482 
least to the most differentiated condition. At the lowest end stands 
Perophora viridis in the buds of which free cells of the blood 
attach themselves to the outer surface of the inner vesicle and form 
definite organs. It has not been proved by direct observation that 
these cells arise from the vesicle itself, but whatever be their source 
whether it is the mesoderm of the embryo or the blood adhere to the 
wall at certain constant points and there become pericardium and 
nervous system for example. I have argued in my paper on Perophora 
viridis that it is much more probable to suppose that the cells 
which build up these structures are all undifferentiated and alike and 
that it is the particular point of attachment which determines what 
kind of organ will proceed from them. It is almost impossible to 
imagine that isolated cells floating freely in the blood could reach 
their proper destination, if they possessed a predetermined nature and 
some were already cells of the nervous system, others of the pericardium, 
etc. But it is much more probable that they all have an indifferent 
character, and their presence at any given place is purely accidental; 
once there, howerer, qualities latent in them are called forth under 
the specific formative influence of the place of attachment. The wall 
of the vesicle is, therefore, only differentiated in so far as it possesses 
the power of determining whether cells of the blood which lodge on 
these particular areas shall give rise to pericardium, nervous system 
or sexual organs as the case may be. 
At a step higher in the degree of differentiation we find 
Ecteinascidia and Perophora annectens. In these forms 
the inner vesicle is known to be one source at all events of cells of 
the blood, which, however, play but a small part in the development. 
But at those portions of the vesicle on which organs are laid down, 
the wall contributes the greater number of cells directly to the 
rudiment in question. 
Finally, as presenting a still higher type of organisation we may 
cite the buds of Botryllus and many other Ascidians, in which 
blood cells are not concerned at all in the development; nor is there 
here a wandering out of cells from the vesicle to form a shapeless 
mass, out of which a definite structure is built up later. 
We may take the origin of the dorsal tube as an illustration; in 
Botryllus, for instance, the roof of the inner vesicle in the region 
which will be cut off as the atrium, directly evaginates to form a 
tubular diverticulum; the posterior connection is lost eventually and 
the definitive dorsal tube which is thus bodily pinched off from the 
inner vesicle, is established. The origin of the pericardium as a 
