REPORT OX THE HYDROIDA. 
XXXIX 
mesembryanthemum. The researches of these last observers have shown that in Tubularia 
mesembryanthemum the sexual cells are derived from that part of the ectoderm which 
forms the deeper portion of the endocodon in the very young gonophore. It is this 
portion which ultimately becomes the ectodermal covering of the spadix, and the place of 
origin of the sexual cells thus coincides with their place of development, so that they do 
not here require as in other instances to wander from one to the other. Weismann and 
Ciamician agree in believing that the great majority of the egg-cells serve merely as 
food-cells, and after attaining a certain size become broken down and dissolved, thus 
affording nutriment for those egg-cells which are destined for development. — a view 
which, as we shall presently see, is supported by the phenomena observed in Myriothelci. 
In the very young gonophore — whether male or female — the sexual cells, as may be 
seen in Tubularia larynx / already surround the spadix, and soon quite fill the cavity which 
is included between the spadix and its now widely separated ectodermal coat, and 
thus serves in the female as a brood chamber for the developing embryo. In the egg- 
cells which at this stage fill the brood chamber in Tubularia larynx may be seen a large 
nucleus with nucleolus (germinal vesicle and germinal spot). 
With the growth of the gonophore the mass of sexual cells which it contains becomes 
more and more voluminous, and we soon find that the place of these egg-cells is taken 
by a plasma-like mass which envelops the spadix, and which still continues to increase 
in volume with the continued growth of the gonophore. We next see that a portion of 
the mass has become detached from the rest in order to undergo a special development 
within the cavity of the gonophore. Whether this detached mass can be regarded as 
a true ovum may be doubted, but at all events it takes the part of an ovum in 
passing through various stages of development. It is destitute of external membrane, 
and it is probable that it is at this stage that the influence of the male element is exerted 
on it, and that the process of segmentation takes place. This process, however, is here 
very obscure, and observations have failed in obtaining satisfactory results as to the 
mode in which it is carried on. 
According to Ciamician the segmentation of the ovum in Tubularia mesembry- 
anthemum is an irregular one, the embryo being formed by an epibolic extension of 
some of the segmentation spheres over the remainder, the former giving rise to the 
ectoderm, and the latter to the endoderm. 
This form of segmentation, by w T hich some of the segmentation spheres rapidly 
multiply and surround the others, closely resembles that of the Ctenophora, and is very 
different from what takes place in the formation of a planula. I have never seen any- 
thing of the kind in the species of Tubularia whose development I have studied, and 
neither Balfour’ nor Kleinenberg have been able to confirm the conclusions of Ciamician. 
1 Gymnoblastic Hydro ids, p. 91, pi. xxiii. figs. 19-24. 
2 Balfour's Works, Memorial Edition, voL ii. p. 154. 
