clap,
and Sphserosyllis Clap.) confused into one lobe, but, although united with each other at the base,
otherwise far apart from each other, as in the Syllis Sav. and many other genera of the Syl-
lides. They are in our animal indeed very small ; nearly rudimentary, but yet distinctly to be
recognised by their being separated by a transversal furrow from the head itself.

2) The 3 tentacles of the head are not ,,placed in a transverse line in the furrow
which divides the frontal lobes from the head", which is according to Claparede the most
important Character of the .genus Exogone. In the Umbellisyllis there are only 2 paired ten-
tacles in or properly speaking close behind that furrow, but the third, (unpaired), is far be-
hind on the head or on the crown, in the middle between the posterior eyes, as in certain
species of Syllis (for inst. S. gracilis Grube) and in the Genus Spherosyllis Clap.

3) Of the 4 eyes, the 2 posterior are larger than the anterior, while the contrary is
the case in the Exogone, and there are moreover immediately in front of the anterior eyes
a pair of extremely small black ocular points, as in the Syllides, Microsyllis, Sphaerosyllis and
Odontosyllis.

4) Underneath the head, or on the bucal segment, there are in the Umbellisyllis two
more tentacles, one on each side of the entrance to the mouth, which in form and size are
precisely similar to the tentacles of the head, and are hitherto only found in one single Syl-
lide, namely Autolytus roseus Clap. (Glanures zootomiques parmi les Annelides pg. 106)
although their position is here not more precisely indicated by Claparede, who nevertheless
in the explanation of the delineations calls them ,,bucal tentacles."

5) A specially characteristic mark of the Umbellisyllis is the peculiar, nearly semi-
lunar membrane or screen, situated at the posterior extremity of the head, to which screen
there is only found something similar in one single genus of the Syllides, namely, in the
analogous Sphserosyllis Clap., the 2 ciliated aliform appendages that appear in the same place.

0) The dorsal filaments are not as in the Exogone small or ,,rudimentary" (Claparede),
but even a good deal longer than the tolerably long feet, and they are not situated on the
feet, but some distance above them on the sides of the body. The ventral filament, which
is quite small (in the Exogone Kefersteinii it is even wanting according to Claparede) is on
the contrary situated close to the end of the foot. The pedal setse differ from those of the
Exogone naidina, in which they arc according to 0rsted sickle-shaped (setse falcate) or

45

curved in their exterior part, and with a similarly curved very short terminal piece; their
(shape is not indicated by Claparede in the E. Kefersteinii) by their shaft being straight, and
the terminal piece also straight, proportionally longer than in the Exogone, linear or knife-
shaped, or in other words, they are not sickle-shaped, but thorny (setae spinosse).

9. Paramphinome pulchella, M, Sars,

nov. gen. et sp.
(Plate IV. fig. 19—35).

Of the extensive group of Amphinomes, which in the tropical seas display so great
riches of genera and species, only very few northern forms have hitherto been noticed, and
these only such as belong to the family of the Euphrosynidse, namely 1 species of the genus
Spinter Johnston and 3 species of the genus Euphrosyne Sav. The typical family Amphino-
midse, was on th'e contrary considered to be exclusively confined to the tropical oceans, until
my Father 1 in a few words announced the discovery of a species of the genus Eurythoe
Kinberg (E. borealis M. Sars) a single specimen of which had been observed by him a long
time ago at Manger in the neighborhood of Bergen. Unfortunately it was not possible to
institute a scrutinising examination of this interesting species; as it has not since been found
again, - and the original specimen was not preserved. Malmgren therefore 2 expresses his
opinion of this species, that it probably does not belong to our fauna, but that the specimen
observed must be- considered as having been accidentally carried by the Gulfstream, from,
warmer seas to our west coast. The discovery of a new form belonging to the same family
is now therefore of great interest; since this form can with full certainty be claimed for our
Fauna, and even seems widely distributed along our coasts. It forms the type for a new
genus, the affinity of which with the typical genus Amphinome is expressed in the name.

The body, which in most specimens is only 9 Mm. long, is (see fig. 19) tolerably narrow
nearly cylindrical, or only a little wider than high, with the ventral and dorsal sides evenly
convex; tapering a little towards both ends although not considerably. The bristle-bearing
segments, the number of which is somewhat variable, and in the largest specimens goes up
to 33, are about twice as wide as long, and rather sharply divided from each other.

The head-lobe is (see fig. 20 & 22) of roundish oval form, and -narrower than the l^1
segment, it is strongly convex in the upper part, especially behind, and it inclines evenly down-
wards in front (see fig. 23). The frontal border is rounded, but in living specimens (see fig.
22) somewhat drawn out in the middle, or conically rounded. The posterior part ofthehead-

1 Christiania Vid. Selskabs Forhandlinger for 1861.
3 Annulata polychteta, pg. 2.

46

lobe is still more strongly drawn out, nearly tongue-shaped, and projects over the first seg-
ment nearly to its posterior edge (see fig. 20 & 22). There is however no trace of any real
caruncula. Eyes are also entirely wanting.

The head is (see fig. 22) furnished with 5 rather short cylindrical, or filiform appen-
dages, of a very similar form and nature; 2 pairs and 1 single. The last appendage repre-
sents the tentacle, and is situated on the middle of the posterior tongue-shaped strongly
convex part of the head-lobe. Its length is somewhat less that the width of the head. The
anterior paired appendages represent the antennae, and proceed, rather near together, from
the anterior border of the head-lobe; they correspond exactly, as well in form as in size
with the tentacle. The 2".d pair, which represents the palpse, project on each side from the
part where the head-lobe has its greatest width, or about the middle of its length, and are
a little longer than the antennae and the tentacle.

The 1s.1 bristle-bearing segment has a dorsal and a ventral filament of the same form
as the appendages of the head; the ventral, also of the same length; the dorsal a little longer.
On the following segments both these filaments become rudimentary (see fig. 31. 32, 33) only
representing a short conical process; on the 3 last segments they become however (see fig.
21) again somewhat longer, although not nearly so long as on the first segment.

The mouth, which forms a longitudinal slit, has (see fig. 24) its place under the head
and 1s.' segment, and is bounded by 4 fleshy lobes or lips, of which the posterior ones espe-
cially are atrongly prominent (see fig. 23).

In spirit specimens one often sees (see fig. 19 & 20) the short fleshy gullet or pro-
boscis evaginated; its borders are irregularly lobed, but without any armament of teeth.

The anus is (see fig. 21) terminal, and covered below by. a half-oval lobe.

The gills are, unlike those of the other Amphinomes, only confined, to the anterior
part of the body, while they are wanting on all the other part. Usually there are only 5
pairs (see fig. 19) which take their beginning on the 4'!1 bristle-bearing segment, and are
always arranged in pairs on this and the 4 following, one on each side of the back (see fig.
22 & 30). They are tolerably large, exceeding in length half the width of the body, and when
extended backwards, partly lap over each other, nearly covering the dorsal side of those seg-
ments on which they are situated. In form they are (see fig. 25 — 29) wide-fan-like, placed
transversely, 2 — 4 times dichotomically divided, with the last branches very short and simply
rounded at the end. Usually they increase in size from in front backward, becoming gra-
dually more strongly divided; the difference between the 3 last being however smaller than
between these and the anterior pairs. In the -more closely examined specimen, each of
the 2 primary halves of the 1s! pair of gills (fig. 25) consisted of 3 terminal branches; only
the interior branches produced by the secondary division having divided themselves fork-wise,
while the other branch was still undivided. In the next pair of gills (fig. 26) both the se-
condary branches were forked on the one side (the gill was thus here 3 times dichotomically
divided) ; and on the other side there was an indication of a 4"; dichotomy, the interior secon-

47

dary branch being tripartite. In the 3r.d pair of gills (fig. 27) the division was considerably
more advanced. On the one side, the gill was 4 times completely dichotomically divided
with 8 terminal branches; on the other side the exterior secondary branch was only bipartite;
while the interior had 4 terminal branches, but was nevertheless not regularly dichotomic;
only the exterior tertiary branch having divided itself, and that so as to leave the interior
terminal branch simple as usual, while the exterior was forked, (the 5'.h dichotomy). A similar
irregularity in the dichotomical division of the gill appeared also in the following pairs of
gills (fig. 28 & 29); and it was observed here in both the primary halves of the gill. Hence
it seems that the gills in the present species may indeed, on the whole, be said to be divided
in dichotomic form; although the dichotomy is not everywhere strictly carried out.

As above stated, 5 pairs of gills appears to be the normal number. Specimens are
however not seldom found, especially younger ones, with only 4 pairs of gills (see fig. 22 &
23); and in some very large specimens, I have sometimes found 6 pairs. But in every case
the gills take their beginning on the 41!1 bristle-bearing segment; so that the first 3 segments,
as also those, coming after the 91!1, are always without gills.

The pedal nodes (see tig. 31) are double, or 2 on each side of the segments, very
small and tolerably far from each other; the lower ones, about in the middle of the height
of the segment; the superior, near to the dorsal side. The rudimentary cirri are situated at
the end of the pedal bristles. The latter are very fine, and about half as long as the seg-
ment is wide. On the upper pedal nodes they, diverge to all sides; on the inferior, they
form a more regular fascicle. The origin of the gills is (see fig. 20 & 30) situated close
behind the superior pedal nodes, and inside of the rudimentary dorsal cirrus.

The pedal bristles (fig. 34) are on the superior nodes simple, but of 2 different- sorts.
Some of them (b) are considerably thicker, and also shorter than the others, nearly thorn-like,
the point becoming suddenly of a very tender structure, and completely hyaline ; when very strongly
magnified, these bristles shew on one edge a slight indication of a peculiar armament of
hooked backward-turned teeth, similar to that described byKinberg in various Amphiuomides
from the tropical 'oceans. The other bristles on these pedal nodes (a) are more than double
as long, and considerably thinner, terminating in an exceedingly fine point, near to which the
bristle is on one side finely serrated. In the inferior pedal nodes, the bristles appear at
first glance similar to those last mentioned. On closer examination however it is remarked,
that most of them (c) exhibit at a short distance from the extremity a tolerably long, forward-
turned, close-lying side tooth, whereby they properly become forked; the end of the bristle,
from this side tooth, is very finely and closely serrated on one edge. Among the lower bristles
attached to this pedal node, some few (d) are remarked which are considerably stronger,
somewhat enlarged at the end, and here furnished with a short side tooth; the end of the
bristle is here much thicker and shorter than on the former, a little curved outwards, and
coarsely serrated on one edge.

On the I5.1 segment there are, besides the bristles described, on the upper pedal node

i: - -: '

T«~jw:l rm th*

with Ut*ral

tksJari? Kreir, a*4 whe» » the least

*rergM* hricdef •• fbe

t**tm, whkh the

«ffl it pereeire* that at Is

\ndj. with a

ft jppejn to he
At f »fotflL where I irst fiswi it

MXW and On m the great deptfc? to 300
I kare «aee «et wtti it again n the Chriftiama fjord at YaD0 at depths from 100—939
fift<Mi aad latel? at Aafemd. IB the last aaned place it was net rare, dose to the tow*
-30 feflMNM M soft mnddr bottom.
The new Oemn and specks may he characterised in the foflowing manner:

Gen.

3L Sars.

Corpos rennifonne. modice elongatmL segmentis panels. Lobus cephalkus parms
prodnetitt. tarantula vero nulla distmcta. Ocob nulli. Tentacnla capitis 5 brevia.
cyKndrka forma et magnitndine snb-aequalia. unum mediannm in parte postica. -2 anteriora
et 2 lateralia. Cirrus dorsalis et rentralis in lr segmento distinctL elongatt forma tentaeulis
eepbalicin similefj. in e«teris rndhnentariL Os fissnram longitndinalem labiis 4 camosis rir-
formans: proboscis breris et crass a apice irregolariter lobato nnllis armato maxillis

;ios terminalis. Pedes biremes. remis mrahnis et longe sejunctis. snperiore
*nb< rjiftgnris capillaribus. aliis multo brevioribus et robustioribus interp-

lateral!. i^rurnque obsoleto bidentatLs. dente altero brevissimo. altero (apice)

-ubtiliter dentato. nonnullis multo brevioribus et robustioribu?

49

prope apicem dilatatis et ut illis insequaliter bidentatis. Hamuli prscterea adsunt 2 validissimi
chitinosi in remo dorsali segment! 1™ setigeri. Branchiae sat magnse dichotomes, totum fere
dorsum tegentes, in segmentis vero paucis corporis per paria obviae in segmentis et anteri-

oribus et posterioribus omnino deficientes.

•

Sp. Parainphinome Pulchella. M. Sars.

Corpus fere cylindricum, parum latins quam altius, et antice et postice leviter attenu-
atum, segmentis setigeris 24—33. Lobus cephalicus segmento Ir angustior subovatus postice
rotundato-productus. Branchiae in segmento setigero 4'? incipientes utrinque 4—6, ter-quater
dichotomo-divisae ramulis terminalibus cylindricis. Color albidus.. Longit. maximorum 9 Mm.

Habitat ad oras Norvegiae pluribus locis in prof. 20 — 300 orgyarum fundo limoso.

The Genus Parainphinome is distinguished from all the other known Amphinomes by
the gills, which are only confined to some few segments of the body, while in the other forms
they usually appear on all. They are also proportionally much more strongly developed than
in the other forms, nearly covering the whole dorsal side of the segments on which they are
situated. The want of eyes, and of a distinctly marked caruncula distinguishes it also from
the other known Amphinomes. With respect to the construction of the pedal bristles, the
present form shews most resemblance to the genus Eurythoe Kinberg: although on the
whole they are much finer than in this, as well as in the other Amphinomes. The short
spiked bristles with backward-turned teeth or hooks, which are characteristic of the Amphi-
nomides, also appear here, although much less developed than usual. On the other baud,
the present genus distinguishes itself remarkably from all other Amphinomes, by the 2 peculiar
and particularly strongly developed chitine-hooks, which are found attached to the dorsal pedal
node of the 1s.1 segment,

7

Anthozoa. .

10, Mopsea borealis, M, Sars,

n. sp.
(PI. V. fig. 1-23).

The group of coral animals which Ehrenberg1 has established as a peculiar family
under the name of Isidea, and which Milne Edwards2 has taken in a somewhat narrower limi-
tation, excluding the genus Corallium as a peculiar group, and considering the 3 other genera
(Isis, Mopsea and Melithsea) as forming a sub-family (Isidinae) of his family Gorgonidse, com-
prises only tropical forms, with the exception of one single species, namely Mopsea elongata
(Isis) Esper, which is found in the Mediterranean. It was therefore highly surprising for
me to find a form belonging to this group living near our coast, and even within the Arctic
circle, namely at the fishing place Skraaven at Lofoten, where in the year 1866 I drew up
in the dredge from the depth of 300 fathoms, a very young bnt quite entire specimen. The
specimen, which is represented in fig. 6, was completely undivided, or only forming a single
stem strongly sinuous at the base, on the exterior half of which the polyps were attached,
while the lower part was sterile and widened at the end to a funnel-shaped 4-bladed root.
My Father as well as myself thought at the time that we had before us a constantly simple
or unbrahched form, which although distinctly stamped as an Isidin by its jointed axis and
peculiar root, yet shewed a remarkable affinity to the Pennatulidae, and therefore seemed
entitled to be considered as a type of a peculiar genus, which my Father named Isidella.
But the next year I found some exemplars which shewed that the present form, like the other
Isidinse, can be branched; although the branches here appear to be unusually few in number
and never again divided. As our coral agrees otherwise in all essential points with the genus
Mopsea established by Lamouroux, there can be no objection to refer it to that genus, which
is thus represented in the tropical seas, in the Mediterranean and in the Arctic seas. It was
also named by my Father, under the above appellation (Mopsea borealis) in his list of deep-
sea animals inserted in the Christiania Videnskabs-Selskabs Forhandlinger for 1868.

The largest tolerably entire specimen of mine (fig. 4) is 85 Mm. high. At the distance
of about 20 Mm. from the root, there proceeds at an acute angle a branch, which was however

1 Corallenthieir drs vothen Mcores. pg. 130.

- Histoirc naturelle des Coralliaire's, Vol. 1. pg. 192.

51

broken at the end; and at the distance of about 60 Mm. from the root, an other branch 25
Mm. long, nearly at a right angle. Both branches are about of the same thickness as the
main stem. Another specimen of the same size, or even a little higher (fig. 3) was, like the
first observed, entirely undivided with a tolerably straight stem. A third specimen (fig. 1)
was considerably larger, namely 105 Mm. high, but yet incomplete; as the whole lower part
was wanting. It had at a distance of 45 Mm. from the point, one single branch 35 Mm. long
issuing nearly at a right angle. A fourth imperfect specimen (fig. 5) probably the exterior
part of an equally large exemplar, had 2 nearly equally large branches issuing close together
from the stem, but in different directions, one at a right angle, and the other at an acute
angle, and besides these one very small branch near the end. This is the most extensive
branching which I have observed.

As will be seen from the above, the coral under consideration seems to be very
slightly branched; the stem being either completely undivided, or only here and there exten-
ding a short lateral branch, which appears never to divide itself again. Hereby the form is
very decidedly distinguished from the other Isidse, which are always extensively, most fre-
quently dichotomically, branched.

As in the other forms belonging to this sub-family, the stem, as well as the branches,
consists of a solid axis formed of alternate calcareous and horn-joints, and a soft and fleshy
envelope (csenenchym), covering this axis, wherein the polyps reside. The lower half or third
part of the stem seems however always to be sterile (see fig. 2. 3. 4. 6); and it is very diffi-
cult to distinguish on this part any sort of envelope, which yet must certainly exist originally
also here in the form of a thin membrane. The whole axis is very slender, cylindrical and
everywhere (excepting at the root, and at the extremity which terminates in a narrow point)
of nearly uniform thickness. In the above mentioned exemplar (85 Mm. high) the axis was at
the basis scarcely 1 Mm. thick. In the upper part, the coral seemed certainly to be some-
what thicker; but this is attributable solely to the exterior polyp-bearing evelope (csenenchym)
which is here rather strongly developed, and in fresh exemplars is often double as thick as
the axis. The joints of the axis are generally extremely elongated, but of somewhat unequal
length, often strongly curved (see fig. 6. 12. 22 &c.) completely calcareous and connected with
each other by very short horny articulations or knees. By means of these latter, the sub-
stance, of which is somewhat elastic, the stem becomes flexible; for the calcareous joints are
completely stiff and inflexible. The lower part of the stem resolves itself into several (3—5)
irregularly curved and folded completely calcareous blades, which fasten themselves down be-
tween, and partly grasp, the harder particles of clay or grains of sand which lie above the
mud that covers the bottom of the sea, 'and thus, as it were like root-leaves, serve to attach
the colony. The latter seems therefore, unlike most of the other corals, to be only slightly
attached.

Most frequently these root-leaves are so arranged (see fig. 3. 4. 6. 18) as to form

together an inverted very short and wide funnel; the stem immediately below the first knee

7*

expanding itself downwards on all- sides, and then dividing itself into a certain number (usu-
ally 4) linear lobes directed obliquely downwards and outwards. These lobes, which are ex-
ternally more or less convex, internally concave, very thin and leaf-like, may again divide
themselves or send out irregular processes, and are frequently strongly enlarged at the extre-
mity, becoming at the same time flatter (see fig. 20 & 21). Externally they are of the same
nature as the calcareous joints of the axis, white and somewhat shining, and shew (in larger
specimens) a distinct longitudinal striping (see fig. 18. 19. 20) while the interior surface (see
fig. 21) is uneven and of a dull greyish color with irregular slightly marked somewhat curved
transverse lines (the convexity of the curve downward) which are doubtless lines of growth.
The whole root with all its lobes forms an entire connected body of completely calcareous
consistency; and it happens not unfrequently that more or less perfect roots of dead exem-
plars are found lying loose in the mud, together with numerous detached axis-joints. Some
of these isolated root-pieces were very large, and shewed, in connexion with the frequently
very thick axis-joints, that this coral certainly attains a considerable size, namely, (judging
from these parts) at least the height of 1 foot. The largest root which I found (fig. 18. 19)
has a diameter between the longest lobes of 57 Mm. but the exterior part of the lobes was
certainly a good deal broken off. On the top, or in the middle of the upper side, it had a
smooth joint-surface to which the first horn-joint of the axis had been attached; below this
joint-surface it becomes wider on all sides, and then divides itself into 5 nearly equally de-
veloped leaf-like branches inclining downwards and outwards, and in semi-circular arrangement,
so that 2 of the branches were diametrically opposite to each other, and the other 3 radiating
all on the same side at equal distances from each other and from the former. A single and
apparently entire root-blade particularly strongly enlarged at the end, which was likewise
found losse in the mud, and which I have delineated fig. 21. & 22, was 50 Mm. long. The
form and the arrangement of these root-blades seem however to be subject to considerable
variations generally, and to be dependent on the form and nature of the objects which are
to serve for the attachment of the coral. They are therefore often found irregularly bent
and twisted, and of unequal development, so that often some few of them are of unusual
length, while others are stopped in their growth (see for instance fig. 22 & 23).

As regards the joints of the stem, they are as stated usually very much elongated
mostly 10—20 times as long as they are thick, and of cylindrical form, but so that they are
somewhat enlarged at both ends, whereby especially the shorter joints acquire a shape some-
thing like that of an hour-glass (see fig. 13 & 16). In smaller specimens (fig. 3. 4. 6. 12)
they are more evenly cylindrical, and apparently quite smooth and somewhat transparent,
while in larger exemplars (fig. 13. 14. 16. 17) they are opaque white and distinctly striped in
a longitudinal direction. This striping, which is formed by raised parallel ribs and inter-
mediate furrows, is usually perfectly straight and regular; I have rarely found joints on which
it appeared to run a little obliquely, as if the joint had been twisted to one of the sides
(see fig. 17). The joint-surfaces are (see fig. 16. & 17) usually more or less salient in the

53

middle, and shew regular concentric lines indicating the different calcareous strata which
have been deposited one outside of the other. As already stated, their length is rather un-
equal; as in the same colony there are often inserted single much shorter joints between the
longer ones (see fig. 4. 22 &c.). These shorter joints seem however to be restricted only to
the sterile part of the stem, while the joints in the polyp-bearing part and in the' branches
are more of one length. The horny joints or knees are everywhere of the same length, par-
ticularly short, scarcely more than half as long as they are wide; in fresh specimens about
of the same width as the calcareous joints, and like these striped longitudinally (see fig. 13).
The proportion of length between the calcareous and the horny joints, is thus in the present
species rather different from the other Isidinae; the calcareous joints being usually 15 — 20
times, and in younger stems (fig. 12), often 30—40 times longer than the horny joints, while
the difference in length between both in the other Isidinae is far less. On more closely exa-
mining the mud drawn up in the dredge, there were found as above mentioned, besides root-
pieces and root-blades, numerous loose calcareous joints, and less frequently more coherent
pieces of stem (fig. 11. 12. 13. 14) of various thickness, and evidently belonging to colonies
of very unequal size. One of these calcareous joints, which I have delineated of the natural
size, fig. 16, was of quite unusual thickness, namely 5V2 Mm. in the middle, but proportio-
nally very short, namely only 10 Mm. and had undoubtedly belonged to the sterile part of the
stem of an usually large specimen. Some few other joints, very short in proportion to their
thickness, were also found; but most were of the elongated form characteristic of the species,
for which reason I must consider such short joints, as those exceptionally inserted between
the longer joints, in conformity with what may be seen in the 2 younger colonies delineated
fig. 4 & 22. It will be seen that the thicker joints, derived consequently from larger colonies,
are on the whole proportionally shorter than the joints of younger colonies, which will imme-
diately strike the eye, on comparing the 3 pieces of stem all represented of the natural size
in fig. 12. 13 & 14. I will further, for better illustration of the unequal proportion between
thickness and length, add the measurements of a number of loose calcareous joints belonging
to colonies of various sizes.

.V thick.

long.

1 V,Mm.

23 Mm.

2 Vs -

16 -

3 Vs -

17 -

4

V -

15 -

5

1

11 -

6

1

15 -

7

1

16 -

8

1V4 - 14 -

9

IV, - 16 -

10

1% - 21 -

JI3

thick.

long.

11

2 Mm.

9 Mm.

12

2

10 -

13

2

13 -

14

2

15 -

15

2% -

4 -

16

3

6 -

17

3% -

22 -

18

3V, -

14 -

19

4

9 -

20

5V2 -

10 -

54

In the 2 entire undivided specimens which I have delineated lig. 3 & 6, the propor-
tions of the joints of the stem were tolerably uniform. The, first knee is (and this seems to
be the rule for all specimens) situated close down to the funnel-shaped enlarged part of the
root. After this comes a very long calcareous joint, then 2 shorter joints, of which the first
is shortest; then again a long joint nearly of the same length as the first. The other joints
(4 or 5 in number) seem to be all about of the same length. Also in the specimen deline-
ated fig. 4 the proportion of the joints appears to have been originally similar, supposing
that the small second joint, which follows on the first side-branch, has properly belonged to
the next following. Likewise a similar proportion will be observed in the piece of stem de-
lineated fig. 22, presuming that the small first joint originally formed part of the next follow-
ing. Finally if we assume that in the 2 single specimens (fig. 3 & 6) the 2 -- - and in the
branched specimen (fig. 4) the 3 jointjs following the first had originally formed a single joint,
it will appear that the stem consists of a series of about equally long joints. It seems to
me very probable that in reality, as the colony grows up, such a division of the calcareous
joints may take place into several smaller joints connected with each other by horny knees,
whereby the joints become gradually more numerous, and shorter in proportion to their thick-
ness. In all cases I have usually found .that the joints in younger colonies .are not only
relatively, but absolutely longer than in the older colonies. This is quite different from what,
according to Steenstrup *, may be observed in the genus Isis, where on the contrary there
occurs during the growth a fusion of joints originally separated, so as to form coherent cal-
careous parts, particularly in the main stem, which on that account often seems to be to a
great extent unjointed. In this coral such a fusion never seems to take place; the stem is
always distinctly jointed right down to the root; and the joints seem just here to be shortest,
as the division of the originally single joints seems just here preferably to take place. This
division of the calcareous joints does not occur equally everywhere, but begins earlier in cer-
tain places, which just seems to account for the frequently great difference in the length of
the joints in one and the same colony. Probably therefore the calcareous joints in very large
colonies, in which this division is further advanced, are on the whole proportionally of much
less elongated form than in younger colonies, even if there may not be so small a difference
between the calcareous and horny joints as in the other species of the same genus.

As regards the branches, their axis is just like the main stem completely divided
into calcareous joints connected with each other by horny knees, about of the same thickness
as the joints of the main axis. They are as already stated, always single or unbranched, and
usually take their origin from the horny joints (see fig. 11), a feature which is precisely fun-
damental in the genus Mopsea, and whereby this genus is distinguished from the allied genus
Isis, where the branches, on the contrary, take their origin from the calcareous joints them-
selves. This rule is however not without exception. I have now and then found calcareous
joints on which, a distinct side-process had formed itself (fig. 15) which seems to be the be-

Om SUegten Isis oj; de under Isis hippuris Linn, sammenblandede Arter pg. 7.

55

ginning of a branch. On the piece of stem delineated fig. 22 there was such a side-process,
issuing from the 2n? calcareous joint, already connected with a horny joint, and the latter
again with another short calcareous joint, being thus clearly a branch; and on the polyp-
bearing piece delineated fig. 5, it was properly only the lowest of the 2 large side-branches
which issued directly from the horny joint, while the other took its origin from the next
calcareous joint near the base.

The so-called bark or the polyp-bearing fleshy • envelope (csenenchym) begins (in the
entire specimens examined) first to be visible about in the middle of the length of the stem,
although really it encloses also the lower sterile part in the shape of an extremely thin mem-
brane. It is whitish, or rather yellowish transparent, so that the axis is clearly seen through
it, (see fig. '8) and full of vitreous spiculse. Its width is in the fresh state about twice the
thickness of the axis, and sometimes even more.

The polyps, which stand out from the stem usually obliquely upwards and outwards,
are situated at some distance from each other, and apparently alternating on both sides of
the stem; but on closer examination they are found to be placed round about the stem, and
thus arranged spirally along the same. Between 2 horny joints there are usually found 5 — 10
polyps. These are of various sizes; as between those fully developed there are often found
very small budding polyps. The 'largest are when extended (see fig. 8) about 3 times as
long as the stem is thick (in the young specimens examined), or about 2 Mm. long; broadly
cylindrical, about twice as long as they are broad; when contracted about half as long, and
nearly globular or shortly cylindrical with rounded extremities. The mouth is surrounded by
8 tentacles, which are nearly as long as the body, conically elongated, or thickest at the
base gradually tapering towards the extremity, and on each side furnished with a series of
about 12 projecting conically pointed pinnules. The short and wide cylindrical gullet, or so-
called stomachal canal, shines through the skin of the polyp with its orange-yellow color, and
extends from the mouth about half way down the body, where the 8 septa or radial lamellae
also shine distinctly through as radial diverging orange-yellow sinuous stripes. The spiculse
in the polyp's body lie very closely packed together in the skin, and mostly in the direction
of the longitudinal axis of the body. They often project with their points somewhat above
the surface of the skin. Some of them, namely in the exterior part of the body, are as long
as half the body and even longer, and project with their points over the skin at, or between
the bases of the tentacles. Also on the exterior side of the tentacles and even of the pin-
nules there are spiculse; but they are here particularly small; they are all (see fig. 9) of
linear form, only somewhat tapering at the ends, and obtusely rounded (fig. 10), also every-
where rough with small irregular projecting nodules.

In the youngest colonies of only 20 Mm. height (fig. 7) there was only one single
polyp developed; and this was situated at the point of the stem. In somewhat larger spe-
cimens there were indications of one or two other polyps below it. The budding of the polyps
seems thus to proceed from above downwards. The axis in these small specimens was par-

56

ticularly thin, almost as fine as a hair, with the calcareous joints extraordinarily long in pro-
portion to the width, and the horny joints (3 or 4 in number) only slightly marked, but still
when highly magnified easily recognisable, by their greater transparency and faint yellowish
color. In the latter there could easily be distinguished a more opaque internal part, in the
form of a denser cord connecting the 2 opposite projecting joint-surfaces of the calcareous
joints with each other, and a more transparent outer part extending some way up along the
base of the calcareous joints. In the- latter there could in some cases quite dearly be seen
an internal canal running through the whole length of the joint. Outside the whole there was
remarked an extremely • thin exterior skin, the not yet polyp-bearing caenenchym (see fig. 70-
Th e color is in fresh specimens pale orange-yellow, which is especially attributable
to the polyps, and partly also to the casnenchym; but the axis is like the root, externally
snow-white, with strongly contrasting dark brown horny joints or knees.

I have only found this interesting coral in the great depths at Lofoten, namely at
the fishing station Skraaven in 300 fathoms, and at the fishing station Risvser down to 400
fathoms, the greatest depth there occurring. It lives everywhere only on soft clay or muddy
bottom, and seems never like most other corals, to attach itself to the naked rock.
The species may be characterised in the following manner.

*
Mopsea borealis. M. Sars.

Polyparium gracile. erectum, indivisum vel parce modo ramosum, ramis indivisis paucis
et divaricatis, stirpe tenuissimo ubique fere latitudine eadem inque tota longitudine distincte
articulata, articulis calcareis plerumque valde elongatis, subcylindricis, ad extremitates tamen
plus minusve incrassatis, longitudinaliter striatis, nonnullis brevioribus in parte imprimis infe-
riore interpositis, nodis corneis (geniculis) brevissimis illis 10ies— 20ics brevioribus ; radice dila-
tata, subinfundibuliformi, in ramos plures (3— 5) foliaceos calcareos ssepe irregulariter flexuosos
et apicem versus dilatatos et applanatos divisa. Cortex polypifera (speciminum juniorum) in
parte modo dimidia superiore distincta, polypis sat magnis spiraliter apicem versus dispositis,
tentaculis 8 sat longis pinniferis instructis. Spicula numerosa insequalia et in cute polyporum
et in cortice polypiferse acumulata linearia, extremitatibus obtusis, tuberculis numerosis minutis
scabra. Color polyporum et corticis polypiferse pallide fulvus; articula calcarea vero et foliola
radicalia nivea, nodis corneis saturate fulvo-badiis. Longitude speciminis junioris completl
90 Mm. latitude stirpis vix 1 Mm.; sed ad magnitudinem longe majorem excrescere potest,

Habitat ad insulas Lofotenses in profunditate 300—400 orgyarum, fundo argillaceo.

The present species stands very near to the Mopsea elongata (Isis) Esper1 (Mopsea
Mediterranea Kisso) also more particularly described by Philippi2, but does not appear to
attain so considerable a size as that, which grows more than 2 feet high and 1 foot wide
between the longest branches. Moreover our species is far from being so strongly branched

1 Pflanzenthierc Vol. 1. p. 47. Tab. 6.

2 Archiv fur Naturgeschichte 1842. 1. p. 38.

57

as the Mediterranean, but is often' quite undivided; and where branches are found, they are
very few in number, simple (not again divided) and extended in different directions (not in
the same plane). Finally the calcareous joints are in our species remarkable from their
unusual length, while the horny joints are quite short. In the Mediterranean species the
calcareous joints are, according to Philippi, at most 8 lines long, and the horny joints 2 lines
long; thus the proportion of relative length is at most but as 4 to 1; while in our species it
is usually as 15—20, and in younger specimens often as 30 — 40 to 1. Further there seems
to be some difference in the spicula. Philippi describes, besides the shorter ones, 8 very
long lancet-shaped spicula, which in dried specimens extend far beyond the calyx in the form
of 8 regular points (according to Risso's figure these points are even more than half as long
as the polyp). Such unusually long projecting and regularly arranged spicula are not found
in our species.

Finally our species seems, by the elongated form of its calcareous joints, to resemble
the species shortly noticed by Pourtales l Mopsea eburnea from Florida; but this is like the
other Isidinee dichotomically branched, and differs besides in the calcareous joints not being
thickened at the extremities, as is always the case with the larger joints in our northern
species. The above described characteristic lobed form of the root has been hitherto, among
the Isidinse only observed in the Mopsea elongata, in which it is described by Philippi in a
very similar manner; although its lobes or branches are according to this naturalist exter-
nally quite smooth, which as above mentioned, is not the case in our species ; the latter
having them very distinctly marked with longitudinal stripes.

I shall in conclusion remark that in all probability we have yet another species of
Isidinse on our coasts. The learned Bishop Gunnerus of Throndhjem has, in ,,Det kongl.
Xorske Videnskabers Selskabs Skrifter" Vol. 4. 1768 Tab. 4. Fig. 8. delineated a piece of a
stem taken up from the sea off Sinolen, which he considers to have belonged to an Isis
hippuris. But as this coral is an evident tropical form, only known from. East India, there
can be no question of our having the same species in our seas. The peculiar form of the
calcareous joints and their unusual shortness in comparison with the horny joints, seem also
decidedly to oppose the notion that this piece should belong to the species here described,
Mopsea borealis, and rather to shew that it belongs to some species of the real genus Isis.

1 Bulletin of the Museum of Comparative zoology at Harvard College. Cambridge No. 7. 1868. p. 132.

58

7

11, Eungiacyathus fragilis. M. Sars,

nov. gen. ct sp.
(PI. V. fig. 24-32).

Of the free TurbJaolidsB, of which most species are fossile, and which in the present
telluric period belong almost exclusively to the warmer seas, we have formerly known only
one single northern form, namely the Ulocyathas uri-licnx described by my Father1 and con-
sidered after later investigations, to be identical with the fossile coral classed by Milne Ed-
wards & Haine2 under the denomination Flabellum laciniatum, which has been found byPhi-
lippi in the tertiary formation of Calabria, It is therefore interesting to state the existence
of another remarkable coral belonging to the same group, and living together with the former
in the great depths off our coast. It forms the type for a proper genus, which was named
by my Father Fungiacyathus on account of a very striking external similarity to the well
known tropical genus Fungia belonging to another family.

My largest specimen has a transverse diameter of 21 Mm.

The polyparium or the calcareous skeleton (fig. 29. 30) has completely the appear-
ance of a Fungia., its base forming (fig. 25. 27) a horizontal flat, nearly circular disc, to the
upper side of which the radiating lamellae (septa) are attached, without being enclosed by
the so-called wall leaf (characteristic for most other stone corals) which forms the outer
boundary of the calyx (theca) and gives it its form. In our coral there can therefore be no
question of any calyx properly speaking ; because the calyx is precisely dependent on a more
or less perpendicular wall-leaf which is here entirely wanting, or strictly speaking exists in-
deed, but by being extended in the same horizontal plane, loses its character, and only re-
presents the base of the coral or a so-called foot-leaf. The coral is therefore below quite
flat, while above it is rather strongly convex (see fig. 29) which however is only owing to
the high salient septa proceeding from the foot-leaf, which from the border rise rather ab-
ruptly, and then again become more gradually lower towards the centre, thus forming a wide
and deep central cavity (fossa calicinalis) in the middle of the upper surface. These septa
are very numerous, particularly thin and brittle, with the free border in its whole length some-
what wavy, and rather regularly folded or corrugated transversely; so that the free edge
seen from above (fig. 31) appears zigzagged. They all remain in their whole length separate
from each other, without any anastomosis ; and have usually a tolerably straight direction
converging from the border to the centre like radii of a circle. Only exceptionally some of
them are somewhat irregularly sinuous in a part, or in the whole length of their course, and
this stands usually in connexion' with an irregularity in the exterior border of the coral. As
usual (! systems can be distinguished, and in each of these 6 — 8 orders of septa. The 6
primary septa (fig. 24. 29. 30. 1) are in most cases easily defined; although it often happens

1 Fauna littoralis Norvcgiae Vol. 2. pag. 73, Tab. 10, Fig. 18—27.
! Annales des sciences naturdles, 3e seric, T. 9, pag. 273.

59

that some of the secondary ones (ibid 2) are nearly as strongly developed. Both these or-
ders of septa are (see fig. 32) particularly high and curved, nearly as high as they are long,
and are continued (see fig. 30) diminishing rapidly in height to the centre, where their border
becomes strongly folded, and as it were covered with small irregular lobes or fringes which are
partly involved with each other, and form in the centre a corrugated part of thin calcareous leaves.
The tertiary septa (fig. 24. 29. 3) are still tolerably high, although considerably lower than
the primary and secondary, and do not extend to the centre; the quaternary and quinary
are however very low, as also those of 6th 7'.h & 8'.h order, which are only confined to the
extreme marginal part. Moreover no little irregularity is observed in the arrangement and
development of the single septa, so that the radial division of the calyx so strictly maintained
in many other corals, is here in many ways disturbed. It will thus be easily observed that
the number of septa in each system in the same coral is by no means always the same ; as
likewise the apparent irregular development, not only of the smaller septa, but sometimes
even of the primary and secondary, makes it rather difficult to define with perfect certainty
the individual systems and orders of septa (see for ex. the specimen delineated in fig. 26).

No real columella is to be found; although the central irregularly curled part, which,
as above stated, is formed by the interior converging ends of the primary and secondary
septa, may be regarded as a slight indication of a columella.

Neither can there be found in our coral any trace of the interior circle of septa (pali)
observable in many other corals.

The border of the foot-leaf, from which the radiating lamellae (septa) take their origin,
is not entire, but closely and irregularly incised or dentated, each of the pointed teeth cor-
responding usually to one of the septa, so that the number and development of the latter
can be tolerably well ascertained, from the number and proportions of the marginal points.

The coral is not completely circular in its general form, but as it appears, rather broader
in one half than in the other. In all the 3 specimens examined by me these wider and nar-
rower halves were indicated by a distinct constriction in the middle, or by 2 irregular lateral
incurvations of the border; and these had, with respect to the bucal aperture, a similar posi-
tion in all. Whether this was a mere coincidence, or a more constant case, I cannot with
certainty decide; as I have not had opportunity of examining any complete specimens excep-
ting these 3.

The foot leaf itself is, as has been stated, quite flat below (fig. 25. 27) without any
trace of attachment. There is only observed in the middle a somewhat corroded part, from
which the growth of the coral has taken its origin, but which was not appreciably salient.
From this spot numerous fine raised stripes or ribs (costa?) radiate towarde the edge. The
most distinct of these correspond to the primary and secondary septa situated above; between
these again there are found numerous others which become finer and finer. An extremely
fine concentric striation, indicating the successive growth of the coral, was also occasionally
perceptible.

60

There was no trace of any external membrane (epitheca) covering this part to be
seen in my specimens. The sott parts are entirely confined to the upper surface of the coral,
where they cover as well the septa as the intervals between them.

The animal is (see fig. 24. 26) single and actiniform like that of the Ulocyathus. The
bucal disk, which is stretched over the central hollow part of the coral (fossa calicinalis) so
that the interior part of the primary and secondary septa is completely concealed, and which
moreover is evenly continued outwards between the septa, shews tolerably coarse radial fur-
rows crossed by finer circular ones. The mouth forms an eliptical opening in the middle,
larger or smaller according to its greater or less contraction. The lip is thickened and crenu-
lated by numerous rounded lobes, which are limited by tolerably deep incisions. From these
lobes there descend, to the very bottom of the stomachal canal, the linear convex bucal folds
which are all nearly alike developed (none notably larger than the others).

The tentacles which always proceed from that point where the septa lose themselves
in the bucal membrane, are very small, about 50 in number, and forming 4 or 5 concentric
rows or cycles. Those in the interior, which are usually 6 in number, corresponding to the
6 primary septa (see fig. 24) but which sometimes appear to be more (see fig. 26) some of
the tentacles answering to the secondary septa coming into the same ring -- are the largest,
although not nearly so large as in the Ulocyathus, and attached at a distance from the centre
of about half the radius. In the following cycles they diminish rapidly in size; so that in
the outer ring they appear only as small tubercles. In shape they are usually cylindro-
conical, somewhat constricted at the base with obtusely rounded point, and everywhere covered
with rather large roundish prominent nodules (see fig. 28) in which are accumulated a great
many small urticary cells.

The color of the animal is, as in the Ulocyathus, a beautiful vermillion, which color is
more intense in the tentacles and particularly in the bucal folds. The coral itself, or the
polyparium, is on the contrary snow-white.

Of this interesting Turbinolide I have only found 4 specimens, all in the same locality,
namely at the fishing station Skraaven in Lofoten at the depth of 300 fathoms on soft clay
bottom, together with Ulocyathus arcticus, which occurs here in quite enormous quantities
and of more considerable size than I have found it in any other place, namely with a trans-
verse diameter of 35 Mm. or 1Y3".

This new Genus may be characterised in the following manner.

Gen. Fungiacyathus. M. Sars.

Polijparinm simplex, liberum, absque vestigio adhsesionis, discoideum vel lamina modo
instructum basali horizontali plana subtus subtiliter radiatim costata, lamina vero murali
(theca) nulla. Calyx subcircularis supra (septis altis) convexus margine crispo. Columella
null a, pali nulli. Septa numerosa, systemata 6 et ordines 6 — 8 formantia, primaria et secun-
daria valde elevata, arcuata, subtiliter transverse plicata, margine superiore subimdulato, usque

61

ad centrum producta, et hie irregulariter lobata et flexuqsa, omnia in tota lougitudine discreta
tenuissima supra marginem. exteriorem calycis prominentia. Fossa calicinalis sat magna et
sequaliter depressa.

Animal simplex actiniiforme sanguineum, ore elliptico plicato, tentaculis parvis conico-
cylindratis, verrucosis, apice obtuso, non retractilibus, 4 — 5 seriatis.

Species unica Fungiacyathns i'ragilis. M. Sars.

Calyx irregulariter circularis, parte dimidia altera latiore, diametro majorum 21 Mm.
Habitat ad insulas Lofotenses in profunditate 300 orgyarum fundo argillaceo.

Among the Turbinolidae hitherto known there are only 2 genera Discotrochus and
Discocyathus M. Edw. & Haine both only known in a fossile state (from the tertiary period)
which have the peculiar disc-shaped Fungia-like form of our coral. The Fungiacyathus is
distinguished from both these genera by the absence of a columella, from the Discocyathus
also by the absence of the inner circle of lamellae (pali) and is hereby evidently most nearly
allied to the Ulocyatlni* arcticus a Turbinolide living on our coast, which however is suffici-
ently distinguished by (inter alia) its short conical or wedge-formed crooked pointed base,
and by the presence of a wall-leaf which is certainly not very much developed in height
though yet distinct, while in the Fungiacyathus it is entirely wanting.

Our coral belongs to that sub-group (of the two in the family of the Turbiuolidse)
which M. Edw. & Haine call Turbinolinse, and which is distinguished from the other group
designated as Cyathinse by the absence of pali.

S p o n g i «e.

12, Trichostemma hemisphiericum, M, Sars.

nov. gen. et spec.
(PI. VI. fig. 1 — 15).

Of this sponge, so peculiar in its appearance, I have seen a number of specimens of
various size; and all these shewed the same characteristic form, which must thus be consi-
dered as perfectly constant for the present species.

The largest specimens have a transverse diameter of nearly 2" and a height of 1".
The form (see fig. 1 & 2) is more or less .high hemispherical, circular with the lower surface
plane, or in the middle a little concave; the upper surface evenly ^convex; the circular edge
rather sharp, and in its whole length closely bordered with a ring of numerous capillary
radiate spicula which together form a corona round the sponge like a nimbus, and in width
about V5 — V7 of the sponge's transverse diameter. The plane or somewhat concave lower
surface of the sponge is sunk in the mud, and the circular rim of radiating spicula contri-
butes essentially to secure its position, by spreading out in the mud on all sides. Of the
whole sponge there is thus only the upper convex surface which projects above the mud.
From this upper surface arise a greater or less number of obtusely conical, fistular processes
curved in various manners, each having at its extremity a circular aperture, the so-called
out-flow aperture (osculum). These fistular oscula, of which the number may vary from a
very few up to 20, are always confined only to the highest part of the convex surface; be-
tween them and the circular rimmed border, there is always a tolerably wide field without
any oscula (see fig. 2) which is naturally accounted for by the sponge being probably sunk
in the mud considerably beyond the border-rim. The latter performs doubtless the same
service as the root-fibres in other sponges ; that is, it serves to fix the sponge in the soft
mud in which it is found. Only in a few specimens I have found that the lower concave
surface had attached itself to "a small stone or to a larger shell, round which it had grown
entirely; so that only a small part of the object was visible externally. The sponge must
therefore properly be said to be free, not attached, but only by help of its border-rim lying,
as it were, at anchor in the soft mud.

The whole sponge is of rather firm consistency, and of a white grey color. The lower
surface is quite smooth, the upper convex part is on the contrary somewhat uneven or cor-
rugated with small roundish tolerably regular elevations and depressions.

63

The smallest specimen which I have found (fig. .3) had already the characteristic he-
mispherical form. The upper convex surface had only one fistular osculum in the centre.
The circular rim of spicula was already completely developed; and its width was equal to
V3 of the transverse diameter of the sponge.

When the sponge is dried, the upper convex side sinks considerably ; so that it be-
comes nearly Hat, which justifies the supposition that its interior is of a very loose texture.
This is also easily ascertained by cleaving a fresh or spirit specimen along the middle. It
then appears (see fig. 4) to enclose a rather large cavity filled with soft parenchym, wherein the
interior support or skeleton formed of spicula is only very slightly developed. But on the other
hand the dermal stratum is particularly thick, and of very firm consistency, owing in a great
measure to the numerous siliceous spicula, with which it is filled. This cortical stratum is
sharply distinguished from the interior parenchym, and has its greatest thickness in the middle
of the lower side, where it occupies about Va °f the height of the whole sponge, but be-
comes gradually thinner towards the periphery; on the upper convex side, the cortical stratum
is on the contrary everywhere of uniform thickness, and about half as thick as in the middle
of the lower side. The inner cavity, filled with sarcode, has therefore a crescent-like section,
as the thickened lower cortical stratum rises in the centre like a watch glass. In this interior
sarcode-cavity may be distinguished many irregular cavities anastomosing with each other,
and bounded by thin lamellae of parenchym, which are partly supported by spicula arranged
in fascicles. These cavities shew on the whole a radial arrangement, and have their out-let
above in a greater or less number of cylindrical tubes which pierce the cortical stratum, and
open each at the end of one of the exterior fistular processes. The whole upper cortical •
stratum seems moreover to be perforated with particularly fine pores, through which the
water is led into the interior of the sponge, to be afterwards emptied out through the
fistular oscula.

The spicula are of 4 different kinds, which yet have all the same general form, namely
the simple needle-shape. In the interior parenchym of the sponge there are found spi-
cula almost exclusively of one sort (fig. 11) long, straight and somewhat fusiform or tapering
at both ends, but so that one end always forms a bulbous enlargement (fig. 13) while the
other terminates in a fine elongated point (fig. 12). These spicula, which thus according to
Bowerbank's terminology must be called ,,fusiformi-acuate" are of very different length (from
0.60 to 5.40 Mm.) and have an evident axial canal. They are connected in long ramified
fascicles which radiate (see fig. 4) from a point lying in the centre of the periphery of the
sponge, but nearer to the lower than to the upper surface, or near the most prominent part
of the lower cortical stratum in the interior of the sponge. These fascicles of spicula sup-
port the thin lamellae of the parenchym which divide the interior cavity, and thence penetrate
into the upper cortical stratum to the greater part of its thickness, yet without reaching to
the extreme dermal layer (see fig. 5). In the lower cortical stratum, these fascicles are still
much closer, and radiate on all sides almost horizontally, penetrating not the thickness, but

64

the length of the stratum, even to the edge of the periphery, where they become gradually
mixed with spicula of another sort (fig. 14) which are much thinner and longer (up to 8 Mm.)
very elastic, bristle-like, not enlarged in the middle, but evenly tapered in their whole length,
and terminating in an extremely fine elongated point. These bristle-like spicula form exclusively
the wide border which gives to the sponge so peculiar an appearance, and radiate on all sides
standing out horizontally from the circular rim. In the upper cortical stratum no trace is to
be found of these bristle-like spicula. On the contrary it derives its firmness chiefly from
spicula of another form, which are also found in the lower cortical stratum near the edge, and
now and then also in the interior parenchym. These spicula, which are particularly closely
packed round about the ends of the long fascicles of spicula that radiate from the inte-
rior of the sponge, and have here usually a transversal position, or stand at right angles
to the exterior surface and with their points turned outwards (see fig. 5) are very short (about
0.24 Mm. long) and pin-shaped (fig. 6 a) or furnished at one end with a globular enlargement
or head (see also fig, 8) while the shaft itself is rather strongly enlarged in the middle, and
at the other end sharply pointed (fusiformi-spinulate Bowerbank). In the exterior part of
this cortical stratum, we find again spicula of a somewhat different form, although most cor-
responding with those last named, from which they differ by the small development of the
head, this being often scarcely perceptible, and by the capillary form of the shaft scarcely
enlarged in the middle (fig. 6, c). These spicula, which are about of the same length as the
pin-shaped ones, are (see fig. 5) extremely closely packed, and stand all side by side at right
angles to the exterior surface which they penetrate with their extreme point. These closely
packed exterior spicula form, as it were, a separate thin outer layer of the cortical stratum
(a— b) distinguished by a great degree of firmness, and by a lighter color than the other part.

The size of the different sorts of spicula is moreover, as already remarked, rather
various; and more casual deviations in the form may not seldom be observed; this is espe-
cially the case with the pin-shaped spicula in the cortical stratum. The head is not always
round in these, but often shews at the extremity a more or less prominent peg-shaped pro-
cess (see fig. 10, a). In one case (fig. 10, b) this peg was quite unusually long and obliquely
inclined to one side, whereby the head came to be situated at the end of the first 41!1 part
of the length of the spiculum. In another case (fig. 9) there appeared in the middle of the
spiculum a few irregular enlargements or nodulous excrescences. Very seldom there were
found, especially in the lower cortical stratum, single spicula of the same sort which were
more than double as long (fig. 7). Otherwise the length of the pin-shaped spicula is tolerably
constant, and only varies within very narrow limits (from 0.21. to 0.28 Mm.). Between these
and the capillary spicula lying in the extreme dermal stratum there were also evident tran-
sition-forms (tig. 6 b) in which the shaft although still extremely thin, shewed itself evidently
enlarged in the middle. In some of these last the head was unusually wide, lancet-like or
top-shaped (fig. 0, d).

This peculiar sponge occurs not rarely at Lofoten (at the fishing-places Skraaven and

65

Guldbrandsoerne) in a depth of 120—300 fathoms on soft clay bottom. I have not yet found
it in any other place ; but it is probable that when the great depths off our coast shall have
been more closely examined, this sponge will also be found in many other places.

As regards the systematic place of this sponge, it is according to the nature of the
spicula easy to see that it is a siliceous sponge ; likewise the arrangement of the spicula
shews that it belongs to the 1s.' group or sub-order established by Bowerbank characterised
by ,,spiculo-radiate skeletons" ; it is further easy to see that it belongs to the group named
by Oscar Smith Corticata and for which the well known genus Tethea is the type. But on
the other hand it will not agree with any of the formerly established genera, from which it
differs as well by its characteristic form, as by its peculiar attachment and fistular oscula.

The genus may be characterised in the following manner.

Gen. Trichostemma. M. Sars.

Spongia silicea, simplex, libera, in limo demersa et hie fimbria spiculorum setifor-
mium, flexibilium, radiantium sustentata, cortice crasso compacto circumdata, interiore multo
minus compacto, parenchymatoso, lacunis numerosis irregularibus trajecto. Oscula numerosa
tubifprmia in facie superiore libere prominentia. Sceletum ex spiculis acuformibus fasciculato-
racliantibus compositum, aliis brevissimis capite globoso in cortice densissime accumulatis,
aliis multo longioribus, fasciculos tenues parenchyma et corticem trajicientes formantibus.

Spec. Trichostemma hemisphaericiim. M. Sars.

Forma regulariter hemisplio?rica, facie inferiore plana vel paulo c-oncava, superiore
sequaliter convexa, limbo distincto circular! spiculis setiformibus tenuissimis et longissimis
ubique radiantibus fimbriam latam et elegantissimam formantibus ornato. Color albido-cinereus.
Diametros speciminum majorum fimbria exclusa 43 Mm. altitudo 20 Mm.

Habitat non infrequens ad insulas "Lofotenses in profunditate 120—300 orgyarum,
fundo argillaceo.

13, Cladorhiza abyssicola, M, Sars,

nov. o-pn. ct spec.
(PI. VI. fig. 16-34).

This remarkable sponge living in the great depths at Lofoten, has also such a peculiar
and characteristic exterior, that it is immediately known from all other sponges; and indeed
one might, at a cursory glance, be tempted not to consider it to be a sponge at all, but rather

some animal of the polyp kind, for instance a Sertularide. or a Polyzoon.

9

, 66

The sponge consists (see fig. 16 & 17) of a main stem, the lower end of which divides
itself into numerous finely ramified roots, by which it is fixed in the ooze. From this main
stem — at a greater or less distance from the root, and at nearly right angles on different
sides — issue undivided lateral branches of the same nature as the stem, and like it, termi-
nating in a somewhat enlarged obtusely rounded point. Everywhere, as well from the stem
as from the side-branches, there issue again secondary continuations in the form of thin
conical pointed and somewhat curved processes, all about of the same length, and usually
arranged, with short intervals, circularly around the stem or side-branches. Towards the ex-
treme end of the latter they become gradually shorter and more obtuse (see fig. 21) and at
last they disappear entirely, so that it looks as if the enlarged end had been produced by a
confusion of these processes with each other and with the branches. As regards the side-
branches themselves they are of somewhat different length, and seem usually to be longest
nearer to the base of the sponge, and to decrease successively in length towards the point
of the stem. The branches as well as the main stem are very thin, of the same thickness,
and mostly quite straight. But on the other hand the secondary processes are as already
stated, always more or less strongly and evenly curved towards the point in one direction or
in another, and strongly compressed laterally.

The color is in fresh specimens yellowish transparent, more or less going over to
reddish yellow; and the height of the largest specimens is about 60 Mm.

Such is the exterior appearance of this peculiar sponge, which, as may be seen, is
very different from anything previously known.

Closer investigation shews that the whole sponge is supported, and receives its form
by means of numerous very long siliceous spicula of the form called by Bowerbank fusiformi-
acuate, which lie longitudinally close together, and are united with each other by a cementing
substance in close and solid fascicles. These needle-shaped spicula form everywhere the
more solid axis, round which the organic parts lie. In the main stem they form a continuous
fascicle (see fig. 18) which at the basis gradually resolves itself into thinner and thinner
fascicles (see fig. 23) continued through the branched roots into which the stem divides itself.
At the base of these roots the spicula are still surrounded by an exterior organic stratum;
but this soon disappears entirely; and the exterior ramifications are formed exclusively by
the spicula themselves, the fascicles of which continue branching, becoming at the same time
thinner and thinner, until the last ramifications only consist of 2 single rows, in which the
spicula are arranged, not side by side but alternating, and so that the ends of the spicula in
each row lap over each other, or wedge themselves in between the preceding ones in the
same row and the corresponding ones in the second row (see fig. 24); each of these terminal
ramifications ends, when in a perfect (uninjured) state, in one single spiculum.

From the fascicle of spicula in the main stem, there issue in the side-branches similar
fascicles, the innermost spicula of which are wedged with their extremities in between the
spicula of the main stem. Also in the numerous tapered processes which are arranged cir-

67

culaily around the stem and the branches, thin fascicles of the same sort of spicula occur in
the same manner; so that their basis frequently rests on several fascicles diverging towards
the main axis (see fig. 18) while the process otherwise (tig. 19) only contains one single thin
fascicle, which is continued centrally up to the very point. In the enlarged end of the bran-
ches these central spicula are very irregularly arranged; although they appear usually to
exhibit a radial arrangement; their points are seen protruding (fig. 21) on all sides through
the outer skin.

On examining more closely these central spicula, we find them, wherever they form
the solid axis, of tolerably similar appearance and size (fig. 25), about 0.60 Mm. long, perfectly
straight, thickest in the middle, and evenly tapering towards both extremities, of which one
is simply rounded (fig. 28) and the other shortly pointed (fig. 27); in the interior there may
be observed a distinct but extremely fine axial canal. Only in the extremities of the branches
there are found, among the normal spicula, some few considerably shorter (scarcely half as
long) which are not perceptibly enlarged in the middle, and the proximal extremity of which
often shews an irregular curve towards one. of the sides, (fig. 26).

The organic substance forms everywhere a tolerably thick stratum, which surrounds
the spicular axis (see fig. 18) and in which there are found imbedded some few needle-shaped
spicula apparently detached from the axis. Otherwise it makes a uniform parenchym, wherein
absolutely no canals or cavities can be discerned. Outwardly this parenchym is covered with
a more solid dermal layer, in which are found imbedded numerous spicula of a figure quite
different from that of the central spicula above noticed. Three different sorts of them are
to be distinguished, all belonging to the group named by Bowerbank ,,retentive spicula". The
largest of these (fig. 29 & 30) the length of which scarcely exceeds 0.10 Mm. are simply
two-hooked (bihamate) and are rather scantily distributed round the stem and the branches
with their processes (see fig. 18), but more plentifully in the enlarged ends of the branches
(see fig. 22). The second form of superficial spicula (fig. 33) which I have found only in the
enlarged ends of the branches, but here in rather considerable number (see fig. 22), are like
the former bihamate, but much thinner and smaller (only 0.04 Mm. long) and decidedly distin-
guished by the incurved hooks not lying in the same plane but inclining to different sides,
often so that they lie nearly opposite, giving to the spiculum the form of a more or less
distinct S. They thus belong to what Bowerbank calls contort bihamate spicula. The 3r.d and
last form (fig. 20. 31. & 32) belongs to the inequi-anchorate spicula; and these are found
in quite extraordinary numbers everywhere in the outer skin, particularly at the ends of
the secondary processes, very closely packed together (see fig. 19). They are remarkably
small (only 0.02 Mm. long) so that their exact figure cannot be represented, unless by appli-
cation of the highest magnifying power. They appear (see fig. 32) to consist of a somewhat
curved stem enlarged at both the bent ends, but much more at one end than at the other.
Both ends seem (fig. 32 a) to terminate in 3 sharp hook-like, somewhat flattened, acutely

pointed teeth; but at the one end these teeth are so small that they can only be discerned

9*

68

with the greatest difficulty, while at the other end they are very large and crooked inwards.
Usually only 3 such teeth appear; but in some (fig. 32 b) it was evident that behind these
there were still a couple of teeth; so that the teeth were on the whole 5 in number. Whether
this applies to all or only to some, the position of which happened to be more favorable, I
am not able with certainty to decide. By means of these innumerable microscopic ,,claws"
which project everywhere from the surface of the sponge, all the more minute animals and
the light floating particles which come into immediate contact with the sponge, become atta-
ched to it, and thus probably fulfil an essential condition for its nourishment. The specimens
which I got up in the dredge shewed also a very peculiar appearance; an infinity of small
creatures adhering to it, especially various sorts of Calanus, small Amphipodes, and young
Anuelides, which adhered so closely that I had the greatest difficulty in cleansing the sponge
from all these extraneous bodies. These small animals came probably for the most part in
contact with the sponge, by being taken up in the dredge together with it; but that a similar
occurrence, even if on a smaller scale, would take place normally in the deep, I consider very
probable. As there appears here no trace of any pores interior cavity or oscula, the ali-
mentation must be supposed to be effected by a process different from that which takes
place where these apparatus exist.

If a small piece from the end of a branch is taken and treated with a weak solution
of caustic potass to make it transparent, there will appear in it under the microscope (fig.
22) a peculiar structure of the parenchym, which is not found elsewhere. It will be observed
that there are imbedded in this part numerous small globular bodies, with finely granular
contents, occurring in greatest numbers near the surface, but which seem also to be found
in the central part. It seems indubitable that we have here the place from which the repro-
duction of the sponge proceeds. In one of the specimens obtained (fig. 16) there were ob-
served in the branches themselves, and partly also in the main stem, imbedded here and there
in the parenchym, between the interior spicular axis and the exterior skin, several globular
bodies many times larger, easily discernible by the naked eye from their darker yellowish red
color and greater opacity. These bodies, which under the microscope exhibited a cellular
consistency (see tig. 34) are undoubtedly the fully developed eggs or gemmules. These had
certainly not developed themselves in the place where they lay imbedded, but had in all pro-
bability been generated at the extremities of the branches, and in the process of development
at last become free or released from the cellular tissue wherein they had originated, and
thence distributed themselves around in the soft parenchym of the sponge.

I have only found this remarkable sponge in one single locality, namely at the fishing
station Skraaven in Lofoten at the great depth of 300 fathoms on soft clay bottom. The
specimens obtained are of very unequal size, but exhibit all the same characteristic shrub-
like form. The largest and most perfect specimen, which is delineated fig. 17, had 14 bran-
ches springing out from the stem at different heights; the lowest about half the length of
the main stem, not reckoning the root; the others successively shorter. Close under the

69

place where the side-branches commenced, a little Area, pectunculoides was attached to the
stem by its byssus (a).

By its highly characteristic form, and peculiar attachment, this sponge differs from
all other known forms. The arrangement of the central spicula in close axial fascicles, from
which again radiate other smaller fascicles in the single processes, seems to shew that it
must be more nearly referable to the 1s.' of the groups of siliceous sponges, established by
Bowerbank, 'and characterised by a spiculo-radiate skeleton. Of the gen'era belonging to this
group, the genus Hijmcdesmioti is that, which in respect of the structure of the superficial
retentive spicula, appears to approach it most closely; but the species of this genus are of
a quite different form, namely consisting of thin crusts over extraneous bodies (shell &c.).
It exhibits moreover some conformity in the structure of the spicula with the genus Des-
m<i<:i<lf>n, which belongs to quite another group, distinguished by a spiculo-fibrous skeleton.
The form is however here entirely different; and the skeleton cannot either be properly called
fibrous; unless the whole sponge, can be considered as one single ramified fibre.

The genus may be characterised in the following manner.

Gen. Cladorhiza. M. Sars.

Spongia silicea ramosa, fasciculis clensis spiculorum acuformium axem solidam forman-
tibus sustentata, radiculis numerosis arborescentibus ex spiculis ejusdem generis formatis in
limo affixa, Parenchyma axem internam corticis instar circumdans spiculis superficialibus
anchoratis et bihamatis ornatum. Oscula et pori nulla. Ova in apicibus dilatatis ramorum
se evolventia.

Spec. Cladorhiza abyssicola. M. Sars.

Spongia ex stirpe composita tenui erecta ad basin in fibrillas radicales numerosas
arborescentes se dividente, de cujus parte superiore exeunt undique radiantes rami laterales
simplices, recti, latitudiue eadem ac stirps ipsa et apice tumefacto terminati. Stirps in tota
fere longitudine et rami processibus numerosis subsequalibus attenuate — conicis, subarcuatis,
lateraliter compressis, fasciculis spiculorum centralium sustentatis circumcirca ornata. Paren-
chyma axem spiculosam ubique (radiculis exc'eptis) investiens, simplex, sine lacunis intersti-
tiisve, membrana vero ornatum externa distiucta spiculis numerosissimis minimis insequaliter
anchoratis in processibus conicis imprimis dense accumulatis iustructa, inter qu« obveniunt
alia multo majora bihamata sparsa in apicibus ramorum tamen frequentiora et cum illis hie
etiam spicula alia multo tenuiora bihamata sed hamulis in directiones diversas vergentibus
insignia. Ova (vel gemmulse) magna spiculis omnino carentia in parenchymate ramorum et
stirpis libere accunmlata. Color dilute pallide-fulvus. Altitude speciminum maximorum GO
Mm., latitude stirpitis et ramorum 1 Mm., longitudo processuum secundarium 3—5 Mm.

Habitat rara ad insulam Lofotensem Skraaven in profunditate 300 orgy arum funclo
argillaceo.

70

14, Hyalonema longissimum, M, Sars.

n. sp.
(PI. VI. fig. 35—45).

Among the deep-sea sponges, which I have occasionally taken up in the dredge at
Lofoten together with other marine animals, I had already during my first stay there (in
1864) noticed one form distinguished hy its peculiar appearance. I therefore kept very care-
fully all the specimens I could obtain, in order to examine them afterwards. My Father
thought at first that we had before us a very anomalous new genus, which he in his notices
called preliminarily Chsetoropalium (Bristle-club). A short time afterwards Professor Loven's
interesting and important treatise on the Hyalonema boreale came out, in which paper the
author describes a very similar form, shewing its near relationship to the problematic ,,glas
roop Spongia" (Hyalonema), long ago well known from the Japan Sea, and completely cor-
recting the numerous misconceptions of the nature of this remarkable sponge. That the form
from Lofoten belongs to the same genus as Loven's species, is sure enough; nay it stands

/

so near to it, that I have even been in great doubt as to the validity of our form as a distinct
species; and the more so because its dissimilarity seemed chiefly referable to the exterior
shape, which in sponges is known to be usually very variable in the several individuals. It
seems however that in some cases, even the exterior form of sponges may exist with great
constancy within certain limits; and to this category the genus Hyalonema seems precisely
to belong, as well as the 2 sponges mentioned in the foregoing lines. I have had occasion
to examine a great many specimens of the Lofoten Hyalonema, and have found in them all
a remarkable agreement in this respect; and I have also had the opportunity, through the
kindness of Professor Loven, of comparing them with other specimens, probably of the same
species, from widely different localities, namely from the Atlantic Ocean, and from the sea
near Greenland; and also in these have found the same differences from the H. boreale Loven
as the Lofoten specimens exhibit. On the other hand there was among the Greenland Hya-
lonema, a magnificent specimen which immediately shewed a very different habitus, and in
this respect very closely corresponded with the specimens described by Loven as H. boreale
which I have also had sent to me for comparison. As there does not therefore appear to be
any evident transition between these 2 northern Hyalonemes, I cannot do otherwise than
consider the form here noticed as a proper species, although very nearly allied to the H.
boreale ; and I note it here under the specific name preliminarily applied to it by my Father.
The length or the height of my largest specimens is 75 Mm. that is considerably
greater than in those described by Loven as H. boreale. But the whole form (see fig. 35. &
36) is far more slender and delicate; and it is specially the very different proportion between
the body or head of the sponge and the stem, that immediately strikes the eye and has also
given rise to the specific appellation. While in the H. boreale, the proportion of length be-

71

tween the head and the stem is as 1 : 3, it is here as 1 :.6 or 8; in other words the stem is
here proportionally double as long as in the H. boreale, and proportionally much thinner.

Usually the stem exhibits a more or less strong but even curve on one side, and is
everywhere cylindrical, increasing quite evenly in thickness towards the base which often forms
a great enlargement (see tig. 46) at the place where the root-fibres begin to appear. This
part of the stem sunk in the ooze, and which may fitly be called the root, is curved in va-
rious ways; sometimes (see fig. 35) even in a right angle, but everywhere so thickly covered
with the numerous sinuous root-fibres full of extraneous matter, that its shape is often not
easily discernible. I have however always found that its extremity tapers again, and gradually
resolves itself into terminal root-fibres.

As to the root-fibres themselves, they are of very solid consistency, and so tough
that they cannot be pulled asunder without difficulty. They are (fig. 41) rather transparent,
of yellowish color, and apparently in their interior of a perfectly homogenous compact struc-
ture. Usually they exhibit a cylindrical form, which is retained even under strong pressure;
they are variously contorted and ramified, and are everywhere covered with extraneous matter
of different sorts (Rhizopod-shells, siliceous spicula, grains of sand &c.) so adherent as not
to be entirely removed without considerable trouble.

The body or head of the sponge (fig. 37) is club-shaped, with the upper surface more
or less flattened, yet most frequently so that such flattening is not at right angles with the
axis but somewhat oblique, From the highest side rises a very remarkable conical process
terminating in an opening, which is bounded by long projecting spicula and leads to the
interior of the sponge. This fistular process, which represents the out-flow aperture (osculum)
is usually situated on the edge of the head, so that its exterior side is flush with the side
surface of the head. Such I have found to be the case in nearly all the numerous specimens
from Lofoten which 1 have examined. Only in 1 specimen in which the head was unusually
thin and elongated, and probably on account of an accidental mutilation rather irregularly
shaped, the osculum was a little within the edge of the upper surface (here more directly
truncated) but still far removed from the centre.

In the H. boreale there is also (at least in one of the specimens examined byLoven)
a single widely gaping out-flow aperture on the upper side of the head; but this aperture is
firstly not fistular nor prominent, and next, it is situated nearly in the centre of the upper
surface, which is also clearly the case in the specimen from Greenland before-mentioned,
which I presume belongs to the same species.

The whole head is moreover everywhere hirsute with rather long projecting fine spi-
cula, which together form, in whatever position the sponge is viewed, a tolerably wide trans-
parent border around the head. These spicula projecting from the exterior skin are also
found in the H. boreale, but are proportionally shorter and less conspicuous.

The color is greyish brown; in very small specimens much lighter or nearly yello-
wish white.

72

As to the interior structure form and arrangement of the spicula, I can refer entirely
to Professor LoveVs masterly treatise; as in these respects both species are very similar. I
shall only notice a few points which possibly might indicate a specific difference.

The spicula of the stem do not exhibit in any of my specimens an evident spiral
arrangement, but only lie (see fig. 40) longitudinally sometimes rather irregularly, so that one
has its point turned a little more to one side, and another more towards the other side.
Their form and length are moreover (fig. 42. a) almost exactly as in the H. boreale. Only
it seems to me that the medial enlargement in the present species is less sharply marked:
often it is (see fig. 43) so slight that it cannot be discovered without attentively tracing the
successive increase of the spiculum from both extremities. In some spicula there may be
remarked (ibid) a very clearly defined interior outline, in some (a) situated nearer to the
axial canal, in others (b) nearer to the exterior surface, and which appears to indicate 2 dif-
ferent strata; in most of them however the substance of the spiculum exhibited everywhere
complete homogeneity.

In a longitudinal central section of the head, it appears (fig. 38) that the spicula of
the stem do not as in the H. boreale run conically into the interior of the club, but spread
themselves out fan-like in close diverging fascicles, whence again the secondary small fascicles
which support the parenchym radiate to the sides and upwards. On making the longitudinal
section through the fistular osculum, it will be seen that the latter forms a very large and
wide cavity, which reaches deep into the body of the sponge and here communicates with
other smaller cavities. The dermal part of the body of the sponge forms (see fig. 36) as it
were a sort of cortical layer, which however is not sharply distinguished from the interior
parenchym, and owes its solidity principally to the extremely numerous minute fusiform spi-
cula (fig. 42, f) packed together without order, which are also characteristic of the skin of
the stem (see fig. 40). From this dermal stratum proceed again (see fig. 39), at right angles
with the exterior surface of the sponge, the numero.us fine terminally curved spicula (fig. 45)
which produce, in the perfect specimens, the very conspicuous silky exterior down that sur-
rounds the whole head.

As to the exterior surface of the head, it appears under the microscope rather regu-
larly punctated; and this is especially plain in younger specimens. Each of the points seems
to be a fine pore leading into the substance of the sponge. In older specimens this fine
porous structure is less evident, chiefly on account of the various extraneous particles always
lodged between the projecting spicula, such as ooze, fine sand &c. which are not easily removed.

I have only met with the present sponge in the great depths at Lofoten. It occurs
there occasionally at the fishing-stations Skraaven and Guldbrandsoerne in 120—300 fathoms
water on soft clay bottom.

It has however in all probability a far wider distribution, namely both in the great
depths of the Atlantic Ocean and in the Arctic Ocean around Greenland, presuming that the
specimens sent to me by Professor Loven for comparison should really turn out, on more

73

minute examination, to belong to this species. We may. characterise our species in the fol-
lowing manner.

Hyalonema longissirnuin. M. Sars.

Forma gracillima et elongata, capite clavato, supine plerumque oblique truncato,
osculo singulo, magno, tubiformi, longe porrecto, marginal! instructo, ubique pluma deusa
spiculorum induto ; stirpe longissima et tenuissima, 6ies — 8ics capite longiore, basin versus sen-
sim latiore, radice plus minusve dilatata et curvata fibrillas numerosas flexuosas et arbores-
centes emittente. Spicula stirpis vix spiraliter disposita in capitis interiore divergentia fasci-
culum formantia magnum et latum, de quo fasciculi tenues spiculorum similium sed minoruin
parenchyma capitis sustendentes radiant. Osculum cum cavo magno interno communicat.
Color fusco-cinereus, Longitudo maximorum 75 Mm.

Habitat non infrequens ad insulas Lofotenses in prof. 120 — 300 orgyar, radice in ar-
gillo immersa.

What therefore distinguishes our species from the. very nearly related H. boreale Lov.
is: the far slighter and thinner form; the very unequal proportion between the head and
the stem; the situation and shape of the out-flow aperture (osculum) the proportionally longer
spicula projecting from the exterior skin, and the arrangement of the axial part penetrating
into the interior of the head, which does not run conically, but spreads out fan-like into a
wide and thick fascicle of spicula.

The characteristic form for the Hyalonema: a club-shaped enlarged part, borne at the
extremity of a long and thin stalk or stem, the other end of which is sunk in the ooze, and
there attached by means of numerous root-fibres, is certainly something quite unusual in this
class, but does not appear to be exclusively confined to the genus Hyalonema. On the
journey which my Father and I undertook together in the summer 1869 to examine the Fauna
of the Hardangerfjord, iny Father discovered in the ooze brought up by the dredge from a
great depth (150 — 300 fathoms) a number of specimens of a minute sponge, which had a
very similar habit to that of the H. longissimum already previously found by me at Lofoten,
but in certain points exhibited a constant dissimilarity; for which reason also my Father
noted it preliminarily as a new species of Hyalonema under the appellation H. parvum. The
closer examination to which I have subsequently subjected this supposed Hyalonema has how-
ever shown that it belongs to an entirely different generic type; the spicula being formed
according to quite a different type; as also the interior structure of the head seems to be
very different. The spicula in the head are proportionally of quite an enormous size, and
mostly belong to the form of connecting spicula which are named by Bowerbank ,,furcated
attenuato-patento-ternate"; among these are found in the exterior skin extremely small star-
shaped siliceous deposits. The Spicula in the stem are not, as in the Hyalonema, fusiform
with a medial enlargement, but simply needle-shaped.

Explanation of the Plates,

Plate I.
Rhabdopleura mirabilis.

Fig. 1. One of the largest connected pieces of a colony found, about 3 times magnified.
&. & the creeping stem, from which the single cells arise.

2. Another, piece, of which the creeping stem is attached on 2 points to extraneous
particles, forming between the two a freely hanging bridge.

3. A part of a living colony magnified about 16 times, a. a, the cells with their po-
lypides in different states of protrusion; b. b, the creeping stem; c. c, the bucal
shield; d. d, the tentacular arms; f. /j the stomach; g, the intestine; h. h, the con-
tractile cord; i. i, the axial cord.

4. The extremity of a cell, with its animal protruding from the aperture, strongly mag-
nified, a, the cell; c, the bucal shield; 'd.d, the tentacular arms, which here are
strongly curved outwards and downwards.

— 5. The animal taken out of its cell, c, the bucal shield; d, the tentacular arms; e,.

the oesophagus'; £ the stomach; g. the intestine; A, the spirally coiled contractile cord.

— 6. The lower end of the contractile cord, still more strongly magnified, shewing every-

where its evidently cellular structure.

— 7. A piece of the contractile cord about in the middle of its length, very strongly

magnified, shewing in the dorsal part a cellular structure, and in the ventral part
a fibrous or muscular structure.
8. One of the tentacles also very strongly magnified.

— 9. A piece of the creeping stem, freed from the adherent extraneous particles, shewing

the irregular arrangement of the circular rings, i. i, .the axial cord.
• 10. A longer piece of the creeping stem, likewise freed from the adherent particles,
together with the basis of the cells with their polypides mostly strongly retracted,
about 20 times magnified, shewing the single chambers into which the stem is di-
vided, c. c, the bucal shield; f. f, the stomach; h. h, the contractile cord; i.i, the
axial cord.

— 11. A small piece of this stem, together with the basis of a cell strongly magnified,

shewing the attachment of the contractile cord, and one of the transversal septa.

75

I. 6, the creeping stem; a, basis of a cell; h. Jt, the contractile cord; i, the axial
cord; k. k, one of the septa between 2 chambers of the stem.
Fig. 12. The broken end of the axial cord (i) together with the cellular marrow (s) issuing

from its cavity, still more strongly magnified.

13. A piece of the stem with 2 completely closed chambers (obliterated cells) i. i, the
axial cord; k. k, the transversal septa.

PL II.
Fig. 14—24. Rhabdopleura mirabilis (continued).

Fig. 14. A piece of a cell with the animal in it, strongly magnified, c, the bucal shield; <?,
the tentacular arms; e, the oesophagus; f\ the stomach; <?, the intestine; A, the con-
tractile cord.'

15. The animal taken out of the cell and slightly compressed, seen from the left side.
c, d, e, f, g, ft, as in the preceding figure; ?, the hyaline semi-lunar border at the
basis of the tentacular arms; m, the under lip; n, the ciliated tubercle at the base
of the tentacular arm; p, the flexor muscle of the tentacular arm; ^, the bucal aper-
ture; r, the cellular body between the end ot the intestine and the oesophagus:

16. The anterior part of the body seen from in front, c, d, I, as in the preceding figure.

17. The animal seen from the dorsal side. The letters as in fig. 15.

18. The animal seen from the ventral side. The letters as in fig. 15.

— 19. A piece of the bucal shield compressed and very strongly magnified, shewing the

extremities of the muscular fibres, which extend through the substance of the bucal
shield, diverging radially.

20. A piece of the creeping stem, shewing 2 polypides (,r & y) in process of formation.
i, the axial cord; k, one of the transversal septa.

21. The earliest stage of development noticed, h, the contractile cord; *', the axial cord.

22. A further developed polypide, seen from the dorsal side. c. c, the bucal shield; <?,
the tentacular arms; h, the contractile cord.

23. The same polypide together with the axial cord (i) seen from the ventral side. The
letters as in the preceding figure.

24. The same polypide compressed and more strongly magnified. The letters as in the
2 preceding figures.

Fig. 25 — 30. Flustra abyssicola.

Fig. 25. The exterior part of a colony, seen from the broad side, about 10 times magnified.
a. a, the avicularia.

— 26. A piece of a colony seen from the narrow side, with the avicularia (a) extended.

— 27. Another colony of narrower form seen from the broad side. a. a, the avicularia.

10*

76

Fig. 28. The end of a living colony, somewhat more strongly magnified, a, a greatly ex-
tended avicularium; b, a polypide fully extended.

— 29. The mandible of an avicularium seen from the inner or concave side.

— 30. A polypide with its cell, a, the ectocyst; &, the.eudocyst; d, the tentacular corona

(the lophophore with its tentacles); e, the oesophagus; f, the stomach with its cae-
cum; g, the intestine.

PI. III.
Fig. 1—3. Triopa incisa.

Fig. 1. A specimen seen from above, about 6 times magnified, a. a, the dorsal tentacles;
I, the anus with the rudimentary gills.

— 2. The same seen from the right side, a, the dorsal tentacle; c, the genital porus.

— 3. The same seen from below.

Fig. 4—15. Dentalium agile.

Fig. 4. The largest specimen found (the annexed line indicates the natural size), seen from
the right side.

— 5. A smaller specimen with evident longitudinal striation in the posterior part of

the shell.

A specimen unusually strongly curved.

The extremity of a specimen viewed from below, shewing the peculiarly ramified

tubes on the surface (extraneous parasitical body) and the terminal aperture with

the protruding supplementary tube.

— 8. The same part seen from the left side.

— 9. A piece of the shell strongly magnified, shewing the anterior ramifications of the

superficial parasitical body.

- 10. A living specimen seen from the right side, with extended foot and captacula.

- 11. The anterior extremity of the shell, with the part of the mantle extended beyond

the aperture, the foot and captacula; more strongly magnified.
12. The posterior spade-shaped appendage seen from the right side.

— 13. The same viewed from below.

— 14. A particularly small specimen, shewing the smooth embryonic shell corrugated

transversely.

- 15. The extremity of the same viewed from above, more strongly-magnified.

Fig. 16—20. Yoldia obtusa.

Fig. 16. A perfect specimen viewed from above, slightly magnified (the annexed line indi-
cates the size).

77

Fig. 17. The right valve seen from the outer side.

— 18. The same, seen from the inner side, shewing the exterior ligament, the border of

the hinge, the pallial sinus, and the muscular impressions.

— 19. The upper part of the same, more strongly magnified, shewing the cardinal teeth

and the muscular impressions.

20. A quite young living specimen with extended foot and siphons, magnified about 8
times, a, the anal siphon; b, the branchial; c, the pallial tentacle; d, the foot;
e, the gills; /', the intestine; y, the exterior ligament; /i, the border of the mantle.

Fig. 21—43. Pecchiola abyssicola.

Fig. 21. A perfect shell viewed from the left side, magnified about 5 times.

22. The same seen from above.

— 23. The same seen from behind.

— 24. The same; front view.

— 25. The right valve seen from the interior side, shewing the pallial line and the mus-

cular impressions.

— 26. The left valve seen from the interior side, a, the small loose ossicle imbedded in

the ligament.

— 27. The upper part of the same, with the umbo, and the ossicle (a).

— 28. The border of the hinge seen from below, with the ossicle (a).

29. A piece of the surface of the shell strongly magnified, shewing the radiating stripes,
and the regularly arranged points.

30. The same in profile, shewing that each of the fine points forms a small conical process.

31. A living specimen with extended foot and siphons, seen from the left side, a, the
anal siphon; 6, the branchial siphon; d the foot.

— 32. The anal siphon strongly magnified, seen from the side.
33. Aperture of the same.

— 34. Some of the superficial conical papillae still more strongly magnified.

— 35. A spirit-specimen with the animal and half-open valves seen from below, d, the

foot half retracted in the oval aperture of the mantle ; t, the thin byssus proceeding
from its base; £, the tentacles placed around the branchial siphonal aperture.

— 36. The same specimen viewed from behind, a, the anal siphonal aperture; &, the

branchial; t. t, the tentacles^ x. x, the conical tubercles arranged circularly around
both the siphonal apertures.

— 37. The animal taken out of its shell and seen from above, «, the anal siphonal aper-

ture; t, the tentacles; x, the conical tubercles; o. o, the ovaries.

— 38. 4 egg-capsules with eggs enclosed, strongly magnified.

— 39. The animal seen from behind, more strongly magnified, a, the anal siphonal aper-

ture; &, the branchial; t. t, the tentacles; x. x, the conical papillae.

78

Fig. 40. On of the tentacles strongly magnified.

— 41. A piece of the same, shewing the elongated conical sicle-papillse.

42. The animal seen from the right side, the mantle being partly removed to shew the
underlying parts, d, the foot; t.t, the tentacles; x. x, the conical papillae ; e, the gill.

43. The exterior part of the gill strongly magnified.

PI. IV.
Fig. l—ll. Oonieolis typica.

Fig. 1. A specimen seen from above, about 4 times magnified, a. a, the dorsal tentacles;
b, the velum; c. c, the labial tentacles; d, the proboscis; e, the penis.

— 2. The same specimen seen from the right side. /", the anus; the other letters as

in fig. 1.
3. Some of the dorsal papillae (the gills) strongly magnified.

— 3. The jaw and soft parts adjacent seen from the left side, a, the tongue; &, the sub-

sidiary bucal cavity.

— 5. The left jaw seen from the outer side.

— 6. The right jaw seen from the inner side.

— 7. The tongue with the rasp-plate, seen from the side (reversed position).

— 8. The tongue seen from the upper edge.

9. A piece of the rasp-plate with the 3 rows of teeth.

— 10. A piece of the middle row of teeth; side view.

11. Some of the middle teeth isolated and in different positions.

Fig. 12—18. Uinbellisyllis fasciata.

Fig. 12. A specimen seen from above, about 3 times magnified.

— 13. The head and the anterior part of the body seen from above, more strongly mag-

nified, a, the semi-lunar membrane.

— 14. The same; front view, a, as in the preceding figure.

— 15. A piece of the body in the middle, shewing the peculiar dark transverse bands

interrupted in the medial line.

— 16. Section of one of the middle segments.

— 17. The posterior end of the body, seen from above.

— 18. One of the pedal bristles, strongly magnified.

Fig. 19 — 35. Paramphinome pulchella.

Fig. 19. A spirit specimen viewed from the right side, magnified about 10 times.

20. The anterior part of the body of the same specimen seen from above, more strongly
magnified, a, the evaginated gullet or proboscis; b. b, the two strong chitine hooks
attached to the first segment; e. e, the anterior gills.

79

Fig. 21. The posterior part of the body seen from below.

22. The anterior part of the body of a living specimen with only 4 pairs of gills, seen
from above.

— 23. The same seen from the right side, shewing the main blood-vessels.

24. The head and first two segments of the same seen from below, shewing the bucal

aperture surrounded by thick lips.
25—29. A gill of If— 5'.h pair, shewing the progressive dichotomic division.

30. Section of one of the gill-bearing segments.

31. Section of one of the middle segments of the body.

32. The dorsal pedal tubercle, with the rudimentary dorsal cirrus, more strongly magnified.

33. The ventral pedal tubercle, with the ventral cirrus.

34. Pedal bristles, a, b of the dorsal; c, d, e, of the ventral pedal tubercle.

35. The double chitine hooks which are attached to the dorsal pedal tubercle of the
1s.1 segment.

PI. V.
Fig. 1 — 23. Mopsea borealis.

(All the figures excepting 7', 8, 9 and 10 of natural size.)
Fig. 1. The largest colony found, of which however the root is wanting.

2. Another specimen, on the top of which are attached 2 specimens of Scalpellum
Stroeraii (a. a).

3. A complete undivided colony, with root.

4. An other colony, somewhat branched.

5. The upper part of a colony, branched in an unusual degree.

6. An undivided living colony, with extended polyps.

7. A quite young specimen, with only a single terminal polyp.

— 7'. A piece of the stem of the same specimen, shewing the exterior skin and a horny

joint, strongly magnified.

8. The upper part of the specimen delineated fig. 6. with partially extended polyps.

9. Calcareous spicula from the skin of the polyps.

— 10. Extremity of such a spiculum, strongly magnified.

— 11. (Above fig. 1). A piece of the axis consisting of 3 joints, of which 2 belong to

the main axis; the third, springing out from the knee, forms the beginning of a
lateral branch.

— 12. A piece of the axis, consisting of 3 elongated calcareous joints.

13. An other piece of the axis of a larger colony, likewise consisting of 3 calcareous
joints connected by knees.

— 14. A third piece of axis, consisting of 6 calcareous joints.

80

Fig. 15. A calcareous joint, with a lateral process in the middle (rudiment of a lateral branch).

16. An unusually short and thick calcareous joint.

17. An other likewise unusually thick but more elongated and somewhat twisted cal-
careous joint.

— 18. The root of an unusually large colony; side view.

— 19. The same, seen from above.

20. A single root-blade greatly enlarged at the extremity, seen from the outer side.

— 21. The same, from the inner side.

22. The lower part of a colony, with abnormally developed root, seen from the side.

— 23. The same, seen from above.

Fig. 24—32. Fungiacyatlms fragilis.

Fig. 24. A specimen with the animal seen from above, not quite twice the natural size.
The ciphers indicate the different orders of septa.

— 25. The same, seen from below.

— 26. An other specimen with more enlarged bucal aperture.
27. The same, seen from below.

— 28. One of the inner tentacles, with a piece of the skin investing the corresponding

primary septum, more strongly magnified.

29. The skeleton of a 3r.d specimen, seen from the side, shewing the high thin septa.
1. 2. 3 the primary, secundary, and tertiary septa.

30. The same specimen seen from above. 1 the primary septa.

31. One of the primary septa, seen from above.

— 32. The same, seen from the side.

PI. VI.

Fig. 1—15. Trichostemma hemispluericum.

Fig. 1. The largest specimen found, seen from above, natural size.
2. The same specimen, seen from the side.

— 3. One of the smallest specimens obtained, with only a single osculum, seen from above.

4. An other specimen, cut through the middle, shewing the interior parenchymatous
cavity and the exterior cortical stratum, with 2 of the fistular oscula in section.

5. A thin lamella, cut out of the sponge perpendicularly to the surface, strongly mag-
nified, shewing the arrangement of the spicula. a-c the cortical stratum ; a-b the
dermal layer of the same; c-d, the interior parenchym.

6. Pin-shaped (fusiformi-spinulate) spicula from the cortical stratum, a, the ordinary
form; 6, a very thin spiculum, which however still shews the fusiformi-spinulate shape ;
c, one of the spicula which are closely packed in the dermal layer; d, the head and
base of the shaft of a somewhat abnormal spiculum of the same sort.

81

Fig. 7. An unusually elongated spiculum, of the fu&iformi-spinulate form, from the lower
cortical stratum.

— 8. The head and basis of the shaft of the same.

9. A fusiformi-spinulate spiculum, with lateral excrescences.

10. 2 spicula of the same sort, with abnormally shaped head.

11. One of the long needle-shaped spicula from th.e radiating fascicles that penetrate
the pareuchym.

12. The pointed end -of the same, strongly magnified.

13. The opposite bulbously enlarged- end.

14. The outer part of one of the marginal bristle-shaped spicula.

15. The thicker end of a similar spiculum.

Fig. 1G— 34. Cladorhiza abyssicola.

Fig. 16. A perfect specimen, with developed gemmules in the interior of the parenchym,
slightly magnified.

17. An other more ramified specimen, also slightly magnified, a, an Arcapectunculoides
attached to the stem.

18. A piece of stem with basis of one of the lateral processes, treated with a weak
solution of caustic potass, shewing the arrangement of the spicula.

19. The extremity of one of the lateral processes, shewing the central fascicle of spi-
cula, and the superficial closely packed inequi-anchorate spicula.

20. Some of the latter, more strongly magnified.

21. The extremity of a branch, shewing the club-shaped enlarged point.

22. A small portion of the extremity of a branch treated with caustic potass and com-
pressed, more strongly magnified; shewing the gemmules in process of formation,
and the 4 different sorts of spicula.

23. A piece of the ramified root, shewing the fascicles of spicula gradually becoming
disunited.

24. One of the extreme ramifications of the root, more strongly magnified, shewing the
arrangement of the spicula in the same.

25. One of the central spicula of the stem.

26. One of the smallest central spicula from the extremity of a branch.

27. The pointed ends of 2 different central spicula.

— 28. The obtuse end of the same.

— 29. One of the simply bihamate spicula; side view.

30. A somewhat smaller Do. seen from the side and below.

— 31. 3 of the inequi-anchorate spicula, magnified to the same degree.

— 32. The same sort of spicula, more strongly magnified, a, as the) most usually shew

themselves ; 6, with 5 distinct hooks (rare).

11

82

Fig. 33. 3 of the contort-bihamate spicula from the extremity of a branch (magnified to the
same degree as the simple bihamate).

— 34. A fully developed gemmule taken out of the parenchym of the sponge.

Fig. 35—45. Hyalonema longissimum.

Fig. 35. 1 specimen with an unusually short stem, and much enlarged rectangularly bent

root, very slightly magnified.
36. 1 other specimen with a longer stem and less enlarged root.

— 37. The head or body with the upper part of the stem of a young specimen; shewing

the fistular osculum, the exterior down of spicula, and the fine pores in the skin.

— 38. The head of an other specimen split along the middle; shewing the fan-like expan-

sion of the spicula of the stem in the interior, the great interior cavity opening
into the fistular osculum, and the parenchym with its lacunes and interstices.

— 39. A thin lamella cut from the head perpendicularly to the surface ; shewing the radi-

ating fascicles of spicula, the dermal layer and the exterior spicula.

— 40. A piece of the stem; the lower part with its exterior skin; the upper part shewing

the arrangement of the spicula in the axis.
41. A root-fibre, strongly magnified and partly freed from the adherent extraneous particles.

— 42. Spicula: a, of the stem; 6, c, d, of the secundary radiating fascicles in the interior

of the head; e, one of the spicula projecting beyond the skin; f, some of the small
spindle-shaped spicula irregularly imbedded in the dermal stratum (all equally
magnified).

— 43. The middle of 2 of the fusiform spicula from the parenchym of the head, strongly

magnified; shewing the very slightly marked medial enlargement, the axial canal
and the 2 different layers in the substance of the spicula.

— 44. The end of .one of these spicula.

— 45. One of the exterior projecting spicula, strongly magnified.

'. ,•.<:„, ,/,/

ON

SOME REMARKABLE FORMS OF ANIMAL LIFE

FROM THE GREAT DEEPS OFF THE NORWEGIAN COAST,

{SEQUEL TO THE UNIVERSITY PROGRAM PUBLISHED UNDER THE SAME TITLE IN 1872, PARTLY FROM
POSTHUMOUS MANUSCRIPTS OF THE LATE PROFESSOR Dr. M. SARS).

II

RESEARCHES d THE STRUCTURE AND AFFINITY OF THE GENUS

BRISINGA,

BASED ON THE STUDY OF A NEW SPECIES:

BRISINGA CORONATA,

BY

GEORGE OSSIAN SARS,

PROFESSOR OF ZOOLOGT .AT THE UNIVERSITY OF CHRISTIANIA.

WITH 4 COPPER PLATES AND 3 AUTOGRAPHIC PLATES.

University-Program for the last half-year 1875.

CHRISTIANIA.

PRINTED BY A. VT. 3H0SQ3R.

1875.

CONTENTS,

Page

I. Introduction 1

II. General description of the external form:

(Physiognomy) 4

III. Special description of the organs:

(Organology) 8

1. The skeleton system 8

a. The ambulacral skeleton of the arms 8

b. The ambulacral skeleton of the disc * . . 11

2. The cuticular system 16

a. The dorsal integument 16

b. The ventral integument 16

c. The calcareous parts belonging to the integument 17

&• The interior calcareous elements of the skin . 17

p. The spines 18

Y. The pedicellariffl . 21

3. The water system . 23

a. The ambulacral vessels 23

b. The water-feet and their ampullae 24

c. The madreporic body and the stone-canal 24

4. The muscular system 25

5. The nervous system 26

6. Organs of sense '. 26

7. The digestive system 28

a. The bucal aperture with its muscular apparatus (the bucal membrane) . . 28

b The digestive cavity 29

c. The radial caeca 30

8. The perivisceral cavity (caeloma) and the blood-system » . 31

9. Organs of secretion 34

10. Organs of generation 35

a. The ovaries 35

b. The spermaries (testes) 36

IV. General arrangement of the various organic systems:

(Topography} 38

V. The vital functions:

(Physiology) 41

1. The animal functions 41

A. Movement 41

B. Sensation 43

C. Functions of the pedicellaries 44

2. The vegetative functions 46

A. Nutrition 46

a. Reception of food • 46

b. Digestion ' 47

c. Circulation of the blood 47

d. Respiration 49

e. Secretion * 50

IV

Page

B. Propagation 51

a. The sexual propagation 51

b. The non sexual propagation 51

3. Faculty of regeneration and tenacity of life 52

VI. History of development;

(Ontogeny) 54

1. Progressive development 54

a. Early stage of the Brisinga 55

b. Development of the disc , 57

c. Development of the arms 58

d. Development of the ambulacral skeleton 60

e. Development of the cuticular skeleton 62

f. Development of the spines 63

g. Development of the pedicellaries ». 64

2. divergent development 66

a. Individual variations 66

b. Accidental variations (monstrosities) 68

VII. Distribution and occurrence:

(Chorology) 70

a. Horizontal distribution 70

b. Vertical distribution . 71

c. External conditions of existence; special occurrence 71

VIII. Remarks on Homology and Affinity:

(Philosophy) 73

A. General Homology:

1. On the scientific significance of the genus Brisinga, considered from the
stand-point of the Darwinian theory 73

2. On the fundamental form of Echinoderms and of their morphological indi-
viduality, as illustrated in the genus Brisinga 75

3. On the phylogenetic relationship of the Echinoderms to other animal types 78

4. On the genealogical relation of the several groups of Echinoderms to each other 81

B. Special Homology:

5. On the relation of the genus Brisinga to the Asteroidea in general ... 85

a. Homology of the ambulacral skeleton 87

b. Homology of the skeleton of the skin

c. Homology of the spines

d. Homology of the pedicellaries

e. Homology of the interior organs

6. Relation of the genus Brisinga to now living star-fishes 92

7. Relation of the genus Brisinga to extinct star-fishes ........ 93

8. Comparison of the 2 species of Brisinga 94

9. On the systematic position of the genus Brisinga, with remarks on the
classification of star-fishes in general ' . 98

10. Diagnoses of Family, Genus and Species 101

Explanation of the plates 103

I.

Intro duction,

In the year 1853 Mr. P. Chr. Asbjornsen, the celebrated Norwegian poet and naturalist*
was dredging in the picturesque Hardangerfjord, and obtained here from the great depth of 200
fathoms a very remarkable star-like Echinoderm quite unlike all that had hitherto been
known. Owing to its highly magnificent appearance, and in the presentiment of its true rela-
tion to forms long ago extinct, he applied to it the very characteristic and excellent mytho-
logical appellation Brisinga1), derived from ,,Brising", the breast-ornament of Freya, which
according to ancient Norwegian tradition was concealed by Loke in the abyss of the primi-
tive Ocean. In 1856 he announced his important discovery in the 3d -Vol. of Fauna littoralis
Norvegise, where the animal in question was briefly described as a new genus and species
of Asteridse under the denomination of Brisinga endecacnemos.

Among the forms of animal life discovered in recent times there are indeed few so
highly interesting as this remarkable Star-fish, distinguished alike by its extremely peculiar
exterior and brilliant color and by its gigantic size. It exhibits in reality an appearance
so totally different from that of all hitherto known forms, that we cannot be surprised, that
the authors have been, and still are, in great doubt as to what place in the system it should
properly occupy. According to its external form we might be inclined to refer it to the
group of Brittle-stars or Ophiurse with which it seems specially to agree in its small circular
disc and the enormously prolonged arms or rays issuing from the same; while on the other
hand there are important features (such as the deep and wide ambulacral furrows running

1 With reference to this name the learned Italian professor Sgn. Angela de Gubernatis says in the 2cl Vol. of
his nMythologie zoologique" pg. 459 : . . . . ,,En effet, soit qu'on les (les mythes) examine en detail, soit qu'on
les prenne dans leur ensemble, on est oblige de convenir que le ciel est le grand rairoir, dans lequel se sont
refletees toutes les images primitives du langage htimain. Ce n'est pas par hasard que le celebre conteur et
naturaliste Norwegien, P. Chr. Asbjarnsen, a donne le nom mythologique de Brisinga a une nouvelle espece
d'animal decouverte parlui au fond de la mer du Nord, et offrant une ressemblance frappante par sa confor-
mation avec le soleil, la perle lumineuse, 1'anneau magiqne, qui tombe dans 1'eau et que retrouve le poisson
mythique II n'y a que les poetes qui soient capables de deviner, de saisir d'instinct certaines verites de la
science, et Asbjernsen est un grand poete."

1

2

along the under-side of the arms, with their 2 rows of strongly developed water-feet or ten-
tacles, the presence of a dorsal madreporic body and of numerous pedicellarise) which tell
decidedly against such an annexation, and point rather to the proper star-fishes (Asterise)
for which reason my Father* also has expressed himself decidedly of opinion that in spite
of its different appearance it ought to be referred to the latter order. Nevertheless the
opinions as to the systematic place of this form seem still to be much divided; and Bronn1
and also in later times Haeckel2 have therefore seen no other expedient than to establish
for it an entirely new order, Brisingastra; as in the opinion of these naturalists it seems too
peculiar to be incorporated in any of the orders previously established in the class of
Asteroidea. At present therefore this remarkable genus stands completely isolated, without
connexion with the forms hitherto known; and even the fossile form Protaster, which is
considered by Asbjornsen as its nearest relative, exhibits in many essential points consider-
able divergences, agreeing in its general aspect much more closely with the normal star-
fishes. A more exhaustive investigation of this remarkable form than that given by As-
bj0rnsen in his preliminary description, must therefore certainly be highly interesting; and
has also probably been long desired and expected by the scientific world, in order that
more light may be thrown on the morphology of this highly enigmatic form; and on its rela-
tion to the other Asteroidea. The great rarity of the species first discovered, and its very
limited occurrence, only on one single secluded spot in the deep Hardangerfjord, where the
use of the dredge is connected with no small difficulty, has however as yet placed insuper-
able obstacles in the way of such investigation, for which abundant material is above all
things absolutely requisite. Every specimen occasionally taken up has been regarded as a
sort of ,,Noli me tangere", a precious object which should escape the reckless attack of
the anatomical scalpel, and shine entire and intact in the museum, as one of its brightest
and most costly ornaments. My Father has indeed subsequently (1. c.) given a number of
very important contributions to the morphology of this star-fish; but the materials at his
disposal, which only consisted of a fragment of the skeleton of the disc (the bucal ring) of
the skeleton of an arm and pieces of a dried arm, were too scanty for any thorough in-
vestigation to be made thereon; and his remarks, which are not accompanied by illustrative
delineations, are probably not very generally known.

In the summer of 1869 my Father and I together undertook a journey (at public
expense) to Hardanger, chiefly in order to procure materials for a renewed study of this
singular animal; this object alone being considered worth all the expenses of such a journey.
- We succeeded, after repeated dredging in the place indicated, in obtaining several speci-
mens, partly in good preservation, of which one or more should have been submitted to ana-
tomical scrutiny. But unfortunately it did not fall to my Father's lot to execute these in-

* Forhandlinger ved dc Skandinaviske Naturforskeres syvende Made i Christiania 1856 Pag. 209

Die Klassen und Ordnungen des Thierreichs.
1 Generelle Morphologie der Organismcn.

vestigations, for which he would certainly have been better qualified than any other. A few
months later his death, so regrettable for the science, took place, whereby these and many
other important examinations which he had just commenced were suddenly interrupted. Al-
ready in the preceding year I had, at the fishing place Skraaven in Lofoten, taken up in the
dredge, from the considerable depth of 300 fathoms, a quite young specimen of a Brisinga
which seemed to exhibit specific differences from the form previously observed, having for
instance only 10 arms, and which my Father, in the reasonable belief that the number of
arms was in this form as constant as in the B. endecacnemos, denominated preliminarily
B. decacnemos. It was however not before 1870 that, having found full grown specimens, I
succeeded in ascertaining the specific differences of the two forms, and also came to the
conviction that the number of the arms in the Lofoten species is, strange to say, and quite
contrarily to what has proved to be the case in the form first discovered, subject to very
considerable variation; for which reason the specific name appropriated to it by my Father
could not be retained. Under the appellation Brisinga coronata I noticed it briefly in the
year following1; reserving to myself to furnish at a subsequent time a more elaborate de-
scription accompanied by illustrative delineations. It is this description which I have now
the honor to lay before the scientific public. I have not so much had in view to estab-
lish by a minute comparative description the independence of the species in relation to that
previously described, as by the most thorough and exhaustive investigation possible of the
whole organisation of the present species, to furnish a contribution to our general knowledge
of the genus Brisinga, in order thereby to gain a more secure stand-point for judging subse-
quently of the morphology of this remarkable animal form, and of its relation to other Echi-
noderms.

The tolerably abundant materials which I have had at my disposal, and which in the
interest of science I have felt bound to use without any limitation for my anatomical in-
vestigations, have enabled me by repeated dissections and preparations, with and without the
help of chemical reactives, to ascertain and verify again and again most of the points in the
finer anatomic structure

I can here add that the present species has been found also during the expeditions
recently organised by the British Government in many places in the great deeps of the
Atlantic Ocean, and has been likewise recognised by the chief leader of these expeditions
Professor Wyville Thompson, whose thorough acquaintance with the Echinoderms is univers-
ally acknowledged, as a species distinct from the form first discovered Brisinga endecacnemos.
In a recent very interesting work by the naturalist mentioned, on the results of these expedi-
tions entitled: ,,The depths of the Sea," Brisinga coronata will be found briefly Described
under this name proposed by me, with a beautiful woodcut of the animal seen from above.

Nye Echinodermer fra den norSke Kyst. Chr. Vid. Selsk. Forhandl. f. 1871.

II.
General description of the external form,

(Pysiognomy).

(See Tab. I, fig. 1, Tab. II, fig. 1 & 2).

In its genera] aspect the present species has on the whole a great resemblance
to the Srisinga endecacnemos Asbj0rnsen as represented in the fauna littoralis Norvegise
2nd volume Tab. IX, fig. 1. From a little circular disc of scarcely more than 1 inch dia-
meter, there issue a number of spreading arms or rays (varying from 9 to 12) of an extra-
ordinary length (more than 1 foot) and armed along the edges with several rows of long
spines. These arms are very sharply defined, and at the base distinctly instricted; after-
wards for a short distance rather strongly enlarged, and then again regularly tapering to-
wards the point; they stand also so close together at the root that there scarcely appears
to be any interval between them. On closer examination there will however be observed
(see Tab. 1, fig. 3) in the angle between 2 arms a small peculiar calcareous plate, which
belongs to the skeleton of the disc, and which when the arm is removed (see Tab. 1, fig. 6,
Tab. Ill, fig. 1) shews 2 semilunar articulating-surfaces separated by an evident interval.
On the upper side this interradial plate rises in a tubercular prominence (Tab. 1, fig. 6 d)
which is very conspicuous in the upper edge of the disc by its lighter color, and by being
for the greater part bare; while the rest of the dorsal surface of the disc is covered with
short cylindrical spines growing close together. In one of the corners, at the issue of 2
arms, there appears on the dorsal surface of the disc a rough button-shaped prominence
(Tab. 1, fig. 3, 4, 6b) which is the so-called madreporic body. The dorsal side of the arms
is very uneven. In their whole length and up to their extreme point (see Tab. I, fig. 12,
Tab. II, fig. 1 & 2) there are usually found at regular intervals, numerous well defined half
cylindrical transverse band-like ridges, which are sometimes variously sinuous, and generally
distinguishable by a more intense color than that of the rest of the arm. These transverse
ridges appear on closer inspection to be of a soft fleshy consistency; although they also
contain calcareous elements in the form of innumerable closely packed microscopic pedi-

cellariae of the same nature as those which in great numbers cover all the cuticular sheaths
of the spines. But besides these soft transverse ridges, there appear in the interior or
adoral part of the arms, to about V* of their length, a number (10 — 14) of transverse ribs
separated by greater intervals of quite a different nature, namely like strongly projecting
solid calcareous beams, which from both sides of the skeleton on the underside of the arms
extend at tolerably regular intervals round the dorsal side of the arm, stretching the soft
dorsal skin over the interior organs that lie beneath. Along these calcareous transverse
ribs (which often (see Tab. II, fig. 4) project very strongly keelshaped on each side, giving
to part of the lateral outline of the arms an irregularly sinuous appearance) there is attached
a row of more or less elevated spines increasing in length outwards or towards the sides,
whereby a certain number of very conspicuous rings of spines are produced over the basal
part of the arms, a feature which is very characteristic of the present species, and has also
given rise to the specific appellation coronata.

If the animal is viewed from the lower side (Tab. I, tig. 2 & 5) there will be seen
in the middle a naked membrane (the bucal membrane) circumscribing in the centre a more
or less ample circular aperture (f) (the bucal aperture) through which the digestive cavity
may be seen. On the sides, this naked and altogether soft part is enclosed in a narrow
circular frame of calcareous consistency. This bucal frame, which is formed of the peripheral
part of the disc, is not even, but exhibits alternately elevated and depressed surfaces. The
former (fig. 5, d. d) which always lie in the direction of the interval between two arms, are
covered with a certain number of rather symmetrically arranged spines of unequal size, which
like the spines on the arms, are enveloped in a cuticular sheath thickly covered with pedi-
cellarise; the innermost of these spines (usually 6 in number) form at the end of the elevated
space a fanlike fascicle turned inwards towards the mouth. The depressed spaces of the
bucal frame (ibid. c. c.) which are always found just before the insertion of the arms, form
the immediate continuation of the wide and deep ambulacral furrows (comp. fig. 2) which
run along the whole length of the arms, and at the bottom of the depression there are 2
pairs of tentacles or water-feet extending forwards between the spines on each side. After
these 2 pairs of water-feet belonging to the disc, come the water-feet belonging to the arms
arranged continuously in the same manner, and at the same distances from each other, and
thus forming with those first mentioned 2 uninterrupted rows. On each side of the ambu-
lacral furrow of the arms, and issuing from the thickened border which forms their lateral
boundary, being itself formed of the lateral parts of the ambulacral skeleton, are the very
strongly developed spines of the arms, in several (usually 4) alternating longitudinal rows
(comp. fig. 14). Those in the exterior row are (see fig. 12) fixed just where the dorsal skin
of the arm connects itself with the ambulacral skeleton, and in the basal part of the arm
always at the extremities of the calcareous transverse ribs; they are the longest of all, and
directed outwards to the sides; while those in the other rows decrease rapidly in length,
and are directed more downwards. Those in the innermost row (fig. 14. 4) are very small,

and directed inwards between the" water-feet which issue and extend to some length from
the ambulacral furrow. All these spines are, as mentioned, enveloped in a sheath of skin
covered with pedicellarise and forming a somewhat enlarged bag beyond their point.

The color of the animal is somewhat variable, but yet so that the dorsal side al-
ways exhibits a more or less deep red tint; while the underside of the disc as well as of
the arms, is always paler, often quite white. On the dorsal side of the disc the color is
usually less pure than on the arms, and frequently goes over to a brownish yellow. In
most cases there appears round the periphery of the disc a series of very conspicuous light
circular spots corresponding in number with the arms, and shewn by closer examination to
be the upper tuberculous projecting part of the above-mentioned interradial calcareous plates
situated between each 2 arms, and on which the spines accumulated over all the other parts
of the dorsal side of the disc are entirely wanting (see Tab. I, fig. 4, Tab. II, fig. 1 & 2).
Only in one place this light spot is absent; as its room is occupied by the strongly promi-
nent madreporic body (b). The color on the dorsal side of the arms goes over in different
individuals through several shades from a light red to a deeper coral red, and more rarely
to an almost purple tint. The numerous soft transverse ridges have usually, as also the
terminal bag of the exterior spines of the arms, a still deeper and often a nearly red violet
color; while the calcareous transverse ribs over the basal region are always lighter than the
ground color of the arms. The spines are otherwise white; but the water-feet are most
frequently of a pale yellowish color.

Such is briefly the exterior appearance of this magnificent animal. It is however
seldom that we can see it as now viewed entire and uninjured, with all the arms in their
natural connexion with the disc. Not a single one of the specimens collected by me have
I been able to preserve entire, although every possible precaution has been employed; but
in all, the arms have been more or less completely separated from the disc, as was also
the case with the specimens hitherto preserved of the other species B. endecacnemos. In
this condition the animal appears like a confused heap of detached parts entangled together,
giving only a very imperfect idea of the splendid exterior of this star-fish; and even the
specimens artificially recomposed of the B. endecacnemos in the Bergen Museum, bear so
evidently the stamp of manufacture, that they can not give us a completely correct represen-
tation of the true original appearance of the animal. I have however sometimes had oppor-
tunity to admire this remarkable Asteride in its full beauty at the moment when it was
taken up from the depths of the ocean. In a few cases I have been so fortunate, when the
dredge at length came up to the gunnel of the boat from the enormous depth of 300 fathoms,
as to see it lying in the dredge-net, whole and uninjured as it had been crawling about on
the bottom, with all its parts in their natural connexion. But the animal is then, as will
easily be understood (after being drawn up from total darkness and from the tranquil depth
of the ocean to the clear light of day and to the disturbed surface of the water) put into
so unnatural a state for all its functions of life, that even the slightest mechanical incite-

ments will have a disturbing influence on its essentially loosely connected organism. How-
ever carefully therefore I have acted in endeavoring to lift up this gem of the ocean, I
have never succeeded in obtaining it entire. On the slightest touch, all the arms usually
break off at their junction with the disc; so that the latter is most frequently quite de-
tached from them. Only in a very few instances, by employing the utmost care, and by
placing the specimen immediately in spirit, I have been able to obtain one or more arms
in natural connexion with the disc. I have thus, as a great rarity, got one single young
specimen preserved, with 4 arms attached to the disc. This is the specimen from which
the 3 first figures in Tab. I have been delineated.

III.

Special description of the organs,

(Organology).

I. The Skeleton System.

(Tab. IV, fig. 1-22 and Tab. V).

The solid support for the disc, as well as for the arms, is composed of numerous
calcareous pieces more or less firmly connected with each other, and forming together the
so-called ambulacral skeleton. On the disc this skeleton forms a narrow exterior frame or
ring (fig. 1, 2 & 3) consisting of calcareous pieces immovably connected; and to the outer
side or periphery of this ring the ambulacral skeleton of the arms is attached. The latter
skeleton (see fig. 15 — 22), which occupies the lower side of the arms from the base to the
extreme point, consists of numerous sections or joints connected, to a certain extent mov-
ably, with each other by elastic bands, and forming the so-called ambulacral vertebrae, of
which each is again composed of a certain number of symmetrically arranged calcareous
pieces. Both the ambulacral skeleton of the arms and that of the disc are externally
covered with a smooth tendinous skin, like a thin envelope, firmly attached by growth, or
as it were a kind of periosteum, which can only be entirely removed by long maceration or
by employing chemical reactives. We shall now regard the ambulacral skeleton of the arms,
and that of the disc each for itself.
*

a. The ambulacral Skeleton of the arms.

(Tab. IV, fig. 15-22).

As already noticed, the ambulacral skeleton of the arms consists, like that of the
other star-fis|ies, of a series of similar successive sections or joints which have been called
vertebrae by reason of a certain general resemblance which the skeleton of the arm exhibits

to the vertebral column of a vertebrated animal; a' resemblance which is certainly more
remarkable in the present Asteride than in any other (comp. fig. 15, 16 & 17). These verte-
brae together form the boundary of the wide and deep ambulacral furrow which runs along
the ventral side of the arms, and in the bottom of which appear the holes or ambulacral
pores situated in pairs and indicating the attachment of the water-feet (see fig. 16). Each
vertebra (Tab. V, fig. 3—6) is composed of 2 ambulacral plates (aa) firmly connected with each
other by a suture in the middle, forming the upper convexity of the ventral furrow, and of 2 thick
cylindrical adambulacral plates (ad ad) which form the borders of the furrow on each side. In
addition to these there are on the 2 interior vertebrae of the arms (fig. 15—17) a pair of peculiar
plates (c c) situated on each side above the adambulacral plates, and which, according to
their place, must be considered as answering to the dorsal marginal plates in other star-
fishes. The connexion between the individual vertebrae is effected by means of smooth
articulating surfaces which in the circumference are partially attached to each other by
elastic muscular bands. Of these joint-surfaces there may be distinguished (see fig. 19—20)
at each end of the vertebra, 4; namely 2 central, situated close together (a a) whereby the
2 dorsal parts of the ambulacral plates which meet in the middle are connected with those
of the preceding and of the following vertebra, and 2 lateral (b b) whereby the adambu-
lacral plates of 2 vertebrae are connected with each other. This last connexion seems to be
the most movable, and is effected with considerably inclined joint-surfaces. On the adoral side
of the interior vertebra of the arm, there is moreover a 3rd pair of articulating surfaces
(c c) connecting the dorsal marginal plates with the skeleton of the disc.

If we regard the skeleton of the arm in its entirety, it exhibits on the upper part
(see fig. 15) along the middle, a rather high ridge or keel with a deep medial furrow. This
ridge represents the central part of the vertebrae of the arm, and is formed by the dorsal
part of the 2 contiguous ambulacral plates. On each side of this ridge the skeleton of the
arm is perforated by a series of circular apertures, which on the ventral side (see fig. 16)
appear at the bottom of the ambulacral furrow as the ambulacral pores, situated close
together in pairs, for the water-feet. These apertures are separated from each other by nar-
row partitions inclining downwards and outwards from the dorsal ridge, with ' their flat en-
larged extremities resting on, and attached to the thickened margin, formed by the adambulacral
plates, which makes the lateral boundary of the ambulacral furrow. At the bottom of this
furrow there is along the middle a narrow and rather deep furrow or semi-canal, which in
some places is covered with connecting ligaments. It is destined to contain the radial am-
bulacral vessel which extends through the whole length of the arm, and from which the
water-feet with their ampollas are supplied with water.

The limits of the individual vertebrae are, both on the dorsal and on the ventral
side, very distinct, and even externally indicated by evident intervals or slits covered with
elastic fibres, which allow the whole arm to be bent or extended. On the dorsal ridge these
slits have (see fig. 15) an inward and somewhat converging direction, and are especially very

2

10

conspicuous at the edges. By these slits the whole dorsal ridge is divided into a series of
successive sections, which are cylindrical or somewhat enlarged at both extremities, and have
the appearance of corpora in the spine of a vertebrated animal. From the middle of each
of these sections there issues on each side -- wonderfully like the transverse processes on
a vertebra -- one of the above mentioned partitions which separate the ambulacral pores
from each other. The flattened enlarged extremity of these partitions, which thus represent
the lateral parts of the ambulacral plates, rests on the thickened lateral margin of the skele-
ton of the arm always at the junction of the adambulacral plates of 2 vertebrae; so that
only a single adambulacral plate lies between them. The above-mentioned ambulacral pores
will thus be bounded outwardly only by one single adambulacral plate, and otherwise by the
ambulacral plates of 2 successive vertebrae. As regards the form of the adambulacral plates,
it is on the whole cylindrical; but the interior surface, which makes the lateral boundary
of the apertures for the suction feet, is rather strongly concaved. They are separated from
each other by distinct inclined slits, which are covered with elastic fibres, and which are,
especially at the extremity of the skeleton of the arm, very wide.

As already noticed, the 2 interior vertebras of the arm are distinguished by a some-
what shorter and broader form of the dorsal part of the ambulacral plates, and by each of
them having, immediately above the adambulacral plates, a distinctly developed dorsal mar-
ginal plate (see fig. 17). On the interior vertebra of the arm this dorsal marginal plate (c1)
is largest, of trapezoidal form, tapering outwards and having a rather large semilunar arti-
culating surface, whereby it is articulated with a corresponding marginal plate on the outer
side of the disc; at the aboral extremity it is firmly connected (often quite grown together)
with the next marginal plate (c2) which is considerably narrower, and of an obtusely conical
form. These 2 pairs of marginal plates, which are also very distinctly apparent on the
exterior of the arm (see Tab. 1, fig. 12 b, Tab. 2, fig. 1 & 2) are certainly intended to effect
a firmer connexion between the disc and the enormously developed arms, than would other-
wise be possible, if the connexion only existed between the ambulacral and the adambulacral
plates.

The skeleton of the arm has moreover in its whole length a tolerably uniform
appearance, excepting only that it tapers gradually towards the extremity, whereby also the
individual vertebrae are successively reduced in size. The extreme point of the skeleton of the
arm exhibits however, when more closely examined, a very abnormal constitution (fig. 21 &22);
the single vertebrae being here fused together in a rather large convex plate, which serves
as a protection for a peculiar organ of sense afterwards described.

11

b. The ambulacral skeleton of the disc.

(The bucal ring).

(Tab. IV, fig. 1-14).

When the skin and the interior soft parts are removed from the disc, there remains
a rather narrow circular frame or ring (fig. 1 — 3), of a firm calcareous consistency, representing
the ambulacral skeleton of the disc. This frame has (see fig. 2) its lower side flattened
horisontally, and divided into regularly alternating elevated and depressed spaces, of which
the former (a) taper a little inwards; while the latter (b) are of more uniform breadth. At
the bottom of each of the depressed spaces there appear 4 circular apertures, situated in
pairs and indicating the attachment of the water-feet of the disc (see also fig. 5). The
upper side of the bucal ring is not horisontal, but rising somewhat obliquely inwards.
From the most salient interior part of the bucal ring, the interior wall is inclined obliquely
downwards; while the much lower exterior wall is more perpendicular. A section of the
ring will thus represent a quadrangle with 2 longer and 2 shorter sides. The 2 longer sides
represent the lower and interior, and the 2 shorter sides, the upper and exterior sides (comp.
fig. 10 & 11). The interior wall of the bucal ring is (see fig. 1, 7 & 8) rather even and
smooth, but has nearer to the lower side a horizontal deep groove or semi-canal (x) ex-
tending in a circle round the whole ring, and having at the lower edge a thickened, and
inwardly somewhat salient border (y) which serves as attachment for the thin strongly con-
tractile bucal skin. This groove is destined to receive the circular ambulacral vessel situated
inside of the disc, and which, by means of the stone-canal, stands in connexion with the
madreporic body on the dorsal side of the disc. At the bottom of the groove there appears,
opposite to the middle of each of the depressed spaces on the under-side of the bucal ring,
a small circular aperture (fig. 8. o.) indicating the place whence the ambulacral vessels ra-
diating to the arms issue from the annular vessel.

The exterior, (as partly also the upper) side of the bucal ring is very uneven, with
numerous elevations and depressions, which however soon appear to have a very regular
arrangement. If lines are drawn along the middle of the alternately depressed and ele-
vated spaces on the lower side, and continued round the whole ring, it will be found that
these uneveimesses correspond exactly to each other on each side of the lines. If the
lines are continued outwards (the arms being considered as in connexion with the disc) then
the lines which bisect the depressed ventral spaces will also bisect the ambulacral skeleton of
the arms; while the lines which bisect the elevated ventral spaces will be continued out through
the intervals between the arms (see Tab. V, fig. 1 — 2). We may therefore, for the sake of facility,
preliminarily call the former the radial spaces of the disc, and the latter the interradial spaces.
On closer inspection of the skeleton of the disc, it will be perceived that the guiding lines

12

above referred to are by no means imaginary, but really exist in the form of evident sutures.
On the radial spaces these medial sutures may be traced round the whole ring; but in the
interradial spaces these sutures are interrupted on the upper side by a narrow single piece,
which we will call the wedge-plate (tig. 4, 6, 7, 8, 9 k) just forming the angle at the issue
of 2 arms, and the upper tubercular salient extremity of which is always, as before remarked,
very distinctly apparent on the exterior of the disc. Besides the sutures mentioned, which
thus bisect the radial and interradial spaces of the disc, there may still be observed on
closer examination a number of other sutures indicating that the bucal ring is composed of
a great .number of single pieces. Of these sutures there are some which go more or less
across the ring; others, which have quite a different direction, running along the upper or
lower side of the ring or parallel with its exterior or interior edge. The last sutures, which
are the most distinctly marked of them all, will be found to lie all in about the same plane,
and are especially very conspicuous on the upper side of the radial spaces, and on the lower
side of the interradial spaces. The result is that each of these spaces is also transversely
divided into 2, or consists of 2 successive sections. If we only consider the sutures running
along the ring, we shall thus find that the latter is composed of 2 concentric rings firmly
attached to each other; the inner being both higher and wider than the outer.

If such a bucal ring, with a part j)f the outer tendinous skin still adhering to it,
be placed in a concentrated solution of potass, the single pieces connected by suture will
after a short time separate completely from each other, and instead of a continuous ring
there will appear only a confused heap of numerous single calcareous pieces of various size
and form. The bucal ring is thus resolved into its simple elements, the appreciation of
which will now present considerable difficulties.

But if the operation be suspended a little before, it may happen that larger or
smaller parts of the bucal ring can be obtained with the single pieces in natural connexion
with each other, and yet connected so loosely that one piece after the other may be sepa-
rated from the rest with the greatest ease.

On examining and comparing a number of such parts, it will soon be found easy to
get some idea of the apparently extremely complicated structure of the bucal ring, especially
on comparing therewith one of the so-called ambulacral vertebrae of the skeleton of the arm.

If we consider the bucal ring in its natural connexion with the ambulacral skeleton of the
arms (see Tab. V, fig. 1 — 2), we shall easily be able to ascertain that it is chiefly composed of
double sets of similar vertebrae to those which form the skeleton of the arms; so that each arm
tinued inwards by two ambulacral vertebrae entering into the formation of the bucal ring is con-
and therefore immovably connected with each other and with their neighbors. The exterior
set of these vertebrae of the disc correspond really in almost every respect with the interior
free vertebrae of the arm, and can therefore, morphologically speaking, be more properly
said to belong to the arms than to the disc; which is also, as will subsequently be shewn,
confirmed by the development. On the other hand, the interior set of the vertebrae exhibit

13

several peculiarities in their form, as also the insertion. of single calcareous plates between
them, to which nothing analogous is to be observed in the other vertebrae, and which are
destined to attach them still more firmly together.

If we once more view the bucal ring from the lower side (fig. 2 & 5) it will
be easily perceived that the radial or depressed spaces make the interior continuation of
the ambulacral furrows of the arms, and consequently are formed in the middle by the am-
bulacral plates of the two vertebra? mentioned; while the elevated or interradial spaces are
formed by the contact of the adambulacral plates of 2 pairs of adjacent vertebrae: If we
then turn the bucal ring, and view it from above (fig. 1 & 4) we shall observe, on that part
of it which lies just over the ventral furrow, the dorsal convex part of the ambulacral plates
(a a) of the 2 vertebrae in contact, continuing the dorsal ridge of the skeleton of the arms
inwards. On each side of this raised part there are (at the extreme outside) 2 deep cavities
which when closely examined are found to extend through the whole thickness of the bucal
ring, and appear on the lower side as the before-mentioned holes, situated in pairs in the
ventral furrow, for the insertion of the water-feet. In this manner the double series of holes
or ambulacral pores that perforate the skeleton of the arm in its whole length, is thus like-
wise continued inwards on the bucal ring. Inside of these holes, the ambulacral plates (a1)
of the interior vertebrae come on each side in contact with a peculiar plate (m) to which
there is nothing corresponding in the other vertebrae, and which we may call the parietal-
plate; as it contributes in an essential degree to form the interior wall of the bucal ring.
It comes here in contact (see fig. 7) with the corresponding plate on the adjacent vertebra,
while these 2 parietal-plates are separated above by a 3rd equally peculiar single plate (k)
wedged in between them, which plate we have previously denominated the wedge-plate.
With the exterior end of this plate (which lies just in the angle between 2 arms) there are
again connected 2 small plates (r) belonging to 2 adjacent vertebrae of the exterior set, each
articulated with one of the interior dorsal marginal plates situated on 2 adjacent arms;
for which reason we may also regard those small plates observed on the exterior side of
the bucal ring as corresponding to the dorsal marginal plates on the disc of other star-
fishes.

The bucal ring is thus in a 10-armed specimen composed of 20 complete ambulacral
vertebrae; 30 peculiar plates inserted between the interior vertebrae, and 20 dorsal marginal
plates. The individual plates of which the bucal ring is composed, are therefore as follow:

14

Plates.

Interior vertebrae.

Exterior vertebrae.

Arabulacral plates .
Adambulacral plates
Parietal plates . . ,
Wedge plates . . .
Dorsal marginal plates

20
20
20
10

20
20

20

70

Together 130 single plates.

60

The more particular relation of the 2 sets of vertebrae which enter into the composi-
tion of the bucal ring, and of the single calcareous plates belonging to and connecting them,
is as follows.

The exterior vertebrae which, as already noticed, should properly be considered as"
belonging to the arms (although they always remain fixed to the skeleton of the disc when
the arms are detached) are considerably shorter and lower than the interior vertebrae
(comp. fig. 10). In their structure they correspond very closely with the interior vertebrae
of the arms, with which they are to a certain degree movably connected in nearly their
whole height. The articulating surfaces which appear distinctly in every detached disc on
the exterior side (see Tab. I, fig, 6, Tab. Ill, fig 1) exhibit therefore also in their form and
arrangement the most exact conformity with those on the adoral side of the interior vertebrae
of the arms (Tab. IV, fig. 20). It is easy to distinguish (see also Tab. IV, fig. 6) the strongly
convey central part (a2), formed by the 2 ambulacral plates uniting in the middle, with its
2 large vertical articulating surfaces. The side parts inclining outwards and downwards from
this central part, form also here the boundaries between 2 pairs of ambulacral pores, of
which the interior pair belongs to the disc; while the exterior pair is destined for the attach-
ment of the 1st pair of water-feet of the arm. At the end of these side parts lie the strongly
developed adambulacral plates (ad 2) which exhibit a large, somewhat inclined, roundly tri-
angular smooth articulating surface, whereby these plates are articulated with the adambu-
lacral plates of the 1st vertebra of the arm. They are considerably wider than the latter,
and are in immediate contact with the nearest adambulacral plates of the 2 adjacent vertebrae,
with which they are firmly connected by suture. Above the adambulacral plates there are
lastly the before-mentioned dorsal marginal plates (rj situated one on each side furthest
outwards, and exhibiting a small semilunar articulatory surface, whereby these plates are
articulated with the corresponding plates on the ls"joint of the arm. These plates also are,
at their interior extremity, in immediate contact with the nearest dorsal marginal plates of

15

the adjacent vertebrae and immovably connected with the same, as also with the extremity
of the wedge-plate (k) belonging to the interior vertebrae.

In the interior set of vertebrae, which are properly those that belong specially to
the disc, we can likewise distinguish (see fig. 9) a pair of ambulacral plates (a 1) and a pair
of adambulacral plates (ad1); while there are no proper dorsal marginal plates. In place
of the latter, there are between each vertebra and its neighbor inserted 3 peculiar plates
which we have denominated the parietal plates (m) and the wedge-plate (k).

The ambulacral plates are not only considerably longer (see fig. 10) but also higher
and broader than on the exterior vertebras ; while the adambulacral plates (see fig. 5 ad ')
are much narrower, and terminate in a somewhat unevenly truncated edge, which projects
freely under the thickened margin (y) and to which are attached the spines radiating to-
wards the mouth (the bucal spines) (see fig. 2). Each of the ambulacral plates is, as before-
mentioned, connected at one extremity by distinct suture with a so-called parietal plate.
This is (see fig. 4, 6, 7, 8 m) quite narrow above, and here contributes to form the boundary
of the interior pair of ambulacral pores, but is considerably enlarged on the interior
side of the bucal ring; so that these plates contribute essentially at this part to form its
interior wall, (see fig. 7), as also the above mentioned inwardly salient margin (y), which
makes the lower boundary of the circular semi-canal for the ambulacral vessel of the disc,
appears to be formed exclusively by these plates. On the other hand the ambulacral plates
themselves (a *) taper here rapidly, as the sutures between them and the parietal plates con-
verge downwards towards the holes, situated in the semi- canal mentioned, for the radial
ambulacral vessels (0). Each parietal plate comes close up on the interior wall otthe bucal
ring to the adjoining one on the adjacent vertebra, yet without being immediately connected
with the same; as there is always between them a very narrow crevice leading into a rather
spacious cavity in the interior of the bucal ring (see fig. 11, 1). These pariietal plates are
however above separated from each other by the insertion between them of a single pecu-
liar plate (k), which may in a manner be said to continue or terminate the series of dorsal
marginal plates inwardly. This plate, which we have called the wedge-plate, is (see fig. 12,
13, 14) quite narrow, in the middle somewhat constricted and nearly cylindrical; while the
upper part is enlarged to a tubercular prominence with corners drawn out on each side. Its
exterior surface is smooth and convex, while its interior surface is somewhat uneven, and
slopes from each side upwards towards the middle (see fig. 13). It is wedged in with its
upper part between 2 parietal plates, so as to extend round the dorsal edge of the bucal ring,
and a little way down on the inner surface (see fig. 7, 8) ; its lower extremity is truncated,
and on each side connected by sutures with the lower lateral extremity of an ambulacral
plate of the interior set of» vertebrae; while it connects itself in the middle with 2 of the
dorsal marginal plates belonging to the exterior set of vertebrae.

16

2. The cuticular system.

The cuticular system or Perisoma of the Brisinga is composed, as in other star-fishes
of the proper integument or skin, and of various calcareous parts which are partly imbedded
in the interior of the skin, partly raised above the same in form of spines ridges or micro-
scopic pincers, and which exhibit both in structure and function very marked differences.
The integument itself also shows a rather different structure in different regions of the body,
and we may thus properly distinguish two principal sorts of integuments, the dorsal and
the ventral.

a. The dorsal integument.

The skin which forms the upper covering of the disc and arms, is of very considerable
firmness, almost leather-like, and shews itself (see Tab. I, fig. 8) to be composed of 2 distinct
layers or strata, a thicker and firmer interior layer, and a thinner and less substantial ex-
terior layer: the former of fibrous, and the latter of cellular structure. In both these
strata of the skin there are imbedded various calcareous parts, which will be more particu-
larly noticed hereafter. In the middle of the disc and of the arms this dorsal skin is ex-
tended quite freely over the subjacent internal organs, and thus forms the immediate boun-
dary of the perivisceral cavity in this part; while it joins the ambulacral skeleton on the
sides, growing together with the tendinous membrane that envelopes the skeleton like a
periosteum.

The internal stratum of the skin is evidently composed of a contractile tissue; white
shining fibres of great tenacity and elasticity being interwoven with each other in different
directions; it is. especially in the dorsal skin of the disc very strongly developed. No di-
stinct pores could be detected in the same, either on the disc or on the arms.

The external stratum or epidermis is of a much softer and more fragile consistency
and partly lined with fine vibratile cilia. It covers as a continuous layer all the uneven-
nesses of the body; and all the spines which issue from the surface of the body thereby
receive their peculiar sheath-like envelope. On the dorsal side of the arms it forms more-
over the peculiar transversal semi-cylindrical fleshy ridges, especially characteristic of this
species, which are traceable even to the extreme point of the arm.

b. The ventral integument.

From the integuments properly so-called we distinguish the cuticular stratum, which
is found on the ventral side inside the ambulacral skeleton, partly lining the bottom of the

17

ambulacral furrows and partly surrounding the oral aperture as the so-called bucal membrane.
This ventral cuticle is of much finer and thinner consistency and without any trace of cal-
careous deposits. In the ambulacral furrows it appears as a thin transparent membrane
without any distinct structure; while the bucal membrane is remarkable for a high degree
of contractility, for which reason also it is interwoven with numerous fine concentric and
radial muscular fibres; it will be more particularly described hereafter.

c. The calcareous parts belonging to the integument.

Of the various calcareous deposits in question we may distinguish 3 principal groups:
1) the calcareous deposits in the interior stratum of the skin, 2) the spines issuing from the
surface of the body and 3) the peculiar microscopic organs found in the exterior stratum
of the skin, which we call pedicellarise.

a. The interior calcareous elements of the skin.

The formations classed in this category, which correspond to the more or less deve-
loped calcareous net in other star-fishes, consist of variously formed calcareous deposits in
the interior stratum of the skin, serving partly for the insertion of the dorsal spines, and
partly as a frame-work to extend the skin over the subjacent interior organs.

In the dorsal skin of the disc, we find these calcareous parts (see Tab. I, fig. 8, 9 a)
in the shape of small truncated cones, situated close together and traversing the interior
stratum of skin in its whole thickness, yet without-being connected wich each other or form-
ing any proper calcareous net. Each of these calcareous parts consists apparently of nume-
rous small cemented calcareous granules (see fig. 9 a); and on the narrower upper extremity,
there is always articulated one of the small dorsal spines which project beyond the skin of
the disc. By means of these close-lying calcareous parts, the dorsal skin of the disc acqui-
res a very considerable degree of solidity; although it retains its elasticity and flexibility
owing to their not being connected with each other.

The calcareous deposits found in the dorsal skin of the arms present quite a different
appearance. These calcareous parts are however essentially confined to the basal part of
the arms; while the rest of the dorsal skin of the arms is quite soft and flexible, and lies
close to the subjacent ambulacral skeleton. But in the basal part of the arm, which con-
tains a number of important and some very strongly developed internal organs, (the radial
caeca and the organs of generation) it was necessary for the skin to have an internal sup-
port, in order the better to protect these organs, and to preserve them from outward pres-
sure. Therefore the calcareous elements deposited in the skin appear in these parts in a
very peculiar form especially adapted for such purpose, namely as continuous solid arched
transverse beams connected on each side with the ambulacral skeleton, and, being placed

3

18

at a suitable distance from each other, extend the whole skin like a sheltering roof above
the said internal organs.

These very strong calcareous transverse beams, which undoubtedly, in spite of their
very different appearance, are to be referred to the same category as the isolated calcareous
particles in the dorsal skin of the disc (both formations corresponding to the calcareous net
of the normal star-fisk) are also (see Tab. I, fig. 12, Tab. II, fig. 1, 2 & 4) very conspicuous
on the exterior as more or less strongly projecting transverse ribs. Like the calcareous
particles in the skin of the disc, these transverse ribs give also issue to spines, which are al-
ways arranged in a single transverse row along the rib. With respect to the extent, number
and form of these transverse ribs, there is considerable variation. Usually they are distinct
on all the baSal fourth part of the arm. and from 10 to 14 in number. They are always placed
at some distance from each other; so that generally there are 2 vertebrae between each of
them and its neighbor. As to their form, they are often more or less sinuous, more rarely
forked on one of the sides. Sometimes they are incomplete or interrupted at parts; and
this seems to be the rule with those furthest from the disc; the sides only being developed,
and one of the above-named soft transverse ridges forming the continuation '^n the middle.
They are, as noticed, firmly attached on each side by a ligament to the ambulacral skeleton;
and their somewhat enlarged extremity gives at this point of connexion issue to an unusu-
ally long spine directed outwards and sidewise, which we will call the marginal spine, as it
proceeds from the margin of the arm. These marginal spines are however, as will be seen,
not confined to the basal part of the arm, but are continued also along the edges of the
whole remaining part of the arm, even to its extreme point; and on closer examination we
shall find that they are also here not attached to the ambulacral skeleton itself, but to
small calcareous pieces, connected with it at regular intervals, which may therefore be con-
sidered as a sort of rudiments of the so strongly developed calcareous ribs on the basal part.

[1 The spines.

The various spines projecting over the surface of the body, issue partly from the cal-
careous particles deposited in the interior stratum of the skin, partly direct from the ambu-
lacral skeleton itself; andihey are always more or less movably connected with the same. Their
form and length vary in the different parts of the body; but the proper calcareous spine is
always enveloped as in a sheath by the exterior stratum of skin, which usually forms a bag
projecting far beyond the point of the spine, and, like the rest of the cuticular sheath,
covered with numerous pedicellaria?.

All the spines articulated with the ambulacral skeleton, issue from the ventral side,
and are here attached to the adambulacral plates which form the lateral boundaries of the
ventral furrows, for which reason also the corresponding spines in other Asterida3 are usu-
ally called furrow-papillae or furrow-spines. They are situated in the Brisinga alternately,

19

forming along the arms on each side of the ventral furrow 3 rather regular longitudinal
rows (see Tab, I, fig. 14). There will most frequently be found on each adambulacral plate

3 such spines (ibid. 2, 3, 4) situated in an oblique transverse row, and diminishing rapidly
in size towards the medial line. Each of these spines belongs thus naturally to a different
longitudinal row. The exterior and largest of them (2) issues from about the middle of the
ventral side of the adambulacral plate, and is here articulated with a very distinct projecting
rounded knob or tubercle; it is usually directed obliquely downwards and a little outwards,
and in the exterior part of the arm exceeds the breadth of the arm. The next spine (3) is
considerably smaller, scarcely '/a as long, and usually pointing straight downward. And last-
ly the interior spine (4) issuing from the interior adoral angle of the adambulacral plate, is
very small, and points inwards towards the ventral furrow between the extended water-feet.
In very young specimens (see fig. 2) these 2 innermost spines are yet undeveloped; so that
there is only a single longitudinal row of furrow-spines along the middle of the adambulacral
areas. But in older specimens, the number of the furrow-spines often increases considerably;
so that along the interior edge of the adambulacral plates there may be found 2, 3 or even

4 small spines, all apparently belonging to the interior longitudinal row (see Tab. IV,
fig. 16).

On the lower side of the disc (see Tab. I, fig. 2 & 5) these furrow-spines are con-
tinued on each side of the ambulacral furrows up to their extremities at the bucal area.
They are also here attached to the 2 pairs of adambulacral plates which form the boundary
of the ventral furrow of the disc; so that a somewhat larger spine always issues from the
middle of each of these plates; while the others stand nearer to the edge, which is turned
towards the ventral furrow (see Tab. IV, fig. 2). The number of these spines is also here
somewhat variable according to age. But from the interior border (turned towards the
mouth) of the innermost adambulacral plate, there always issue a certain number (usually 3)
larger spines pointing horisontally inwards towards the mouth, and which together with the
corresponding spines on the adjacent adambulacral plate, form at the end of each of the
interradial spaces a regular fan-like outspread fascicle of usually 6 spines, corresponding to
the bucal spines in other star-fish.

All the spines mentioned are in their structure completely similar. The calcareous
spine itself is (see Tab. I, fig. 15—21) slender, shaped like an awl or like a needle; not
echinulated, but smooth and shining and moreover distinctly striped longitudinally. These
stripes, which often go somewhat obliquely, or rather spirally, are raised smooth ribs; and
in the intervals between them, there may be observed, when strongly magnified, rows of
transversely oval holes traversing the hyaline calcareous net of which the spine is composed.
Only at the extreme point there appear when strongly magnified (see fig. 17) some short
irregularly placed points;' otherwise the spine is quite smooth. At the base the spine is
somewhat enlarged, and has a depressed articulating surface, whereby it is articulated with
a corresponding elevated knob or tubercle on the ambiilacral skeleton. The articulating

3*

20

surfaces are in the periphery connected by tendinous muscular ligaments, whereby the spines
can be moved in various directions. The thick sheath formed by the exterior stratum of
the skin (cuticle) which envelopes the spine, terminates always in a somewhat enlarged bag,
often prolonged far beyond the extremity of the spine, and not unfrequently curved in various
ways, sometimes nearly at right angles (fig. 18).

The very long spines, attached along the sides of the arms, which with reference to
their position we have called marginal spines, are also of exactly the same structure. These
spines are however, as already noticed, not attached to the ambulacral skeleton, like the fur-
row spines, but, in the basal part of the arm, to the extremities of the calcareous trans-
verse ribs, and further out, to peculiar small calcareous plates connected by ligaments with
the adambulacral plates. They form (see Tab. II, fig. 1 & 2) on each side a single row
from the base of the arms to their extreme point.

If we consider these marginal spines together with the furrow-spines, the arms (see
Tab. I, fig. 14) will appear furnished on each side with 4 rows of spines, of which those in
the exterior row (the marginal spines) (1) are the longest. As the marginal spines do not,
like the furrow-spines, correspond in number with the adambulacral plates, but most fre-
quently are to the latter only in proportion of 1 to 3, their number is also always smaller
than that of the furrow-spines belonging to a single row. At the base of the arms they
are only slightly developed, but increase rapidly in length outwards until towards the middle
of the length of the arm, where they are 3—4 times as long as the arm is broad. Further
outwards they then gradually decrease in length, until they become rudimentary again to-
wards the point (see Tab. II, fig. 6).

The dorsal spines of the arms, are, as noticed, only attached along the calcareous
transverse ribs, and thus confined to the basal part alone.

They form here always only single rows, and usually increase in length outwards,
but without attaining to nearly the length of the marginal spines and of the exterior furrow-
spines. They seem to be of somewhat stouter structure than the other spines of the arms,
with a less developed cuticular sheath and fewer pedicellarise.

The spines attached to the dorsal skin of the disc are however rather different.
These are quite short, situated vertically, all about of the same length and close together,
whereby the whole of the dorsal side acquires, as it were, a hairy appearance, (see Tab. I,
fig. 4, 6). They are of cylindrical form with a bluntly rounded extremity. This form is
however chiefly attributable to the thick cuticular sheath in which they are enveloped. On
applying a solution of potass, which renders the exterior cuticle transparent, the interior
proper calcareous spine (see fig. 8, 9 b) will appear attached by a somewhat enlarged base
to one of the calcareous pieces (tig. 9 a) imbedded in the interior stratum of the skin. It is
finely reticulated, narrowest in the middle, while the extremity is somewhat enlarged, termi-
nating in a greater or less number of fine points (fig. 9, 10, 11).

21

y. The Pe die ell arise.

These organs, which are very characteristic of the proper star-fishes, and which are
entirely wanting in the brittle stars or Ophiurre, appear in the Brisinga in the greatest
abundance, as well on the disc as on the arms. They are observed in quite enormous masses
in the cuticular sheath of all the furrow-spines and marginal spines, giving to it, as it were, a
granulated appearance ; likewise in extraordinary abundance on the soft cuticular ridges which
surround the dorsal side of the arms in their whole length; they are observed in much
smaller numbers also, distributed over the remaining part of the skin, and on and between
the dorsal spines of the disc, (see Tab. 1, fig. 8 a).

They are in the present species of Brisinga extremely small, scarcely more than
Viomm Jonc» an(l exhibit a remarkably complicated structure. Of all the known forms of
pedicellarise they seem in their structure to resemble most those described by my Father
in the genus Pedicellaster; and, like them, belong to the principal form designated by
Joh. Miiller with the denomination Pedicellariae forcipatse, characterised by 2 pincer-like
branches provided with teeth that fit to each other and enabled by means of a peculiar
muscular apparatus to move towards each other, and thereby to grasp and to hold small
objects.

On each pedicellaria there may be distinguished (see Tab. IV, fig. 23 & 24) a thicker
exterior part or capitulum, and a much thinner flexible stem whereby the pedicellaria is
attached to the skin. The exterior enlarged part which represents the proper body of the
pedicellaria, exhibits usually an oval form with the greatest thickness at the base, and the
extremity laterally somewhat compressed. On closer examination this part appears to con-
sist of 2 broad more or less gaping lobes, each of which envelopes the extremity of one of
the jaws of the forceps; the dentated edges here advancing freely from the fleshy envelope,
so as to work directly against each other. If a diluted solution of potass be applied to^
such an isolated pedicellaria, whereby the exterior cuticular sheath will be rendered trans-
parent, the calcareous pieces which form the foundation of the pedicellaria will be seen very
distinctly inside the sheath in the upper enlarged part or capitulum; these calcareous pieces
(see fig. 25—29) are always 3 in number, namely 2 uniform side-pieces (a) which represent
the 2 movable jaws, and a 3rd unpaired calcareous piece (c), which is inserted between them
and serves both as articulation for the side pieces and attachment for the muscles (m) that
move them. All 3 calcareous pieces are hyaline and more or less perforated with small
circular or oval apertures. The side pieces (a) are thickest in the middle, and have in that
part on the inside a vertical dentated edge, which, when the jaws of the forceps are bent
towards each other, meets the corresponding edge on the other side-piece. On the pedi-
cellariae of the arms (fig. 27 — 29) this edge is quite short and evenly rounded, while on the
pedicellaria; of the disc (fig. 25, 26) it is considerably longer, quite straight, and projecting
at the anterior extremity in a short lobe. Immediately below this dentated edge, the side

22

piece is attached to the middle piece (c) so as to form with it a sort of hinge, which allows
of a very considerable mobility in a fixed direction. The side-pieces are continued forward
into the forceps, which exhibits a strong curvature inwards and has a kind of short cylindrical
neck, after which it is enlarged at the extremity almost in the form of a gouge. Along the
upper nearly straight truncated and sharp edge and a little way down on each side, the
extremity of the jaw of the forceps is armed with a row of pointed teeth bent inwards, of
which that situated at each corner is largest, (see fig. 30). When both jaws are curved to-
wards each other, (see fig 25 - 27) these teeth fit accurately between each other, and can
thereby hold the smallest objects fast. In like manner the above-mentioned vertical dentated
edges proceeding from the middle, meet in the middle line, while there always remains be-
tween these and the extremity of the forceps a considerable open space. Herein the pedicel-
larise in the Brisinga differ essentially from those of the genus Pedicellaster and Asterias,
where the forceps closes in its whole length. From each side-piece there issues downwards
a large plate-shaped process (b) directed obliquely inwards, yet not lying in the same axis
as the proper forceps, but forming with it an obtuse angle (see fig. 26). At the base this
process in somewhat instricted, and distinctly marked out from the rest of the side-piece
by a small sinus in the exterior edge, in which one extremity of the middle piece (c) fits.
Inside of the exterior convex border there are a certain number of larger and smaller oval
apertures which have a somewhat constant arrangement on all the pedicellarise. At the en-
larged plate-like extremity of this process there is attached an evident muscle (fig. 26 m)
the other end of which is inserted in the opposite similarly plate-formed part of the middle
piece (c). By the contraction of this muscle the jaws of the forceps are moved inwards
and towards each other, turning round the extremities of the middle piece as on a hinge.
When the muscle is relaxed, the jaws separate again from each other, whereby the side-
pieces are brought nearly into a horisontal position (see fig. 28). The middle piece (c) in-
serted between the two side-pieces and serving as an articulation for them, is much smaller;
it is situated transversally, and exhibits on the under-side 2 plate-shaped enlargements sepa-
rated by a deep sinus in the middle, and like the lower process of the side-pieces, perforated
with small apertures, which are usually arranged in a double row along the lower curved
edge. These plate-like enlargements, which serve for the insertion of the muscles that move
the jaws of the forceps, do not however lie exactly in the same plane, but have in relation
to the axis of the middle-piece a somewhat oblique position. Viewed from above (fig. 29 (
this piece exhibits therefore a narrow nearly cross-like shape with its greatest width in the
middle, and the rapidly tapering ends, curved each to its side, partially embracing the side-
pieces in the middle. From the upper side there issue 2 short obtuse processes directed
obliquely outwards and leaning on each side against the inner vertical dentated edge; they
partially cover each other, when the middle piece is viewed from the broad side, and together
appear like an elevation in the middle of the upper edge, (see fig. 25, 27).

23

3. The water system.

The water system in the Brisinga exhibits in the main the same normal features as
that of the proper star-fish. We have here to distinguish 1) the ambulacral vessels, and the
exterior parts which stand in connexion with the same ; 2) the water-feet with their ampollse,
and 3) the madreporic body with the stone-canal.

a. The ambulacral vessels.

The main trunks of the ambulacral vessels consist of the circular vessel belonging
to the disc and of the radial vessels issuing from it into the arms. The circular vessel
(Tab. II, fig. 12 f) has its place in the circular groove or semi-canal which is situated on
the skeleton of the disc (the bucal ring) round the interior wall and bounded below by the
thickened edge whereto the bucal skin is attached. The periosteum-like tendinous membrane
covering the ambulacral skeleton goes uninterrupted over this groove; so that the circular
vessel thus becomes enveloped in a completely closed canal. On the side of this- canal which
is turned towards the perivisceral cavity, there appear a number of peculiar small globules
(fig. 11, 12 d) which immediately attract the eye by their yellowish color. They are usually
attached in pairs (more rarely 3 together) close below all the wedge-plates, and have the
form of small thin-skinned vesicles (fig. 14) containing in their interior a finely granulated
matter (fig. 13). Of the nature of these globules, which also occur in the same place in
other Asteridoe, I can not state anything positively. According to their structure they might
appear to form a sort of secretive apparatus. Of the arapolla-like enlargements of the an-
nular vessel (the so-called Polish bladders) so strongly developed in other star-fishes, there
is indeed no trace to be seen in the Brisinga; but 1 have reason to presume that they
nevertheless really exist, and that the before mentioned rather wide cavities in the interior
of the bucal ring, bounded by the parietal plate and the wedge-plate, are destined to receive
such reservoirs of water for the circular vessel.

The radial ambulacral vessels, the number of which corresponds to that of the arms,
issue from the circular vessel at right angles; traverse the groove wherein this vessel lies,
and then extend along the ventral side of the disc and the arms, close in to the ambulacral
skeleton, occupying a corresponding narrow groove, which runs along the whole arm at the
bottom of the ventral furrow, and is partially covered over with a tendinous ligament. On
account of its concealed position this vessel is very difficult to extract. It appears however
very distinctly in all the sections of the arm (see Tab. Ill, fig. 3) as also on the exterior

24

side of the isolated disc, where the arms have been attached (fig. 1 b). From these radial
vessels there issue on each side, at short intervals, lateral canals, which supply the paired
water-feet and their ampollse with water.

b. The water-feet and their ampollae.

The water-feet or ambulacral tentacles, which, as in other star-fishes, represent the
most important locomotive organs of the animal, are very strongly developed, and are in
structure and arrangement quite similar to those of other star-fishes. They issue in a similar
manner from the bottom of the deep ventral furrows which run along the ventral side of
the arms and the disc; and they are here arranged in a double row. At the base of each
of them there is on the inside a large membranous bladder or ampolla protruding freely
through the so-called ambulacral pores into the interior cavity of the arms, and forming
along the upper side of the ambulacral skeleton a double row (see Tab. Ill, fig. 3 c, fig. 4
and 17). These ampollse represent contractile reservoirs of water, by the help of which the
water-feet can at any time be made to swell or to contract. Similar ampollre are also ob-
served projecting from the ambulacral pores (fig. 1 c) which correspond with the water-feet
of the disc. When fully extended, the water-feet in the Brisinga are 2 — 3 times longer than
the arm is broad, and they are curved and twisted in various manners (see Tab. I, fig. 12 — 14).
Even in spirit-specimens they will always be found extended far beyond the ventral furrows,
whence it seems evident that the water-feet in the Brisinga are not so highly contractile as
those of other star-fishes. In form they are (see fig. 22) cylindro-conical or broadest at the
base, tapering regularly towards the extremity, where they have a muscular circular suction,
disc. In the interior, these water-feet exhibit as usual a wide cavity filled with water; their
exterior skin is interwoven with numerous close muscular fibres longitudinal as well as cir-
cular; but there is no trace of calcareous particles in the terminal disc, nor in any other
part of the skin. The before mentioned double rows of water-feet, which extend along the
whole ventral side of the arms, are also continued (see Tab. I, fig. 2) immediately on to the
underside of the disc itself; so that there are for each arm 4 water-feet which specially be-
long to the disc. In a 10-armed specimen therefore the number of the water-feet belonging
to the disc will be 40.

c. The madreporic body and the stone-canal.

The Madreporic body is in the Brisinga, as in the other star-fishes, situated on the
dorsal side of the disc in the interval between 2 arms at their issue. It is however here,
by reason of the small breadth of the skeleton of the disc, moved out quite towards the
edge, where it covers the tubercular projecting upper end of one of the wedge -plates
(see Tab. 1, fig. 3, 4, 6 Tab. II, fig. 1, 2, 8, 11, 12 a). It is in full grown specimens rather

25

large, strongly convex, and of a more or less circular form. The surface is as usual very
uneven, having numerous irregular narrow raised ribs, which radiate from a medial depres-
sion; but there is here in our present species not any trace of spines as in the Brisinga
endecacnemos. From the madreporic body, the so-called stone canal (Tab. II, fig. 12 e)
extends somewhat obliquely downwards along the interior wall of the bucal ring connecting
itself with the circular ambulacral vessel (f). This so-called stone-canal is rather thin, but
of a very firm consistency, on account of the numerous porous calcareous particles contained
in it; its upper extremity is in great part covered by the wide cuticular sheath (c) which
projects into the cavity of the disc and serves to envelope a very problematical organ, the
so-called Heart, while its lower extremity lies quite free on one side of the sheath, and con-
nects itself with the circular ambulacral vessel, after forming a small S-like curve. The
water sucked in by the madreporic plate is still further filtered through the stone canal,
and then conducted over to the annular vessel to be conveyed thence further into the radial
water-vessels which run along the bottom of the ambulacral furrows of the arms.

4. The muscular System.

We have already several times had occasion to mention this system. In the Bri-
singa there are a very great number of muscles, but they are mostly very small, and of an
extremely simple structure. Not only are all the single vertebrae or joints in the ambulacral
skeleton of the arms movably connected together by short muscles going from one to the
other in the periphery of the articulating surfaces of the single calcareous plates, but even
every one of the spines situated along the sides of the arms and on the underside of the
disc, is provided with such short muscles which move the spine in different directions; nay
every single one of the microscopic pedicellarise, which in immense numbers cover the cu-
ticular sheaths of all the spines and the soft transversal ridges on the dorsal side of the
arms, has, as already noticed, its special muscular apparatus. All these muscles seem
however, as before stated, to be of extremely simple structure, and should properly be re-
garded merely as contractile ligaments, the smooth fibres of which are all parallel. But the
muscular system seems to be more developed in the water-feet, and especially in the
extremely contractile bucal membrane. Here we find several distinct layers of muscles,
the fibres of which cross each other in different directions, and thereby bring about a more
complete mobility of the said parts.

26

5. The nervous System.

In the nervous system we find again the same radial arrangement as has been al-
ready mentioned in reference to the ambulacral system. In the main features it corresponds
completely with what we know in other Asteridse. The chief parts, or the central parts, are
formed by the ambulacral nerves belonging to the arms, or as we may call them the ambu-
lacral brains. They have each (Tab. VI, fig. 4, 5) the form of a narrow band lying quite
superficially, and extending along the middle of the ambulacral furrow in the whole length
of the arm. In the middle this band forms a distinctly elevated ridge, arching over a narrow
longitudinal canal or sinus which is divided by a medial septum (a) in two compartments.
During its course it sends out on each side fine converging nerves to the water-feet, and in
the spaces between the latter it expands on each side, insensibly merging in the cuti-
cular lining. It terminates at the point of the arm in a somewhat claviform enlargement
(fig. 12 b) which lies at the base of a peculiar organ of sense herein-after more particularly
described. The whole nervous band is of a light yellowish color and is particularly brittle
and fragile, lying so closely enveloped in the thin membrane which lines the ambulacral
furrow, that it can only with extreme difficulty be dissected out and isolated from the adja-
cent parts. On removing the water-feet however, it may be distinctly perceived along the
whole length of the ambulacral furrow. These nervous bands belonging to the arms are
continued immediately inwards on the underside of the disc, where they likewise occupy
the middle of the ambulacral furrow, and when the water-feet are removed, appear very
distinctly (Tab. VI, fig. 1 b). At the interior extremity of the ambulacral furrows, all
the ambulacral nerves are connected by a somewhat broader band-like circular com-
missure (ibid a) situated close in against the bucal ring at its junction with the bucal
membrane. This commissure lies, like the ambulacral nervous band itself, quite superficially
extending along the lower side of the circular rim which projects from the inner wall of the
bucal ring, and it is, only at the end of the interradial spaces, partially covered over by
the edges of the inner adambulacral plates with their diverging spines. Like the ambulacral
nerves this commissure also seems to arch over a narrow circular sinus, which stands in con-
nection with the above-mentioned radial canals running along the ventral furrows of the arms.

6. Organs of sense.

Under this category we must class a peculiar organ which is situated at the extreme
point of all the arms. The organ exists there, as well in the previously observed form B.

27

endecacnemos as in the present species. That it has not been noticed before in the Bri-
singa, is owing to the extreme point of the arm being so easily broken off when the animal
is taken. In a perfect arm the organ is always present, and easily distinguishable; the
skeleton of the arm forming here a peculiar enlargement, covered with spines, and in the
form of a somewhat bent plate, which is destined precisely to protect the organ in question.
It is well known that in many star-fish there is at this place a very conspicuous red pig-
mentary spot, in which there has been noticed a peculiar structure of the extremities of the
nerves completely answering to the so-called chrystal cones in the compound eyes of Articu-
lata, for which reason these pigmentary spots have been confidently designated as real appa-
ratus of vision. It was therefore reasonable to anticipate that we should find the organ
under consideration in the Brisinga likewise adapted to serve in some manner as an organ
of sight. Closer inspection has however not confirmed this presumption. The 2 essential
parts which characterise the apparatus of sight in other AsteridaB, namely pigment and
chrystal cones, are wanting. But it corresponds in its structure more with an other far
less conspicuous part, situated close above the organ of sight, in the form of a simple ten-
tacle, which in appearance is very little different from one of the ordinary water-feet,
excepting that it is unpaired. In the Brisinga the organ in question (see Tab. II, fig. 5, 6 c,
fig. 7 a) has the form of a soft cylindrical plug evenly rounded at the extremity, and of
relatively considerable size, many times larger than the nearest very rudimentary water-feet;
like these it is directed downwards; and it is sheltered as under a vaulted arch, by the be-
fore mentioned peculiar calcareous plate (see fig. 5) which is thickly covered with pedicel-
lariax In its structure it seems on the whole to correspond pretty nearly with the water-
feet. Like them it is hollow in the interior, and is filled with water from the ambulacral
vessel. Its walls are rather thick, but seem to want the layers of muscle peculiar to the
water-feet. At the extremity, it is surrounded by a thin transparent membrane a little raised
from the proper wall, and here, a fine radial striation is observed. The structure of this
organ is, as may be seen, very simple, and it might easily be taken for an unpaired water-
foot, simply closing at this place the double row of these organs which runs along the ven-
tral furrows of the arms, if its unusual size, its peculiar protection, and the evident enlarge-
ment of the ambulacral nerve at its base, did not indicate an entirely special function. It is
for this reason that I must presume it to be an organ of sense. But what sort of sensation
is transmitted through it, is difficult to say. It can scarcely be a pure and simple feeler, a
mere tentacle. In this capacity the numerous water-feet far-reaching and movable in all
directions might naturally be presumed to suffice; while on the other hand the isolated
position, slight mobility, and peculiar protection of the organ here considered, would evi-
dently make it less well adapted for such a function. Probably the animal receives through
it a more specific conception of its surroundings than it receives by means of immediate
contact; and we might perhaps rather be justified in regarding it as a sort of organ of smell
analogous to the so-called olfactory papillae in the Arthropods.

4*

7. The digestive System.

The digestive system in the Brisinga exhibits in its general structure the typical
organisation of the proper star-fishes. We have to distinguish 1) The bucal aperture with its
muscular apparatus, 2) the digestive cavity or stomach and 3) the radial caeca. We shall
consider each of these principal parts separately.

a. The bucal aperture with its muscular apparatus (the bucal membrane).

In the middle of the under-side of the disc in the Brisinga, there appears a more
or less wide or gaping circular opening, which is the mouth (Tab. I, fig. 2, 5 f, Tab. II, fig.
11 c). It is surrounded and limited by a naked membrane attached along the inner peri-
phery of the skeleton of the disc (the bucal ring) to the circular projecting edge which is
formed by the lower parts of the parietal plates. This bucal aperture is of a very variable
appearance. It is sometimes quite small and narrow, only occupying the centre of the under-
side of the disc, and far removed from the interior border of the skeleton; sometimes it is
so much enlarged that it occupies nearly all the ventral space within the bucal ring, and
thereby becomes apparently bounded only by the skeleton of the disc itself. Its form is
also rather variable; as it is sometimes more irregularly shaped (see Tab. II, fig. 11) or
drawn out in one direction or another, to an oval or elliptical shape (see Tab. 1, fig. 2).
This variability in size and form of the bucal aperture, depends on the great contractility
of the membrane which limits it (the bucal membrane). On closer examination it will also
be found that this membrane (Tab. II, fig. lib) is interwoven by a complicated system of
fine muscular fibres which have partly a radial and partly a circular arrangement. The radial
fibres diverge from the border of the mouth, which often exhibits an extremely fine crenula-
tion (see Tab. II, fig. 11) and attach themselves round about to the skeleton of the
disc. By the action of these fibres, the bucal membrane can be contracted from all sides;
the consequence of which will be, that the bucal aperture will be enlarged. The circular
fibres act antagonistically; as they contribute together, like a sphincter, to restrict the bucal
aperture and thus extend the bucal membrane. Also in other star-fishes, there is a similar
bucal membrane; but it is so little developed that the bucal aperture seems to be limited
directly by the intruding angles of the skeleton of the disc with the spines on the same;
these spines are often of a peculiar form, and have therefore been distinguished by the
appellation bucal spines; they may indeed also act partly, as a sort of masticatory apparatus.
If we imagine in the Brisinga the bucal membrane in its highest degree of contraction,

29

whereby consequently the bucal aperture will occupy the whole of the ventral space within
the bucal ring; then the ends of the interradial spaces formed by the adambulacral plates
of the interior vertebrae will make the limits of the mouth; and the spines diverging inwards
from these plates in fan-like fascicles may also partly be able to act as bucal spines; for
which reason we have also thus named them in the foregoing.

b. The digestive cavity.

Through the more or less gaping bucal aperture in the Brisinga, there is an uninter-
rupted view into the digestive cavity (Tab. 1, tig. 2 & 5) which occupies the greatest part
of the cavity of the disc. Its walls form below the immediate continuation of the bucal
membrane, as the latter (see Tab. II, fig. 9 b) suddenly curves itself round outwards at the
border of the bucal aperture, acquiring at the same time a different quality and structure.
There is thus no proper section which might be compared to the aesophagus; and immedia-
tely within the border of the mouth the digestive cavity or stomach begins. It has the form
of a flat and wide sack, the walls of which are tolerably thick, and consist of 2 distinct
layers; an outer fibrous, and an inner cellular, which latter forms numerous irregular folds
and sinuosities projecting inwards (see fig. 9). The exterior convex surface of the stomach,
which likewise exhibits a more or less well defined longitudinal folding, is attached to the
bucal ring by numerous very strong fine tendinous fibres (fig. 9 a). These tendinous fibres,
which form an immediate continuation of the tendinous membrane that covers the skeleton
of the disc, issue (see fig. 11 c) close together from the upper edge of the latter, just before
the insertion of the arms; that is from the dorsal side of the ambulacra! plates of the in-
terior vertebrae; they extend themselves horisontally inwards, diverging from each other, and
attach themselves at the same height, round the exterior wall of the stomach.

The stomach is moreover fastened in the middle to the dorsal skin of the disc, by
a central ligament which will be hereafter described. By this ligament, and by the tendi-
nous fibres before mentioned, the stomach is thus kept in its place and, as it were, suspended
in the central cavity which it fills almost entirely; only leaving below a very small space,
like a canal surrounding the lower part of the stomach, limited below by the upper surface
of the bucal membrane, and on the outside by the interior wall of the bucal ring. This
space, thus representing the lower part of the perivisceral cavity, stands round the peri-
phery in connexion with a similar narrow upper space, which above the radial spaces of the
bucal ring, communicates again with the cavities of the arms.

The upper part of the stomach is separated, by a distinctly marked annular fold,
from the lower proper digestive cavity, with which it nevertheless communicates by a wide
circular aperture. From the periphery of this narrow upper section of the alimentary cavity
there radiate (see Tab. II, fig. 8 c and fig. 10) on all sides the strongly developed so-called
radial caeca, hereafter more particularly described, extending far into the cavities of the arms.

30

The upper wall of this section, which thus forms properly the bottom of the digestive cavity,
and can also be seen more or less completely through the gaping bucal aperture (see Tab. 1,
fig. 2, 5 f) has the form of a flat extended circular disc of half the diameter of the proper
stomach. Its structure seems to be very peculiar. On the interior or lower side turned to-
wards the digestive cavity, it exhibits (see Tab. II, fig. 10) an appearance of being, as it
were, granulated, having on its surface numerous small wart-like elevations or sinuosities
of a glandular structure, which, especially towards the centre, are crowded together and
projecting, and in this part frequently arranged in larger or smaller groups separated by
more or less distinct sinuous furrows; towards the periphery these elevations become more
indistinct, and disappear at last entirely where the radial caeca take their issue. The upper
side (see Tab. II, fig. 8 c) which is in great part covered by a glandular apparatus (f) at-
tached here in the centre, (hereafter more particularly noticed) seems to exhibit an entirely
different structure. We remark here a very conspicuous and peculiar striation, which on
closer examination appears like a number of raised ribs radiating from the centre to the
periphery and divided by intermediary furrows into a great many finer and finer ramifications.
The whole has at the first glance a striking resemblance to a ramified complicated system
of canals in the walls of the digestive cavity; and a corresponding structure in other star-
fishes seems also to have been considered by an earlier naturalist (Ti4emann) as a part of
the supposed blood-vessel system (stomach veins). Nevertheless on more minute investi-
gation, and especially by examining the transverse section of this part, the conviction is
soon acquired that here is no real system of canals, but only a peculiar radial corrugation
of the roof of the digestive cavity itself, which only appears distinctly on the upper side of
the same ; while on the lower side it is completely hidden by the numerous wart-like papillae
projecting in the digestive cavity.

I have not been able to find any evident central aperture in the upper wall. The
whole digestive cavity appears to be quite closed, without any part being discoverable that
can properly be called an intestine. There may indeed be remarked on the outside of the
dorsal skin of the disc a fine opening or pore (Tab. 1, fig. 4, 7 a) which according to its
position might easily be taken for an anal aperture; but I consider it certain that this is
not the case. It is, as will be shewn in the sequel, only the exterior outlet for a strongly
developed glandular organ which has its place in the middle between the stomach and the
dorsal skin, and which answers to the so-called interradial caeca in other star-fishes.

c. The radial case a.

These organs, which are peculiar to the proper star-fish and entirely wanting in the
Ophiurse, and which must be considered as special sections of the digestive cavity destined
for the individual arms, are also very distinctly developed in the Brisinga. Even if they do
not, as in other star-fishes, extend right into the points of the arms, they are yet in pro-

31

portion to the central part of the digestive cavity situated in the disc, at least quite as
strongly developed as in the others, extending about to the end of the 1st third part of the
arms, and are thus more than 4 times as long as the diameter of the disc (see Tab. VI,
fig. 35). They take their issue as single rather wide and somewhat flattened tubes in the
periphery of the upper section of the stomach (see Tab. II, fig. 8 c c, fig. 10) and radiate thence
on all sides, extending horisontally over the lower folded section of the stomach and over
the ambulacral plates of the disc into the cavities of the arms. Immediately on entering
into the arms, they divide themselves fork-like into 2 main trunks (see Tab. II, fig. 8 c on
the right; Tab. Ill, fig. 25 a) which then run side by side in a tolerably straight direction
along the dorsal side of the arms. On each side, these main trunks send out short and
closely placed thin-skinned lobes or secondary caeca, variously folded and contorted, giving
to the whole organ an elegant ramified appearance. The radial caeca, the number of which
will thus correspond with that of the arms, are firmly attached to the inside of the dorsal
skin in their whole extent by 2 longitudinal parallel ligaments. At a short distance from
the issue of the radial caeca from the stomach, each of these ligaments is connected with
the adjacent one by a thin arched commissure which is likewise firmly attached to the
dorsal skin. Then they separate themselves from this skin, and follow the radial caeca to
their issue from the stomach, going over into the upper wall of the latter, after having
again connected themselves with their neighbors by a thin membrane, which on the out-
side is deeply concave. The stomach is hereby as it were suspended from the dorsal skin
by a sort of rather complicated mesentery pierced at the angles between the issues of the
radial caeca, by large semilunar apertures, through which the central space between the roof
of the digestive cavity and the dorsal skin communicates with the other perivisceral cavity
(see Tab. VI, fig. 6).

The fluid contained in the cavities of the radial caeca is of a more or less clear yel-
lowish color, and contains a considerable quantity of fat or oil which collects itself in larger
or smaller bubbles on the surface of the water, when the caeca are opened and the contents
allowed to flow out.

8. The perivisceral cavity (caeloma) and the Blood system.

The cavity, immediately limited by the skin and the interior skeleton, which contains
the various viscera, is in the Brisiuga well developed; and it is not, as in the Ophiurae, only
restricted to the disc, but extends also, as in the proper star-fishes, into the arms and even
to their extreme points. In the basal section of the arms, it is (see Tab. Ill, fig. '3) very
wide; so that besides the radial caeca (d d) it can here contain also the frequently very
strongly developed organs of generation (e e). In the periphery of the disc, the cavities of
the arms are by a proportionally very narrow aperture connected with the cavity of the

32

disc, which uninterruptedly surrounds the central part of the digestive cavity or stomach,
without being here, as in other star-fishes, divided by radial septa into several distinct spaces.
The whole perivisceral cavity is, as already mentioned, filled with an aquseous fluid which
thus immediately surrounds the interior organs. This fluid is however, as has also been
shewn in the case of other star-fishes, not purely and simply water, but contains also nume-
rous small cellular elements freely suspended in the same. According to their whole nature,
these cellular elements are real blood globules; and the said fluid must therefore, even if
mixed with considerable quantities of water taken in from without, be considered as real
blood. The whole perivisceral cavity represents therefore in the Brisinga -- and as I must
presume also in the other star-fishes, one single great reservoir of blood.

I have not succeeded, even with the utmost attention, in discovering any actual
blood vessels in the Brisinga; not to speak of the complicated system of veins and arte-
ries which, according to the indications of earlier naturalists, is said to be found in some
star-fishes. I have endeavored, by repeated dissections and preparations, to discover in the
Brisinga the widely ramified dorsal or venous system of blood vessels so minutely described
by Tidemann, but always without any result. Whatever may at first have appeared to re-
semble such a system has always, when more closely examined, shewn itself to be some-
thing quite different; namely sometimes the ligaments and tendinous fibres whereby the in-
terior organs, and especially those of the digestive system, are attached to the dorsal skin,
sometimes a peculiar folding in the walls of these organs. Neither have I been able to
confirm the observations of the earlier naturalists with respect to the ventral or arterial
system of blood vessels. Real blood vessels are indeed not to be discovered here, but only
a narrow longitudinal canal-like sinus, without any special walls, extending along the ventral
furrows of the arms, immediately above the radial nerve-bands, and communicating with a
similar narrow circular sinus enclosed between the 2 lamellae of the bucal cuticle nearest to
the bucal ring. My investigations of other native star-fishes have given the same negative
result. Real blood vessels are no where to be found; according to which I must suppose
that the earlier representations of the blood system of star-fish are inaccurate and founded
on a less careful examination. There is however one part that has been likewise referred
by earlier naturalists to the blood-vessel-system, and which is very distinctly developed and
strongly prominent in the Brisinga, that is namely the peculiar membranous organ which in
other star-fishes extends from the ventral to the dorsal wall of the disc accompanying the
stone canal in its whole course, which is usually regarded as the central organ of circu-
lation or heart, whereby the supposed ventral (arterial) and the dorsal (venous) circular
vessels are said to be connected. This organ lies, as we know, in other star-fishes together
with the stone-canal enclosed between the double lamellae of one of the radial septa which
divide the cseloma of the disc like a fan. As there are no such radial septa in the Brisinga,
the organ is here enclosed and kept in position by a sheath-like duplicature of the tendi-
nous membrane which lines the inside of the skeleton of the disc, and thereby firmly at-

33

tached to the interior wall of the bucal ring. When the stomach is removed, the said sheath
appears (Tab. II, fig. 11, 12 c) like a short and thick pillar projecting convexly inwards and
extending vertically along the interior wall of the bucal ring, connecting the bucal membrane
with the dorsal cuticle of the disc at the place where the madreporic body (a) is situ-
ated; it also partly incloses the upper end of the stone-canal (e). On cutting through the
exterior tendinous skin, the membranous organ in question (comp. Tab. VI, fig. 6 h & fig. 8)
will immediately be discovered partly attached on the upper side by a ligament to the
stone-canal (s), and with its upper end attached to the dorsal skin at the same height as
the stone-canal, but somewhat inside or on the inner edge of the madreporic body. Its
other extremity extends considerably further downwards than the stone-canal, curving itself
round the thickened edge of the bucal ring, which below bounds the half- canal or duct
wherein the circular ambulacral vessel is situated. The organ is here attached by a broad
base to the skeleton of the disc. With regard to the more particular structure of the organ,
it appears (fig. 8) like a membranous tube variously folded and contorted; its walls exhibit
under the microscope (see fig. 9) a fine net-like structure; thickened and thin places alter-
nating with each other. There may be plainly perceived in them light longitudinal fascicles
of fibres, partly anastomosing with each other and surrounded by a more opaque fine
granular tissue. In spirit specimens, the walls of the tube will usually be found collapsed
or in contact with each other; but in fresh specimens, this does not seem to be the case.
In the interior of it is found a fluid, which appears to be of the same nature as that con-
tained in the perivisceral cavity. The connexion of the upper part of the tube with the
dorsal skin might justify the assumption that it here terminates blindly. I have however
reason to suppose that there may yet be found here a sort of communication with the
perivisceral cavity; although I have not yet succeeded in indicating any such communication
with certainty. The lower end of the tube -lies close to" the before mentioned circular
sinus, which is inclosed between the 2 lamellae of the bucal cuticle; and the tube stands
here, as I believe, in free communication with the sinus.

Having as already mentioned, not been able to demonstrate any real ventral or
dorsal system of blood-vessels, nor to find in the Brisinga any special organs of respiration
which could make such a distinction of arteries and veins probable, I cannot attribute to
this apparatus the importance which has been ascribed to it. In any case it appears to me
that many circumstances forbid the assumption that we have here the real central organ
for the circulation of the blood. Of the probable distination of the organ I shall say some-
thing more in a subsequent section (Physiology).

CAI

34

9. Organs of secretion.

Under this category may be reckoned a rather voluminous and according to its whole
structure exquisite glandular apparatus (Tab. II, fig. 8fg) which has its place in the middle,
immediately above the roof of the stomach between it and the dorsal skin of the disc. The
apparatus, which evidently enough corresponds according to its situation to the so-called
interradial caeca or rectal caeca in other star-fishes, consists of 2 unequal parts united in the
middle, of which one is usually twice as large as the other. Their arrangement is always
constant, so that the smaller section turns to that side of the periphery where the madre-
poric body is situated, but their size varies in different individuals. In the middle,, where
both sections unite, the whole apparatus is attached by a ligament, as well to the roof of
the stomach as to the inside of the dorsal cuticle, on which the ligament is continued in
the form of 2 thickened stripes proceeding from the centre in different directions, and then
both curving towards one side of the periphery, uniting themselves each with one of the
longitudinal ligaments by which the 2 radial caeca situated nearest on the right of the madre-
poric body are attached (see Tab. VI, fig. 6). The central part of the ligament is perfo-
rated by a narrow canal, the excretory passage, which ascends and has its issue on the
dorsal side of the disc in a fine pore surrounded by a somewhat raised elliptical border
(Tab. I, fig. 1 a). Although the point where the ligament is attached to the inside of the
dorsal skin lies exactly in the centre, it will always be found that the said pore has an
evidently excentric position, being always found nearer to that side of the periphery where
the madreporic body has its place (see Tab. I, "fig. 3, 4 a, Tab. II, fig. 1, 2). This is caused
by the excretory passage perforating the thick dorsal skin in a very oblique direction, which
also quite naturally accounts for the peculiar elliptical form of this pore (Tab. I, fig. 7 a).

With respect to the structure of the apparatus in question, it seems to be, as be-
fore mentioned, completely glandulous. Both the main parts which compose the apparatus
consist of a rather firm and compact mass externally divided into a greater or less number
of large irregular lobes, and each of these lobes again into a great number of smaller
rounded lobules, whereby the whole apparatus acquires the appearance of a real acinous gland.
On the outside there is a thin but tolerably resistent membrane, wherein there may be seen,
more or less distinctly, various fibres that partly cross each other (muscular fibres), and in-
side an opaque white granular substance, which on the side where both sections meet
appears more loose and transparent. The whole apparatus rests immediately on the roof of
the stomach, but so that a greater or smaller part of the latter always remains uncovered
(see Tab. II, tig. 8).

35

10. Organs of Generation.

In full grown specimens we often find, as before mentioned, that the arms at some
distance from the base exhibit a very remarkably strong enlargement; the dorsal skin rising
high" above the subjacent ambulacral skeleton, while it otherwise lies more or less closely
on the same (Tab. I, fig. 12, Tab. II, fig. 1). On slitting the skin it will be found that
the cavity of the arms (coelom) is here stretched by a very voluminous mass divided into
numerous lobes and branches, which in different individuals exhibit a somewhat different appear-
-ance. This mass represents the organs of generation, which in some individuals appear as
ovaries and in others as spermaries. Both sorts of organs of generation are easily distin-
guished from each other., partly by the color and partly by the manner of ramification; but
else they occupy the same place in the Cavity of the arms, without ever extending into the
coelom of the disc. By spreading on each side the dorsal skin when slit in the middle,
and pushing asunder the various lobes or ramifications .from each other (see Tab. Ill,
fig. 4, 17) it is easy to ascertain that what appeared at first glance to be a confused mass
of lobes, belongs to 2 organs, lying symmetrically on each side of the medial line, each of
which is attached only by one point to the interior of the skin. On examining the skin at
this place more closely, an evident exterior opening will be discovered (see Tab. I, fig. 12 a,
Tab. Ill, fig. 25 c c) surrounded by tuberous projecting borders, which is the issue for the
sexual products. These exterior apertures for generation are usually situated quite sym-
metrically, one on each side of the arm nearer to the lower side, and distant about one
diameter of the disc from the base of the arm. When once their position has been noticed,
it will never afterwards be difficult to find them on any arm, without proceeding to any dis-
section.

a. The Ovaries.

The ovaries have usually (see Tab. Ill, fig. 4) the form of more or less elongated
and ramified cylindrical tubes which at last unite themselves on each side to a common
short trunk attached to the inside of the skin and containing the short exit passage (oviduct)
through which the mature ova are ejected from the body. Sometimes several such ovarian
tubes are blended together in the form of large irregular sack-like enlargements, whereby
the whole ovary acquires a more crowded and compact appearance (see fig. 5, 6). These
enlargements may then either begin with ovarian tubes in the usual form (tig. 7) or they
may exhibit only small rounded lobes (fig. 8) or even only form simply rounded reservoirs
(see fig. 5). Even in the smallest specimens I have been able to examine, I have found

5*

36

evident incipient ovaries in the usual place. Notwithstanding their small size, they exhibited
already distinctly their characteristic form and ramification (see fig. 9, 10). The fully deve-
loped ovaries are of a more or less intense reddish yellow color, and are already hereby
immediately distinguishable fr,om the spermaries, which are always of a lighter whitish
color. The skin which surrounds the various ovarian tubes is rather firm, fibrous and semi-
transparent; so that the ova-cells inside may be seen through them more or less distinctly
(see fig. 4, 11).

On examining the contents of such a tube, there will be found egg-cells very diffe-
rently developed. The smallest of them (fig. 12, 13), which can only with difficulty be di-
stinguished from the epithelial cells that line the inside of the wall of the tube, are per-
fectly pellucid, often of an irregular form, and without any distinct exterior membrane, but
all furnished with a large and well defined nucleus (germinative vesicle) wherein there may
usually be distinguished a single small nucleolus (germinative point). The larger egg-cells
(fig. 14, 15) appear more isolated in the interior of the tube, and are sometimes quite
detached from the walls. In them there may already be discerned a distinct light enveloping
membrane (corion) and a greater or smaller quantity of reddish yellow yolk deposited in the
cell-plasma, which makes them less transparent. Finally in the largest egg-cells (fig. 16)
which always lie completely free in the interior of the ovarian tubes, this mass of yolk has
increased so much, that the whole egg-cell, with exception of the clear enveloping membrane,
becomes quite untransparent, and the germinative vesicle is usually quite hidden. In this state
the egg seems to have arrived at maturity, and is then carried out by the ovarian ducts.

b. The spermaries (testes).

The spermaries (fig. 17, 18, 19) correspond, in their general form and arrangement on
the whole, to the ovaries; but are easily distinguished from the latter by their lighter whitish
color, and by the single tubes or caeca not being simply cylindrical, but in many ways bent
and lobed, whereby the whole organ exhibits a still more complicated appearance than the
ovaries. In young specimens (see fig. 18, 20) the single cseca may usually be traced in
their whole length, as narrow tubes nearly of uniform thickness, but bent in and out at
short intervals or zig-zagged, and gradually uniting themselves to a certain number of con-
verging main-trunks. These are at last connected in a single short exit, which issues at
the same place as the ovaries in the females (fig. 19 a). In older specimens the single cseca
often form (see fig. 17) as is the case with the ovaries, large sack-like enlargements, which
externally exhibit innumerable small rounded lobes. If one of the extremities of the cseca
is viewed under a microscope, it will be found (see fig. 21) that their interior is, as it were,
divided into a number of irregular spaces, wherein are found numerous small clear elliptic
bodies with a great number of interior nuclei. These bodies (fig. 22) represent the deve-
loping cells of the spermatozoa. In the larger sack-like extensions we find the fully deve-

37

loped spermatozoa, which lie here together in large clusters or heaps (see fig. 23). The
single spermatozoa consist (fig. 24) of two sharply distinguished parts, head and tail. The
head is more or less globular, completely pellucid, and contains one or more clear nuclei.
The so-called tail, which can often attain a very considerable length, has the form of an
extremely fine thread or bristle. I could not discover any independent undulating movement
in them. But there was during the dissection an evident constant movement of the whole
mass of semen, probably caused by the elastic tails or bristles of the spermatozoa gradually
straightening themselves out from their original convolved position.

38

IV.

General arrangement of the various organic systems,

(Topography).

If the various organs described in the foregoing lines are considered in connexion
with each other, we shall find that each arm (antimeron) contains in itself all the most im-
portant organic systems (vital organs). In the disc these organic systems are so connected
that the whole complex of arms forms a common organism. We have thus in fact in the
single arm a complete expression for the whole organisation. The complete star-fish is in
other words to be considered as a multiple of arms or rays arranged around a common centre
and (in the disc) brought into organic connexion with each other. In the disc we find there-
fore chiefly connecting parts (commissures) between the various organic systems of the arms;
while the real principal parts of the organs have their place in the arms themselves. To
enumerate separately the particular organic systems above noticed: the commissure of the
ambulacral skeleton is formed by the so-called oral ring, or, (as the ambulacral vertebra
contained in the same are flush with the skeleton of the arm, and in reality represent the
interior continuation of the same) more correctly speaking, only by the connecting plates
peculiar to the disc viz. the wedge-plates and the parietal plates; the commissure of the
cuticular system is formed by the dorsal skin of the disc; the commissure of the water
system, -by the circular ambulacral vessel; the commissure of the nervous system by the
circular nervous band lying in the periphery of the oral cuticle; the commissure of the
digestive system, by the central stomach, with its single oral aperture; the commissure of
the blood-system, by the cavity of the disc surrounding the stomach and by the circular
blood-sinus contained between the lamellae of the oral cuticle.

The special organic systems of the disc are only the apparatus of secretion and the
so-called ,,heart", if we do not reckon this last as belonging to the blood-system.

The special organs of the arms are the organs of generation and the terminal
organ of sense.

The foundation of each single arm consists again of a series of similar consecutive
joints or metamera, each of which contains its own section of the skeleton, water-vessel,
nervous and muscular systems, with a corresponding pair of external members (Water-feet).

39

The terminal joint of the arm is however distinguished', as stated, from the others by a
remarkable development, and by the absence of water-feet, in stead of which it is furnished
with the above-noticed special organ of sense. This joint is, as will be shewn hereafter,
the oldest of them all; and it is from the basis of this that all the others develop them-
selves little by litle during the growth of the arm. There is a common cuticle for all the
metamera and a common blood-cavity. Of the interior organs contained in the latter, the
digestive system (the radial caeca) is limited to the basal third part of the arm; and the
organs of generation are limited only to a certain section of this last part.

If we now endeavor to picture to ourselves the manner in which these different
organic systems are topographically arranged in relation to each other, we shall obtain the
best view by examining a section of an arm. If we take this in the basal part, and at the
place where the organs of generation are situated, we shall find all the most important
organs of the arm' in the section and in their relative positions, (see Tab. Ill, fig. 3). Above
and on the sides there appears the highly convex dorsal cuticle with its 2 layers; on the
outside, the soft epidermoidal layer with its pedicellariaB ; on the inside, the fibrous layer
with the calcareous ribs contained therein, from which rise the spines ranged in circlets
along the same with their cuticular sheaths formed by the epidermis. Limited above and
on the sides immediately by the cuticle, appears the perivisceral cavity or the blood-cavity
with the interior organs inclosed; highest up and intimately connected with the inside of
the cuticle, the double radial caeca (d d) with their lateral sinuosities ; further down and on
the sides, the well developed organs of generation (e e) which here fill the greater part of the
perivisceral cavity, with their numerous ramified caeca; below, and here limiting the peri-
visceral cavity, appears the ambulacral skeleton (b) with the dorsal ridge (projecting upwards
in the middle into the interior of the arm), on the sides of which ridge the thin-skinned
ampollse (e) for the water-feet project from the ambulacral pores. The ambulacral skeleton
is connected on each side with the dorsal cuticle, and thus closes completely the cavity of
the arm below; the long marginal spines (ff) project externally at this place on each side.
Immediately below the ambulacral skeleton, at the bottom of the ventral furrow formed by
the same, there appears in the middle a narrow opening (see fig. 1 b) which is the section
of the radial ambulacral vessel.; this opening is in the lower part separated by transversal
ligaments from another narrow opening (fig. 1 a) which represents the radial blood-sinus;
this last is again covered by the band-shaped radial nerve, which nerve again is covered by
the thin membrane lining the ventral furrow and going over at the sides immediately into
the tendinous cuticle which clothes the ambulacral skeleton. From the bottom of the ambu-
lacral furrow there project to a greater or less distance the 2 water-feet (fig. 3 a) and from
the adambulacral plates which limit the ventral furrow on the sides, there issue in various
directions the so-called furrow-spines.

By examination of a section of the disc (see "Fab. VI, fig. 36) we may ascertain that
the various connecting parts of the organic systems have on the whole a similar arrange-

40

ment. In the lower part of the interior wall of the oral ring, we shall thus find the sec-
tion of the circular ambulacral vessel, of the circular blood-sinus and of the commissure of
the nerves, arranged precisely in the same manner and in the same serial order. In the
middle of .the disc, we have a longitudinal section of the central part of the digestive appa-
ratus extended freely in the cavity of the disc; in the lower part, the oral aperture (m),
limited by the horisontal naked oral membrane and leading immediately to the lower strongly
corrugated section of the stomach (st), the periphery of which lies close in to the upper
margin of the oral ring and is attached to it by fine tendinous fibres; above this section is
the much narrower upper division of the stomach, which in its periphery is suspended by
ligaments from the dorsal cuticle, and the cavity of which goes over on the sides into the
radial caeca (r). Above the roof of the stomach, there appears finally in the centre the
apparatus of secretion (n) with its short exit-duct issuing on the dorsal side of the disc.
If the section is carried exactly through the interradial space in which the madreporic body
is located, we have at the same time a longitudinal section of the latter (m) and of the two
asymetrical parts standing in connexion with it, the stone-canal and the so-called heart (h).
The oral ring (m r) will then on this side be cut through, so that the wedge plate will be
bisected, and the 2 contiguous dorsal marginal plates, the 2 contiguous parietal plates and
2 pairs of contiguous adambulacral plates, will be separated from their connexion with each
other. Immediately below the cut wedge-plate, the triangular cavity which exists in the in-
terior of the oral ring at this part will appear. If the section is made on the opposite side
of the oral ring precisely in a radial space, the oral ring will here be cut, so that the 2
ambulacral vertebrae corresponding to one arm will be longitudinally bisected; the section
will thus coincide with the vertical sutures between the 2 pairs of ambulacral plates be-
longing to these vertebrae; and at the lower side (the bottom of the ventral furrow) we shall
have a longitudinal section of the radial ambulacral vessel, of the radial blood-sinus and of
the radial nerve. If we now consider an arm here in connexion with the disc, the skeleton
of the latter will also be divided in the same manner; so that the consecutive vertebrae of
the arm will all be bisected in the vertical sutures between the ambulacral plates.

41

V.

The vital functions,

(Physiology).

I. The animal functions.

It is of course very difficult to acquire any complete idea of these vital functions
in the Brisinga; because the animal, by being brought up from the enormous depth at
which it lives, is evidently placed in very abnormal exterior circumstances which must neces-
sarily have a highly obstructive influence on such functions. But as I have been able to
preserve specimens, or rather those parts of which the specimens consist, alive some time
after capture, I have still succeeded in making a few observations on this subject which
may not perhaps be without interest.

A. Movement.

The principal acting parts during the locomotion of the animal are certainly, as in
other star-fishes, the numerous water-feet situated along the ventral furrows, by help of
which the animal creeps slowly along the bottom in different directions. I have not how-
ever been able directly to observe the animal's true locomotion; and that for the very
natural reason, that in all the specimens, as has been said, the arms have been at the time
of capture more or less completely detached from the disc.

However in the arms thus separated one may see, as in other star-fishes, that the
feet move in different manners, now contracting now extending themselves and turning in
various directions, partly also attaching themselves by their suction-discs to the objects with
which they come in contact. This changing play of the water-feet seems however in the
Brisinga to be far from taking place with that liveliness and intensity that we have occasion
to observe in other star-fishes, for instance in the genus Asterias ; while we must not omit to
allow for the abnormal circumstances in which the captured specimens of the Brisinga will
always find themselves. As regards the proper movement of the water-feet, it is effected

6

42

apparently only by help of the muscles imbedded in their walls; not as is usually snpposed
by any independent contraction of the ampollaB belonging to the water-feet These ampollae
are in any case in the Brisinga so extraordinarily thin-skinned, that they scarcely could have
any other destination than to take up the superfluous water as simple reservoirs when the
water-feet are contracted, and again to deliver out the water required wheji the water-feet
are extended.

I have not been able directly to observe any independent movement of the arm-
spines; but according to the frequently very different direction which they, especially the
long marginal spines, shew in the captured specimens, sometimes standing out from the arm
at nearly right angles, sometimes lying close to the sides of the arms, we may conclude that
they really have a voluntary movement; an inference that seems also to be warranted by
their peculiar articulation to the skeleton, and by the muscular skin surrounding their base.

Also the whole arm is to a certain extent movable; all the ambulacral vertebrae
being, as already mentioned, movably connected with each other by elastic muscular liga-
ments. This movement, which is easily observed in the specimens recently captured, can
take place both in a horizontal and in a vertical direction. In the last named movement
the whole arm can even be curved upward nearly in a complete circle. The first named
movement is chiefly observed in the exterior part of the arm only. All these movements of
the arm are however far from being effected with the same force and rapidity as in the
Ophiurae, but in an extremely slow nearly imperceptible manner; it must however be re-
marked that in all probability all the movements, and consequently also these, would, while
the animal is creeping on the bottom in its normal state, be effected with greater liveliness
than noticed in the captured specimens. The greatest mobility seemed to exist between the
exterior arm-joints; for which reason it is also especially this part of the arm which will
be found variously bent and twisted, in the captured specimens. Nearer to the base the
mobility becomes more and more limited, owing to the much shorter muscular ligaments
between the single vertebrae, until, at the junction of the arm with the disc, it is nearly
reduced to nothing. I have therefore reason to presume, that in the normal state the basal
parts of the arm always retain unaltered their relative positions during the movements of
the animal. When the animal is captured, and I think also in another case which will be
subsequently noticed, an abnormal convulsive movement of the whole arm takes place, which
causes the complete separation of the arm from the disc. Such a separation, always taking
place just where the first vertebra of the arm connects itself with the immovable skeleton
of the disc, may easily be explained, partly by the movement of the whole arm operating
just at that point with the greatest force, and partly by the skin at that part being more
easily broken than at any other.

Any movement of the disc itself is however made impossible by the firm attachment
of the calcareous pieces composing the bucal ring. It is only in the softer parts, stretched
within the calcareous ring, that phenomena of movement can be observed. Apart from the

43

internal organs, these movements consist indeed in an essentially passive rising and sinking
of the dorsal skin, accordingly as the fluid contained in the perivisceral cavity is more con-
centrated in the cavity of the disc or in that of the arms, whereby the roof of the disc
becomes alternately rather convex or vaulted, and then quite flat again; moreover there is
a contraction and extension of the bucal membrane, which takes place within very wide limits,
and whereby the form and size of the bucal aperture can be materially altered.

i
B. Sensation.

I have already described as true instruments of sense the peculiar terminal organs
of the arms (Tab. II, fig. 5, 6 c, fig. 7 a) the structure and peculiar protection of which
distinguish them from all the other appendages of the body. I have previously developed
my reasons for not supposing them to be simply and solely organs of feeling or tentacles.
Although it is of course very difficult to decide as to the nature of the sensation received
through these organs, there is however reason to presume that it must be of a more
specific sort than the more indifferent sensation of feeling. If I have been disposed
to consider them as something like organs of smell, it was only because in their struc-
ture they seemed to me to correspond to what in other invertebrate animals are usually
called olfactory papillae, but I do not therefore mean to say that the sensation conveyed
through these organs must exactly correspond to what we understand by smell. There is
naturally no objection to our assuming that in several of the lower animals there may be
senses of an intermediate sort impossible to class precisely as any of the 5 human senses.
With regard to the position of these organs of sense in the Brisinga, is is evident that in
occupying the extreme point of the arms they have got the most convenient place which
they could have had. The animal is hereby enabled to explore the immediate environs in
all directions at the same time, and indeed over a very considerable area (more than 4
square feet); and whether the presence of suitable food or of inconvenience or danger be
thus discerned, the animal will easily receive information through one or other of these
organs and regulate its movements accordingly.

As real organs of feeling or tentacles (exclusively of their functions as organs of
locomotion previously noticed) we must undoubtedly consider the numerous water-feet at-
tached along the ambulacral furrows, which by their great contractility and mobility in all
directions, and by their fine cuticular covering, seem to be admirably adapted for their
functions in such capacity. All naturalists are also agreed in attributing both the functions
mentioned to these organs in other star-fishes.

It is also extremely probable that the general cuticular covering, and especially the
exterior layer of skin in the Brisinga, may be susceptible • of a rather fine sensation of a
more general kind.

6*

44

C. Functions of the Pedicellaries.

With regard to the nature and destination of these remarkable small organs, which,
as is well known, are only peculiar to the proper star-fishes and to the Echinidse, and which
in the Brisinga exhibit a remarkably complicated structure, appearing also in greater number
than in any other known Echinoderm, many and various conjectures have been set forth,
without apparently elucidating their homology or their functions. Without flattering myself
that I can solve the difficult problem in a perfectly satisfactory manner, I think that I ought
here to set forth my views as to the latter point. The homology of the pedicellaries will
be noticed more particularly in a subsequent section.

P'rom the structure of the pedicellaries it may certainly be inferred that they are
destined to seize and hold fast the objects which come in contact with them. Whether
however the intention is to protect the tender and sensitive exterior cuticle, or to retain
such particles as may serve the animal for food, is a question about which the opinions
of the authors do not yet quite agree. In the Echinidse it has certainly been ascertained,
by direct observations on the living animal, that the pedicellaries play an important part in
removing the excrements expelled from the anal aperture, which else might easily become
entangled among the numerous spines and thereby hinder the free play of the water-feet,
and also pollute the fine layer of skin between the spines. For this purpose, only the
pedicellaries situated in the upper half of the shell are required. But pedicellaries in just
as great abundance are found also on the lower side of the shell, close up to the nearest
environs of the mouth, where they even appear in quite unusual numbers. These pedicel-
laries must certainly have a widely different destination in the economy of the animal. It is
here in my opinion most natural to think of a retention of such particles as should serve
for food, and which probably would be brought within reach of the mouth in a manner
similar to that in which the excrements on the other side are conveyed away down the
sides of the shell.

A similar difference in the pedicellaries' functions may in my opinion also be pre-
sumed to exist in the Brisinga. The pedicellaries situated on the dorsal side of the body
and especially on the numerous soft transverse ridges of the arms, are probably chiefly in-
tended for the removal from the dorsal surface of various extraneous matter (particles of
mud &c) which might come in contact with them; thereby protecting the interjacent naked
cuticle. But on the whole, we must certainly presume that those pedicellaries have a different
destination, which are found in masses on the cuticular sheaths of the spines, attached to
the skeleton of the arms and of the disc, especially those at the extreme ends of the same.

We might indeed imagine, as regards the innermost furrow-spines, that the pedicel-
laries attached to them were intended to prevent extraneous particles from penetrating and
irritating the thin sensitive skin which covers the ambulacral furrows; and according to

45

the position and direction of these innermost furrow-spines, in between the far extending
water-feet, this is also in a high degree probable. But such cannot be supposed to be the
intention of the pedicellaries attached to the other spines. I presume on the contrary that
their function, like that of the pedicellaries on the lower part of the shell of the Echinidse,
is principally exercised in the service of alimentation. If we consider a perfect arm (see Tab. If,
fig. 1 <fe 2) it will be seen that these spines form together a broad brim extending along
the sides of the arms from their basis to the extreme point. The width of the space hereby
occupied, namely from the point of the marginal spines on one side to the corresponding
point of the spines on the other side becomes as much as 6 times that of the arm itself;
and consequently so much greater an area is commanded by the animal than if the spines
were away. Xow as soon as any particle of matter comes within reach of these spines, it
is retained by the numerous pedicellaries situated on them; and at the same time the animal
no doubt becomes aware of its presence. How far the particle of matter retained may be
serviceable for the nourishment of the animal or not, may be probably more particularly
investigated by the sensitive and flexible water-feet, which are extensible on all sides. In the
latter case the object is rejected and removed; but in the former it must necessarily be
brought within reach of the mouth. This may be imagined feasible in two manners; either
by the animal moving its centre towards chat point; or by the object being, through the
joint instrumentality of the spines, and partly doubtless also of the water-feet, successively
conveyed nearer to the centre, until it can be seized by the pedicellaries attached to the
spines which proceed from the environs of the mouth (the bucal spines) whereupon it can
be immediately taken into the bucal aperture, which is extensible as far as to these spines.
It is probable that both these cases can occur, accordingly as the object is of greater or
less bulk. As in all probability the movements of the animal, even when it is living in
completely normal circumstances, are very sluggish and slow, and as the arms themselves
cannot possibly act as instruments of prehension, it could scarcely be imagined possible that
the animal without the help of these small organs would be able to provide itself with suf-
ficient nourishment. But by this peculiar equipment, the animal is enabled in an extremely
easy manner, to secure any prey that may be found within reach of the arms. It is indeed
not rare to find in the captured specimens a whole fauna of various sorts of small animals
(Annelides Crustaceans &c) hanging fast in different places to the arms, and often so firmly
entangled between the spines as not to be extricable without much difficulty. Every little
animal, which in its course along the bottom of the sea is so unfortunate as to come in too
close contact with the slowly-gliding magnificent star, is immediately seized and retained by
numerous microscopical forceps, and stopped in its career without being able by any exertions
to liberate itself.

On the other hand this peculiarity in the Brisinga makes it comparatively easy to
capture. No complicated tackle is required. Not even the ordinary simple dredge. It is
quite sufficient to fasten to a heavy lead a thick rope's-end, the single strands of which are,

46

as in the ,,swabs" used at sea, separated from each other, and only kept together at the
part where the lead is attached. When such a swab is dragged • slowly along the bottom,
the loose strands spread themselves out like a fan over a considerable space, and sweep
the bottom. If any part of the swab comes in contact with a creeping Brisinga, the latter
will immediately become attached to it by the action of the pedicellaries, and entangled in
the strands with its spiny arms, and can thus simply be drawn up, together with the swab,
to the surface of the water. There is very seldom any risk that the animal should fall off.
More frequently it will be found so well attached, that the greatest difficulty will be experi-
enced in extricating the spiny arms uninjured from the threads. This very simple instru-
ment was, as is well known, first systematically employed in the Porcupine's expedition under
the name of ,,hempen tangles"; and has since been very extensively used in examining the
bottom of the ocean.

2. The Vegetative Functions.

What I have to state of these functions is based less on direct observations of the
living animal, than on the conclusions to which I have come by a more minute examination
of the structure and relation of the organs herein concerned. Although there can generally
here be no question of any thing ^Ise than what relates to the whole group of animals to
which the Brisinga belongs, I feel bound for the sake of completeness to touch briefly on
this part of the natural history of the Brisinga, especially as my conception of it is in
several points rather different from that usually adopted.

A. Nutrition.

We have under this head briefly to notice: The reception of food, digestion, circu-
lation of the blood, respiration and secretion, all which functions must on the whole be
supposed to proceed exactly in the same manner as in all other star-fishes. As however there
is much that in connexion with this subject is still very obscure; and as it is highly neces-
sary to subject many points to renewed investigation, I think that it will not be out of place
to develop here more particularly the views which I have been led to adopt by more minute
examination of the present form.

a. Reception of fo o d.

The food on which the Brisinga lives seems to consist of all sorts of smaller and
larger deep-sea animals, annelides, crustaceans, rhizopodes &c. On the whole it does not

47

seem to be dainty in its choice, but to utilise all the organic particles that come within its
reach, and into its pover. Even such animals as are provided with a firm calcareous crust,
for instance shells and foraminifera are swallowed with equal avidity, and the organic par-
ticles of the same rapidly extracted. How, in all probability, the animal obtains its prey
we have already considered. That the pedicellaries play an important part in such capture
I think is indubitable. I consider it very probable that the animal does not only attack
smaller organisms, but also animals or masses of organic matter which are too large to pass
through the bucal aperture, even when it is extended to its greatest circumference. In this
case the folds of the stomach are protruded more or less beyond the bucal aperture, and
laid round the object or a part of it; so that the digestive cavity may in this case be said,
in a manner, to lie outside of the body. It is by no means rare, especially in the other
species Br. endecacnemos, to find specimens in which the folds of the stomach are in this
manner protruded through the mouth, in a very remarkable degree, so as even to project
far beyond the spines attached to the lower side of the disc, and thereby entirely to conceal
the proper border of the mouth (see the delineation in Fauna littoralis Tab. IX, fig. 2).

b. The digestion.

From the walls of the stomach there is secreted, during the reception of the ali-
ments, a fluid which seems to have a quickly dissolving and decomposing influence on the
food; so that in a comparatively very short time the nourishing matter is extracted even
from such organisms as, like shell-cased molluscs and foraminifera, are enveloped in a hard
calcareous shell. The indigestible residuum of the aliments is, together with the extraneous
parts, (the particles of mud) which enter the stomach at the same time, simply thrown out
by the same way they came in. But the fluids, which can be more easily assimilated, are
received into the upper stomachal cavity, whence they pass into the radial ca3ca, which will
always be found filled with a greater or less quantity of oily juice (chylus). From the main
trunks of the radial caeca this juice enters again into thin-skinned lateral recesses, in which
probably the absorption or the transfer of the nourishing parts of this juice to the rest of
the body, or conversion into blood, takes place. According to my conception, this is effec-
ted simply by a transudation through the walls of these lateral recesses, whereby the juice
is thus immediately mixed with the fluid contained in the perivisceral cavity.

c. The circulation of the blood.

As is well known, Tidemann supposed that he had found in the great Mediterranean
star-fish Astropecten aurantiacus a completely developed blood-vessel system of veins and
arteries, and assumes therefore in this case a completely developed circulation of the blood.
The central organ of the circulation of the blood, or the heart, is considered by him to be

48

the peculiar membranous organ which accompanies the stone canal, and which in his opinion
connects itself with a dorsal and a ventral annular vessel whence the other blood vessels
issue; the dorsal annular vessel with its ramifications representing the venous system, and
the ventral annular vessel with the branches issuing from it, the arterial system. This
complicated arrangement of the blood-vessel system described by him, which would also
necessarily presuppose a complete circulation of the blood, has been, as it reasonably might
be, considered referable to all star-fishes, without any one, so far as I know, having subse-
quently been able by direct investigation to demonstrate anatomically the existence of this
system of vessels in other star-fishes, and thereby to confirm Tidemann's statements. Neither
have I, as already mentioned, been able to discover this system of vessels in the Brisinga.
The only part which in this animal, as well as in other star-fishes, may easily be indicated
is the so-called heart, the real nature and destination of which still appear to me to be highly
problematical. Also in the Brisinga this organ (Tab. VI, fig. 6 h, fig. 8) accompanies the
stone-canal, extending from the ventral to the dorsal side of the disc, where it terminates
at a short distance from the stone-canal on the inner side of the madreporic body. The
complicated system of veins, which according to Tidemann's description is supposed to stand
in connexion whith the heart I consider to be something quite different, namely simply
ligaments serving to attach the stomach and the radial cseca to the dorsal skin. By repeated
careful dissections I have succeeded in obtaining preparations of all these ligaments in
their natural connexion with each other and with the adjacent parts (see Tab. VI, fig. 6)
whereby a somewhat complicated structure is exhibited, in the chief features of which it
is easy to recognise the arrangement indicated by Tidemann for the supposed veins. I have
however been able at the same time, to ascertain by satisfactory evidence the real nature
of all, these parts, differing widely from that of blood-vessels. As regards the supposed
ventral, or so-called arterial blood-vessel system, my investigations in the Brisinga and other
star-fishes have led me to the conclusion that we have only here to do with a system of
narrow canal-like sinus, without any particular walls, covered immediately below by the chief
parts of the nervous system, and not in any proper sense entitled to be called blood-vessels.
The blood is thus in the Brisinga, and as I have reason to believe also in other star-
fishes, not inclosed in a special system of vessels, but freely suspended in the whole peri-
visceral cavity (caeloma) where it immediately surrounds and laves all the interior organs.
That there should also be found in the perivisceral cavity a considerable quantity of water
taken in from without, I do not regard as forming any objection to this opinion; on the
contrary I hold such a mixture of blood to be quite necessary for its maintenance and puri-
fication; as special breathing apparatus is, at least in the Brisinga, entirely wanting. The
locomotion of the blood-fluid contained in the perivisceral cavity is brought about by a
lively ciliary movement of the whole interior surface of the skin, partly no doubt also by the
contractions and expansions of the skin itself and of the central part of the digestive cavity
or stomach. There can thus be no question here, properly speaking, of any real circu-

49

lation of the blood in the usual signification of the expression; as the blood contained in
the perivisceral cavity always surrounds immediately the organs enclosed in the same and
thereby effects their alimentation.

There is however an important system of organs, the anomalous position of which
in the star-fishes necessitates a peculiar arrangement of the blood-system; and that is
the nervous system. While all the other vital organs have their place in the common peri-
visceral cavity (cceloma) the chief parts of the nervous system, the so-called ambulacral nerves,
are, as already indicated, situated quite superficially immediately under the thin skin which
lines the ventral furrows, and entirely separated from the perivisceral cavity by the inter-
jacent ambulacral skeleton. In order that the nervous system might be alimented in the same
manner as the other vital organs, it was therefore necessary that apart of the blood-fluid should
be conducted to it. This is effected by means of the before mentioned system of canal-like
sinus which is immediately bounded below by the central parts of the nervous system, and which
has formerly, and doubtless erroneously, been regarded as composed of real blood-vessels of
an arterial nature. How the blood-fluid from the perivisceral cavity is conducted into these
canals, and whether the so-called heart really plays any part herein, is not yet so com-
pletely clear to me that I can venture to express any decided opinion. I have as above
mentioned, not been able to indicate with certainty any direct communication between the
upper end of the ,,heart" and the perivisceral cavity; and although its lower end seems to
be in immediate contact with the circular sinus that lies inside of the bucal membrane.
I have yet not succeeded in ascertaining by direct investigation that there is any actual
communication in both cases. The question might probably be best decided by experi-
mental injection in large living star-fishes, for which unfortunately I have no opportunity at
present. That there really exists a communication between the said canals and the peri-
visceral cavity, I consider in any case as given; as the fluid contained in them corresponds
entirely to that which is found in the perivisceral cavity, and like it contains small cell-like
globules (blood cells) which are set in motion by an interior ciliary arrangement, Likewise
I consider it to be very probable that at the point of the arms there exists a direct con-
nexion between this latter and the canal-like blood-sinus which run along [the ventral furrows.

d. The respiration.

No special organs of respiration are found in the Brisinga. I have sought in vain
in the dorsal skin for the so-called respiratory tentacles which in most other star-fishes are
so distinct and numerous. They are quite certainly wanting in the Brisinga. The respira-
tion may therefore reasonably be supposed to stand in connexion with the reception of
water into the interior of the body, which appears to take place continually. Of the parts
of the water-system, the ampollae belonging to the water-feet project more or less from the

7

50

ambulacral pores into the perivisceral cavity, where they form along the upper side of the
skeleton of the arm a double row from the base to the extreme point (see Tab. Ill, fig. 4
and 17). The extremely thin-skinned nature of these ampollse in the Brisinga seems to
make them well adapted for effecting a transmutation of gas between the water enclosed in
them and the fluid of the perivisceral cavity. Moreover the direct reception of water which
takes place in the perivisceral cavity itself, chiefly as we may suppose through the terminal
section of the ambulacral vessels, must no doubt contribute in an essential degree to the
purification and maintenance of the blood.

e. Secretion.

I have previously noticed as an apparatus of secretion, the peculiar and strongly
developed glandular organ in the Brisinga, which is situated in the middle above the
stomach between the latter and the dorsal skin, and from which there issues a short canal
terminating on the dorsal side of the disc in a fine pore with a slightly excentric position.
It is this pore which in the Brisinga, and indeed also in other star-fish, has b£en called
the anal aperture, the narrow canal connected with it having been considered as a terminal
section of the alimentary cavity or a real intestine. I have already remarked, in describing
this apparatus, that in my opinion such a view is erroneous. Between the said canal and
the cavity of the stomach, I have not been able to discover any direct connexion whatever;
and even if any such connexion really existed, it would still be difficult to understand how
this canal could act as a terminal intestine; for it is so extraordinarily narrow, that even
the smallest solid particles could not possibly be expelled through it and through the like-
wise extremely fine dorsal pore. Neither have I ever seen real excrements expelled in this
manner; all the indigestible remains of the aliments are on the contrary ejected again through
the mouth. I can not therefore consider this canal in the Brisinga as any thing else than
a simple issue for the glandular apparatus previously noticed; and I think it highly probable
that many of the star-fishes which have been presumed to possess an anus, are really without
it, but have only a similar small secretory pore.

The product of secretion is a colorless fluid developed in the organ which, ac-
cording to its whole structure, is evidently glandulous. Concerning the nature of this secre-
tion, it seems reasonable to regard it as a sort of urinary product, on which supposition
the whole glandular apparatus would have to be considered as a blood-purifying organ
similar to the kidneys in the higher animals. Such a function has also usually been previ-
ously ascribed in other star-fishes to the so-called interradial or rectal cseca, to which the
organ in question evidently corresponds.

51

B. Propagation.

I am inclined to suppose that the Brisinga has two different means of propagation,
a sexual and a non sexual process. Such an assumption will not preliminarily appear to
be so entirely unreasonable; since we have (mainly through Ltitken1) recently become ac-
quainted with some Asteridse in which, collaterally with the usual sexual propagation, we
can distinctly demonstrate a multiplication by simple division. A non sexual propagation,
somewhat different indeed from this, may I think be claimed for the Brisinga collaterally
with the ordinary process.

a. The sexual propagation.

As in other star-fishes, only one sort of sexual organs, the ovaries, or the sperma-
ries can be developed in the same individual. The sexes are therefore always separate.
At the time when the propagation takes place, the ovaries, as well as the testes, swell
very considerably; so that the dorsal skin of the arms comes into a state of extremely
strong tension. The ripe sexual products issue from the sexual apertures, and are mixed
together in the water without any actual copulation taking place. In the captured specimens
the sexual products will often be observed issuing from the sexual apertures in the form of
a more or less projecting plug, which only little by little is dissolved or becomes diffused
in the water. The emission of the sexual products takes place thus as it seems very slowly;
and the whole act of propagation extends therefore over a tolerably long period. Only
when the specimens have attained their full size, have I found the sexual products mature.
In younger specimens, we can indeed always discern both ovaries and testes, but they are
still quite solid, with undeveloped contents. In the ovaries there are only found simple clear
cells without any deposit of the opaque yellowish red yolk-mass peculiar to the mature
ovum; and in the seminaries only the small elliptical bodies whence the characteristic fili-
form spermatozoa are subsequently developed. With regard to the time for propagation it
appears chiefly to be limited to the summer months.

b. The non sexual propagation.

I have certainly not succeeded in ascertaining by direct observation the faculty in
the Brisinga of such multiplication by non sexual means; and what I have to state on this
subject will be for the present chiefly hypothetical. I feel however bound to produce the

l) Beskrivelser af nogle nye eller mindre bekjendte Slangestjerner, med nogle Bemcerkninger om Selvdelingen hos
Straaledyrene. (Oversigt over d. Kongl. Danske Vid Selsk. Forhandl. 1872).

7*

52

reasons which have led me to adopt the notion that such a non sexual propagation does at
times actually take place in the Brisinga.

There is one thing which must immediately be remarked by any one who has exa-
mined a sufficient number of specimens of this form; and that is the frequently very une-
qual development of the arms or rays. In nearly all the specimens examined by me one
or more of the arms, and sometimes even most of them, proved to be in a more or less
rudimentary state, and evidently only quite recently regenerated from the disc. That these
newly formed arms had grown in the stead of fully developed arms which had been sepa-
rated from the disc, was evident from the structure of the disc, and admits of no doubt.
The question is now to explain in a reasonable manner the so frequent detachment of fully
developed arms in the Brisinga. To assume that such detachment is caused by accidental
external violence, or hostile attack of larger sea-animals, might certainly be admissible if
the question was only of isolated cases. But where the occurrence proves to be so frequent,
as my investigations have taught me, that it may even be considered as the general rule,
while it is the exception to find specimens with all their arms equally developed; I should
be more inclined to recognise a more normal phenomenon, namely a real voluntary act of
the animal, a spontaneous detachment of one or more of its symmetrical principal parts
(Antimera) that is to say a peculiar sort of division, the object of which is to effect
a non sexual propagation. To state briefly my opinion, I think that every arm thus detached
does not perish, but is destined to form a new individual by reproducing from its adoral
extremity a new disc, which then, by a sort of budding, sends forth the other Antimera
or arms belonging to a completely developed individual. I think I may be allowed to as-
sume the possibility of such a propagation, when we consider the great self-substanciality
of the arms of star-fishes generally, and especially of those of the Brisinga; and likewise
that direct observations are not wanting which appear to prove that such regeneration may
in isolated cases take place even in much more centralised star-fishes, for instance in the
genus Asterias and Ophidiaster.

3. Faculty of regeneration and tenacity of life.

That the star-fishes possess an extraordinary power of regeneration is sufficiently
well known. This is especially easy to ascertain in the forms that are found in the litoral
belt, and therefore exposed to many external injuries, often bearing the marks of the most
diversified damage, which yet does not appear to have any appreciable influence on the
vital vigor of the animal. From the injured parts it may be observed that a very lively
new formation takes place, whereby not only the wounded surfaces heal rapidly, but also
the lost parts, with all the organs belonging to them, are in a shorter or a longer time
completely regenerated. In the deep-sea star-fishes it is however much more rare to find

53

traces of accidental mutilation, and that is by reason of the calm and quiet which always
reign in the great deaths of the sea, and which rarely exist in the literal zone where so
many sorts of external perturbations exercise their influence. It is however presumable that
also the deep-sea star-fishes possess the same faculty and facility of repairing by regene-
ration all sorts of injuries. This appears very evidently, in the Brisihga. That lost arms
can be completely reproduced again from the disc, has been already noticed; and likewise
that we often find in the same specimens a great number of such small arms simultaneously
sprouting forth. In the specimens delineated by Wyville Thomson (1. c.) there are repre-
sented no less than 5 such rudimentary arms in one uninterrupted series. Also in the
middle of broken arms regeneration takes place in the same manner very easily. I have
once observed such a case in the arm of a full grown male specimen where the reproduced
part of the arm was evidently and sharply distinguished from the original basal piece.
I shall have occasion in a subsequent section to notice a highly remarkable case, where one
of the arms in an otherwise normally developed individual was bifurcate at the extremity
(see Tab. II, fig. 3). This monstrous formation of the arm owes its origin without doubt to
an accidental lesion and consequent regeneration. In this case the reproducing power shows
even such a redundance that it has not confined itself to reproducing the lost piece only, but
has done it doubly. It is, as already noticed, my conviction that even a single separated
arm will be able to reproduce a completely new disc, with the other arms belonging to the
same, as has already been stated with great probability with reference to the genus Ophidi-
aster and Linckia (see the above cited memoir of Mr. Lutken).

That the animal, in otherwise normal circumstances, possesses great tenacity of life
under external injuries, is undoubted. It is a different question whether the animal would,
under circumstances which may be regarded as exercising an injurious influence on all the
vital functions, exhibit any unusual tenacity of life. Now the animal must undeniably find
itself so situated when brought up from the deep, as is also clearly indicated by the con-
vulsive agency whereby all the arms will almost always in this case detach themselves from
their connexion with the disc. Nevertheless I have been able to keep such specimens divi-
ded into their separate parts, alive even for many days by supplying them with fresh sea-
water in sufficiently large vessels.

I will mention one more apparently important fact in support of my hypothesis
above set forth of the non sexual propagation, namely that the arms always shewed signs
of life during a somewhat longer time than the disc itself whence they were detached. The
chief seat of vital force seems thus to be rather in each single arm than in the disc itself,
whence again it appears to be matter of inference that it would be more difficult for a
disc detached from all its arms to reproduce a single arm, than for a single detached arm
to form both the disc and all the other arms belonging to it.

54

VI.

History of development,

(Ontogeny).

I must presume it very probable that the embryonic development of the Brisinga
may exhibit peculiarities of very great interest, and perhaps will open for us a still clearer
view of the nature of the Echinoderms in general; and I must so much the more regret that
I can give no information as to this important point in the natural history of the Brisinga.

How the ova, when ejected from the ovaries, progress after fecundation; whether
they produce, as seems to be the rule for star-fishes, a bilateral symmetrical freely swim-
ming larva in which the first foundation of the future star-fish is found, or whether the
development takes place more directly, as is the case with a much more limited number of
them, must be decided by future investigation.

We may however by anticipation, and with some degree of confidence, assume the
existenc'e of the first named mode of development. My direct investigations as to the deve-
lopment do not begin until after that the animal has acquired the radiary form peculiar
to the class.

I. Progressive development.

Under this head we have to treat of the changes in the appearance of the whole
animal, and in the form and structure of the single parts which they undergo in the normal
progressive growth, from the time when the animal has acquired its radiary form, until it
attains the fully developed state. These changes are in general such as are common to
all individuals without exception, and must be well distinguished from the changes in parti-
cular details which usually occur later, partly determined by purely, accidental causes partly
in consequence of a more or less abnormal development.

oo

a. Early Stage of the Brisinga.

From the same locality whence I obtained my supply of Brisingas, I once chanced
to get out of the mud brought up in the dredge a little tiny 10-armed young star-fish
which, in spite of its very different appearance at first sight, may nevertheless with full
certainty be classed as a Brisinga in a very early stage of development, exhibiting in more
than one respect points of very special interest.

The specimen (Tab. IV, fig. 38, 39) has only a disc-diameter of 2.50 Mm., and has all
its arms 10 in number in natural connexion with the disc. The length of the arms cannot
accurately be ascertained; as in all of them the extreme point is broken off. They do not
however appear to have exceeded in length the double diameter of the disc. Now what
immediately appears remarkable in them is that they proceed from the disc apparently
quite in the ordinary manner like simple prolongations or expansions of it, without any
indication on the exterior of any sharp limitation as in the adult Brisinga. Altogether this
young animal is in its whole habitus a completely normal star-fish resembling most the
young Solaster papposus; and it would not easily have occurred to me that the little animal
was a real Brisinga if I had not noticed in it a peculiarity which induced me to examine it
more minutely. Quite unlike the young Solaster, it remained clinging with great tenacity
to every object it come in contact with, exactly in the same manner as a Peclicellaster,
which already would at once indicate a very specical development of its pedicellaries. On
examining the latter microscopically, I found to my surprice that they corresponded even in
the minutest details to those of the adult Brisinga, and, as in the latter, a great many of
them were crowded together on the skin-sheaths of the arm-spines. A further examination
of the specimen brought to light many other characteristics, demonstrating with absolute
certainty that we have here really a young Brisinga before us.

Let us therefore consider this young animal somewhat more closely. Viewed from
above (fig. 39) it exhibits, as before mentioned, all the characteristics of a completely normal
young star-fish. In the dorsal skin there appear already to be formed various calcareous
deposits. In the interior there may be remarked an extremely fine hyaline calcareous net,
perforated with regular holes, and here, strangely enough, forming large connected divisions
of irregular form. Each of these divisons consists however only of the most primitive cal-
careous elements widely different from the complicated calcareous net which is found in
developed star-fishes. From these calcareous nets in the skin there issue at regular inter-
vals larger and smaller spines, a couple of which were of quite unusual length and extre-
mely thin, almost like bristles, but every one enveloped in its separate cuticular sheath.
In the angle between two contiguous arms there may be observed a more regularly formed
plate, which without doubt is the rudiment of the wedge-plate found here in fully deve-
loped specimens. But no distinct trace is yet to be seen of the marginal plates that stand

56

in connexion with the wedge-plate. Neither can there be certainly distinguished any evident
madreporic body. The arms issue like immediate prolongations from and round the disc;
they are broadest at the base, and thence rapidly and evenly tapering towards the extre-
mity which however as before mentioned was broken off in them all. Through the trans-
parent dorsal skin of the arms, in which every trace of calcareous keels or transverse ridges
is still wanting, there appears the subjacent ambulacral skeleton with its medial elevated
crest formed by the dorsal part of the ambulacral plates. The single vertebra are of a
particularly narrow and elongated form. Along each side of the arms there is seen a single
row of spines, which are all provided with a distinct skin-sheath thickly larded with pedi-
cellaries. According to their position, they appear to answer to the outer furrow-spines in
the developed individuals.

If we turn the specimen round, and view it from the ventral side (see fig. 40) the
naked skin-like area which occupies nearly the whole of the underside of the disc, with the
widely gaping circular oral aperture (o) in the middle of it, strikes the eye immediately.
Only in the extreme periphery there appear those calcareous parts which support the disc
and the arms (the ambulacral skeleton) which as regards the disc forms an extremely nar-
row frame, composed of a single set of vertebrae, whence the skeleton of the arms takes its
issue. In conformity herewith we see only a single circle of 20 water-feet belonging to the
disc. These water-feet are placed in pairs at the extremities of the ambulacral furrows
which run along the ventral side of the arms; each pair separated from the next by an
obtusely conical calcareous piece (ad1) projecting inwards from the angle between 2 arms,
and formed by the 2 contiguous adarnbulacral plates belonging to the 2 adjacent vertebrae.
The ventral furrows of the arms taper rapidly outwards, and are, between each successive
pair of water-feet, distinctly instricted. The adambulacral plates which limit the ventral
furrows on the sides are long and narrow, yet still indistinctly separated from each other,
and only furnished each with a single spine in the middle. Likewise each of the innermost
adambulacral plates (ad *) belonging to the disc and joined together in one piece, is fur-
nished in the middle with a long spine directed obliquely downwards and outwards. Of the
oral spines, spread out fan-like towards the mouth, which in fully developed specimens are
so distinct, there is yet no trace to be seen. Immediately before the interior extremity of
the ambulacral furrows, and partly covered by the point of the 2 contiguous innermost
adambulacral plates, there extends round the whole ventral sid» of the disc a narrow calca-
reous, ring, to which the oral membrane (m) is attached. It answers to the circular ridge
projecting inwards from the oral ring in adult animals, and consists similarly of 20 single
pieces, which, immediately before the ambulacral furrows and the point of the interraclial
spaces, are connected with each other by evident sutures.

It will be seen frdm the above description that the disc in the young Brisinga men-
tioned, although externally appearing to be rather large in proportion to the arms, is yet
really, with due regard to the skeleton, still more reduced than in the full grown animal;

57

as only one single set of vertebrae enters into the composition of the disc; while the fol-
lowing set, which in the adult animal always continues in firm connexion with the oral ring,
does here evidently not belong to the ring but to the arms. As regards this innermost set
of vertebrae, their form and structure appear here to correspond so perfectly with the proper
vertebrae of the arms, that we may really also consider them as belonging to the arms.
In this manner there will thus remain only the narrow calcareous ring to which the oral
membrane is attached, that can be considered as a special skeleton-formation belonging to
the disc; while all the remaining skeleton is only formed by the 10 arms, which are so
arranged around a common centre that their interior vertebrae are contiguous and grow
together with each other. The important inferences which may be hence deduced for a
correct appreciation of the nature of the star-fishes, will be noticed more particularly in a

subsequent section.

\

b. Development of the disc.

The structure of the disc has been already above noticed in a very early stage of
development, whereby it appears evidently enough that the exterior set of vertebra, which
in fully developed specimens enters into the composition of the calcareous ring, does not
originally belong to the disc, but represents the interior vertebrae of the arms. In the be-
beginning there is also between both these sets of vertebra?, a movable muscular connexion
similar to that between the other vertebrae of the arms; so that it is in fact only the
interior set of vertebras which serves as support for the disc itself. During the subsequent
growth and enormous development of the arms, it becomes however necessary that the
skeleton of the disc should acquire a greater solidity than could be produced by only a
single set of vertebrae; for which cause the interior vertebrae of the arms have to give up
their mobility little by little, and, by connecting themselves with the original calcareous ring,
to contribute to its more powerful development.

In quite young specimens with a diameter of disc of 10 — 12mm and a length of arms
of G0--70mm these vertebrae still retain in some degree their original self-substantiality. They
are (see Tab. V, fig. 13 & 15) comparatively to the yet slightly developed interior vertebrae,
considerably more prominent, and even project distinctly beyond the periphery of the calca-
reous ring, being bounded on each side by a rather deep sinus in the calcareous ring, which
indicates the place, where the wedge-plate rises. The connexion with the interior set of
vertebrae has however already acquired so much firmness that these vertebrae, as in adult
specimens, always remain attached to the disc when the arms are separated from it.

With- respect to the exterior form of the disc in such young specimens (the smallest
having only a disc-diameter of 10mm) it differs from that of the full grown, in the more or
less evidently projecting radial parts, as well as in its considerably greater flatness or thin-
ness, and in its generally weaker and less compact appearance. It is like the arms, of a

8

58

very pale reddish color, and when held up against the light so transparent that several of
the interior organs may be distinctly seen through the dorsal skin.

The dorsal spines are proportionally smaller and more dispersed; and only a few of
those attached nearer to the centre may sometimes be distinguished by a peculiar strongly
elongated bristle-like form, evidently a relict from the embryonic state; the madreporic
body occupies the same characteristic place as in the adult specimens, but is always very
small and only slightly convex. The spines attached on the lower side of the disc (to the
adambulacral plates) are relatively much shorter than in full grown specimens, and fewer in
number; and of the oral spines directed towards the mouth, there are only 2 or 4 developed
for each interradial space.

c. Development of the arms.

In the little juvenile Brisinga previously mentioned, the extreme points of all the
arms were, as already stated, broken off; so that their exact form and length could not be
accurately ascertained. I have noticed as characteristic the less distinct demarcation between
them and the disc; the entire absence of calcareous ribs and transverse ridges, and the
presence of only a single row of spines on each side (the exterior furrow-spines).

We have however -even in fully developed specimens of the Brisinga the best oppor-
tunity for studying the development of the arms, and for tracing their changes to a yet
much earlier stage of development than in the young one mentioned. There is always
taking place in the Brisinga, as before stated, a continually repeated reproduction of new
arms, in place of those previously detached from the disc. But it is here to be remarked
that these arms, formed by a sort of germination, develope themselves apparently in a
manner somewhat different from that in which the arms originally belonging to the young
animals are developed.

In the earliest stage of development these new arms sprouting from the disc
(see Tab. VI, fig. 1 x) have only the appearance of inconsiderable conical processes, issuing
nearer to the ventral side from the middle of the radial spaces. But on closer examination
the rudiments of the most important parts may already be distinguished. The extremity
always exhibits a distinct button-like enlargement, from the lower side of which there pro-
jects in a horisontal direction a relatively large cylindrical appendage (b) which is the
previously mentioned peculiar terminal apparatus, the first evidently developed organ that
appears. On the upper side, and especially on the terminal enlargement of the arms, there
appear (see fig. 11, 12, 13) already several pedicellarise developing themselves, but still quite
in a cellular form, and distributed without any distinct order over the skin. Below there
may be seen very distinctly the rudiments of a limited number (12-14) water-feet (w) in
the form of small button-like warts arranged in 2 longitudinal rows, and rapidly increasing
in size towards the base of the arm, with a distinctly marked raised line (n) running between

59

the two rows, which line is the ambulacra! nerve that forms a well defined club-shaped
enlargement '(b) at the base of the organ of sense (a). Of spines no trace is yet to be
seen. Only after that the nascent arm has increased somewhat more in length, there may
be observed on each side of the water-feet a single row of regularly arranged sinuosities in
the skin, wherein the calcareous particles begin to develop themselves (fig. 13 p). These
are the growing furrow-spines. Nearer to the dorsal side there may be seen a similar row
of bag-like enlargements in the skin, inclosing calcareous particles in a state of progressive
development, but more irregularly arranged. They represent the growing marginal spines,
and on those situated nearest to the base a few pedicellariaB have already begun to develop
themselves (see fig. 15 p), The pedicellarise which appear in the dorsal skin have in the
mean-time developed themselves further, and now begin to be arranged • in distinct raised
transverse stripes or bands. The rudiments of water-feet have increased considerably in
number; and those placed nearest the base have already acquired a cylindrical form and
commenced their activity (fig. 13w). The terminal organ of sense has during the progress
of this development retained approximately its original form and size, but is now partially
overhung by a distinct protecting calcareous plate with accompanying spines and pedicel-
larise (see fig. 14 & 15).

The further development of the arms consists now essentially in a successive in-
crease of joints or metamera proceeding from the base of the terminal enlargement, with a
corresponding increase in the numner of water-feet and spines; while the arm grows rapidly
in length (see Tab. I, fig. 4, 5. 6 e, Tab: II, fig. 1 & 2).

The transverse bands over the dorsal skin, which are covered with pedicellarise,
appear now both in greater number and with greater distinctness and regularity; and the
marginal spines are more and more elongated, especially at the base. Even after that the
arm has attained a length of 30— 40mni there is still no trace to be seen externally of the
spined calcareous ribs so characteristic of the full grown animal. Only in a much later stage
namely after that the organs of generation have begun to develop themselves, these ribs
begin to appear as quite narrow slightly raised calcareous stripes, on which the spines are
still only short and indistinct (see Tab. I, fig. 1). The marginal spines have already long
ago attained their full development, as also the exterior furrow spines; but the interior
furrow spines do not begin to develop themselves until much later; nay, it appears that
even after the animal has attained its full development, a constant increase takes place in
the number of these small spines ranged nearest to the ventral furrow. The arm which in
the beginning was thickest at the base and thence rapidly tapering towards the end, begins
now to b*e somewhat instricted at the base and thereby to be more sharply distinguished
from the disc; as the dorsal skin lies here close on the subjacent ambulacral skeleton;
while at some distance from the base, this skin begins to be more and more elevated above
it and stretched to the sides. The characteristic almost fusiform appearance, which the
basal part of the arm often exhibits in adult specimens (see Tab. I, fig. 12, Tab. II, fig. 1)

8*

60

is due to the strongly developed organs of generation lying within it. If these organs, as
is the case in other seasons than in the summer, are less developed, then neither will the
arms in full grown specimens exhibit such a remarkable enlargement; although they are
always at the base considerably thinner than they are further out from it.

d. Development of the ambulacral skeleton.

To examine the development of the ambulacral skeleton, it is only necessary to
isolate from the disc a recently formed arm-shoot, and to put it in a solution of potass, which
will quickly make all the organic parts transparent. On examining the arm-shoot under
the microscope with sufficient magnifying power (see Tab. VI, fig. 13, 14, 15) the calcareous
particles which are to form the ambulacral skeleton will then be distinctly seen deve-
loping themselves in the interior. As the formation of the various joints or metamera, of
which the arm is composed, always proceeds from the base of the terminal knob-like en-
largement of the arm, the youngest or earliest foundations of the vertebrae will always be
found nearest to the same, and the nearer to the base of the arms the more developed
they will be found. Thus in one and the same preparation, we have a whole series of suc-
cessive stages of development before us. Every vertebra is formed of numerous reticular
ramified hyaline calcareous staves, which take their issue from 4 different points or centres
answering to the single calcareous plates of which each vertebra is originally composed.
These points of ossification of the vertebrae are situated somewhat nearer to the ventral
than to the dorsal side of the arm, and they are arranged in pairs slightly varying in height,
and regularly alternating with each other. The pair situated nearest to the medial line, and
in a more dorsal position (fig. 14 a), forms the foundation for the ambulacral plates; that
more remote from the middle, and nearer the ventral side (ad), represents the adambulacral
plates. The former pair appears first, and afterwards successively the other. Originally
these points of ossification lie widely separated from each other, and only during the further
development and ramification of the calcareous particles approach each other little by little,
so as at last to become united partly by suture, partly by muscular ligaments.

The manner in which the ramification of these calcareous parts composing the verte-
bras proceeds, may be traced very distinctly in a recently formed arm-shoot. First of all
there appears only a little angular calcareous granule (tig. 17, 18) which then shoots out a
certain number (usually 6) radiating processes (fig. 19) situated in the same plane. Each of
these processes divides itself then at the extremity, or shoots out in 2 diverging branches
(fig. 20) which are again subdivided, and at last partially coalesce with those adjacent, and
thereby, together with the original processes, form the boundaries of a single ring of oblong
apertures (fig. 21). By a dichotomic subdivision of the outermost processes continued in
the same manner, there is little by little formed a thin calcareous disc (fig. 22) perforated
with numerous holes ranged with tolerable regularity in concentric order. This disc begins

61 •

likewise subsequently to increase in thickness; as processes issue also from the surfaces of
the disc, ramifying themselves in a similar manner. At last the ramification of the calca-
reous staves becomes so manifold and dense, that the whole forms a compact calcareous
mass, in which however the original reticular or lattice-work formation may easily be ob-
served by applying sufficient magnifying power (fig. 23, 24).

The ambulacral plates, which thus originally, like the adambulacral plates, had the
form of thin circular discs, increase relatively very rapidly in volume, and soon acquire
(see fig. 14, a) a transverse position on each side of the medial line. Both ends, the exterior
as well as the interior, become gradually enlarged, and meet the nearest adjacent ambu-
lacral plates belonging to the same side, whereby these plates acquire a somewhat fusiform
shape. Between them there will thus remain oval apertures, which are the ambulacral pores,
whence the water-feet project. The interior extremity of the ambulacral plates developes
itself then more strongly than the exterior, and begins to raise itself up in the interior of
the arm to form the dorsal ridge of the arm-vertebrae which is afterwards so prominent.
Still however the ambulacral plates of each pair are separated in the medial line by a
distinct fissure widened between the vertebrae (see fig. 16, a). The adambulacral plates (ad)
grow relatively much more slowly, and remain therefore in respect of size far behind the
nearest ambulacral plates; they are a little elongated, and ranged in a somewhat oblique
direction in the intervals between the exterior ends of the ambulacral plates, and soon come
in contact with these extremities.

The further development of the skeleton of the arm consists in each pair of ambu-
lacral plates belonging to a vertebra growing together with each other and with the respec-
tive adambulacral plates. Afterwards the whole vertebra becomes gradually elongated during
the further growth of the arm (fig. 16). In the young Brisinga above described even the
interior vertebra) of the arm were as long as they were broad; and in young specimens with
a disc-diameter of 10mm the joints of the arm (see Tab. V, fig. 15) are still quite unusually
elongated in comparison with those of full grown individuals, in which, in the whole basal sec-
tion, they are scarcely half as long as they are broad. As to the ambulacral skeleton of the
disc, it has been already stated that it consists originally (in the earliest stages of develop-
ment) only of the interior arm-joints that are connected with each other round a common
centre (see Tab. V, fig. 11, 12). Besides the ambulacral and adambulacral plates belonging
to these vertebrge, there are still only the so-called parietal plates (p) distinctly developed ;
these are however yet very small, and appear indeed only to 'contribute to the formation of
the circular border (x) to which the oral membrane is attached. Of the wedge-plate only
slight indications are observable (w), and these seem to shew that they ought rather to be
reckoned among the calcareous parts of the cuticular system than as belonging to the ambu-
lacral skeleton; although they are subsequently firmly connected with the latter. The same
may be said with even greater certainty of the dorsal marginal plates, which are connected
with the wedge-plate and which in like manner are, in fully developed specimens, firmly

62 •

connected with the oral ring, but in the early stage appear only in the form of 2 thin
bone pegs (r) loosely attached to the oral ring. Of the 2 following pairs of marginal
plates belonging to the arms there is still, in the young animal above described, not the
smallest trace to be seen.

In specimens with a disc-diameter of 11 — 12mm the oral ring (fig. 13) is still only
very slightly developed; although it already consists of all the calcareous pieces which, in
fully developed individuals, enter into its composition. It is especially very thin and flat;
so that its upper surface is nearly horisontal, and lies about level with the skeleton of
the arm; while in full grown individuals it is strongly convex, and towards the interior
considerably raised above the skeleton of the arm. This comes mainly from the slight deve-
lopment of the interior set of vertebrae, which still retain much of their original appearance
similar to that of the arm-vertebrae. It has been before mentioned that the exterior set "of
vertebrae, which likewise belong to the oral ring, and which in full grown individuals are
considerably reduced in size, correspond in the early stage completely with the arm-vertebrae
both in size and form, as well as by their looser connexion with the interior set of vertebrae,
being moreover distinctly salient in the periphery of the oral ring. -In consequence of the
peculiar form of the oral ring hereby produced, the wedge-plates (fig. 14, 15 w), which in
full grown specimens are more or less vertical, lie here in almost perfectly horizontal posi-
tion; and on each side of them there appear, on the upper surface, the 2 pairs of ambu-
lacral pores clearly and distinctly limited, the one behind the other.

e. Development of the cuticular skeleton.

The development of the calcareous particles enclosed in the dorsal skin proceeds,
exactly in the same manner as the development of the ambulacra! skeleton, by a radiary
ramification of originally simple isolated calcareous granules. In the dorsal cuticle of the
disc, this development of the calcareous pieces to which the disc-spines are attached, is
easily observed in quite young specimens, in which these calcareous pieces have still parti-
ally the form of small thin perforated circular discs. In the young specimen above described
there is in the dorsal cuticle a more continuous extraordinarily thin calcareous net with wide
circular meshes (Tab. VI, fig. 33), evidently a relict from the embryonic state, and which is
afterwards replaced by the usual limited calcareous parts corresponding with the spines.
Just opposite to the angles between the bases of the arms there can however be discerned
10 somewhat more distinctly limited plates, but of the same fine reticular structure. These
plates (see Tab. IV, fig. 39) belong here most undoubtedly to the skin, but enter subse-
quently into connexion with the rudimentary wedge-plates, and form the tubercular promi-
nence afterwards so conspicuous on the same.

The dorsal transverse ribs are indeed not outwardly visible until after that .the arm
has already attained a considerably development; but their foundation is laid very early,

63

and simultaneously with the formation of the ambulacral skeleton. Even in the very earliest
foundations of the arms (see Tab. VI, fig. 15) there may be observed in the dorsal cuticle,
immediately behind the terminal enlargement, 3 longitudinal rows of calcareous particles (r)
forming themselves in the same manner as the ambulacral and adambulacral plates. These
appear in the beginning (next to the terminal enlargement of the arms) quite separate, in
the form of the usual regularly perforated circular discs; but in the subsequent progressive
ramification of the calcareous staves connecting them, the 3 calcareous plates lying in a
transverse row approach each other little by little until they coalesce, and thus in connexion
form continuous transverse bands. At each end of these transverse bands there is deve-
loped already at a very early stage, a spine (p) which is the marginal spine. In the be-
ginning, such calcareous particles in the dorsal skin develop themselves constantly from- the
basis of the terminal enlargement. There are thus formed more and more of the calcareous
transverse bands. But after that the arm has attained a certain length this formation ceases.
The arm continues however to grow by a constantly continued formation of new joints, which
gradually force the first formed joints nearer to the base of the arm. Hence it is easily
explained that these calcareous particles, which afterwards form the transversal spined cross-
ribs, although they originally proceeded from the extremity of the arm, at last become far
removed from it and confined only to the basal section.

It appears that besides the above-noticed calcareous particles specially imbedded in
the skin, there are also a number of calcareous plates which belong to the cuticular skele-
ton, and which subsequently enter into intimate connexion with the ambulacral skeleton,
namely the wedge-plates and dorsal marginal plates standing in connexion with the skeleton
of the disc, also the 2 pairs of marginal plates which are found at the base of the arms,
and finally the calcareous plate which covers the terminal organ of sense. This last (fig.
13 — 15, pi) is formed already very early, simultaneously with the ambulacral skeleton, as a
net of thin calcareous staves ranged in a half-circle, whence subsequently thin spines take
their issue (fig. 25). The dorsal marginal plates of the arms appear on the contrary not to
form themselves until considerably later (see fig. 16, r, r).

/

f. Development of the spines.

The first spines which appear distinctly on the arms are, as before mentioned, the
exterior furrow spines. They form (see fig. 13, p) first, close to each side of the ambulacral
skeleton, a regular row of short cylindrical cuticular processes, each answering to a separate
adambulacral plate. In the interior of these processes there appears first a little angular
calcareous granule (fig. 26), which then, in the usual manner, begins to shoot out radiating
processes whereby there is formed a simple perforated disc (fig. 27). From the middle
of one surface of this disc, there arise a certain number of thin calcareous staves, close
together, which rapidly grow in length (fig. 28). After these calcareous staves have attained

64

a certain size, they shoot out, at right angles and at regular intervals, short lateral processes
which grow together with the nearest calcareous staves (fig. 29, 30). Accordingly as the origi-
nal number of staves may have been greater or smaller, there is at last formed a more or
less complicated latticed process which is the proper calcareous spine; while the basal disc
forms the enlarged joint-surface whereby the spine is subsequently articulated with the
adambulacral plates. When the spine has attained a certain size, there begin to appear,
at the end of the cuticular sheath, some few cell-like bodies projecting more or less from
the same, and having small calcareous granules contained in them (fig. 31) ; this is the first
development of the pedicellaries. All the other spines are formed precisely in the same
manner. The marginal spines on the growing arms and the spines situated on the terminal
enlargement of the arm appear nearly at the same time. The last named (fig. 32) are
usually remarkable for their particularly slender shape, and for being originally formed of 2
or 3 strongly elongated calcareous staves connected with each other at regular intervals by
horisontal intermediary bars. Of the spines belonging to the' calcareous ribs of the arms,
no trace is to be observed before the arm has attained a considerable development. They
shew themselves then first as insignificant processes which however grow rapidly, and in
adult specimens exhibit the circlets of spines, repeated at certain intervals around the dorsal
side of the basal section of the arm, so specially characteristic of this species.

Among the spines situated on the dorsal side of the disc, we may sometimes ob-
serve in very young specimens a spine here and there of a quite unusual form strongly
elongated and slender almost like a bristle. In the little young animal above described all
the spines seem to have been of this nature; but most of them were already, when the
animal was captured, more or less injured and some quite broken off. I succeeded however
with some certainty in getting an idea of their original appearance. They are, as has been
said, very strongly elongated and slender, quite hyaline and finely latticed like the ordinary
spines, but very distinguishable by not being smooth, shewing several consecutive distinctly
enlarged sections, each of which shoots out in a certain number of sharp points (fig. 34).
The whole spine is hereby distinctly echinulated at regular intervals from the very base;
while on the ordinary spines only the extreme end exhibits some few microscopic points.
These peculiar spines belong, as it appears, only to the embryonic state; they afterwards
disappear entirely, and are replaced by the usual spines which form themselves in the same
manner as the arm-spines.

g. Development of the pedicellaries.

On the dorsal side of the very earliest foundations of the arms, as also on the
cuticular sheaths of the spines which afterwards grow out of the arms, there appear, as has
already been noticed, a greater or less number of these peculiar small organs, but mostly
very far from being developed or in activity. We have thus here occasion to study the first

65

formation and further successive development of these organs. It is however by no means
necessary to depend only on such growing arms for opportunity of studying the develop-
ment of the pedicellarise. This investigation may be executed quite as well, and much more
easily, on any developed arm; as, throughout the whole life of the animal, there seems to
be a continual new formation of these organs, especially on the cuticular sheaths of the
spines of the arms.

On examining more closely the cuticular sheath of one of the long spines of the arm. it
will be found that the pedicellarire in this exterior part are all fully developed and of the struc-
ture above described; while those nearer the base become gradually more rudimentary, until at
last one can only observe a layer of simple cells surrounding the root of the spine (see fig. 31).
This simple stratum of cells is the starting point for the progressive development of the
pedicellarise from the base. In the interior of the cells the development of the 3 calcareous
particles above described can be very distinctly traced. They appear first (fig. 32) as 3
small widely separate granules situated in one row, and all nearly of the same size. Then
the 2 outside granules which represent the side-pieces develope themselves more rapidly
than the middle-piece, and soon assume a distinct triangular form (fig. 33) while the middle-
piece still appears only like a little indistinctly angular granule. 2 of the 3 corners of the
side-pieces now increase (see fig. 34) rapidly in length, developing themselves each in its
manner; so that it can soon be discerned which of them is to form the proper forceps, and
which shall become the lower plate-shaped process; but the 3rd corner is relatively less
developed, and only increases a little in breadth so as to form the vertical dentated edge
projecting inwards from the middle of the side-pieces. The middle-piece now begins also
to develope itself, by the extrusion in different directions of 4 cylindrical processes, 2 longer
and 2 shorter, whereby it acquires, when viewed from above, the form of an irregular cross.
In the subsequent development (see fig. 35) the 2 processes issuing from the side-pieces
increase more and more in length, bending inwards towards the medial line. They are
however quite simple, without teeth or plate-like enlargements; the upper one truncated at
the extremity; the lower, conically pointed. The upper process, or the proper forceps, soon
begins however (see fig. 36) to assume its characteristic spade-like form; and from each side
of the truncated extremity, there protrudes a sharp corner, forming the exterior tooth of
those which subsequently appear. The middle-piece now changes its form rapidly; small
processes beginning to grow from the lower side, and again connecting themselves by
their extremities, forming thereby the commencement of the perforated plates which
subsequently appear, and which serve for the insertion of the motor muscles of the side-
pieces. In the same manner the hitherto simply conical processes of the 2 side-pieces
afterwards begin also little by little (see fig. 37) to assume their definitive plate-like form;
as small processes issue from their exterior sides, and finally connect themselves by the
extremities, so as to circumclude the characteristic oval apertures, in the periphery of which

9

66

the before mentioned muscles are attached. All 3 pieces come while undergoing this change
closer together; and the articulation between the side-pieces and the middle-piece is effected
in the manner which characterises the developed pe'dicellarise. The changes described affect
only the calcareous particles which develop themselves in the interior of the pedicellarise, and
which therefore are to be considered as a secretive product of the plasma of the original
cell. The soft surrounding parts, or the cuticular sheath, are formed immediately by a simple
readaptation of the whole cell. Already in a very early stage of development, there may be
observed (see fig. 34), on the exterior side of the cell, a slight insinuosity like an incipient
division. This insinuosity becomes little by little deeper and deeper, and represents at last
a deep incision dividing the cell, into 2 broad lobes, each of which includes the anterior
extremity of one of the side-pieces. These lobes unite themselves, during the further deve-
lopment, directly with the 2 movable flaps of the pedicellaria. At the same time the whole
cell, formerly sunk in the skin, rises little by little above the surface, so as at least only to
remain attached to it by a thin continuation, which becomes the flexible stem whereby the
pedicellaria is fastened to the skin. When this stem is fully formed, then all the other
parts belonging to the pedicellaria have also attained their definitive development; and it
now begins to participate in the common function of all these small organs, to seize and
hold fast all objects that come in connexion with them.

B. Divergent development.

The phenomena of development noticed in the preceeding lines are such as are
common to all individuals, and generally do not only apply to the whole genus Brisinga,
but partly also to all the other star-fishes. We come now to a series of changes which
usually do not take place until the animal has attained a much more advanced age, and
which go in a more divergent direction. These changes manifest themselves as more or less
distinctly marked differences in the external form of the body, and in the particular details,
between the different individuals. We have again to distinguish the more .ordinary deviations
from those dependent on quite accidental causes, and therefore to be regarded as more
abnormal deviations.

a. Individual variations.

On examining a series of individuals, it will soon be found that the individual varia-
tions in the present species range within tolerably wide limits. The variations apply partly
to the number of the arms or ambulacra, partly to the form and arrangement of the diffe-
rent appendages belonging to the skin, and partly to the color.

67

Firstly as regards the ambulacra, the present species is in this respect distinguished
in a remarkable degree from the other species of Brisinga. While in this latter the number
of arms in all the specimens hitherto examined, has been always without exception 11, an-
swering to the specific denomination ,.endecacnemos", I have found in the specimens of the
present species examined by me a remarkable inconstancy in this respect; so that it is very
difficult to say what number should be regarded as the normal number.

Of the 22 more or less perfect specimens examined by me, 7 were 9-armed; 9, 10-
armed; 5, 11-armed, and 1, 12-armed; hence it appear that 9-armed, 10-armed and 11-armed
specimens are about equally numerous; while it seems that specimens with more than It
arms are very rare; since out of 22 specimens, only one was found with 12 arms. I do not
however consider it improbable that, on examination of a greater number, it may be found
that the inconstancy in the number of antimera ranges within still wider limits; and that
specimens may be found with fewer than 9 and more than 12 arms. The specimen deli-
neated by W. Thomson (1. c.) has thus, according to the drawing, no less than 13 arms.
If this is really correct, then the number of arms in the present species would vary at least
from 9 to 13; a very remarkable inconstancy in an animal of radiate construction. This
variation in the number of antimera or ambulacra, does not however appear to be the
result, as might be supposed, of a subsequent divergent development. On the contrary, I
have reason to presume that the number of arms in each individual is already fixed in the
embryo or larva body of the star-fishes, and continues unchanged during the whole life of
the animal; for I have never seen on the disc the smallest sign of any new formation of
ambulacra. All were, independently of the greater or smaller development of the arms,
exactly of the same appearance and uniform development. Neither are the younger indi-
viduals distinguished by having fewer arms than the older ones. The little tiny young Bri-
singa above described had 10 arms distinctly developed, just as in most adult specimens;
and of the 3 young individuals with a disc-diameter of 10 — 12mm, which I have had occasion
to examine, only one was 9-armed, while the other two were 10-armed.

As regards the length of the arms in proportion to the diameter of the disc, it is
far from being constantly the same in all individuals; nay even in one and the same indi-
vidual, we may often find a considerable difference in the length ot the arms, which in
great part is caused by the -regeneration of new arms taking place at various times,- in stead
of the older arms previously detached from the disc.

With respect to the number and arrangement of the calcareous transverse ribs on
the basal part of the arm, we find an infinity of variations; and even in one and the same
individual there will often be some difference on the different arms. The number in adult
specimens seems to vary at least from i> to 14. The number of the soft transverse bands
covered with pedicellaria.1 and arranged in the intervals between these calcareous ribs, is
likewise very variable. Sometimes there is only one (see Tab. II, fig. 2) sometimes — and

9*

68

this seems to be the most usual case — 3 (fig. 1); more rarely the number is greater, up
to 7; but in this case the transverse bands are usually interrupted in several places,
and rudimentary (fig. 4). In like manner the spines attached along the calcareous ribs
are very differently developed. Sometimes they are proportionally small, and in great
numbers (fig. 4) sometimes the number is more limited, in which case they usually
especially towards the sides — attain a very considerable length (almost like the exterior
furrow-spines).

Also the disc itself is subject to several variations. Its form is seldom perfectly
circular, but often broader on one side, and irregular in a greater or less degree. Its dorsal
spines are likewise in different individuals more or less developed, sometimes higher and
more closely packed, sometimes lower and more spread. Finally the form and size of the
madreporic body is subject to divers variations, even in equally developed individuals. It is
however always naked, and never, as in the other species endecacnemos, covered with spines.
The color of the animal varies, as already stated, through a series of tints, from light orange
to deep purple red.

b. Accidental variations.

(Monstrosities).

Under this head I would class 2 cases which I have observed, and in which one of
the arms in an otherwise normally and fully developed individual exhibited an appearance
very different from what is usual.

Both cases were evidently caused by accidental mutilation, and consequent regene-
ration of the arm. In one instance the arm was very short, scarcely half as long as the
others, but nevertheless in its basal section quite as much developed as the others, with
swelling organs of generation, and well developed calcareous ribs and spines; while the ex-
terior part of the arm was very slightly developed, and sharply distinguished from the rest
of the arm, being in the form of a little thin lash-like appendage with rudimentary spines
and water- feet. The arm had evidently been quite lately broken off at this place; and
from the wounded surface, there had sprouted a nascent new arm to replace the lost part.
I have also observed a perfectly similar instance in the Brisinga endecacnemos (see Tab. VII,
%. 1 a).

In the second case the arm was of the usual length, but highly remarkable for
having the extremity bifurcate, or going out in 2 nearly equally long branches completely
normally formed with all appertaining parts, and having at each end the terminal organ of
sense (Tab. II, fig. 3). The ambulacral furrow running along the arm divided itself, at the
issue of these two branches, precisely in the same manner as in a Pentacrinus or an An-
tedon; and the ambulacral skeleton seemed likewise to have been similarly affected. That

69

an accidental mutilation of the arm had also in this case been the disturbing cause, is
undoubted; but, instead of the single regeneration of the lost part, the reproduction has
taken place in a double form; so that from the wounded surface there have sprouted forth
2 separate nascent arms, which have been further developed each for itself. Monstrosities
analogous to these have already been observed in some exotic star-fishes, the Oreaster
gigas and Astropecten aurantiacus.

70

VII:

Distribution and occurrence,

(Chorology).

With regard to the distribution and occurrence of the present species, much valu-
able information has been obtained by the great English expeditions, that may in connexion
with my own observations, help us to form tolerably accurate notions on this subject.
It appears to be everywhere a true deep-sea form, and, like most deep-sea animals, to
have a very wide geographical distribution.

a. Horisontal distribution.

As already noticed, I have hitherto found the present species of Brisinga, on our
coast, only on the fishing ground Skraaven in Lofoten in about 68 V2 ° N. Lat.

This point is at present its most northern limit; although I do not doubt that on
further investigation it will be found in the arctic zone much further towards the pole.
Its distribution in a southern direction extends at least to the coasts of Ireland, where it
was taken during the expedition of the ., Porcupine" in the year 1869. Already in the previous
year it had been found during the expedition of the ,,Lightning" in a much more northern
locality, namely between Scotland and the Fseroe islands. Its distribution as at present known
thus extends through no less than 15° of Latitude. The other species Br. endecacnemos seems
to have a still greater distribution. It was found during the 2nd expedition of the ,,Porcupineu
off the coast of Portugal. Its most northern locality is Flor0 13 miles (91 miles Eng.) north
of Bergen, where 2 years ago I took up a perfect specimen. According to this different di-
stribution of both species, it appears that there is some ground for supposing the Brisinga
coronata to be on the whole a more northern form than the Br. endecacnemos, which also
seems to be confirmed by the physjcal conditions under which both species are found.

71

b. Vertical distribution.

The specimens which I have examined of the present species of Brisinga were all
taken from about the same depth, namely 250 — 300 fathoms. Nevertheless it goes, as has
been proved, considerably further down into the deep. In the Fseroe channel it was taken
form a depth of 500 fathoms ; and off the west coast of Ireland it was found at the depth
of even 800 fathoms. In all probability 200 fathoms must be considered as its highest
limit. Whether it goes even lower than 800 fathoms, is not yet certainly decided; but yet
it seems to me very probable that it goes at least as deep as the other species Br. ende-
cacnemos, which off the Portuguese coast has been taken in 1000 fathoms. The highest limit
for this last species seems to be about the same as for Br. coronata, namely about 200
fathoms.

c. External conditions of existence.
Special occurrence.

All my specimens of Brisinga coronata were taken at a considerable distance from
the shelving bank of the coast, on an even and flat bottom covered with a more or less
thick layer of clay or mud. It appears to have been taken under the same conditions
during the atlantic expeditions; but the occurrence of the other species seems to be very
dissimilar. It has been always found by Asbj0rnsen, as well as by myself, only nearer to
the coast on the steep bank which rises from the deeps "outside, where the bottom is rocky
or stony without any deposit of mud or sand. Whether this dissimilarity in occurrence
between the two species is the same everywhere, I do not venture to say; as this is not
sufficiently clear from the reports of the atlantic expeditions which we possess. But on our
coast the conditions appear to be perfectly constant. The temperature in which the Br. coro-
nata lives at Lofoten is, according to the investigations made by me in the summer of 1869,
about -f- 4 ° C.; but it is necessary to remark that I had at the time only a very imperfect
deep-sea thermometer at my disposal. It is thus very possible that the temperature may
have been in reality somewhat lower, which would also agree with the statement of tempera-
ture communicated in the report on the „ Lightning's" expedition. It occurred, according to this
report, only in the cold area in the Faeroe channel, where the temperature, at the depth of 500
fathoms, was only a little above the freezing point.. Off the west coast of Ireland it seems
however to have been taken by Jeffreys in a much higher bottom-temperature. It appears never-
theless from other circumstances of the animal's occurrence here, that the Fauna exhibited on
the whole a decidedly northern character. In Lofoten it was found together with Astropecten
Andromeda and tenuispinus, Echinus norvegicus, Stichopus natans, an enormous number of
Ulocyathus arcticus. various Molluscs, among which Admete viridula, Trophon barvicensis,

72

Dentalium agile and Nesera obesa were the most numerous, and with various Annelides and
Crustacea. It occurred chiefly in a single confined area, at the distance of about 1 Nor-
wegian mile from land, but here in quite unusual abundance. In one single haul I once got
up in the dredge no less than 5 more or less perfect specimens. As often as I removed a
little from that place, which could only be found with great trouble, by reason of its great
distance from the shore, the specimens became immediately very scarce. Whether this
particular spot on the bottom of the sea offered any specially favorable points for the thri-
ving of the species, I have not been able to Ascertain. I should be more inclined to pre-
sume that this congregation of individuals was only temporary, and mainly to be considered
as connected with the propagation of the species taking place just at that time.

73

VIII.

Remarks on Homology and Affinity,

(Philosophy).

A. General Homology.

i.

On the scientific significance of the genus Brisinga, considered from the stand-point of the Darwinian theory,

When we now, after having gone through the natural history of this remarkable
star-fish, finally review the most important features of its organisation, we shall certainly
be forced to acknowledge that its structure is in reality by no means so entirely anomalous
as its very peculiar external appearance might induce us to suppose, and as it has indeed
been hitherto considered; for we can now, without any hesitation whatever, assign to it its
systematic place in the order of the Asteriae or star-fishes, without being forced to regard
it as a connecting link between the Asterise and the Ophiura. Nevertheless this star-fish
exhibits in its structure some highly remarkable and significant peculiarities, which can not
but lead us to several reflexions on the Asterides in general, and on their relation to other
Echinoderms; and we shall hereby come quite naturally to the important question of the
relation of the whole type of Echinoderms to other animal types. It is also now my opinion
that this remarkable star-fish opens to us on many points a clearer view of the nature of
the Echinoderms than we can hitherto have had. Still it is, here as everywhere, only by
Darwin's reformed theory of descent, that these reflexions can bring us to more general,
and therefore also in a scientific point of view more important conclusions. Without Dar-
win's theory we should only see in the Brisinga an abnormally developed star-fish, which in
a remarkable and inexplicable manner seems to depart from the conventional type of Echi-
noderms (the radiary hemispherical form) thereby, as it were, forming one of the extreme
points of possible modification within the limits of the type. In regarding the matter from
the stand point of the Darwinian doctrine, our reflexions take quite a different direction and

10

74

a wider range; and we come also to a very different result with respect to the relation of
the genus Brisinga to the other Echinoderms. With Darwin's doctrine in view, we can not
imagine the type of Echinoderms as anything perfectly defined and given from the very
first; but we must imagine this type, like the other higher animal types, as having been
produced through successive divergent developments of lower animal forms. We have there-
fore first and foremost to examine the relation in which the type of Echinoderms stands
to the lower animal forms known to us, and seek to discover which of these may be con-
sidered as the starting point for the series of developments observed in the Echinoderms.
To settle this important point, it is first expedient to decide which of the now living Echi-
noderm-forms may be presumed to have undergone the least change, and therefore may be
considered as the oldest or the most original forms. For the solution of this question
Palaeontology will naturally furnish the most important data. The first Echinoderms which
occur, are those in which we may expect to find the phylogenetic origin of the type best
illustrated. For a long time it has been supposed that the Crinoidse or stone-lillies were
the oldest Echinoderms, and especially the so-called Cystidea which already appeared in
great number in the palaeolithic time. It appears however on more recent examination, that
the Asterides are of still much greater antiquity, for which reason we may also consider
this group of Echinoderms as the eldest, from which all the other groups of Echinoderms
have been developed. Of the now living Asterida3, those forms may be considered as the
eldest or least changed, which in their habitus agree most with the first forms of remote
antiquity. This is now precisely the case with the Brisinga, in a greater degree than with
any of the other known star-fishes.

Its habitual resemblance to the oldest known star-fish Protaster is unmistakable;
and Asbj0rnsen has already drawn -attention to this important point. Several other circum-
stances will now confirm us in the opinion that we have before us in the Brisinga a very
ancient form, an isolated surviving representative of the Echinoderms of primitive times,
which, confined to the great depths of the ocean, where the physical circumstances have
been in all times somewhat similar, has managed in undisturbed peace to preserve the
original structure, without requiring to keep pace with the remarkably extensive transfor-
mation and diversity of development observed in the Echinoderms generally, and specially
in the Asterides. Now the Asterides being, as already stated, considered as the oldest
race of Echinoderms, from which the other types of Echinoderms have only been ramified
in later' times, the genus Brisinga acquires a double importance, as the starting point for a
correct appreciation of the nature of the Echinoderms and of their relation to other ani-
mal types.

75

2.

Of the fundamental form of the Echinoderms and of their morphological individuality as illustrated

in the genus Brisinga.

As is well known, Cuvier referred the Zoophytes and Echinoderms to one and the
same animal type, the so-called radiates; and he has been also followed in this respect by
the American naturalist (Agassiz) while here in Europe. We have generally adopted the
views first developed by Leuckart, according to which both these groups are sharply distin-
guished from each other as belonging to essentially different original types. It was the
radiary primitive form presumed to be common to both groups, which Cuvier laid so much
stress on; and the American naturalists have also held this characteristic to be of such pro-
minent importance, that they still consider the Zoophytes and the Echinoderms, in spite of
the many essential differences in their organisation, as belonging to one and the same ani-
mal type. If however we follow the method of investigation above indicated, according to
which we must take the Asterides, and among these especially the genus Brisinga, as the
most original or least changed form of star-fish, for our type, and from this point of view
institute our comparison between the Zoophytes and the Echinoderms, we shall find that
in reality there is a very essential difference between the strictly radiary structure of the
former and the so-called radiary structure in the star-fish, a difference which is of so vital
importance that we cannot do otherwise than reopen the question which might seem to
have been settled long ago, namely whether the Echinoderms, when all is considered, can
be held to be real radiates in the same sense as the Zoophytes. If we now examine the
so-called rays or arms in the proper star-fishes, from which again all the other Echinoderms
may naturally be derived, we shall find at first that they exhibit a far greater self-sufficiency,
a far more self-contained organisation than we ever can find in the Zoophytes, a fact which
in the Brisinga is more prominently evident than in any of the other known Echinoderms.
While in other Asterides there is always found a more or less strongly developed central
section or disc, this is in the Brisinga so extraordinarily reduced, that the whole body may
be said to consist of a certain number of arms connected at the base. If we further con-
sider a ray or arm of a star-fish by itself, we shall find that besides exhibiting a perfect
bilateral symmetry, it consists of a series of consecutive joints or sections (metamera), a
division which is not only expressed in the ambulacral skeleton, but also in most of the
other organic systems. If we as formerly, consider the arms of the star-fishes to be real
corresponding parts (antimera) analogous to the antimera of Zoophytes, we come to the
paradoxical result that an antimeron can be composed of metamera, which according to
Hiickel l is the next highest order of individuality. On the other hand the matter is explained

1 Generelle Morphologic der Organismen.

10*

76

in a natural manner, if we regard the rays of the star-fishes not as real antimera, but as
morphological individuals of the 5th order, or persons.

The whole star-fish may be thus properly regarded as a colony (cormus) a collection
of articulated persons which have arranged themselves in a radiary form round a com-
mon centre, and here formed for themselves a common ingestive aperture, in the same
manner as the individuals of the Botryllus colonies arrange themselves radially, and form
for themselves a common egestive aperture. But as in the case of other colonially orga-
nised animals, we do not define the original form according to the form of the colony itself,
but according to the structure of the constituent individual animals or persons, so we come,
in the case of the star-fishes, to the result that the fundamental form of these animals is not
the radiate form, but the bilateral symmetrical (eudipleural), and that the apparently radiate
structure is of secondary origin produced by concrescence of a number of eudipleural per-
sons arranged around a common centre. In this manner the last reason would fall away
for the union (in itself so unnatural) of the Zoophytes and Echinoderms under one and the
same animal type. We might as well consider the star-shaped Botryllus colonies among the
Ascidise as radiate animals; which probably no one at present would think of doing.

If we accept the above theory (first clearly formulated by Hackel) of the compound
structure of the Echinoderms, as properly representing complexes of animals or cormi, it
must be admitted that this cormus in the Brisinga is less centralised than in any other
Echinoderms; the arms (in this case according to Hackel persons) • • as well by their con-
siderable size and loose connexion with the insignificant disc, as by their possessing besides
the other organic systems, also a perfect apparatus of generation with corresponding geni-
tal apertures — exhibiting an independence that is without parallel in any other Echinoderm.
In examining a Brisinga it will therefore be much easier to recognise a real colony or
complex of animals, than in examining most other Echinoderms, where the already far
advanced centralisation of the cormus undeniably makes such recognition in many cases
very difficult. The preponderance of the arms over the insignificant disc, is in the Brisinga
so great, that the conception hitherto prevailing of the disc in star-fishes as representing
the principal part of the body, of which the arms are only radiating expansions, must here
be completely reversed; as we must necessarily consider the arms in the Brisinga as the
principal parts, to which the disc only stands in the relation of a sort of appendage. It has
also been previously stated, that I have been able to convince myself by direct experiments,
that the single arms of the Brisinga, detached from the disc, continue to live and to exer-
cise their ordinary vital functions, even long after the disc itself has ceased to live; and
that there is likewise a very great probability for their being capable, under favorable cir-
cumstances, of continuing their life, each for itself, and little by little reproducing the other
parts belonging to a complete colony, the disc as well as the rest; I have even felt bound
to state, as something which has the highest degree of probability for it, that such a sue-

77

cessive detachment of the arms takes place in the normal state as a voluntary act of the
animal, conducive to non-sexual propagation (by divisio radialis).

There is another feature in the species of Brisinga here noticed which is worthy of
remark, as essentially supporting the theory above developed; and that is the contrast to
the real radiates (Zoophytes), in the great inconstancy of the number of the rays (arms),
which even makes it very difficult to decide what number shall be considered as normal or
typical. A similar, although certainly not so extensive, inconstancy is found only in some
few forms within the group of star-fishes, which we must consider as the oldest and most
original Echinoderms. Thus the number of arms varies not a little in our two species of
Solaster, as also in several exotic forms of the genus Asterias, Luidia, Ophidiaster. In most
of the star-fishes however the number of arms has with the increasing contralisation of the
cormus become fixed at the number which is typical for the Echinoderms, namely five. The
same is the case with the Ophiurae, which stand next to the star-fishes, where in like man-
ner the number five, with very few exceptions (some species of the genera Ophiacantha,
Ophiactis acd Ophiothela with 6 arms) is completely constant; and in the 2 most divergent
groups of Echinoderms, Echinoids and Holothurians, in which the centralisation of the body
has attained its highest degree, not a single instance is to be found of deviation from the
normal number of five ambulacra. It is also only in these groups that, collaterally with the
complete concrescence of the single individuals (persons) originally composing the body, new
and peculiar relationships appear, which seem still more to disguise the original polymerous
composition. Thus we see in the so-called irregular Echinides (Spatangus) and in divers
Holuthuriffi (Psolus), that there proceeds from the whole cormus, by means of one of the
sides acting as a sort of creeping disc, (the ventral side) an apparently single bilaterally
symmetrical individual; and in the footless Holothurise (Synapta) where proper ambulacrals
are wanting, the last remnant of the complex character has disappeared, and the whole
body seems even to return to the lowest (monaxonous) fundamental forms. If we consider
one of these anomalous forms by itself, it will certainly be difficult to recognise in it any
-real complex or cormus. But if we go through the whole series of Echinoderms, we shall
soon see that there can be no question of any other than phenomena of secondary adaptation;
and these abnormal relations can therefore by no means disprove the above noticed theory
of the original composition of the body of the Echinoderms, as derived from the oldest and
most original forms of star-fishes.

78

3.

On the phylogenetic relationship of the Echinoderms to other animal types.

If we accept the theory noticed in the preceding paragraph with respect to the compo-
site individuality of the Echinoderms for which, as will be shewn, the examination of the Bri-
singa seems to afford essential support, we shall quite naturally come to judge of the generic
relationship of the Echinoderms to other animal types in a very different manner from here-
tofore. We thus see in' the star-fishes the most original and least altered Echinoderms, which
therefore must be selected as the starting point for our comparison with other animal types.
Moreover it is not the whole body of the star-fish, but the single rays or arms which must
here form the subject of our investigations; as these rays or arms do properly represent the
original echinoderm individuals or persons. We have in the first place before us a comple-
tely bilaterally symmetrical body, with a dorsal and a ventral side, right and left side, oral
and aboral extremity, the interior organs of which shew the same strict symmetry. Next
we shall find that these internal organs have a fixed regular and peculiar position relatively
to each other: below, the central parts of the nervous system; then, the vascular system
with the parts appertaining to it; then, the perivisceral cavity with the digestive system and
organs of generation contained therein; and externally, the skin, consisting of 2 distinct
layers, the interior muscular, and the exterior cellular. We find moreover that this body
exhibits a series of consecutive similar sections or metamera, not only expressed in the
skeleton but also in several of the interior organic systems: the muscular system, the
nervous system, the vascular system. These are 'all things which we only find again in
the great race of the articulata and particularly in the worms (Vermes). We are naturally
led hereby to the conclusion that the nearest relations of the Echinoderms are not. as
hitherto generally supposed, the Zoophytes, but the Vermes; and that this last ancient and
extensively ramified trunk, in which we trace, as it were, the very first rough sketches of
all the higher animal types (even including the vertebrata) must also be regarded as the
trunk from which the Echinoderms, although apparently of a very different fundamental
structure, have had their origin. This theory of the phylogenetic development of the Echi-
noderms, which is likewise most sharply and most clearly represented by Hackel, acquires
also an essential support from the Brisinga, the so-called rays or arms of which exhibit
more evidently than in any other Echinoderm a worm-like appearance, and the great self-
sufficiency of which is not only expressed in the exterior, but also in the interior organi-
sation; for, as has been already observed, even the organs of generation, which else, at least
in great part, are confined to the central section or disc, are here entirely separated from
the same, and symmetrically imbedded in the cavity of the arms. The relationship of the
Echinoderms to the Vermes has indeed long ago been recognised by some naturalists;

79

although certainly other views have partially prevailed. That some Holothuriae (Synapta) in
their exterior habitus, have a striking resemblance to worms, is a well-known fact, and has
given rise to the popular denomination ,,worm-cucumbers". We have likewise a peculiar
class among the Vermes, the so-called trunk-worms (Gephyrea), the organisation of which
exhibits several remarkable points of resemblance with the Holothuriaus, and therefore
also has formerly been referred to the type of the Echinoderms. Since however, as above
stated, we have in the HolothuriaB really the most divergently developed group of Echino-
derms, it cannot actually be here that we should seek for the original ancestry of the Echi-
noderms ; although it can not be denied that we may also here observe a remarkable retro-
gression to the same type which may be assumed as basis for the oldest and most original
Echinoderms, the star-fishes.

In opposition to the theory above stated as to the phylogenetic relationship of the
Echinoderms to the Articulata, the ambulacral or water-system has been adduced as presum-
ably exclusively peculiar to the Echinoderms. It can certainly not be denied that this organic
system affords, by its extremely peculiar and complicated development and functions, one of
the most distinctive characteristics of the Echinoderms; but it is by no means on that
account decided that nothing of the kind is to be found in other animal types. We have
frequent instances of one and the same organic system developing itself in different groups
of animals in a very different manner, and exercising quite dissimilar functions; so that the
originally common fundamental form may be difficult to recognise. We have in the Vermes
a very extensive organic system in the form of tortuous canals, opening partly externally
partly in the cavity of the body, and appearing decidedly in certain cases to be water-ducts;
while in other cases they have secretory functions. Where the body is evidently devided
into segments, these vessels appear to be arranged symmetrically in each segment, and
have therefore in the annelides been indifferently denominated segmental organs. It may
be assumed that we have here the analogon of the ambulacral system in the Echinoderms;
at least it appears to me that there cannot be adduced any decided proof to the contrary.
There is another characteristic which seems to be foreign to the Vermes, and which is so
prominent in the Echinoderms that even the whole type has derived its name therefrom,
namely the cuticular skeleton more or less strongly developed by calcareous secretions, to
which we may add that in the Astericles and Crinoides there is even another sort of interior
skeleton which exhibits a so striking habitual ressemblance to the vertebral column in the
Vertebrata that the single segments even bear the same name (vertebrae). In examining
the real Vermes we certainly do not find anything analogous; but this characteristic is by
no means foreign to the whole tribe of Articulata. In the Crustaceans we have frequent
instances of the integuments, by absorbing lime, assuming quite as firm a consistency as in
the Echinoderms; and by interior processes of this cuticular armor, there is also formed
here (Decapods) a sort of interior skeleton, which serves partly for the insertion of the
limbs with their muscles, and partly also as support for interior organic systems (the ner-

80

vous system). That the ambulacral skeleton in the Asterides, in spite of its habitual resem-
blance, has no analogy whatever with the vertebral column in the Vertebrata, but properly
belongs to the cuticular skeleton - - is satisfactorily ascertained, and is clearly evidenced
(inter alia) by its development being perfectly conformable to that of the cuticular skeleton.
We have not yet touched on a point which, just where there is a question of phylo-
genetic relations, must occupy a very important place, and will often alone be able to give
us the most certain and significant indication in this respect; I mean the history of deve-
lopment. We shall find also in considering this important part of the natural history of
the Echinoderms, an essential support for the theory above noticed; at the same time as
many hitherto quite unintelligible points connected with the subject will, by help of this
theory, find a natural explanation. As is well known, most Echinoderms go through a most
remarkable and peculiar metamorphosis, which has first been made the subject of particular
and minute research by the celebrated German naturalist Johannes Miiller. From the egg
there proceeds a creature of most romantic appearance, furnished with various lobes and
processes, swimming away by means of sinuous bands of cilia, and having no resemblance
whatever to the respective Echinoderin, being a perfectly bilaterally symmetrical animal,
which according to its whole structure would most naturally be referred to the group of
Vermes, and which also exhibits a striking resemblance to the so-called larvae of certain
Nemertina and Gephyrea. Only at a later period, there is laid, in a limited part of the
interior of this worm-like animal, the foundation of the future echinoderm; and as the latter
is developed, the original so-called larva-body shrivels up little by little or becomes resor-
bed. There is much to forbid considering this peculiar process of development as a real
metamorphosis. It is indeed usual for the whole larva-body to be resorbed, or as it were
taken up in the formation of the Echinoderm; but we have still instances of this not being
always the case. Thus the Bipinnaria-larva continues to swim about even after the young
star-fish has detached itself; and it may even be imagined possible that such a larva can
again give origin to a new star-fish. This instance alone is sufficient to distinguish very
sharply the development of the Echinoderms from the ordinary development by metamor-
phosis. We must on the contrary bring it under the great law of alternate generation, as
a peculiar modification of the same, chiefly characterised by the cycle of generation being,
as far as we yet know, only represented by 2 Bionta. It is this form of alternate gene-
, ration which has been called by Hackel, Metagenesis successiva. We have consequently in
the peculiar so-called Echinoderm-larva properly speaking not an incomplete stage of deve-
lopment, but one of the 2 fully developed alternating generations, namely the non sexual
generation. The apparently paradoxical and unintelligible fact that the 2 Bionta in the same
cycle of generation are constructed in totally different fundamental forms, one in the racliary
and the other in the bilaterally symmetrical (eudipleurous) form, can only find its natural
explanation, if we consider the developed Echinoderm as an individual of a higher order, or
a colony, the apparently corresponding parts or antimera of which represent the original

81

individuals (persons). Each of these is then, as it were, the individual which represents the
sexual generation, constructed according to the bilaterally symmetrical fundamental form,
and, as it were, most naturally referable to the class of Vermes. The so-called develop-
ment of the Echinoderms in the body of the larva, must then be considered as a sort of
budding of several worm-like individuals, which arrange themselves radially round a common
centre, and, by concrescence of their oral extremities, form an individual of a higher order,
a colony or a complex of animals. A rather analogous case has already long been known
in the Salpse, in which similarly the 2 alternating generations represent morphological
individuals of different orders ; the non-sexual generation always consisting of single indi-
viduals (persons), which again, by budding, produce in their interior coherent complexes of
individuals or colonies (the so-called Salpae-chains) representing the sexual generation.
If we examine the manner in which an arm or ray is reproduced de novo in an otherwise
developed star-fish, we shall similarly find cases which decidedly remind us of the Vermes.
The various joints or metamera are formed, as already noticed above, successively from the
base of the extreme first-formed joint, exactly in the same manner as the joints of a tape-
worm-chain, or like the segments of a nascent annelide. We come thus, in considering the
history of the development of the Echinoderms in general, to the same result to which the
minute investigation of the organisation in a fully developed star-fish has already led us.
The theory already set up by Duvernoy, afterwards developed in greater detail by Huxley,
and recently finally elucidated by Hackel, as to the composite individuality of the Echino-
derms and their phylogenetic connexion with the Vermes, — which theory was formerly
generally regarded as a wild fancy — must in reality, on more minute examination, be recog-
nised as having very much to support it; and so much the more, as we hereby obtain a
quite unexpected explanation of some hitherto very obscure cases in the natural history of
the Echinoderms.

4.

On Ihe genealogical relation of the several groups of Echiuoderuis to each oilier.

We have above considered the proper star-fishes as the oldest, or original Echino-
derms, from which therefore all the other Echinoderms may be derived. We are led to
assume this in advance, quite simply by the fact that the star-fishes are, so far as we know,
of all the Echinoderms those which go furthest back in time, namely to the sub-silurian
formation. This assumption is also in the best harmony with the above-noticed theory of
the composite individuality and phylogenetic relations of the Echinoderms. If we consider

11

82

the Echinoderms as original cormi or complexes of vermicular individuals, it is clear that
those forms must be the oldest or least changed, in which this cormus exhibits the least
degree of centralisation, and in which therefore the component individuals (persons) evidently
shew their original selfsufficiency as such. This is precisely the case with the star-fishes,
the rays or arms of which give, each for itself, a complete image of the whole organisation.
Among the star-fishes again, it is the genus Brisinga which in this respect most distin-
guishes itself; for which reason we must thus consider this form as the oldest or most
original of all the Echinoderms. To prevent misunderstanding, it must be expressly remar-
ked that this refers chiefly to the general combination of the body, not to the organisation
in its details. In this respect the Brisinga is in exact conformity, as has been shewn, with
the star-fishes now living; and it cannot fairly be assumed that these have been able to
preserve unchanged the self-same organisation of the ancient worm-like creatures from which
the whole tribe of Echinoderms proceeds.

In the genealogical development of the tribe of Echinoderms, which may be sup-
posed to have progressed uninterruptedly in the immense periods during which the animals
hereto belonging have existed upon the earth, there appears chiefly to be manifested a
tendency towards greater and greater centralisation of the originally independent individuals
composing the cormus. The Echinoderms in which this has been in the highest degree
successtul are undoubtedly the Holothurians; and these must therefore be considered as the
newest, most divergently developed and most altered Echinoderms; this conclusion is also
corroborated by Paleontology; as we do not find any traces of these animals until the Jura
period; while all the other groups of Echinoderms are referred to far earlier times. Between
the 2 extreme points, the Asterise and Holothurise, there are a great number of in some
cases widely divergent series of developments, all of which may however be naturally refer-
red to the Asterise as the proper fundamental forms. In examining the various groups of
Echinoderms, we shall find that the said tendency in the cormus, after this is centralised as
an independent physiological individual, takes 2 courses, both leading to the same end,
although in rather different manners; both are already partially indicated in the group of
Asteroidea, but only come to their full manifestation, on the one side in the Echinoids and
Holothurians, and on the other side in the Ophiurans and Crinoids.

One course consists in more or less extensive concrescence of the individuals (the
rays, antimera) originally connected only by their oral extremities, whereby the central
connecting part or disc gains in circumference, as it were, at the expense of the arms,
which at last seem entirely to disappear or to be absorbed in the disc. This change is
manifested chiefly in the exterior, without, at least in the commencement, being accompanied
by any corresponding alteration of the interior organisation. Only at a later period there
occur other alterations, chiefly those connected with the digestive system, the main parts
of which, originally belonging to each single individual (person), alter being thus brought
mechanically nearer together, at last undergo a transformation more convenient for the ali-

83

mentation of the whole colony. Among the known star-fishes now living, we have numerous
instances of such a centralisation manifested in the exterior. Between the forms provided
•with long arms and particularly small discs (Asterias, Pedicellaster &c.) and the short-armed
star-fishes (Porania Pteraster) in which the arms are only insignificant processes of the power-
fully developed disc, we have all possible transitions. At last the arms appear to be en-
tirely wanting and only indicated by the pentagonal form of the disc (Goniaster) nay, in
the genus Culcita even the angulosity of the disc is indistinct; and we have here before us
a disc-like body in which only the 5 ambulacrals still indicate the original arms or per-
sons. The further change next consists in an incipient diminution of the antiambulacral
area, whereby the ambulacral area, originally confined to the ventral side, gains a constantly
greater and greater extension over the surface of the body; until at last the antiambulacral
area is still only visible as an insignificant space on the superior pole. In the periphery of
this space, there lie, in the regular Echinoids, the 5 peculiar ocellary plates with their pig-
mentary spots, exactly corresponding to the organs of sight situated in many star-fishes at
the extremity of the arms, whence again it appears that we have in the ocellary plates* the
counterpart of the extreme arm-joints of the star-fishes. In several Ecbinoids the body exhi-
bits moreover a very evident pentagonal form as indication of the original 5 arms. In like
manner we have among the Echinoids many forms in which the body is quite as disc-like as
in the star-fishes. Between these flat Echinoids and the high hemispherical, including the
pyramidal forms, we have all the transitions. If we imagine the high pyramidal form still
further developed in the same direction, it is not difficult to derive the cylindrical form,
which we find in the Holothurians. Here also the originally ventral (ambulacral I side occu-
pies the whole surface of the body; and of the antiambulacral area every trace has dis-
appeared. The peculiar development of the water-feet situated nearest to the mouth of the
Holothurians into capturing arms, is, like the complicated dental apparatus of the Echinoids,
of secondary origin; just as we can also see in the peculiar apparently bilateral symmetry
of the irregular Echinoids and of certain Holothurians (Psolus) only formations determined
by particular processes of adaptation.

The other course of development, which likewise leads to considerable centralisation
of the cormus, is not properly speaking accompanied by any extensive concrescence of the
original individuals (arms, rays) while it is here essentially an internal transformation which from
the first takes place, a dislocation of the internal organic systems, whereby their principal
parts retract themselves, as it were, from their original place in the arms to the centre of
the colony, or to the disc, which hereby becomes the most essential and most organised
part; while the original individuals, the arms, are reduced to the rank of mere organs. The
organic system first acted on is also in this case the digestive system, the most important
really digesting sections of which in the star-fishes have still their original place in the
cavities of the arms as the so-called radial caeca, but which, in the course of development
here treated of, centralise themselves little by little in the cavity of the disc itself. This

11*

84

becomes possible by means of a peculiar development of the antiambulacral side of the
disc which hereby, quite contrarily to what is the case in the other course of development,
acquires a preponderance over the ambulacral side; a phenomenon especially manifested in
the Crinoids, in which the so-called calix, with the articulated stem proceeding from the
same, is attributable to such a development. In connexion with this centralisation of the
digestive system, there stands also a centralisation of another organic system, namely the
blood-system; as the cavities of the arms become at the same time narrower and narrower;
so that the perivisceral blood-cavity is likewise, for the greater part, limited to the disc.
We may already see evident signs of the centralisation referred to in real star-fishes, namely
in the remarkable fossil form Protaster; but we do not see the full development of it until we
come to the Ophiurans and Crinoids, in which the arms, by the above-mentioned reduction
of 2 of the most important organic systems, have lost so much of their original indepen-
dence, that they have become mere appendices of the disc, or simple organs. The greatest
centralisation of the cormus attained in this manner, is probably exhibited in the Ophiurans
and- in their relatives the Euryalae, in which one more important organic system has aban-
doned its original place in the arms and centralised itself in the cavity of the disc, namely
the generative organs. But in the Crinoids these organs are still quite separate from the
disc; for which reason we must also consider these Echinoderms as less completely centra-
lised than the Ophiurans, and therefore of more ancient origin, which is also in some degree
indicated by the rather variable number of arms in the sea-lilies; and we are likewise led by
Paleontological deductions to the same conclusion; as evident Ophiurans do not occur until
long after the Crinoids; although the course of development which seems to tend towards
the ophiurean type is found manifested at a very early period, namely in the ancient star-
fish Protaster. The articulated stem peculiar to the Crinoids must be assumed to be a
characteristic acquired at an early period in consequence of special conditions of existance,
and continued in the younger stage even in the Antedon, which is free when fully deve-
loped. The genus Antedon must therefore, in spite of its habitual resemblance to the
Euryalse among the Ophiuridse, be considered as a further developed divergent branch of
the genuine Crinoids.

It is thus evident that in the first course of development, which we are able to trace
through the groups of star-fishes, sea-urchins and sea-cucumbers, the centralisation of the
body is chiefly manifested in a reduction of the arms brought about by concrescence. In the
other course of development we have on the contrary an instance of even further develop-
ment of the arms, without however any decentralisation of the cormus taking place. This
further development of the arms may consist partly in an increased reproduction of joints
or metamera, whereby the arms may often attain a really extraordinary length in proportion
to the diameter of the disc (for instance in the genera Amphiura and Ophiopeltis belonging
to the Ophiuridse) and it may. consist partly in a ramification of the arms, which may either
be a more or less extensive dichotomic splitting, or a development of alternating lateral

85

branches (Pinnulse); of both these sorts of ramification we have instances in the Crinoids,
and of an extremely extensive dichotomic splitting in the Euryalse. In these last the arms
may be said to have attained a stronger development in this direction than in any other
Echinoderm. Nevertheless such a Euryala with its widely ramified arms, the mass of which
surpasses many times the mass of the disc, is far more centralised than even the most
short-armed star-fishes. As regards the fossil groups Blastoidea and Cystidea these, are
most nearly related, as divergent branches, to the Crinoids. The arms have here apparently
undergone, by some peculiar processes of adaptation, a retrograde development. We may
perhaps so far most properly consider them as degenerate Crinoids, just as the footless
sea- cucumbers (Synapta) may be supposed to be degenerate Holothurians.

The genealogical relationship of the various groups of Echinoderms to each other
may henceforth, with special r,egard to the various modes of centralisation of the cormus,
be graphically represented as follows:

Centralisation by continued concrescence with reduction of the antiambulacral area.

Centralisation without
continued concrescence, with special
development of the antiambu-
lacral area.

I

(Pentacta)

Holothurida

(Eupoda).

Centralisation without

continued concrescense, with special

development of the arms.

Antedon.
(Pentacrinus).

(Palseechinns).

Euryalae.
(Amphiura).

C r i n o i d a .

(Articulata).
(Archasterias?).

E c h i n i d a .

(Dermostichaj.

(Asterias).

O p h i u r i d a ,
(Ophiura).
(Protaster).

A s te r i d a.
(Brisinga).

Colony-forming
Vermes.

5.

Of the relation of the genus Brisinga lo the Asleroidea in general.

After having made the above more general remarks on the organisation of the Echi-
noderms and their relationship to other animal types, as exemplified in the genus Brisinga,

86

we return more specially to this genus to examine the relation in which it stands to the
other known forms of the class Asteroidea. As is well known, the general view has hitherto
been that the genus Brisinga formed the type of an entirely separate group or order of
Asteroidea (Brisingastra Hackel) which as it were, formed the transition between the Asteridse
and the Ophiuridae. And to this conclusion we may really be easily led by considering the
animal's exterior habitus. But we shall now, after having become better acquainted with
the organisation of this remarkable form, be forced to come to an entirely different result.
Its relationship to the Ophiuridse is in reality very distant; while its conformity with the
proper star-fishes (Asterida) in all essential characteristic points is so great, that in spite
of its anomalous appearance, it must take its place in this order. In illustration of this
I will here briefly point out the characteristics whereby it connects itself with the Asteridae
and distinguishes itself from the Ophiuridse.

1) The considerable and variable number of arms in the Brisinga has only its parallel in
certain Asteridae; while the number of arms in the Ophiuridse, in conformity with the
advanced centralisation, has mostly already become permanent as the typical number
for the Echinoderrns 5.

2) The ambulacral skeleton is in the Brisinga in its main features constructed exactly
according to the same type as in the proper star-fishes and differs essentially from
the ambulacral skeleton of the Ophiuridae.

3) Also the cuticular skeleton shews a much greater conformity with the Asteridse than
with the Ophiuridae, and especially the connecting calcareous ribs over the basal sec-
tion of the arms are entirely foreign to the Ophiuridse; while it is easy to recognise
in them the analogon of the wide-meshed calcareous net found in the skin of many
Asteridse.

4) Pedicellaries are never observed in Ophiuridse, while we find them in many Asteridse
of a very similar structure to those of the Brisinga.

5) The structure of the madreporic body and its dorsal situation characterise the Bri-
singa as a genuine Asteride.

6) Likewise the wide and deep ambulacral furrows, which are never found in the Ophiu-
ridse, where, on the contrary, the ventral side of the arms is covered with peculiar
cuticular plates (ventral plates).

7) The water-feet of the Brisinga correspond in their strong development, and in their
whole structure, completely to those of the star-fishes ; and especially we do not know
any Ophiuridse in which they terminate in a clearly developed sucker.

8) Neither do we find in the Ophiuridse any separate Ampullae for the water-feet; while
these Ampullae are in the Brisinga developed precisely in the same manner as in the
proper star-fishes.,

9) The perivisceral cavity is in the Ophiuridae confined to the disc only; while the arms
represent solid parts only containing a narrow canal for the reception of the radial

87

water-vessel and the radial nerve. In the Brisinga on the contrary, the perivisceral
cavity extends, as in the Asteridae, through the whole length of the arms between the
ambulacral skeleton and the dorsal skin, containing in its basal part various internal
organs.

10) Radial continuations of the digestive system (radial cseca) are never found in the
Ophiuridae; while in the Brisinga they are developed precisely in the same manner
as in the Asteridae.

11) No special masticatory apparatus (teeth, dental papillae) are to be found in the Bri-
singa any more than in other Asteridae. But on the other hand the furrow-spines
situated on the interior contiguous adambulacral plates assume a particular develop-
ment, just as in other star-fishes, forming the so-called oral spines.

12) The organ of secretion noticed in the Brisinga. with its issue on the dorsal side of the
disc, is something which we find only in the proper star-fishes.

13) The so-called ,,heart", to which nothing corresponding has yet been noticed in the
Ophiuraus, agrees, in its structure and in its relation to the other parts, entirely with
the same organ in the Asteridae, as is the case also with the situation and structure
of the stone canal.

14) As the organs of generation in the Brisinga are confined to the arms only, we do not
of course find the genital fissures on the lower side of the disc which characterise the
Ophiuridae.

15) The terminal organs of sense in the Brisinga find their most complete homologon in
the Asteridaj; while nothing corresponding is found in the Ophiuridae.

16) The manner in which the Brisinga moves, is the same as in the Asteridae; that is to
say, essentially by the play of the water- feet only; and not as in the Ophiuridae by
the flexion of the arms.

Considering all these important characteristics, the points of agreement with the
Ophiuridae which the Brisinga exhibits, namely the little rounded disc plainly distinguished
from the long thin arms, and the absence of cuticular pores and cuticular tentacles (respira-
tory tubes), must go for nothing.

In order still better to understand the great conformity of the genus Brisinga with-
the proper star-fishes, we will once again briefly consider some of the most important
organs in the Brisinga, and institute a detailed comparison with the same in other star-fishes.
We will dwell chiefly on the solid calcareous parts which support the body and determine
its form.

a. Homology of the ambulacral skeleton.

(Tab. V).

We have previously divided the ambulacral skeleton of the Brisinga into 2 parts :
the skeleton of the disc, and the skeleton of the arms; and this is so far correct, as the

88

skeleton of the disc in the Brisinga really forms a connected whole, a solid calcareous
frame, to the exterior side of which the skeleton of the arms is attached. It has been
however shewn that this frame is actually composed of the same elements as the ambu-
lacral skeleton of the arms, namely of 2 sets of ambulacra! vertebrae, in a line with the
series of vertebrae in the arms, and properly representing the interior continuation of these
series. It has likewise been shewn that the exterior set of vertebrae in very young speci-
mens, come out of their connexion with the oral ring, and thereby evidently connect them-
selves immediately with the skeleton of the arm. Properly therefore there remains in the
skeleton of the disc only a single set of vertebras, connected with each other by interposed
calcareous plates, so as to form a connected ring which surrounds the oral aperture.

If we now examine how the case stands in other star-fishes, we shall also find that
the proper ambulacral skeleton of the disc forms here only the immediate continuation of
the series of vertebrae of the arms. Only the interior set of vertebrae are in immediate
contact, or form a closed calcareous ring round the oral aperture; while the other part of
the series of vertebrae, reckoned as belonging to the disc, is only indirectly connected by
the skin of the disc and by the calcareous plates imbedded in the same. Accordingly as
the arms are longer or shorter, the number of vertebrae entering into the composition of
the disc is extremely variable in the various star-fishes. In certain forms (Goniaster, Cul-
cita) where the arms are only indicated by obtuse angles, the whole ambulacral skeleton
may properly be said to be absorbed in the disc ; in the long-armed star-fishes, for instance
Asterias, the part of the ambulacral skeleton belonging to the disc is, on the other hand,
only insignificant; while the greater part belongs to the arms. Between these extreme
points, we find again all possible transitions. It is now easy to see that the skeleton of
the disc in the' Brisinga can by no means be compared with the whole skeleton of the disc
in other star-fishes, but only with its innermost part (the interior or the 2 interior sets of
vertebrae which immediately surround the mouth). The apparently very anomalous structure
of the skeleton of the disc in the Brisinga, will then be found in complete conformity with
what we know of other star-fishes; only with the difference that the interior vertebras are
necessarily more firmly and intimately connected together, in order to support the enor-
mously developed arms. In Tab. V, fig. 1 & 2 there is represented the oral ring of a 9-armed
specimen of Brisinga coronata, with the basis of the skeleton of 3 of the arms in natural
connection with the disc; and for the sake of comparison, there is also represented in
fig. 7 & 8 a corresponding part of the ambulacral skeleton of a likewise 9-armed spe-
cimen of Solaster endeca. We shall here recognise homologous parts everywhere. As in
the Brisinga, so also in the Solaster and other star-fishes, the interior set of vertebras
acquire a peculiar development; the vertebras connecting themselves with each other so as
to form a sort of oral ring. The connexion takes place here also by means of peculiar
connecting plates perfectly corresponding with the 2 sorts of interposed plates in the oral
ring of the Brisinga. The wedge-plates especially are plainly developed, without however

89

connecting themselves immovably by suture with the adjacent vertebrae. To their exterior
side are attached the radial septa, which in a fan-like arrangement divide the coeloma of
the disc; and in these septa there are aften developed peculiar calcareous plates, the inner-
most of which are articulated with the wedge-plate. In one of the radial septa which
include the stone-canal and the so-called heart, this calcareous plate is double; and its
homology with the first pair of dorsal marginal plates in the Brisinga seems thereby to be-
come highly probable. The so-called parietal plates, which .in the Brisinga contribute
essentially to form the interior wall of the oral ring, are also found again in the Solaster
and other star-fishes. They are however here of much slighter development, and separated
from each other by a considerable interval; on the interior side they form a horisontal fur-
row for the circular ambulacral vessel, limited below by a projecting sharp edge, which
completely corresponds to the circular rim projecting inwards on the oral ring in the Bri-
singa. While in the Brisinga it goes uninterruptedly round the whole interior of the oral
ring, in the Solaster and other star-fishes it is interrupted at each vertebra in two places,
where the circular ambulacral vessel is only circumcluded by ligaments. In fig. 9 & 10,
which represent corresponding p^arts of the ambulacral skeleton seen from above in the Bri-
singa and Solaster, the corresponding calcareous plates are indicated by the same letters,
in order that their complete resemblance may be more easily observed.

If we -now turn the ambulacral skeleton, and view it from the ventral side (fig. 2 & 8),
it will likewise be at once evident that in both forms there are everywhere found comple-
tely homologous parts. The deep ambulacral furrows, at the bottom of which the median
longitudinal furrow for the radiary water-vessels, and the holes placed in pairs for the water-
feet are seen, are bounded in both on each side by a row of plates (adambulacral plates),
which are connected with each other by elastic muscular bands, and to which the so-called
furrow-spines are attached; the innermost of these plates being, in both, united immovably
by suture with its neighbors into one continuous piece projecting under the oral ring.
To this piece are attached the so-called oral spines directed in fan-like arrangement to-
wards the mouth. That there are apparently in the Brisinga no oral angles to- be found, is
a consequence of the slighter development of the piece, in connexion with the greater
breadth and more solid composition of the oral ring. It will be evident from what has
previously been written, that if we imagine the basis of the arms in the Brisinga connected
to a certain extent above and below by skin, we shall have a tolerably normally developed
star-fish; just as, if we imagine in a Solaster the incisions between the arms continued up
to the last ambulacral vertebra but one, we shall obtain a form pretty nearly corresponding
to the Brisinga. That the coeloma of the disc in the Brisinga is not continued beyond the
oral ring or the innermost contiguous vertebra, and that the dorsal skin attaches itself im-
mediately in the periphery of the oral ring, is again a natural consequence of the reduction
of the skeleton of the disc. The interior organs enclosed in the cavity of the disc, the

12

90

stomach and the apparatus of secretion, are in consequence of this limited to a much
smaller space than in other star-fishs, (comp. Tab. VI, fig. 36 & 37).

b. Homology of the skeleton of the skin.

The integuments in the Brisinga exhibit, as well with regard to consistency as to
structure, the most complete resemblance to those of the proper star-fishes; and also the
calcareous parts belonging to the skin may be referred to corresponding parts in other star-
fishes. It has thus been previously shewn that the calcareous parts imbedded in the dorsal
skin of the disc, as well as the very peculiar calcareous ribs in the dorsal skin of the arms,
can very naturally be referred to the same category as the so-called calcareous net in the
skin of other star-fishes. Also with regard to the so-called marginal plates, likewise ber
longing to the skin, we have in the Brisinga something analogous in the small plates,
arranged at the basis of the arms or in the angles of the arms, of which the innermost,
connected with the wedge-plates, have, in fully developed specimens, entered into the com-
position of the oral ring, while the 2 others on each side belong specially to the arms, as
evidently developed dorsal marginal plates. Also along the whole of the rest of the arm,
there may be noticed, as previously remarked, rudiments of similar dorsal marginal plates,
which however in the basal part of the arm go in one with the transversal calcareous ribs,
and represent the exterior somewhat enlarged ends of the same. On the other hand, no
trace is to be found in the Brisinga of ventral marginal plates ; as also, owing to the com-
plete absence of real interbrachial space, there can of course be none of the so-called
intermediary ventral plates (interambulacral plates) imbedded in the ventral skin.

c, Homology of the spines.

We find in the Brisinga, as in other star-fishes, dorsal spines, lateral or marginal
spines, and so-called furrow- spines; the latter are attached immediately to the ambulacral
skeleton (to the ambulacral plates) while the others issue from the calcareous parts deve-
loped in the skin.

That the spines in the Brisinga are surrounded by wide cuticular sheaths, seems at
first glance to be something quite special and peculiar for this genus. Nevertheless on more
minute examination we shall find also in other star-fishes a similar, although slighter and
less remarkable cuticular sheath round the spines. Even if this cuticular sheath can not in
isolated cases be indicated; it is yet certain that it has originally existed, and has only at
a later period been removed.

As to the arrangement of the spines, this will naturally be regulated by the arrange-
ment of the calcareous parts to which they are attached. The different situation of the
dorsal spines on the disc and on the aims is quite naturally determined thereby. Of mar-

f* o

/ UNIVERSI
N^OALiFC.

91

ginal spines there are in the Brisinga only the dorsal; ventral marginal plates being, as
previously noticed, entirely wanting. The dorsal marginal spines have however precisely the
same arrangement along the sides of the arms as in other star-fishes. The furrow-spines
have likewise, apart from their number, on the whole the same arrangement as in other
star-fishes, those on each adambulacral plate forming a (certainly not quite regular) trans-
verse row, and those on the innermost contiguous adambulacral plates assuming also in the
Brisinga a peculiar form and direction as real oral spines.

d. Homology of the pedicellaries.

The pedicellaries in the Brisinga are of a very similar structure to those of several
other star-fishes, for instance Asterias, Pedicellaster. On the other hand they differ by their
comparatively insignificant size, their enormous number and their different arrangement.
While those of the other star-fishes are more dispersed on the skin, or at most group
themselves in some quantity around the base of the larger spines (Asterias) those of the
Brisinga are present in enormous numbers on the cuticular sheaths of the spines them-
selves; and those which are attached immediately on the dorsal skin of the arms are col-
lected, likewise in enormous numbers, into plainly defined transverse stripes, which can
even assume the form of semicylindrical transverse ridges (Br. coronata). As regards the
general homology of the pedicellaries, their position on the cuticular sheaths of the spines
shews clearly enough, that they cannot be, as Agassiz* has tried to demonstrate, originally
homologous with the spines. These latter are developed from the interior cuticular layer
(corium) while the pedicellaries are, as the development also shews, (see above p. 64) exclu-
sively an epidermis-formation. It may indeed in a certain sense be said that the so-called
pedicellaries of the Echinidae are homologous with the spines. But what we call pedicel-
laries in the Echinidae are, it must be remembered, essentially different from what we call
by that name in the star-fish. A pedicellary of the former is properly a compound organ
including a spine, beneath the proper pedicellary, (the so-called stem of the pedicellary) ;
such a pedicellary is in other words a real spine, on the cuticular sheath of which there
is only developed one single terminal pedicellary; and it cannot therefore be compared
with a single pedicellary of a star-fish, which never has any such actual stem as in the
Echinidae.

e. Homology of the interior organs.

With regard to the other organic systems in the Brisinga: the water-system, with
the exterior parts belonging thereto, (the water-feet, the madreporic body); the nervous
system with its terminal apparatus of sensation; the digestive system; the blood-system ; the

* Revision of the Echini, Part IV, pg. 668.

12*

92

»

secretive system; the complete resemblance of all these to the same in other star-fishes,
has been previously sufficiently demonstrated; for which reason it is unnecessary here again
to notice each one of them separately. But with respect to the generative organs, an obser-
vation may yet not be out of place. In the arrangement of these organs, the Brisinga appears at
first glance to be very remarkably distinguished from all other Asteridse; as the generative
organs are in the Brisinga quite separated from the disc, and limited to the arms only;
while in all other star-fishes they take their origin trom the disc itself, even if in some few
forms they also extend more or less into the cavities of the arms. This apparently very
anomalous case may yet be quite naturally attributed to the excessive reduction of the disc
in the Brisinga, which only corresponds to the central part of the disc in other star-fishes;
while the peripheral part of the disc in these, is in reality homologous with the basal part
of the arms in the Brisinga. If we admit this, it will easily appear that the organs of
generation (as described in the Brisinga coronata) really correspond,, not only in structure
but also in position, to what is normal in star-fishes generally (comp. moreover the 2 sche-
matic Figures 36 and 37 in Tab. VI).

6.

Relation of the genus Brisinga to now living slar-fishes.

Among the now living star-fishes, the genus Brisinga seems indeed to stand rather
isolated, as well with regard to its exterior habitus as in respect of particular points of its
interior organisation. There are however certain genera of star-fishes with which it seems
to exhibit a closer affinity than with others. Among the better known native genera, there
are chiefly 2 which we may take as instances of this, namely Asterias and Solaster. The
former is, next to the Brisinga, that one of all our Asteridse in which the disc is the most
reduced relatively to the arms; and it is probably also, next to the Brisinga, that one which
we must regard as the oldest or least altered star-fish form, a conclusion which is comple-
tely corroborated by paleontological indications. The Brisinga shews affinity to the genus
Asterias in the form of its pedicellaries, partly also in the structure of the cuticular skele-
ton, and finally in the more solid composition of the interior set of vertebras (the oral ring).
The genus Asterias has however 4 rows of water-feet; while the Brisinga has only 2, in this
respect resembling the genus Pedicellaster, which was established by my Father, and
which in other respects is in close affinity with the genus Asterias. The Brisinga is how-
ever distinguished both from the Pedicellaster and Asterias (at least from our native spe-
cies) by its considerable number of arms or rays, in which respect again it agrees better

••

93

with the genus Solaster. Still neither of these genera exhibits a disc reduced in any thing
like the same degree as that of theBrisinga; and even in those species of the genera Asterias
(A. glacialis) and Pedicellaster which are furnished with a relatively very small disc, we
shall yet find that a rather considerable number of vertebrae have entered into the formation
of the disc; while the number of such vertebrae belonging to the disc in the Brisinga is
properly speaking zero. It is this which properly determines the peculiar exterior habitus,
so different from that of other known star-fishes, that distinguishes the genus Brisinga, and
gives to it at first glance a certain habitual similarity to the Ophiurae.

7.

Relation of the genus Brisinga to extinct star-fishes.

Asbj0rnsen has already (1. c.) drawn attention to the habitual conformity of the genus
Brisinga with the oldest known fossile star-fish Protaster. The small round ophiura-like
disc, with the sharply defined arms issuing from it, in the Protaster, reminds us decidedly
of the Brisinga; as also the skeleton of the arms, by its shape and composition of ambu-
lacral and adambulacral plates only, exhibits in both forms considerable resemblance. We
shall however find, on more minute examination, that the body even in this ancient form, is
already considerably more centralised than in the Brisinga; a considerably greater number
of vertebrae enters into the composition of the disc, notwithstanding its small dimensions.
On the whole we shall find in the Protaster evident indications of the course of develop-
ment diverging from the star-fish type and tending towards the Ophiurae. type, which is by
no means the case in the Brisinga. Antecedently to this divergent development we must
naturally presuppose a number of more indifferent stages, wherein the nature of the deve-
lopment would not yet be clearly indicated, in other words, wherein the cormus would shew
a slighter degree of centralisation; and these stages must certainly be supposed to have been
in some degree more corresponding to the Brisinga. If we admit the theory above noticed
as to the composite individuality of star-fishes, or their original value as colonies of worm-
like persons, we have actually in the now living Brisinga a still more primitive form than
even in the Protaster, wherein this colony has already attained to a considerable degree
of centralisation, and in which the number of persons (arms) has become permanently the
typical number for the Echinoderms, namely 5. The Brisinga may henceforward be assumed
to exhibit a still greater conformity with the supposed ancestors of the genus Protaster

94

than with this genus itself. In any case we must consider the genus Brisinga, in respect
of the general composition of the body, as the most primitive and therefore the oldest of
all Echinoderms.

8.

Comparison of the 2 species of Brisinga.

The two hitherto known species of the genus Brisinga, Br. endecacnemos and Br.
coronata, stand in close connexion, but exhibit nevertheless certain distinctive characteristics
which make their specific distinction necessary. The latter of these species has already
been exhaustively described in the foregoing pages. As regards the first species discovered,
Br. endecacnemos, its general appearance and structure are pretty well known from Asbj0rn-
sen's description and figures. It might possibly however not be without interest here to
notice again this species, and to compare it with the Br. coronata. I have therefore added
a plate (Tab. VII) in illustration; and will now briefly point out the characteristics wherein
the two species differ from each other.

In size, color and general habitus they are (see Tab. VII, fig. 1) nearly similar. But
an important distinctive characteristic is the number of the arms or rays. While this num-
ber is in the Br. coronata, as previously noticed, extremely variable, the number of rays in
the Br. endecacnemos has already become completely permanent, and is constantly 11. In no
single one of the numerous specimens which have been examined of this species, (and their
number far exceeds that of the specimens taken of the other species), has there ever been
noticed any deviation from this rule; for which reason also the specific denomination Ende-
cacnemos is fully justified. The manner in which the arms are inserted on the disc is
completely similar in both species. Still the interval between the bases of the arms
is in the Br. endecacnemos still more restricted; so that the arms nearly touch each other
at the base; while we always find in the Br. coronata an evident, even if only a small
interval.

The disc is, viewed from above (see fig. 1) in both species circular; seen from the
side (fig. 2) it appears however in the Br. endecacnemos somewhat higher than in the other
species, and has its exterior border more perpendicular. The disc-spines, (fig. 6) are nearly
similar to those in the Brisinga coronata. They stand usually still more closely together;
and the proper calcareous spine exhibits at the extremity only a few secondary spines. The
shape of the madreporic body (see fig. 1, 2 & 4) is characteristic for the Br. endecacnemos.
It is always remarkably prominent; so that when the disc is viewed in profile (fig. 2) it pro-

95

jects frequently like a more than semiglobular, or even like a nearly pyramidal tubercle.
On closer inspection it will be found (see fig. 4) to consist as it were of 2 parts; only the
exterior, more salient part having the meandric furrows; while the other part is covered
with strong pointed spines, which also (in annular arrangements) encircle the more salient
part. The secretory pore situated on the dorsal side of the disc (the so-called anal aper-
ture) has the same somewhat excentric position as in the Br. coronata; and the disc exhi-
bits at this point a slight elevation (see fig. 2). I have not succeeded in discovering any
traces of pedicellaries on the dorsal side of the disc, where in the Br. coronata they are
always plainly apparent. The spines attached to the underside of the disc seem also to be
somewhat different from those of the Br. coronata, The furrow-spines on the lower side of
the adambulacral plates (see fig. 2, 3 & 5) are remarkably large, usually 12 for each inter-
radial space, of which 4 especially, on the interior adambulacral plate, are remarkably long
and strong, and directed inwards towards the oral aperture, whereby they appear to have
assumed the functions of the proper oral spines. These latter, 6 in number, are however
here (see fig. 9) extremely small and quite covered by the other spines; so that they do
not become apparent until the others are removed. The oral membrane resembles comple-
tely that of the Br. coronata (see fig. 5) but is in most captured individuals difficult to per-
ceive; as usually the folds of the stomach are evaginated a long way out of the mouth,
and the oral aperture itself strongly enlarged (see fig. 3).

The arms (see fig. 1) have about the same length in proportion to the diameter of
the disc, and exhibit in their general form a great resemblance to those of the Br. coronata,
but are immediately distinguished by the absence of the elevated spines in annular arrange-
ment on the transversal ribs, which are so characteristic of the latter species. Likewise we
do not usually find any trace of the soft transverse ridges covered with pedicellaries, which
are so distinctly marked in the Br. coronata. Only in a few unusually large specimens
(see fig. 1) I have found slight indications of such ridges between the interior calcareous
ribs; while in the exterior part of the arm no such indications were apparent. With respect
to the calcareous ribs, they have in the Br. endecacnemos a relatively far greater extension
on the arms than in the Br. coronata, being apparent on all the interior half. Their num-
ber is therefore also much greater, namely 30—40, or about twice as great as in the Br.
coronata. In their shape and arrangement they correspond otherwise perfectly to those of
other species; and also in this they appear frequently variously sinuous and anastomosing
with each other. If we now examine these calcareous ribs with sufficient magnifying power,
we shall find on them, besides numerous pedicellaries, also a single row of extremely small
spines (see fig. 13). These spines are (fig. 14) compressed, lancet-formed, and, as it appears,
entirely without any cuticular sheath; neither do they appear to be movably articulated to
the calcareous rib, like those of the Br. coronata in the same place, but merely to repre-
sent simple processes of the same. The whole dorsal cuticle of the arms between the ribs
will also be found covered with similar, although somewhat longer and thinner, small micro-

96

scopic spines — quite in contrast to what is the case in the Br. coronata — whereby the sur-
face of the arms acquires a peculiar rough, or, as it were, finely chagrined quality (see fig. 11).
In the interior, at the base of the arms, these spines stand most closely; further out on the
arms they are arranged in larger irregular spaces, mostly in a transverse direction and with
bare intervals. All these minute spines, of which there is no trace to be seen in the Br.
coronata, rise (see fig. 12) from small thin circular perforated calcareous plates in the skin;
and among these spiniferous plates there are besides a great number of still smaller disc-
like plates, which are without spines, but from which spines are no doubt subsequently
gradually developed. By these innumerable small cuticular plates in connexion with the
numerous calcareous ribs, the cuticle of the arm in the Br. endecacnemos acquires, on the
whole, a considerably firmer consistency than in the Br. coronata; a fact which stands in the
closest connexion with the very different arrangement of certain internal parts (the organs,
of generation) in the latter species, as will be further noticed in the sequel. With regard
to the proper arm-spines, they are nearly like those of the Br. coronata. They are pro-
portionally even somewhat longer than in this species, especially the so-called exterior
furrow-spines. which are attached in the middle of the adambulacral plates (comp. fig. 11).
The number of the interior furrow-spines is moreover also rather variable in different indi-
viduals, and seems usually to increase with age. The marginal spines are in the basal part
of the arm very small; but outwards they increase rapidly in length; so that in the middle
of the arm they become 3 times as long as its transverse diameter. They are also movably
articulated to the outer extremities of the calcareous ribs, which form very distinct plate-
shaped enlargements furnished with an evident joint (see fig. 13); in the exterior part of
the arm, where the calcareous ribs are wanting, they are attached to special small plates
(dorsal marginal plates). At the point of the arm there is an organ of sense quite like
that in the Br. coronata, and likewise arched over by a peculiar calcareous plate furnished
with long marginal spines (see fig. 17).

In quite young specimens the disc (see fig. 10) is very thin and transparent;
and the exterior set of vertebrae come forward here in the same manner as in equally
small specimens of Br. coronata. Of the dorsal spines, a great many have still the
peculiar bristle-like form which is described in the young of the Br. coronata, and which
in all- probability is a remnant from the embryonic or larva-state. On the other hand the
madreporic body already shews completely the form and arrangements which are charac-
teristic of the species. The arms are, in proportion to the diameter of the disc, conside-
rably shorter than in adult specimens, evenly tapering from the very base, and as yet
without any evident calcareous ribs.

In the structure of the ambulacral skeleton there is certainly a great resemblance
to the Br. coronata, nevertheless, on a more minute investigation, it will be easy to dis-
cover, even here, several specific differences.

The oral ring (fig. 7) is, in conformity with the greater height of the disc, of a

97

somewhat different shape from that of the Br. coronata, with a more sharply projecting
dorsal ridge and more vertical exterior sides. The vertebrae of the exterior set are con-
siderably shorter above and nearly disc-like; so that, when the oral ring is viewed from
above, (fig. 7) they appear like an extremely narrow rim around the strongly developed
interior set of vertebrae. As a consequence of this, the wedge-plates (see fig. 8) rise also
more perpendicularly than in the other species. Their upper enlarged end is less tuber-
cularly salient, and regularly formed; and the dorsal marginal plates attached to their lower
extremities, stand so closely together that their exterior articulating surfaces nearly touch
each other. The very narrow, scarcely perceptible, fissure between the parietal plates in the
interior wall of the oral ring of the Br. coronata, is here (see fig. 7) widely gaping and
spanned over with extended soft brownish ligaments. On closer comparison of the single calca-
reous plates which compose the oral ring, with the corresponding plates in the Br. coronata,
there will likewise be found several points of dissimilarity in shape, on which however it
would not be expedient here to enter. The greatest difference appears to be in the interior
adambulacral plates. They are here (see fig. 9) much narrower, and have a very conspi-
cuous instriction in the vicinity of the adoral extremity, which is thus divided, as it were, by
a furrow of demarcation, from the rest of the adambulacral plate.

With regard to the ambulacral skeleton of the arms (fig. 15 & 16), the individual
joints or vertebrae are on the whole shorter and broader than in the Br. coronata, especially
the interior ones; and their dorsal ridge is more regularly semicylindrical; not, as in Br.
coronata, distinctly enlarged at the extremities. Of the dorsal marginal plates there are
always 3 distinctly developed, and rudiments of 1 or 2 others (see fig. 11, 15, 16). It is
still more clearly to be seen here than in the Br. coronata, that the marginal plates further
out on the arm are in reality represented by the enlarged extremities of the calcareous
ribs (see fig. 11 & 13).

As to the internal organisation, we must further notice a remarkable difference
between the two species of Brisinga. This remark applies to the organs of generation.
While in the Br. coronata they only form 2 widely ramified symmetrical organs situated in
the basal part of each arm, each organ with its single issue, they represent in the Br. ende-
cacnemos (fig. 18 & 21) a great number of separate glandulous bodies, which are arranged
on each side of the medial line in a single row, extending to about half the length of the
arm. Each of these bodies has its separate issue, opening nearer to the dorsal side of the
arms in a fine pore. Therefore, instead of the 2 symmetrical apertures of generation in the
Br. coronata, we find here for each arm a double row of numerous apertures, as has already
been remarked by Asbj0rnsen. This wonderful arrangement of the generative apparatus, to
which nothing corresponding is found in other star-fishes, is in so far of great interest, as
we may see herein a characteristic of very ancient origin, an inheritance from the worm-
like ancestors of the Brisinga, which has still been only partially preserved in the class of
Crinoidse. The ovaries (see fig. 18, 19, 20) are very slightly ramified, and often appear like

13

98

perfectly simple.' oval, stemmed vesicles, in the interior of which the reddish yellow ova are
shining. Only nearer to the base of the arm, they exhibit in large specimens some few
short and broad lobes (fig. 20). The spermaries (fig. 21, 22) are more ramified, and look
like small bunches of grapes, with considerably smaller rounded lobes of a whitish color.

The color of the body is in both species of Brisinga tolerably uniform; but the red
color of the dorsal side in the Br. endecacnemos, seems generally to be lighter or less in-
tense than in the Br. coronata.

The remarkable difference in the occurrence of both species has been previously
noticed. On our coasts it appears to be quite constant.

Br. endecacnemos was first discovered by Asbj0rnsen in the interior of the Har-
dangerfjord, where its occurrence is not so very rare in one locality (Hesthammer). It is
found here, as in every other place, only on a rocky bottom, on the steep incline from the.
outside deep (400 fathoms) towards the shore. Subsequently it has also been met with in
a few other places in the same fjord under similar circumstances. But on the other hand,
it was not known among us as occurring elsewhere than in the Hardangerfjord, until I found
it again some years ago at Flor0, 12 miles north of Bergen. The figures here given are
taken from the fine specimen then obtained. According to reports of the Atlantic expedi-
tions organised by the English government, it is found at least as far towards the south as
the west coast of Portugal, where several specimens were taken, partly at very considerable
depths, during the Porcupine's expedition under the guidance of Mr. Jeffreys.

9.

On the systematic position of the genus Brisinga, with remarks on the classification

of the star-fishes in general.

We have previously mentioned that the genus Brisinga, according to its whole orga-
nisation, is a genuine star-fish, and has nothing in common with the Ophiurse, which latter
must be considered as a diverging branch from the group of the Asteridse, wherein the course
of development has taken a very peculiar direction different from that of the star-fish now
living. In the general composition of the body, 'and especially in the extremely slight cen-
tralisation of the cormus, the Brisinga certainly occupies a rather indifferent position; and it
may in this respect be considered as more nearly agreeing with the ancient primitive forms
from which all the other echinoderms have sprung. But on the other hand, the organisation is
otherwise completely developed in the manner characteristic of the star-fishes; for which

99

reason it must in a natural system be necessarily referred to them. Thus there only re-
mains the task of assigning to it its systematic place in the order of the star-fishes.

In the division of the star-fishes it has, as is well known, been usual according
to Miiller & Troschel1 to have regard chiefly to 2 characteristics, namely the arrangement of
the water-feet in 2 or 4 rows, and the presence or absence of the anal aperture. To neither
of these characteristics can there however in my opinion be attributed such a systematic
importance, as that any natural division could be established on that basis. According to
the first mentioned characteristic, the star-fishes generally are divided into 2 large groups
or sub-orders: those with 4, and those with 2 rows of water-feet. To the former group
there belong only a proportionally small number of forms; while the majority of known
star-fishes belong to the latter group, or those with only 2 rows of water-feet. Also the
genus Brisinga would, on account of its biserial water-feet, have to be classed in the last of
these groups. The unnaturalness of this binary division of the star-fishes, will however on
further consideration appear manifest.

We have thus in the genus Asterias not only forms with 4, but also with more
rows of water- feet; and of those star-fishes provided with 2 rows of water-feet, there are
several which decidedly in all other respects approach extremely nearly to those with 4
rows. Thus the genus Pedicellaster, with 2 rows of water- feet, stands so particularly
near to the genus Asterias with 4 rows, that in a natural system it must even be referred to
the same family. The case is similar with the 2 genera Cribrella and Stichaster, which
are undoubtedly nearly related to each other, but which have been formerly, on account of
the before-mentioned difference in the arrangement of the water-feet, placed in quite diffe-
rent sub-orders. Finally we have in the interesting form Pteraster multipes, discovered
by my Father, an instance of considerable variation occurring in respect of the arrangement
of the water-feet, even within the limits of one and the same generic type. This star-
fish, which otherwise, as well according to its exterior habitus as according to its whole
organisation, is a well defined genuine Pteraster, has for each ambulacrum 4 distinct rows
of water-feet; while all other known species of this genus have only 2. According to the
former method of division, this star-fish would therefore not only be separated generically
from its near relatives Pt. militaris and Pt. pulvillus, but would have to be classed in a quite
different sub-order, where it would stand together with forms with which it only exhibits a
very distant affinity. It appears already, from this one instance, clearly enough, that the
characteristic derived from the arrangement of the water-feet, must be entirely abandoned
as systematic characteristic having the value formerly attributed to it.

Neither does the other characteristic, the presence or absence of an anal aperture,
appear to me to be applicable as a principle of division; and that chiefly because the aper-
ture which has been called anal, is frequently, perhaps in most cases, not an anal aperture,
but only, as in the Brisinga, a more secretory pore. This seems especially to be the case

1 System der Asteriden.

13*

100

in the star-fishes whose anal aperture is indicated as sub-central. But to ascribe to the
presence or absense of this insignificant secretory pore, such a prominent importance as to make
it the basis for the formation of larger groups, seems to me to be quite unreasonable. We are
thereby forced to separate evidently related forms from each other. As an illustrative in-
stance of this we may mention the 2 genera Archaster and Astropecten, which accor-
ding to the earlier division belong to different groups of star-fishes: the former to those
furnished with an anal aperture, and the latter to those without it. But these 2 genera are
in reality so extraordinarily closely related, that even some species which now are usually
referred to the genus Archaster (for instance A. Andromeda and A. Parelii) have been taken
for genuine Astropectens, even up to the most recent times. Only after that the previously
overlooked anal aperture (here undoubtedly a mere secretory pore) had been observed in
these species, it has been found necessary, in regard to the formerly adopted method of.
division, to transfer them not only to another genus, but even to an entirely different group
of star-fishes. Another distinctive characteristic has indeed been indicated for the genus
Archaster, namely the structure of the water-feet; but that it is not possible from this either,
to get any trustworthy mark of recognition, has lately been demonstrated by Ltitken (I.e.).
The 2 above-named species of the genus Archaster agree also in this respect completely
with the species of the genus Astropecten.

As has been shewn, the previous division of the star-fishes is scarcely tenable any
longer, being based on characteristics of very doubtful systematic value. As principles of
division in a new classification of the Asterides, the structure of the ambulacral skeleton,
the structure of the cuticular skeleton, and the relation of the other calcareous particles,
spines and pedicellaries belonging to the skin, will chiefly have to be considered. Especi-
ally the pedicellaries seem in this respect to deserve our attention. They are entirely wan-
ting in a great number of star-fishes; while in others (for instance the genus Hippaster)
they exhibit a very peculiar structure and arrangement, totally different from what we have
become acquainted with in the Brisinga.

With respect to the structure of the pedicellaries, the Brisinga exhibits most resem-
blance to the Asterias and Pedicellaster, which also agree with the Brisinga in some
other points, and may therefore be considered as the nearest relatives of this genus among
the star-fishes of the present day. Yet the genus Brisinga exhibits so many peculiarities,
that it can scarcely be placed together with those two genera in one and the same family.
The Brisinga must therefore in any case be considered as the type for a particular family
among the Asterides. With respect to the place of this family in relation to the other
families of the Asterides, if we are to follow a definite serial order consistent with the sup-
posed phylogenetic development, it must be placed not at the end of the order of the
Asterides, but at the very beginning, as representing the oldest and most primitive of all
the Asterides.

101

10.

Diagnoses of Family, Genus and Species.

Familia: Brisingidae.

Habitus externus Ophiuridarum, structura vero Asteridarum.

Discus minimus ammlo sustentatus calcareo solido e vertebris modo adoralibus fir-
miter inter se conjunctis composite.

Corpus madreporiforme singulum tuberculiforme prope marginem disci dorsaliter situm.

Brachia perlonga a disco bene definita, tessellis ambulacralibus et adambulacralibus
distinctis marginalibus vero obsoletis, sulcis ventralibus profundis, tentaculis magnis disciferis.

Cavitas intestinalis in brachia extensa ibiqve caeca radialia bene evoluta bifurcata et
organa generations continens.

Genus : B r i s i n g a , Asbj0rnsen.

Discus orbicularis, sat depressus supine cute coriacea dense spinigera tectus, in-
ferne nullis angulis oralibus prominentibus sed cuticula nuda valde contractili os circum-
dante instructus. Porus secretorius subcentralis in cute dorsali perspicuus. Brachia nume-
rosa semicylindrica, parte adorali in adultibus plus minusve subfusiformi apicem versus
sensim attenuata, apice tenuissimo filiformi organo sensorio distincto terminato, ad basin
utrinqve tessellis marginalibus duabus tribusve rudimentariis instructa, cute dorsali in parte
basali costis calcareis transversis vario raodo flexuosis ad intervalla firmata spinis margi-
nalibus longissimis et ut ambulacralibus vaginis cutaceis magnis obvelatis.

Vertebra lraa brachiorum in adultibus annulo calcareo disci firmiter adnata; tessellse
connectentes annuli inter qvamque vertebram 3, media dorsaliter tuberculi instar prominente.

Pedicellarise numerosissimse forcipatse apice fortiter dentato et in vaginis spinarum
accumulate et in superficie dorsali brachiorum per fascias transversas plus minusve per-
spicuas distributee.

Tentacula ambulacralia biserialia e sulcis ventralibus longe porrecta. Tentacula
respiratioms riulla.

Spec. 1: Brisinga endecacnemos, Asbj0rnsen.

Brachia semper 11 costis calcareis numerosis (20-30) per dimidiam brachii longi-
tudinem et ultra occurentibus spinulis perparvis microscopio modo visibilibus obsitis, cute
dorsali inter easdem spinulis numerosis minimis aspera, fasciis pedicelligeris parum vel
plane non perspicuis.

102
Corpus madreporiforme valde prominens, interdum fere conicum, ex parte spinulis

**

rigidis armatum.

Pedicellarise magnitudine inaeqvali nonnullis in vaginis spinarum ceteris plus duplo
majoribus et structura multo fortiore.

Organa generationis numerosa per dimidiam brachii longitudinem dispersa, seriem
duplicem formantia, unoqvoqve poro externo discrete prope superficiem dorsalem instructo.

Spec. 2: Brisinga coronata, G. 0. Sars.

Brachia numero valde inconstantia (9—13), costis calcareis circiter 12 in parte modo
basali tertia occurrentibus, spinis altis coronam conspicuam formantibus armatis, cute dor-
sali inter costas nuda, fasciis vero pedicelligeris distinctissimis, semicylindricis, per totam
brachii longitudinem usqye ad apicem conspicuis.

Corpus madreporiforme parum elevatum nullis spinulis armatum.

Pedicellarise omnes eadem magnitudine, minimse sed portentoso numero.

Organa generationis cujusqve brachii solummodo duo valde ramosa in parte basali
symetrice disposita poro externo utrinqve singulo a basi circiter diametro disci remoto.

103

Explanation of the Plates.

Tab. I.

Fig. 1. A very young 10-armed specimen of Brisinga coronata, seen from
above, natural size; on the left there appears a recently regenerated arm.

Fig. 2. The disc of the same specimen, with the base of 4 arms in their natural
connexion, viewed from below, somewhat magnified.

fig. 3. The same viewed from above: 6, the madreporic body.

Fig. 4. The disc of a full grown 9-armed specimen viewed from above, very slightly
magnified: a, the secretory pore; b, the madreporic body; c c, the radial spaces of the
disc; d d, the interradial spaces, with the wedge-plate inserted in the angles of the arms;
e, a recently regenerated arm.

Fig. 5. The same viewed from below: c d e, as in the preceding figure; f, the
oral aperture strongly contracted.

Fig. 6. The same viewed from the side. The letters as in the 2 preceding figures.

Fig. 7. A piece of the dorsal skin of the disc, with the secretory pore (a) more
strongly magnified.

Fig. 8. Perpendicular section of the dorsal skin of the disc, still more strongly
magnified, shewing the 2 layers of the skin, the spines with their cuticular sheaths (b) and
the pedicellaries (a).

Fig. 9. A single disc-spine with the basal plate belonging to it, strongly magnified:
a, the basal plate contained in the dorsal skin; 6, the spine itself; c, its cuticular sheath.

Fig. 10—11. Various forms of dorsal spines.

Fig. 12. An arm of a full grown specimen, viewed from the side, natural size:
a, the left genital aperture; &, the dorsal marginal plates at the base.

Fig. 13. A piece of the same about in the middle of the length of the arm, viewed
from the side, slightly magnified: 1, marginal spine; 2, 3, furrow-spines.

Fig. 14. The same piece viewed from below. The water-feet are removed, with
exception only of 2 pairs: 1, marginal spine; 2, 3, 4, furrow-spines; a, water-leet.

Fig. 15. A marginal spine, magnified, and treated with a solution of potash, whereby
the cuticular sheath has been rendered transparent.

104

Fig. 16. One of the exterior furrow- spines with its cuticular sheath.
Fig. 17. The point of the calcareous spine itself, strongly magnified.
Fig. 18, 19. Interior furrow-spines.
Fig. 20, 21. 2 of the furrow-spines of the disc.

Fig. 22. A water-foot magnified, shewing the numerous muscular fibres crossing
each other. •

Tab. II.

Fig. 1. A full grown 10-armed specimen of Br. coronata, viewed from above, natural
size; only one of the arms drawn.

Fig. 2. An other specimen with 12 arms, of which 2 are drawn; one recently
regenerated and still without evident calcareous ribs.

Fig. 3. The extremity of a monstrously developed bifurcated arm, viewed from
above, natural size.

Fig. 4. The basal part of an arm, with strongly projecting calcareous ribs, and
unusually numerous and interrupted bands of pedicellaries, natural size.

Fig. 5. The extremity of an arm viewed from the side, strongly magnified, shewing
the dorsal bands of pedicellaries (a) the terminal organ of sense (c) and the peculiar plate
(b) which shelters it.

Fig. 6. The same seen from below: 6, c, as in the preceding figure.

Fig. 7. The terminal organ of sense, still more strongly magnified, seen from the
side: &, the 2 exterior rudimentary water-feet.

Fig. 8. The disc of a full grown 11-armed specimen seen from above, slightly
magnified. The dorsal skin is cut open round the periphery and turned back, in order to
shew the subjacent interior organs: a, the madreporic body; 6, the dorsal skin turned back;
c c, the basal trunks of the radial caeca (to the right a bifurcation of one these indicated);
d, the peripheral part of the stomach; e, the upper vault of the stomach, with its radial
folds; f, the apparatus of secretion; g, section of excretory duct for the same; h, the
porus secretorius which perforates the dorsal skin.

Fig. 9. A piece of the peripheral part of the stomach viewed from within, shewing
the numerous sinuous longitudinal folds: a, tendinous fibres whereby this part is attached to
the oral ring; &, the oral membrane.

Fig. 10. The upper vault of the stomach viewed from the under or interior side,
shewing the numerous papillae issuing from the same, and the terminal trunks of the radial
caeca proceding from the periphery.

105

Fig. 11. The disc of another likewise ll-armed specimen seen from above, after
that the dorsal skin and all the interior organs situated in the cavity of the disc have been
removed, whereby the skeleton (the oral ring), with its tendinous lining membrane, is en-
tirely exposed: a, the madreporic body, with the sheath preceding downwards from the
same, in which the so-called ,,heart" is inclosed; 6, the oral membrane; c c, the fan-like
extended tendinous fibres issuing from the upper border of the oral ring and attaching the
peripheral part of the stomach; d d, glandulous corpuscules attached to the periphery of
the circular ambulacral vessel; e, the oral aperture.

Fig. 12. A piece of the oral ring with the tendinous membrane that lines the same,
viewed from the interior sides a, the madreporic body; 6, the oral membrane; c, the tendi-
nous sheath in which the ,,heart" is inclosed; d d, the glandulous corpuscules attached in
the periphery of the circular ambulacral vessel; e, the stone canal.

Fig. 13. Some of the granulous contents of the corpuscules (d), strongly magnified.

Fig. 14. One of the glandulous corpuscules isolated.

Tab. III.

•

Fig. 1. Part of the disc from the side, magnified, shewing the articulating surfaces
for an arm and the adjacent parts: a, section of the radial nerve with the blood-sinus
above it; 6, section of the radial ambulacral vessel; c, ampullae for the exterior water-feet
of the disc ; d, section of one of the radial caeca.

Fig. 2. Part of the disc viewed from below. The spines and the water-feet are for
the greater part removed, in order to show the arrangement of the nervous system: a a, the
radial nerves; b 6, the circular commissure of the nerves (see also Tab. VI, fig. 1).

Fig. 3. Transverse section of the basal part of an arm of a full grown female specimen:
a, the water-feet; b, vertebra; c c, the ampullae for the water- feet, projecting into the cavity
of the arm; d d, section of the 2 branches of the radial caeca; e e, section of the ovaries;
f f, marginal spines.

Fig. 4. The basal part of an arm of a full grown female viewed from above, a
little magnified. The dorsal skin is cut along the middle and extended on each side, to
expose the strongly ramified ovaries situated in the cavity of the arm. Along the middle,
the skeleton of the arm's appears; and on each side of it, we see the ampullae for the
water-feet arranged in pairs.

Fig. 5. An ovary of more compact form, isolated and viewed from the interior side.

Fig. 6. The same viewed from the exterior side: a, a piece of the skin of the
arm, with the exterior genital aperture (b) situated in the same.

Fig. 7. A branch of an ovary, with (terminal) ramified cylindrical caeca.

14

106

Fig. 8. Another branch, on which the terminal caeca only form small vesicular
enlargements.

Fig. 9. An ovary of a very young specimen, viewed from the exterior side, strongly
magnified.

fig. 10. The same viewed from the interior side.

Fig. 11. A terminal ramification of an ovary drawn from a fresh living specimen.
Inside of the half-transparent ovarial cuticle, there appear numerous egg-cells in different
stages of development.

Fig. 12-1*5. Egg-cells in different stages of development, with an evident germinal
vesicle.

Fig 16. A fully developed egg, in which the germinal vesicle is not visible; the
yellowish red mass of yolk is surrounded by a thick pellucid chorion.

Fig. 17. The basal part of an arm of a full grown male specimen, viewed from
above. The dorsal skin is, as in fig. 4, cut along the middle and extended on the sides, in
order to expose the strongly developed seminaries.

Fig. 18. A seminary of a younger specimen, isolated, viewed from the interior side*.

Fig. 19. The same, seen from the exterior side, a, the genital aperture.

Fig. 20. A branch of the same seminary, isolated.

Fig. 21. The extremity of a branch of a fully developed seminary, drawn- from a
fresh living specimen.

Fig. 22. Some of the developing cells of the spermatozoa, strongly magnified.

fig. 23. A fascicle of fully developed spermatozoa.

Fig. 24. 3 spermatozoa isolated, very strongly magnified.

Fig. 25. An arm of a fully developed female specimen, viewed from the lower side.
The skeleton of the arm, together with the parts in connexion, has been removed, in order
to shew the organs lying above it in the cavity of the arm in their natural position:
a a a, the radial caeca with their 2 branches; b &, the ovaries; c c, the 2 genital apertures
situated symmetrically.

Tab. IV.

Fig. 1. The skeleton of the disc (the oral ring) of a fully developed 10-armed spe-
cimen, seen from above, slightly magnified: a a, the ' radial spaces; & 6, the interradial
spaces; c, oral spines.

fig. 2. The same viewed from below.

Fig. 3. The same viewed from the side.

Fig. 4. A piece of the oral ring, more strongly magnified, seen from above.

107

Fig. 5. The same seen from below.
Fig. 6. The same viewed from the exterior side.
Fig. 7. The same viewed from the interior side.

Fig. 8. The same viewed obliquely from above and from the interior side, the better
to see the furrow for the circular ambulacral vessel, with one of the holes through which
the ambulacral vessel passes.

Fig. 9. A piece of the oral ring, from which the exterior set of vertebrae are removed.
Fig. 10. Transverse section of the oral ring in the middle of a radial space.
Fig. 11. Transverse section of the oral ring in the middle of an interradial space.
The following indications serve for the last 8 figures:
a l, interior ambulacral plates.
a 2, exterior ambulacral plates.
ad l, interior adambulacral plates.
ad 3, exterior adambulacral plates.

f, furrow for the radial ambulacral vessel.
&, wedge-plates.

I, interior cavity bounded by the wedge-plates and the parietal plates.
m, parietal plates.

o J, apertures for the interior water-feet of the disc.
o 2, apertures for the exterior water-feet of the disc.
r, dorsal marginal plates.
x, furrow for the circular ambulacral vessel.

«/, the circular rim, projecting from the interior wall of the oral ring, which below forms
a boundary for the circular ambulacral vessel.
Fig. 12. A wedge-plate isolated, viewed from the exterior side.
Fig. 13. The same, viewed from the interior side.
fig. 14. The same, viewed in profile.

Fig. 15. The base of the skeleton of an arm, viewed from above, somewhat magni-
fied: a a, the dorsal ridge formed by the interior part of the ambulacral plates; & &, the
lateral parts of the ambulacral plates; ad, adambulacral plates; c, the 2 connate dorsal
marginal plates at the base of the arm; r, rudimentary marginal plates for attachment of
marginal spines; o, holes for the water-feet.

Fig. 16. The same seen from below: f, the furrow for the ambulacral vessel;
p l, marginal spines; p 2, p 3, furrow-spines; the other letters as in the preceding figure.

Fig. 17. The same, viewed from the left side: a, ambulacral plates; ad, adambu-
lacral plates; 6, the lateral parts of the ambulacral plates; r\ rz, the 2 connate dorsal
marginal plates at the base.

Fig. 18. Base of an ambulacral skeleton split along the middle and viewed from

the surface of fracture: f, furrow for the ambulacral vessel.

14*

Fig. 19. The aboral terminal surface of an arm-vertebra: a, ambulacral plates;
&, adambulacral plates; n, the furrow for the ambulacral vessel.

Fig. 20. The oral terminal surface of the first free arm-vertebra: c, dorsal mar-
ginal plates; a, &, w, as in the preceding figure.

Fig. 21. The extremity of the skeleton of the arm, strongly magnified, viewed from
above: a, the terminal spine-covered calcareous plate.

Fig. 22. The same viewed from below.

Fig. 23. A complete pedicellary from the cuticular sheath of one of the oral spines,
strongly magnified, viewed from the broader side. The exterior cuticular sheath has been,
by means of a solution of potash, rendered transparent; so that the enclosed calcareous
particles appear distinctly.

Fig. 24. The same, viewed from the narrower side.

Fig. 25. The calcareous skeleton of the same pedicellary, still more strongly magni-
fied, viewed from the broader side: a&, the side pieces (a, the proper forceps or jaw;
&, the lower perforated plate for insertion of the adductor-muscles); c, the middle piece
(the articulation-plate).

Fig. 26. The same viewed from the narrower side: m m, the adductor-muscles;
a, &, c, as in the preceding figure.

Fig. 27. The skeleton of a pedicellary from the cuticular sheath of one of the arm-
spines, viewed from the broader side, with closed jaws.

Fig. 28. Another with widely opened jaws.

Fig. 29. The same viewed from above : for all 3 figures the indication a, &, c, as in
figures 25 and 26.

Fig. 30. The extremities of the closed jaws, viewed from above, to shew their
toothed edges.

Fig. 31. Some of the developing cells of the pedicellaries from the base of the
cuticular sheath of an arm-spine.

Fig. 32 - 37. Pedicellary-cells in 6 different stages of development, strongly magnified.

Fig. 38. A litle young Brisinga coronata, viewed from above, natural size.

Fig. 39. The same magnified, viewed from above.

Fig. 40. A part of the same, still more strongly magnified, viewed from below:
ad\ the 2 adambulacral plates meeting in the angle of the arms; m, the oral membrane;
o, the oral aperture.

Fig. 41. Skeleton of a pedicellary from the cuticular sheath of one of the oral
spines in the Brisinga endecacnemos, for comparison with fig. 25.

Fig. 42. Skeleton of one of the larger sort of pedicellaries from the cuticular
sheath of one of the arm-spines of the same species, for comparison with fig. 27.

109

Tab. V.

Fig. 1. skeleton of the disc (the oral ring), with the base of the skeleton of 3
arms in their natural connexion, of a 9-armed specimen of Brisinga coronata, viewed
from above.

Fig. 2. The same viewed from bolow.

Fig. 3. A complete arm-vertebra seen from above.

Fig. 4. The same viewed from the left side.

Fig. 5. The same viewed from the aboral extremity.

Fig. 6. The same viewed from below.

The following indication will serve for the 4 last figures:
a, interior part of the ambulacral plates.
&, lateral parts of the ambulacral plates.
ad, the adambulacral plates.

Fig. 7. The part of the skeleton of a 9-armed specimen of Solaster endeca
corresponding to fig. 1, viewed from above.

Fig. 8. A part of the same viewed from below, for comparison with fig. 2.
Fig. 9. An interradium with 2 half radii of the skeleton of Brisinga coronata repre-
sented in fig. 1.

Fig. 10. The corresponding part of the skeleton of the Solaster endeca represented
in fig. 7.

In both figures the letters indicate corresponding parts:
1, 2, 3, 4, 5 ambulacral plates in their order:
ad, adambulacral plates.
r, dorsal marginal plates.
p p, parietal plates.
w, wedge-plates.

o, aperture for the passage of the radial ambulacral vessel.

x, the border of the oral ring, which below limits the furrow for the circular ambu-
lacral vessel.
I, ligaments.

Fig. 11. A part of the skeleton of the disc (the oral ring) with the base of 3 arms
of the young Br. coronata represented in Tab. IV, fig. 38-40, viewed from above, strongly
magnified: ad, the adambulacral plates of the disc; r, rudimentary dorsal marginal plates;
iv, wedge-plate.

Fig. 12. The same viewed from below: ad2, the second pair of adambulacral
plates (here evidently belonging to the arms); s, embryonic furrow-spines; x, the circular
rim which proceeds from the interior wall of the oral ring, and to which the oral membrane
is attached.

110

Fig. 13. The oral ring of a young specimen of Br. coronata, magnified, viewed from above.

Fig. 14. Part of the same, more strongly mangnified, seen from the exterior side:
w, wedge-plate.

Fig. 15. Part of the oral ring of the same specimen, with the base of an arm in
natural connexion, viewed from above: w, wedge-plate; r, dorsal marginal plates; x, the
circular rim preceding from the interior wall of the oral ring.

Tab. V.

Fig. 1. A part of the disc (a radial space) seen from below, strongly magnified.
The water-feet and the spines are removed, in order to expose the central parts of the
nervous system: a a, the circular nerve-commissure; 6, the radial nerve; c, insertion for
the water-feet; #, a recently formed germ of an arm, on which the terminal organ of sense
and the 2 rows of water-feet are already evidently traced.

Fig. 2. The part of a radial nerve belonging to the disc, together with a part
of the circular commissure, isolated and viewed from the upper (interior) side: a, the longi-
tudinal septum which divides the radial blood-sinus.

Fig. 3. The same vieved from the side: a, the septum.

Fig. 4. A piece of a radial nerve from the basal part of an arm, shewing 2 of the
enlargements that correspond to the water- feet, viewed from above: a, the septum.

Fig. 5. Another piece with 3 enlargements viewed from below.

Fig. 6. The dorsal skin of the disc, viewed from the inner side, with the interior
parts attached, chiefly to shew the complicated arrangement of the ligaments by which the
stomach and the radial caeca are fixed: a, the secretory apparatus; h, the so-called ,,heart";
s, the stone-canal.

Fig. 7. The madreporic body viewed from above, magnified.

Fig. 8. The ,,heart" isolated and viewed from the side.

Fig. 9. A piece of the skin of the same, more strongly magnified, shewing the
reticulated muscular fibres.

Fig. 10. Cells in the skin of the ,,heart", very strongly magnified.

Fig. 11. A recently formed germ of an arm, magnified, viewed from above: a, the
terminal organ of sense.

Fig. 12. The same viewed from below: a, the terminal organ of sense; 6, the
enlargement of the radial nerve at the base of this organ; n, the radial nerve; w, rudi-
ments of the water-feet.

Fig. 13. Another germ of an arm, treated with a solution of potash to shew the
interior skeleton (s) forming itself, seen from the right side: j», nascent furrow-spines;

Ill

pi, radiate calcareous bodies, forming the foundation 'for the terminal calcareous plate of
the arm.

Fig. 14. A somewhat further developed germ of an arm, viewed from below. This
preparation had been lying in maceration for a long time in a solution of potash; so that
all the organic parts are quite transparent, whereby the calcareous skeleton forming itself
is made to appear very distinctly: a, nascent ambulacral -plates; ad, adambulacral plates;
p, furrow- spines; p*, marginal spines with rudiments of pedicellaria?; p 2, thin spines issuing
from the terminal calcareous plate; 6, the terminal organ of sense.

Fig. 15. The extremity of the same germ of the arm, viewed from above, to shew
the formation of the cuticular skeleton: b, the terminal organ of sense; r, calcareous con-
cretions in the dorsal skin forming the foundations for the transversal calcareous ribs;
i, the growing terminal calcareous plate with its spines (p1); p, marginal spines with
nascent pedicellaries.

Fig. 16. The basal part of the skeleton of a recently formed arm, magnified, viewed
from above: a, ambulacral plates; ad, adambulacral plates; r r, the 2 interior dorsal mar-
ginal plates (here distinctly separated) ; r ', rudimentary marginal plates further out on the
arm; p, furrow-spines; p ', marginal spines.

Fig. 17 — 23. Successive development of an ambulacral plate, strongly magnified.

Fig. 24. A recently formed adambulacral plate.

Fig. 25. One of the radially ramified calcareous concrements which form the foun-
dation of the terminal calcareous plate, strongly magnified.

Fig. 26 — 31. Successive development of an arm-spine.

Fig. 32. One of the thin spines attached to the terminal calcareous plate, isolated.

Fig. 32 (bis). A recently formed water-foot.

Fig. 33. A piece of the embryonal cuticular skeleton of the dorsal skin of the
disc, strongly magnified.

Fig. 34. One of the echinulated spines attached to the same.

Fig. 35. Schematic representation of the digestive system (the stomach with radial
caeca) of a 10-armed specimen, viewed from above, natural size.

Fig. 36. 'Section of the disc, and base of an arm of Brisinga coronata (schematic
figure). The section goes to the left, through an interradial space; and to the right, through
a radial space. The dotted line refers to the next figure.

Fig 37. A similar section of Solaster endeca,

The following corresponding indications will serve for both figures:
ft, the furrow for the radial ambulacral vessel.
g, genital organ.

h, the ,,heart" with the stone canal.
m, the madreporic body.
m (bis), the oral aperture.

112

w, r, the oral ring.

n, secretory apparatus.

p, oral spines.

r, radial caeca.

st, stomach.

Tab. VII.

Brisinga endecacnemos.

Fig. 1. A full grown specimen viewed from above, natural size. One of the arms
is completely developed; another is broken off near the middle, and the point recently
regenerated; a 3rd is just newly formed.

Fig. 2. The disc of the same specimen, somewhat magnified, viewed from the side.

fig. 3. The same viewed from below. The folds of the stomach are strongly pro-
truded; so that the oral membrane is not visible.

Fig. 4. The madreporic body, with a piece of the dorsal skin, strongly magnified,
viewed from above.

Fig. 5. A piece of the disc of another specimen, viewed from below, to shew the
oral membrane and the arrangement of the spines.

Fig. 6. 3 of the dorsal disc-spines, with basal plates belonging to them imbedded
in the skin.

Fig. 7. The oral ring viewed from above, magnified.

Fig. 8. A piece of the same, somewhat more strongly magnified, viewed from the
exterior side.

Fig. 9. The same viewed from below. The spines are removed, with exception of
the rudimentary oral spines preceding from the extremity of the interior contiguous adambu-
lacral plates.

Fig. 10. The disc of a very young specimen, viewed from above.

Fig. 11. The base of a dessicated arm, viewed from the left side, slightly magnified.

Fig. 12. Calcareous bodies and small spines from the dorsal skin of the same,
strongly magnified.

Fig. 13. Extremity of one of the transversal calcareous ribs, with a marginal spine.

Fig. 14. One of the small spines situated on the calcareous ribs, strongly magnified.

Fig. 15. Base of the skeleton of the arm, viewed from above.

Fig. 16. The same viewed from below.

Fig. 17. The extremity of an arm, viewed from the left side and strongly magni-
fied. The preparation is treated with a solution of potash; so that the interior skeleton is
distinctly seen through the skin.

Fig. 18. The basal half of an arm of a full grown female, viewed from above,
natural size. The dorsal skin is cut open along the middle and turned back on the sides,
to expose the numerous ovaries attached along both sides.

Fig. 19. An ovary consisting only of a single csecum, isolated.

Fig. 20. Another ovary, with incipient dichotomic division.

Fig. 21. Base of an arm of a full grown male, viewed from above. The dorsal skin
is, as in the preceding preparation, cut open along the middle; its left half is removed,
while the right is spread out to expose the cluster-shaped seminaries.

Fig. 22. A seminary isolated and more strongly magnified.

'^^V-

.

i

I

•. ! /•SV'

N -r

RETURN TO the circulation desk of any
University of California Library
or to the

NORTHERN REGIONAL LIBRARY FACILITY
Bldg. 400, Richmond Field Station
University of California
Richmond, CA 94804-4698

ALL BOOKS MAY BE RECALLED AFTER 7 DAYS
2-month loans may be renewed by calling

(415) 642-6753
1-year loans may be recharged by bringing books

to NRLF
Renewals and recharges may be made 4 days

prior to due date

DUE AS STAMPED BELOW

MAY 28 1991

APR 04 1991

LD 21-loOm-8,'34

,ot> *c**" •**-*'**

**.'.

5?^

£3

3ft