M. C. Mereschkowsky on the Hydroida. 239 
beneath very pale, almost white. Five dark violet stripes, 
a mesial and two pairs of lateral, extending along the entire 
length of the dorsal surface. The mesial stripe narrow and 
linear, the succeeding pair broad and band-like, and the 
outermost pair again linear. The outermost pair placed at a 
short distance from the lateral margin of the upper surface, 
and the band- like pair at half the distance between these and 
the central stripe. Just behind the head the two lateral 
bands on either side fuse together, and form a pair of broad 
dark patches. 
Faint and narrow violet stripes mark the margin of the 
ambulacral line on the under surface of the body. 
Length of the single specimen 9 inches ; extreme breadth 
of the body \ inch, of the head -g- inch. 
Exeter College, Oxford, 
Feb. 18, 1878. 
XXIX. — Studies on the Hydroida. By C. Mereschkowsky. 
[Plates XIII., XIV. & XV.] 
I. Morphological Considerations. 
The human mind has not the power of retaining in its 
memory the representations of all the concrete objects which 
are presented to its five senses ; for the number of these objects 
and of facts is too immense for its faculties, which are still so 
imperfectly developed. But, at the same time, the mind de- 
sires to be in possession of as many facts as possible ; hence 
the tendency to generalization and the double character of 
every science : on the one hand, we have concrete facts 
without any bond between them, without any idea, serving 
only as raw material ; on the other, generalizations, more or 
less abstract ideas. Not only every science, but even every 
branch of each science, every group of events or facts, may 
therefore have its philosophy — that is to say, its generaliza- 
tions, its ideas, its laws which govern the facts. 
The usefulness of these laws or generalizations, even in the 
case of small groups of events, cannot be doubted ; in reality 
it is often only by taking advantage of them that a thinker can 
arrive at generalizations of a higher degree, without the neces- 
sity of busying himself in the midst of thousands of little facts 
and minute details. 
In the following pages I shall speak of a group of 
facts which may be observed among the Hydromedusfe, 
240 
M. C. Mereschkowsky on the Hydroida. 
and which may be generalized into a single idea, a single 
law of metamerism {Metamerengesetz) or oi articulation. This 
law may be formulated as follows : — The Hydroid may he 
composed of two or several metameres^ similar or not : each 
metamere in its turn is composed of several antimeres. 
Fig. 1 shows diagrammatically a Hydroid belonging to 
what I call the articulate type ; it will be seen that it con- 
sists of three very distinct metameres, each of which is in 
its turn composed of four antimeres*. 
This law governs a considerable number of forms among 
the Hydromedusse ; we may recognize it in the species be- 
longing to various genera — for example, in Stauridium^ Coryne, 
Fig. 2. 
Fig. 1. 
Syncoryne^ Millepora^ Cladonema, Tuhularia, Cordylophora 
Gemmariaj &c. ; but all the cases in which we remark the arti- 
culate type among the Hydromedus^e belong exclusively to 
the order of naked Hydroids (Athecata). There is not a 
single Hydroid belonging to the order Thecaphora which 
* According to M. E. Hackel we should have to call such a type 
^[forma staurasfoma cfiplopola articulafa,'" 
M. C. Mereschkowsky on the Hydroida. 241 
has the least normal tendency to the production of meta- 
meres*. 
The number of metameres is very variable in different genera 
and species. We know several Hydroids (as, for example, 
Cladonema radiatum) which have only two metameres, usually, 
in this case, very distinct and well marked. Up to the pre- 
sent time we do not know a single Hydroid which has three 
distinct metameres ; but we know one with four very 
clearly developed ; this is Stauridiuni 'productum^ which I 
have found in the White Sea. After it come the forms 
which have more than four metameres ; but in these cases they 
are not very distinct, and their number is no longer constant, 
but varies with the age and development of the individual. 
At the same time, this variation in the number of metameres 
takes place within certain limits ; for there are species in which 
the metameres never attain the great number met with in 
other species. 
Among the forms which are very rich in metameres I can 
cite several, but especially Coryne pusilla and Gemmaria 
implexo^ in which we may see a very great quantity of meta- 
meres. It is true that in such cases it is impossible to fix 
clearly the boundaries of two metameres, and that it is often 
impossible to decide whether two tentacles belong to one or 
to two different metameres ; but nevertheless it is easy to see 
that we have to do with the same articulate type as in Stau- 
ridiuin and Cladonema ; only here the order of arrangement of 
the tentacles, in consequence of their great number, has be- 
come very much effaced. 
As to the nature [das Wesen) of the law of metamerism, the 
cause which has produced the articulate type among the Hy- 
droids, I think I am justified in explaining it in the following 
manner : — The Hydroid, in consequence of a great abundance 
of nourishment, or from some other cause unknown, began to 
grow in the direction of the primary axis of its body. Growth, 
as we know, does not differ generically from the process of 
multiplication ; the latter is only a particular case of the 
former ; and the two processes depend greatly upon each 
other. 
The growth of the Hydroid beyond its specific limit causes 
* A single anomalous fact is known to me in the Tliecapbora, in which 
there appears a tendency, although a very feeble one, to take on the arti- 
culate form. This is the Clytia poterium, Agassiz (fig. 2), in which one 
hydrotheca is placed above another, which has produced it, no doubt, by 
division. But this case can only be regarded as an anomaly, the nor- 
mal individuals never having any trace of metamerism. This anomaly 
has been described by Agassiz, ' Contributions to the Natural History of 
the United States,' iv. p. 303, pi. xxix. fig. 1. 
242 M. C. Mereschkowsky on the Hydroida. 
multiplication by means of incomplete transverse division 
{imvollstdndige Quertheilung) — that is to say, the appearance 
of one or several new systems of secondary axes, which 
are all, at first, in accordance with the law of heredity, equal 
among themselves and to the first axial system from which 
they have proceeded. But instead of separating from each 
other and entering upon a free and independent life, as we 
see in a very analogous process of gemmation in the Scy~ 
pMstoma of the Discophorous Medusae (fig. 3)*, each system 
of secondary axes remains connected with a small community 
and leads a social life (fig. 4) . 
In order to demonstrate that this view is correct, and that 
the articulate type is nothing more than the product of an 
incomplete transverse division, we may consider the singular 
anomaly presented by Glytia ])oterium, Ag. (fig. 2). There 
is no doubt that this form is produced by increase of growth, 
which for its part produces a transverse division analogous to 
that which takes place in the ScypMstoma^ but with the dif- 
ference that here the superior articulation does not separate 
from the colony, because the division is incomplete. But if 
we imagine the Hydroid deprived of its calycle we shall have 
fundamentally the same picture that is presented by the digram- 
matic figure of an articulate Hydroid in fig. 1. 
Let us now consider fig. 5, which represents, after Mr. 
Hincksf, the interesting Hydroid Vorticlava proteus. Owing 
to its great contractility it can take on different forms ; and 
one of them (fig. 5), in which the superior metamere is 
removed to a great distance from the inferior metamere, and 
in which the two articulations are united only by a long and 
very thin peduncle, proves very clearly that the metameres 
are true articulations produced by incomplete division. Just 
the same thing (that is to say, the great individuality of each 
metamere, united only hj a fine peduncle) occurs also in Cory- 
morpha pendula, Agass.J ; only the individuality of the meta- 
meres is unequal, being easy to see in some Hydroids (such 
as Vorticlava, Stauridium, and Gladonema) ^ and more or less 
effaced in others with many metameres (Zanclea, Co?'yne, 
&c.). But in any case we must regard the articulate type as 
a small colony. 
The number of individuals in such a community may be 
* Each tentacle (each antimere), or rather each pair of antimeres, is 
nothing but an axis vertical to the axis of the body (principal or primary 
axis), which is ordinarily called a secondary axis. The different types of 
Hydroids have I, 2, 3, 4 n, ... . secondary axes, t. e. 2, 2x2, 2x3, 2x4, 
2 X n tentacles. 
t Mon. Brit. Hydr. vol. ii. pi. xxiii. fig. 2, d. 
J Contrib. Nat. Hist. Un. States, vol. iv. pi. xxvi. figs. 14 and 17. 
M. C. Mereschkowsky ow the Hydroida. 
243 
considerable, as we have already seen, about ten or even 
more ; and, in fact, it is most usual for their number to ex- 
ceed four or five. 
Fig. 4. 
Fig. 3. 
Fiff. 5. 
In these cases, and especially where we have to do witli 
about ten raetameres, the explanation of the origin of these 
articulated forms which I have just given cannot suffice ; 
and hence it must be supplemented by a very in- 
teresting law, which Prof. N. Wagner of St. Peters- 
burg has denominated tlie law of ijliysiological inertia^ 
and has so happily applied to the explanation of the 
incredible number of metameres (articulations) with 
which various worms (Annelida) are furnished. 
According to this law, some cause having origi- 
nated two or three metameres, the appearance of the 
following raetameres may be brought about without 
the further aid of the primary cause, but solely under 
tlie influence of a tendency that the organism has to repeat 
the process of the appearance of metameres (a process at first 
induced by some external influence \choque\^ such as abun- 
dance of food) by inertia^ as it were, until finally resistance, 
under different forms, may put a stop to it. 
The difi^erent qualities and properties of an organism are 
often retained, by force of heredity, without interruption and 
without modification during a long series of generations, 
even when the cause which has induced these qualities has 
long disappeared. It is so in the case in question : a certain 
244 M. C. Mereschkovvsky on the Hydroida. 
cause has induced iu the organism the tendency to grow 
constantly in length by incomplete transverse division ; and 
if it happens that this cause acts for a long time, through a 
long series of generations, it is easy to understand that this 
tendency may acquire so great a persistency, and may become 
so powerful, that it will continue to manifest itself even after 
the disappearance of the original cause. Considered from this 
point of view, the laio of physiological inertia appears simply 
to be a particular case of another more general law — the law 
of heredity ; and I believe that if we apply this law (without 
which the phenomena of the Annelida are perfectly obscure 
and incapable of explanation) to the group of Hydroids, and 
especially to the articulate type, we shall attain the possi- 
bility of explaining and understanding the appearance of such 
forms as Coryne pusilla and Gemmaria implexa. 
In the articulate type there is a peculiarity which is very 
interesting, especially because it can be very clearly explained, 
and to which I wish to call attention, namely the form of the 
tentacles. One of the most characteristic features which 
always accompany the law of metamerism in the Hydroids, is 
the capitate form of the tentacles, which in this case, are 
always very short (figs. 1, 4). This peculiarity of the articu- 
late forms is especially observable in the species with nume- 
rous metameres, in which the tentacles are excessively short. 
There are very few exceptions to the rule that articulation is 
combined with the capitate form of tentacles ; and nearly all 
these exceptions can be perfectly well explained. 
In seeking to explain this fact, and to find the cause of its 
occurrence {raison d^etre)^ we must first of all call attention to 
the coexistence of the two facts, articulation and capitate ten- 
tacles, and inquire whether this singular and invariable coex- 
istence is not due to a causal relation between the two facts. 
It is more than probable that this is the case ; and, as we shall 
see immediately, it is the articulation that is the cause 
of the form of the tentacles. 
The articulate type has no doubt originated from the non- 
articulate type with 4, or, in general, Ixn filiform tentacles. 
The tentacles of this general type (2 x w) are: — 1, usually 
rather long and slender, not capitate, endowed with great con- 
tractility ; and, 2, covered over all their surface with a quan- 
tity of thread-cells. Such an organization is adapted to sub- 
serve two functions at once — namely, (1) seizure of food, 
and (2) defence against enemies by means of the veno- 
mous thread-cells. When the Hydroids furnished with meta- 
meres began to be developed from this type thus constructed 
and non-articulate, and, at the same time, the length of the 
M. C. Mereschkowsky on the Hydroida. 
245 
Fig. 6. 
whole animal increased considerablj, so that the long, slender, 
original tentacles, which previously extended beyond the apex 
of the body (where the mouth is placed), became relatively 
shorter, they would no longer reach the buccal orifice, which 
would deprive them of all power of acting as organs of nutri- 
tion. This must certainly take place, especially with the 
lowest tentacles. The part which they performed being thus 
diminished, and their significance in the economy of the 
animal changed, the tentacles would no doubt undergo, if not 
complete atrophy, at least a consi- 
derable diminution in their develop- 
ment. This is, in fact,what we ob- 
serve. In such articulate forms as 
Gladonema radiatum (fig. 6), for ex- 
ample, which is furnished with two 
very distinct metameres, we remark 
that the four lower tentacles, belong- 
ing to the inferior metamere, are too 
short to reach the mouth, and conse- 
quently cannot possibly assist in the 
process of nutrition; at the same time 
they are much less developed, much 
shorter and more delicate than the 
other four tentacles belonging to the upper metamere, which 
can very easily reach the mouth. 
The form wnth four metameres [Stauridium productum) 
shows us the same thing. In this also the four tentacles of 
the inferior metamere are, to a very great degree, atrophied*. 
The same thing takes place in all the other articulate forms, 
even when the number of metameres is very considerable, as, 
for example, in Coryne pusilla and Gemmaria implexa. In 
all cases the inferior tentacles are less developed, half or one 
third of the length of those of the superior metameres, and the 
more they approach towards the basal extremity of the body 
the shorter they are, so that in the lowest regions the length 
of the tentacle often does not exceed its thickness ; but the 
superior tentacles, as well as the inferior, are comparatively 
much shorter in the articulate than in the non-articulate type. 
This atrophy of organs evidently depends upon a diminution 
* We know no form governed by the law of metamerism having 3 
metameres ; but it is easy to see that such a form must once have existed, 
and that it, perhaps, still exists in some little-iuvestigated sea. If it be 
found some day, we may predict with great probability, from the evidence 
of the forms with 2 and 4 metameres, that it will also have 4 tentacles 
(or 2xw) belonging to the inferior circle more atrophied than the rest. 
The genus Triridium, to which this hypothetical Hydroid must belong, 
is repre-sented in fig. 1. 
246 M. C. Mereschkowsky o« the Hydroida. 
of their utility to the organism ; for in the articulate type the 
tentacles, instead of fulfilling two functions at once, only per- 
form one, namely that of defence against enemies. At the 
same time it has become possible for the organ to adapt itself 
better to the single function of defence than before, when it 
required also to capture food ; it has attained the possibility 
of retaining the characters which are only useful for defence 
and which are even injurious to prehension. It is precisely 
this possibility of adapting themselves to the single function 
of defence that is the cause of the tentacles in the articulate 
type being very short and capitate at their extremity. 
Imagine now a Hydroid reposing after a full meal, with 
its tentacles quietly expanded in the water and gently moved 
to and fro by the waves. When any enemy approaches it 
with hostile intentions and is inclined to attack it, the assailant 
must most certainly strike against the ends of the tentacles 
before it can touch the body of the Hydroid. Upon the effect 
produced by this first contact with the ends of the tentacles 
will depend all the subsequent actions of the enemy : if it 
receives a very strong charge it will be killed on the spot, 
or will make its escape as quickly as possible ; in the contrary 
case, when the pain caused by the thread-cells arranged in 
the ends of tentacles is too insignificant, the enemy may arrive 
at the very body of the Hydroid, which is then menaced with 
great danger. We see, therefore, that, for the purpose of 
self-preservation, it is very important that the first line of 
fortifications, so to speak, should be as strong as possible — in 
other words, that the ends of the tentacles should be as for- 
midably armed as possible, that there should be as many 
thread-cells as possible in these ends ; for those which are 
placed in the other parts of the body and tentacles are not of 
equal importance for the purpose of defence. To fulfil all 
these conditions it is clear that the tentacles must be inflated 
at their extremities, in order that a great quantity of thread- 
cells may be accumulated in the enlargement. When once 
these tentacles have ceased to act as organs of prehension, it 
is no longer necessary that they should be long, fine, supple, 
and movable ; this is why in the articulate type, at the same 
time that they acquire the capitate form, they also become 
much shorter than usual. 
It generally happens that in those cases in which the body 
becomes very much elongated it acquires great flexibility and 
the faculty of twisting about very briskly, and so assists the 
proboscis in the capture of food, whilst, on the other hand, 
this organ in such cases also becomes strongly developed and 
very mobile. This flexibility of the body consequently re- 
M. C. Mereschkowsky on the Hydroida. 247 
places the want of tentacles for the function of alimentation, 
as is veiy well shown in fig. 5, p. 64, of the ' Histoiy of British 
Hjclroid Zoophytes' byMr.Hincks, as well as by the description 
which accompanies it*. In the Hydroids without metameres, in 
Avhich the body is consequently very short and not flexible, 
the tentacles are always filiform, long, fine, and very supple ; 
their length sometimes even becomes very great, as, for ex- 
ample, in Monobrachium parasitum^ mihif, which has only a 
single tentacle. 
This, then, is the explanation that, I tliink, may be given 
of the fact that the articulate type of the Hydroids is asso- 
ciated with short and capitate tentacles. 
This view is further supported by the fact that the capitate 
tentacles are exclusively met with in the order Athecata, or 
the Gymnoblastic Hydroids — that is to say, among the naked 
Hydroids, — and that, on the contrary, in the order Thecaphora, 
in which each hydranth is furnished with a hydrotheca or 
calycle of chitine within which it can entirely withdraw itself, 
and which often may even be closed by a small operculum, 
we only find filiform tentacles. This is very easily explained, 
seeing that these Hydroids, which are very well defended from 
all attacks of their enemies by the hydrothecaa, within which 
they can conceal themselves in case of danger, have no neces- 
sity for organs so well designed for defence as are the capitate 
tentacles. On the other hand, as the Thecaphora grow in 
very numerous colonies, the number of individuals sometimes 
exceeding 1000, it is necessary for them to adapt themselves 
to the possibility of procuring food in sufficient quantity for 
so great a number of individuals living together. This adap- 
tation in the case in question consists in the number of long, 
fine, filiform tentacles appropriated to prehension with which 
each individual is provided becoming very great, greater than 
it usually is in the naked Hydroids. (There are generally 
not fewer than 16, most frequently 20, 22, 24, and sometimes 
30, 32, or more.) 
Finally, I may mention another fact, which will serve in 
* In fact the flexibility of the body of such Hydroids as Cladonema 
radiatum, Stauridium p7-oductum, and others is excessively developed, and 
may very -well compensate for the want of filiform tentacles in the func- 
tion of prehension of nourishment. But it is especially in Clavatella 
prolifera that the length, contractility, and flexibility of the body have 
attained their maximum : and it is, I think, by this cause that we may 
explain why it also has capitate tentacles, although not belonging to the 
articulate type. It furnishes the only example of capitate tentacles in a 
non-articulate type. 
t See my paper, " On a new Genus of Hydroids from the White Sea," 
in this journal for September 1877, ser. 4, vol. xx. p. 220. 
248 
M. C. Mereschkowsky on the Hydroida. 
support of the explanation above given of the forms with 
capitate tentacles. I refer to the blastostyles of the genus 
Hydractinia. It is well known that the gonophores, or sexual 
individuals, appear upon the surface of the body of the tropho- 
somes, or nutritive individuals, which are furnished with 
several filiform tentacles. When these gonophores appear, 
the individual upon which they are seated, and which is then 
called a " blastostyle," becomes much thinner and smaller (the 
material of the animal being absorbed by the sexual bodies), 
the mouth closes, and the tentacles (which, from this moment 
lose their importance as organs subserving the purpose of 
nutrition, since the mouthless individual cannot feed) become 
shorter and shorter and more and more insignificant. Soon 
we can only perceive a few knobs or tubercles furnished with 
a great quantity of thread-cells, greatly resembling the dila- 
tations with which the capitate tentacles are furnished. At 
the same time these tentacles or tubercles only retain the 
function of defence from enemies. 
But I have said that there are exceptions to the rule that 
articulation is associated with capitate tenta- 
cles, and that these exceptions are not con- Fig. 7. 
tradictory to the explanation that I have 
given ; on the contrary, it is possible to 
explain these exceptions only by admitting 
all that I have said above. 
There are some forms, evidently belonging 
to the articulate type, which do not possess 
capitate tentacles, but, on the contrary, have 
those organs slender, filiform, and very long. 
For example, Clava'^, Cordylophoraj &c. 
(especially Cordylojphora) have tentacles 
longer than in any other species. This is 
to be explained as follows : — In becoming 
developed into the articulate type the Hy- 
droid became more and more elongated, 
whilst the tentacles remained the same, which 
rendered them relatively shorter ; and it is 
precisely this that induced their capitate 
form. But if we assume that all the time 
the elongation of the tentacles proceeded Cordylophora 
side by side with the elongation of the body, lacustris. 
we shall see that their original significance, 
as aiding in alimentation at the same time as for defence, 
* Especially Clava leptostyla, Ag. (A. Agassiz, Illustr. Cat. Mus. Comp. 
Zool. ii. p. 170, tig. 274). In the cases in whicli the tentacles are not 
too long we maj' admit that these species have only become articiilate 
M. C. Mereschkowsky on the Hydroida. 
249 
Fig. 8. 
must remain intact ; for in proportion as the body became 
elongated, the tentacles lengthened likewise, so that they 
could always reach the mouth and convey food to it (fig. 7). 
But if the functions of the tentacles did not undergo any 
change, we need not expect them to change their form, except 
perhaps to become longer. This is what we remark in such 
forms as Cordylojjiioraj Clava, &c., in which the lowest ten- 
tacles are not in the least shorter or less developed than the 
upper ones; on the contrary, they are sometimes a little longer; 
and in all cases they are all, without exception, longer than 
the body, owing to which they all have the faculty of assisting 
in the capture of prey, as has been very well described by 
Van Beneden in the case of Cordylo^jhora* . 
I must still mention an articulate type, represented by the 
genera Tubular ia, Acliaradria^ Gorymorpha^ Pennaria^ &c., 
in which the superior metamere has 
the tentacles capitate, but much less 
developed than those of the other me- 
tamere. All these forms are derived 
from a non- articulate form with tenta- 
cles so well developed, so long, and in 
such great quantity^ that when the 
formation of the second metamere was 
induced by some cause, the tentacles 
belonging to it were perfectly useless 
to the organism, which caused them 
to become atrophied, and at the same 
time capitate — that is to say, adapted 
solely to the defence of the organism 
(fig. 8). As the Medusa may be re- 
garded as a hydranth reversed, and in 
which the tentacles (radial canals) are united together by a 
gelatinous substance (ectoderm or bell), it may be understood 
that the Medusse which have tentacles at the extremity of 
the manubrium belong to the same articulate type as Tuhu- 
laria^ Acharadria^ &c., with two metameres, the inferior of 
which is more strongly marked than the upper. 
To complete this morphological chapter I propose to explain 
in a few words a point of view from which I regard all the 
Acharadria larynx. 
quite recently, and that they are in process of forming capitate tentacles, 
or of lengthening them. Thus Mr. Allman remarks, "Some Hydroids 
with filiform tentacles show, like Clava squa?nata, a tendency to the ter- 
minal enlargement of the tentacles in certain states of contraction" 
(Allman, Mon. Gymnobl. Hydr. p. 245). 
* Van Beneden, ' Faune littorale de Belgique,' Polypes. 
An7i. & Mag. N. Hist. Ser. 5. Vol i. ' 17 
250 M. C. Mereschkowsky 07i the Hydroida. 
forms and in general all the morphological facts presented by 
the Hydroids, and which I think may contribute somewliat to 
the better comprehension of the idea of the Hydroid, so to 
speak, and to concentrate all the differences presented by this 
group in a single representation. 
Every one at present regards a hydranth, with its tentacles, 
as a single individual, furnished with organs radially ar- 
ranged. For my part, I believe we must regard such an 
organism, not as an individual, but as a colony composed of 
two kinds of individuals — the one kind appropriated to the 
function of seizing food, with the gastral cavity but slightly 
developed, without a mouth, very flexible and thin {tentacles) ; 
the other destined exclusively to nourish the whole colony, 
furnished with a mouth, and Avith a large cavity in the body 
which is but slightly flexible [the actual body of the hydranth). 
We should thus have a polymorphic colony after the fashion 
of the Siphonophora ; and this polymorphism is explained here 
also by adaptation to different functions by the division of 
labour. Certainly before this division of labour was effected 
the colony only consisted of similar individuals, produced from 
the parent individual by gemmation ; and it was then that the 
individuality of each individual must have been most strongly 
marked ; but in course of time, in proportion as the division of 
labour was effected, this individuality was effaced, and the 
whole acquired more or less the character of a single individual 
furnished with several organs. This is what we see now-a- 
days. Therefore, in saying that the hydranth must be re- 
garded not as an individual but as a polymorphic colony, T 
do not wish by any means to say that each tentacle is a true 
individual, but only that it has been so formerly, and that it 
has retained [this character] in part even to the present day. 
I may, moreover, urge the enormous difference which exists 
between the organ tentacle and the organs of other animals 
— a foot, for example, and still more the hand of man ; 
this difference is profound and primordial {'principielle)^ be- 
cause a hand (or, in general, any organ) is not homologous 
with a tentacle, and is only analogous to it in its physiological 
function. 
Such a view as this would perfectly explain the origin of 
the organ tentacles, which would be merely the result of the 
reproduction of a Protohydra^ Leuck., or rather an Archhydra^ 
Hack., by the process of gemmation. From this point of 
view, therefore, I must give the name of individual to each 
axisoi cylindrical form, composed of ectoderin and endoderm ; 
and it is very remarkable that any Hydroid, however com- 
plicated it may be, appeared at first precisely in the form of a 
M. C, Mereschkowsky on the Hydroida. 251 
cylinder with a single diplopolar axis, in this respect differing 
in no way from the first appearance of a tentacle upon a 
hydranth, or of a medusa. In fact it is impossible to distin- 
guish a tentacle, a medusa, and a hydranth at the first moment 
of their appearance (fig. 9) ; each of them 
is merely an Archhydra or a Protohydra^ ^^S- ^• 
which, if we accept the biogenetic law, C^'\_^ 
leads us to believe that they are all dif- ^ i^ 
ferent modifications of a single primitive 
organism, and that they are all homolo- r\ 
gous. u\ \c\ a 
At any rate, I believe that to regard ' 
a (stauraxonic) hydranth as a colony of «^ Young hydrotheca. 
(monaxonic) Archydrce is to look at the ^- X°^^°^ tentacle. 
£c ■ -i. • i.' 1 i. n f- i oung medusa, 
aiiair as it is lundamentally. ° 
Let us farther remember two interesting Hydroids, namely 
Ophiodes mirahilis^ and Ophiodes parasiticus^ Sarsf, which, be- 
sides the tentaculiferous individuals (colonies according to me), 
have monaxonic individuals, without tentacles, and absolutely 
presenting no difference from the tentacles of certain Hydroids. 
And this case proves further that the tentacle (an individual), 
which cannot serve the colony either by procuring or by 
digesting food, only remains useful to it by serving to defend 
it, a function which induced the capitate form of the tentacles. 
In fact, the monaxonic individuals have no mouth, and there- 
fore do not aid in nutrition ; and, at the same time, they are 
often placed so far from the colony-individuals that they can- 
not serve for seizing food. The function of defence, therefore, 
alone remains for them ; and we find that they have acquired 
the capitate form, which we have seen to be appropriated to 
defence. 
Il, Remarks on the Reproduction o/'Obelia 
flabellata, Hincks. 
Among about forty species of Hydroids that I have ob- 
served and collected in the White Sea, Ohelia Jlabellata, 
Hincks, is very frequently met with. At the end of the 
month of June I found it with a great quantity of gonothecse, 
all filled with young Medusje in various stages of development. 
Although in all other respects the Ohelia flabellata of the 
* Hincks, Mon. Brit. Hydr. Zooph. pi. xlv. fig. 2, p. 231. 
t Gr. Sars, "Bidr. tilKundsk. om Norges Hydroider," Forh. i Vidensk/ 
Selsk, i Christiania, 1873, p. 109, pi. iv. figs. 5-8. 
17* 
252 M. C. Mereschkowsky on the Hydroida. 
White Sea did not differ at all from that described by Hincks*, 
the gonotheca was distinguished by the absence of the little 
tubular elevation placed upon its flattened summit, of which I. 
never observed any trace ; nevertheless it can only be re- 
garded as a feeble arctic variety of the British Obelia jiabel- 
lata. 
The development of the Medusae has been studied princi- 
pally by L. Agassizf, who was the first to publish some im- 
portant facts, and by F. E. SchultzeJ, as well as by Mr. All- 
man (the last on Gorymorpha nutans)^ who have made some 
alterations in the views current before their time. But as most 
attention has been paid to the Meduste belonging to the order 
Athecata, I have thought that it might not be altogether with- 
out interest to have their results confirmed by a Medusa be- 
longing to the quite different order Thecaphora. 
Plate XIII. fig. 1 shows the first commencement of a Medusa, 
which only consists of a protuberance (expulsion) of the walls 
of the blastostyle, composed, like the latter, of two layers, the 
ectoderm and endoderm, separated from one another by a very 
thin layer, which is not so distinctly contoured as in the buds ; 
it is, no doubt, the hyaline intermediate layer between the 
ectoderm and endoderm which Schultze calls the " Stlitzla- 
melle," In form, this bud differs in no respect from a young 
hydranth developed from a planula ; and both have exactly 
the same form as Protoliydra Leuckartii, Greef, and the same 
as must have been possessed by Hackel's Archhydra^^ i. e. if 
we choose to accept Hackel's biogenetic law. The next 
stage is represented in fig. 2 (PL XIII.) ; we see that the bud 
has considerably widened, and that the ectoderm (which is 
here also distinctly separated from the endoderm by the double- 
contoured line) has become much thicker at the summit of 
the bud than elsewhere. The thick part, which is in the form 
of a cone, is turned downwards towards the endoderm, in which 
the cone buries itself; the inner surface of the endoderm and 
also the general cavity of the body retain their original form : 
they do not form any expulsion ; and their apex is always 
hemispherical. But, at the same time, we already remark 
that the depression of the outer surface of the endoderm is not 
a regular cone, but, on the contrary, that the edges of this 
depression are dentate. There are four teeth formed by the 
superior layer of the endoderm, and between these teeth four 
depressions occupied by the inferior layer of the ectoderm. It 
* 'History of British Hydroid Zoophytes/ i. p. 157. 
t ' Contributions.' 
X ' Ueber den Bau der Syncm-yne Sarsii,^ 1873, p. 27. 
§ It is his form " monaxonia diplopola inarticulata." 
M. C. Mereschkowsky on the Hydroida. 253 
is easy to see that the first stage consists in the depression of 
the ectoderm (it is the latter that is active here), and that 
the first indications of radial canals are not the result of the 
expulsion of the endoderm into the ectoderm, but, on the con- 
trary, of the impulsion of the ectoderm into the endoderm. 
The latter continues quite passive ; it submits to the action 
of the ectoderm, which of itself begins to thicken, and by this 
means to bury itself in the endoderm. The following stages 
(figs. 3, 4, and 5) are only the more advanced stages of 
the process which we have already analyzed : the ectoderm 
becomes more and more developed, at the same time that it 
buries itself in the interior, leaving in their place only the 
four spots, which thus become converted into radial canals. 
In fig. 3 the apex of the cavity is already excavated, and there 
are faint indications of the four canals (only two are repre- 
sented) ; but speedily this cavity again becomes convex, and 
it is from this moment only that the endoderm becomes active ; 
it thickens in the middle (fig. 4), and begins in its turn to 
penetrate the ectoderm in order to form the manubrium. In 
fig. 5 we see the same stage with the four canals. It will be 
seen that between the two neighbouring canals there is only a 
uniform layer of ectoderm in which they are immersed, which 
proves that F. Schultze is right in not admitting any interme- 
diate layer between the canals and distinct from the ectoderm, 
as was done by Agassiz. After all this, according to F. 
Schultze, the ectoderm itself splits into two layers, one of 
which produces the muscular sac belonging to the umbrella, 
and the other forms the superior layer of the manubrium*. 
Unfortunately various circumstances drew me away from 
these observations ; so that I have not seen the stage inter- 
mediate between fig. 5 and fig. 6, in which the Medusa is 
I'cady to detach itself. 
As the Medusa of Ohelia Jlabellata is, so far as I know, 
undescribed, I will give a short account of it. PI. XIII. fig. 7 
represents a mature example, and shows that this Medusa 
differs very little from other Medusse of the same genus, as, 
for example, that of Ohelia dichotoma figured by Hincks t- 
The umbrella is very flat, but slightly campanulate, without 
thread-cells, with four radial canals, four oval sporosacs filled 
with ova and placed at the extremities of the four canals, 
where they unite with the circular canal, and where conse- 
* By this the development of the Medusa of Ohelia fiahellata differs 
from that of Corymoi-jyha nutans, in which the division of the ectoderm 
takes place sooner, as early as the first indications of the radial canals 
(AUman, Monogr. Gynmobl. Hydr. p. 77). 
t TjOc. cit. pi. xxviii. fig. 1, c, d. 
254 M. 0. Mereschkowsky on the Hydroida. 
quently the nutritive material attains its maximum abun- 
dance ; for here the current of the radial canal unites with the 
current of the circular canal. The margin of the Medusa is 
furnished with eight lithocysts and a great quantity (more 
than thirty) of short tentacles, which are only sixteen in 
number at the moment of liberation. The manubrium is 
short, changes much in form, and is furnished at its orifice 
with four rounded lobes. The size is very variable, but it is 
usually about 6 millims. in diameter. It is completely colour- 
less, whitish ; the sporosacs are slightly yellowish. By 
leaving in a marine aquarium a branch of Ohelia Jlahellata 
with gonothecse, one can always obtain as many Medusae as 
one wants. Fig. 7 a shows a Medusa of the natural size. 
The ova are large, of irregular form (PI. XIII. figs. 8, 9, 10), 
with a very thin membrane and granular contents. In the 
middle, or more frequently near the margin, we always ob- 
serve very distinctly a large, clear and non-granular nucleus, 
more regular than the ovum itself. In the nucleus we always 
observe one or several nucleoli, and in each nucleolus a nu- 
cleolulus. All these formations are distinguished from each 
other by their behaviour with transmitted light ; when the 
first of them is lighter, the second is darker, and the third 
again lighter. On changing a little the focal distance of the 
microscope all is changed ; what was dark becomes lighter, 
and vice versa. In the youngest ova we see only one nucleo- 
his and one nucleolulus (fig. 8) ; the latter is usually very 
variable in its form, whicli is most frequently irregular. 
Sometimes it is very large (fig. 11, representing a nucleus 
very much enlarged). Further, we see ova in which the 
nucleolus has acquired a biscuit-shape, in each half of which 
we observe a nucleolulus which has evidently divided into 
two (figs. 9, 14) . A subsequent stage may be seen in fig. 10, 
in which the nucleolus is completely divided and each half 
contains a nucleolulus, A still more advanced stage shows 
(fig. 12) a nucleus with four nucleoli, each containing a 
nucleolulus, which is very large and variable in form. The 
form changed before my eyes with considerable rapidity, and 
the whole moved like a little Amoeba. Lastly, the succeeding 
stage that I have been able to observe (fig. 13) is furnished 
with a nucleus with a great quantity (about twenty) of 
nucleoli, almost every one of which contained a very small 
nucleolulus, which, however, it was sometimes impossible to 
define. 
It is evident that all these nucleoli have originated from a 
single one by division, and that this division was always 
preceded by the division of the nucleolulus into two. Only 
M. C. Meresclikowsky on the Hydroida. 255 
once I observed in a perfectly round nucleolus more than one 
nucleolulus, or rather a single nucleolulus in the centre sur- 
rounded by five or si:?^ very small granules forming an aureole 
round the centre. 
As I observed all these stages of development in ova which 
had not issued from the sporosac, and, moreover, there was 
not a male individual in the neighbourhood, all the processes 
described took place in ova not yet fecundated. It would be 
interesting to know wliy there is all this enormous complica- 
tion. But as yet the facts are too few to permit us even to 
think of an explanation. 
Lastly, this hydroid has offered me another interesting fact 
which I will mention. For the purpose of observing the 
development of the Medusae I placed a branch of Obelia 
jiabellata upon a slide, and laid over it a covering-glass ; the 
sea water by evaporating became salter and Salter, which 
(and perhaps also the want of oxygen) appeared to affect the 
organism in a singular manner. In a short time J remarked 
that the coenosarc of the stem detached itself in fragments of 
different sizes, especially near the end. First of all there 
appeared a constriction in a particular spot ; this constriction 
became deeper and deeper ; and finally the two parts separated 
entii-ely, so that the end of the stem formed a fragment quite 
independent of the colony. The two parts contracted, moved 
away from each other, and became rounded at their ends, so 
as to leave no trace of their lesion (PL XIII. fig. 16); the 
ectoderm and the endoderm recurved at the newly formed end 
just in the same way as in the end of the stem : the cavity 
was very distinct ; and I could even observe the movement of 
nutritive granules, which I also saw in perfectly fresh 
specimens. Except wanting the mouth and cilia, this little 
cylinder much resembled a planula. In fig. 17 is repre- 
sented a hydrotheca not yet completely developed, closed at 
its future aperture, in which the coenosarc is contracted, not 
into a cylinder, but into a perfectly regular globular form, 
with a cavity, and surrounded by two layers. At one spot a 
very fine and colourless membrane, evidently produced by the 
ball, is seen to detach itself ; this, no doubt, is a w&sv layer of 
perisarc formed by the ectoderm ; and I believe that in tlie 
other cases the perisarc is also present, but that in them it 
adheres very closely, for which reason it cannot be perceived. 
What is the signification of these structures ? For what 
purpose are they formed? Is it not a sort of encysting, 
analogous to the process so often met with among the Infu- 
soria ? Are not these fragments of the Hydroid the result of 
its dismemberment, caused by the evaporation of the water ? 
256 Geological Society. 
and is it not their function to survive these unfavourable con- 
ditions and thus serve, not only for the preservation of the 
individual, but also for propagation ? 
I think we may answer all these questions in the affirma- 
tive, and regard these fragments as formations analogous to 
what is known to us from Prof. Allman's * observations on 
the spontaneous fission of Schizocladium ramosum and Cory- 
morpha nutans, as a means of reproduction by fission. In 
Schizocladium ramosum the upper portion of a branch becomes 
detached as a little cylinder, just in the same way as in Obelia 
flahellata ; and then, after having ruptured the perisarc, this 
free portion departs from the colony, ibrms the perisarc again, 
and becomes transformed into an individual. 
[To be continued.] 
PEOCEEDINGS OF LEARNED SOCIETIES. 
GEOLOGICAL SOCIETY. 
April 11th, 1877.— Prof. P. Martin Duncan, M.B., F.R.S., 
President, in the Chair. 
The following communications were read : — 
2. " The Bone-caves of Creswell Crags." — Third Paper. By the 
Eev. J. Magens Mello, M.A., P.G.S. 
In this paper the author gave an account of the continued ex- 
ploration of these caves, and of the completion of the examination 
of the Robin-Hood Cave, noticed in his previous communications. 
Five deposits could be distinguished in the Robin-Hood Cave, 
namely, when aU present :— 
1. Stalagmite, 2 ft. 
2. Breccia, with bones and flint implements, 1 ft. 6 in. 
3. Cave-earth, with bones and implements, 1 ft. 9 in. 
4. Mottled bed, with bones and implements, 2 ft. 
5. Red sand, with bones and quartzite implements, 3 ft. 
Variations both in thickness and in character occur in different 
parts of the cave. The surface-soil yielded traces of Romano-British 
occupation, such as enamelled bronze fibulae, fragments of pottery, 
&c. The most important discoveries were made in the cave-earth ; 
and chief among these was a fragment of bone, having on it a well 
executed outline of the head and neck of a horse, the first recorded 
discovery of any such work of art in this country. The cave-earth 
also yielded a canine of Machairodus latidens, hitherto obtained in 
England only in Kent's Hole. Numerous remains of the Pleistocene 
Mammalia already recorded were found, together with a great 
* AUman, " Reproduction by Fission in Hydroids," Brit. Assoc. Report, 
1870; and Quart. Jom-n. Micr. Sci. 1871, pi. ii. figs. 2, 3.