AND DEVELOPMENT OP MYRIOTHELA. 
569 
mass originates as a cellular bud from the inner surface of the ectodermal layer of the 
gonophore, and that this pushes itself into the endoderm and becomes afterwards cut 
off from its attachment to the ectoderm, I paid great attention to the gonophores 
of Myriotliela from the earliest moment when they became recognizable, but entirely 
failed to detect any process resembling that described by the Belgian zoologist as taking 
place in Hydractinia. In the very earliest stages of the gonophore which I could find 
the gonogenetic cavity had been already formed and filled with the primitive generative 
matter, and I failed to meet with any thing which would lead me to believe that this 
had its origin in the ectoderm. It is true that in Myriotliela a difficulty occurs in the 
observation which we do not meet with in Hydractinia ; for while the complete separa- 
tion of the sexes on different colonies in Hydractinia will enable us at all times to say, 
no matter how young may be the gonophore under examination, whether this be male 
or female, in Myriotliela we have no certain sign by which to decide as to the sex of 
the gonophore in its youngest stages, gonophores of both sexes being here borne on the 
same blastostyle. It is scarcely possible, however, that among the many cases of 
extremely young gonophores which I examined there were not both male and female 
examples ; and in no case did I find any thing which would lead me to believe that the 
origin of the generative elements in one was different from what it was in another. 
The facts here noted have thus led me to maintain that both male and female elements 
have their origin in the endoderm. Still, considering the difficulty of the observations, 
and the fact that the appearances lie possibly open to another interpretation, I do not 
desire to insist on the impossibility of the generative elements being in one or both 
sexes primarily introduced from the ectoderm into the endoderm ; and I am willing to 
wait for the confirmation which may be expected from further investigations. 
As is well known, all the fixed hydroids pass through a free locomotive stage before 
finally attaching themselves. I have elsewhere* pointed out that this free stage shows 
itself under one or other of two forms, namely, that of a planula (as in the great majority of 
hydroids, Campanula, Sertularia, Coryne, See.) and that of an actinula (as in Tubularia). 
The free hydroid planula is a closed sac in whose walls an endoderm and an ectoderm are 
differentiated, not by a process of invagination, but by one of dilamination, and in which 
an oral orifice is afterwards formed by a perforation of its walls, the planula thus 
becoming the gastrula of Haeckel. The external surface of the planula is almost 
always clothed with vibratile cilia. 
The actinula represents a form more highly organized than either the planula or the 
gastrula ; for not only is a mouth always present in it, but locomotion is effected not by 
vibratile cilia, but by means of external appendages in the form of tentacles or arms, 
which may be either transitory or permanent. 
It must not, however, be supposed that the planula stage does not exist in hydroids 
whose free phase is that of an actinula. It is, on the contrary, as truly a phase of 
their development as it is of that of the others : but the planula stage is then, if we 
* Gymnoblastic Hydroids, p. 85. 
