THE METAMORPHOSIS OF ECHINODERMS, 95 
but here again the negative evidence ought, it seems to me, 
to weigh very heavily with those who rely, as much as MacBride 
does, upon ontogeny as a repetition of phylogeny. At present 
the positive evidence is exceedingly weak—only in Asterids, 
and not even in all of them, has this fixation been found, in 
spite of most diligent search for it ; and so weak is the atavistic 
tendency to recover this supposed phylogenetic character that 
Amphiura squamata, though fixed to the body of the mother 
in its young stages, is fixed by the posterior, not by the anterior 
end. Here the fixation is clearly secondary, and until further 
evidence is brought forward I am strongly disposed to regard 
the fixation of Asterina as also secondary, and quite inde- 
pendent of the fixation of Antedon. Fixation by the preoral 
lobe is no uncommon thing, so that it may easily arise over and 
over again ; and when we consider the apparent difficulties of 
the transition from the bilateral to the pentamerous stage in 
other larvee (as evidenced by the rapidity of the metamorphosis 
and the frequent obliteration of the cesophagus), we can easily 
see the advantages of such fixation, especially to a shallow- 
water form exposed to wave action. ‘This, however, is pure 
speculation ; what is really important is, I repeat, the strength 
of the negative evidence, which to my mind is so great as to 
make it unwise to assume the fixation of the ancestor so long 
as any Other explanation is possible. That the phenomena 
can be accounted for without this assumption I shall endeavour 
to show in the following pages. Iam aware that the proof of 
my views is still far from complete, and for that reason I shall 
not attempt to follow out the details so far as some of my 
predecessors have done; but the large number of speculative 
suggestions as to the origin of Echinoderms which have 
appeared in the last few years, almost all assuming original 
fixation in some form or other, seemed to make it advisable 
that I should attempt to show that such an assumption 1s 
neither embryologically sound nor necessary as a basis for 
phylogenetic speculation. 
