THE METAMORPHOSIS OF ECHINODERMS. 91 
probably primarily in the same interradius as the water-pore. 
Again, on purely theoretical grounds he assumes that the 
closure of the water-vascular ring in ontogeny ought to take 
place where we find it in Asterina; and he therefore boldly 
denies (p. 157) the accuracy of the observations tabulated in 
my last paper (5, fig. 28), except in the case of Ophiurids, in 
which he assumes that the hydrocel has been turned com- 
pletely round in ontogeny—its anterior end in the larva being 
the original posterior end! I leave the facts in these cases to 
speak for themselves ; but I cannot help expressing a regret 
that Biitschli should have allowed his theories to carry him so 
far without taking the smallest pains to find out for himself 
where the truth lay. 
MacBride’s hypothesis (19) is in many respects more nearly 
in accordance with the facts of embryology than its predecessors, 
and contains much that is suggestive ; but though it is perhaps 
unfair to criticise it while only an abstract of it is before us, 
yet it gives one the impression of being based too completely 
upon the ontogeny of Asterina, the only form, apparently, in 
which MacBride has personally followed the metamorphosis. 
He assumes that the bilateral ancestor possessed two hydro- 
coels (“‘ collar-cavities”’), of which one at least (the left) had 
five tentacles. Whether there were five more on the right side 
I cannot certainly gather, but at least they are not represented. 
This bilateral form became fixed by the proral lobe, and then 
the left hydroccel and left posterior enteroccel grew round the 
cesophagus, the former embracing the base of the preoral lobe 
in Echinozoa, but not in Crinoids. This encircling of the 
cesophagus is regarded as brought about by “ the curious, and 
as yet unexplained, peculiarity of Echinoderms, the predomi- 
nance of the left side (left hydroccel and left posterior body- 
cavity”). On these views I offer the following comments: 
(1) Whatever may be thought of the existence at an early 
period of a second (right) hydroccel (and to this question we 
shall return later), it is improbable that it could have retained 
enough importance to bear tentacles at the stage under con- 
sideration, without leaving more trace of them in ontogeny. 
