Eimbryogeny of the Bryozoa. 401 
Thus, if we would now inquire what signification must be 
ascribed to the mode of development just explained, we can- 
not do better than take as our exclusive basis the develop- 
ment of the Entoprocta, since it is this which serves as a 
starting-point for all the rest. An ideal type, based upon a 
mixture of the essential characters observed in the different 
types, would furnish us only with a sort of mean between the 
characters of the different families. The larvee of Entoprocta, 
on the other hand, represent the primitive type from which 
all the others are derived; and this is what we want if we 
desire to appreciate the nature of the embryogenic phenomena 
and arrive at comparisons with neighbouring groups. 
Without wishing to prejudge any thing with regard to the 
affinities which it is desirable to attribute to the Bryozoa, it 
may be remarked (and this no one can deny) that in the 
whole extent of the subkingdom Vermes the Bryozoa are 
perhaps the only forms, with the exception of the Rotifera, 
in which the éelostomian arrangement is manifested in a con- 
stant manner throughout the whole group, whether during 
the course of development or in the adult state. It may be 
said, taking up a general point of view, that the Bryozoa, as 
also the Rotifera, are organisms constructed upon the telosto- 
mian plan—that is to say, in which the division of the body 
recalls the primitive division of the gastrula, with an oral and 
an aboral pole. These are, apparently, the only two groups of 
Vermes in which this arrangement continues so permanent. 
Moreover, among the primitive Bryozoa (Entoprocta), as 
in the Rotifera, the aboral surface forms from the first the 
entire integument, while the oral surface is retracted into a 
vestibule, surrounded by a more or less developed circlet of 
cilia. ‘ 
As a matter of fact (whatever may be the value of this 
approximation, whether it constitutes a real relationship or is 
only a simple analogy), an Entoproct larva, at least in the 
free state, represents a form constructed on the same plan as 
a Rotifer; it is a perfectly comparable organism as regards 
the great divisions of the body. ‘This being established, we 
shall come to conceive of the original formation of a Bryozoan 
at the expense of its larva as resulting from a simple change 
of life in an organism resembling a Rotifer. We know that 
frequently the larve of Entoprocta, instead of swimming 
through the liquid in the manner of a Rotiter, reverse their 
position and proceed to creep upon their oral surface, as 
shown in fg. 9. Nowif this habit becomes more frequent, 
the larva gradually abandoning its first mode of life to adopt 
the second, this change of manners might be the source of the 
Ann. & Mag. N. Hist. Ser. 5. Vol. x. 27 
