322 Miscellaneous. 
of the stem before the appearance of any gonophores ; they are true 
endodermic cells differentiated ; and we find all the transitions be- 
tween an ordinary endodermic cell and a well-developed ovum. 
They are conveyed, with the neighbouring endoderm and ectoderm, 
into the gonophore which is in course of development, and which 
is at first only a simple diverticulum of the body-wall of the Hydroid 
olype. 
. it a variety of Plumularia echinulata that I found at Roscoff I 
observed the same phenomena: the stem is filled with ova before 
the appearance of the gonophores ; then the perisare is perforated 
at a certain point in the stem where a gonophore is to be deve- 
loped ; the endoderm and ectoderm of the stem form, as it were, 
a hernia through this fissure, and carry the ova along with them 
as they advance. A new chitinous envelope is secreted, and we 
get a young gonophore, in which the ova complete their develop- 
ment. My observations upon Sertularia pumila have led me to the 
the same results. 
Thus in these three species, which have their sexual generation 
represented by gonophores which always remain attached to the Hy- 
droid polype, I was last year led to conclude that the ova are deyve- 
loped, not in the gonophore as was supposed, but in the actual stem of 
the polype itself, which has been regarded as the asexual gene- 
ration. 
This year I wished to profit by my sojourn at the Laboratory at 
Roscoff, in order to extend my observations to the species which 
have not gonophores fixed through their whole life, but free 
Meduse. 
I commenced with a species which has semimeduse (that is to 
say, a gonophore with an umbrella, tentacles, and well-developed 
canals, but which remains constantly attached to the polype upon 
which it buds), namely Gonothyrea Lovent. Here, again, the ova 
originated from the endodermic cells of the differentiated stem ; they 
are carried away with the neighbouring tissues into the blastostyle 
and the gonophore, to the interior of the gonangium, and they 
complete their development in the semimedusee which are met with 
at the summit and exterior of the female capsules. 
To study what takes place in the species of which the sexual 
generation is represented by free meduse, I selected Podocoryne 
carnea, which lives parasitically upon the shells of Nassa, and Obelia 
gemculata. The former belongs to the Tubularian and the second to 
the Campanularian group. 
In the region of the body of the Hydroid polype upon which the 
Medusz are to bud we find ova in course of development. They are 
modified cells of the endoderm. When the Medusz begin to bud 
they are at first only a simple diverticulum of the two layers which 
form the wall of the body of the polype; the ova are carried into 
the interior of this young bud: they are then much smaller than 
in the species cited above; but from this moment they are rapidly 
developed ; soon the peduncle of the medusa ruptures, it separates 
from the Hydroid polype upon which it has budded, swims about 
freely, and completes its metamorphoses, while, at the same time, 
the ova it contains attain maturity. 
