THE METAMORPHOSIS OF ECHINODERMS. 127 
evidence that the latter ever had a second anterior cavity, 
though there are some grounds for believing it; but if they 
had, its importance must have been subordinate before the 
divergence of the two lines of descent, for we can hardly 
suppose that the predominance of the left cavity has been in- 
dependently acquired. 
The same applies to the second pore discovered by Field, 
and present in some Enteropneusta ; if it is really an ancestral 
feature (which is not yet fully proved) it must have been lost 
before the separation of the two groups, or we should not be 
likely to find only one (the left) in the adult form of both. 
Spengel’s suggestion, that the left collar-cavity of Tornaria 
may be comparable with the hydrocel of Echinoderms, is 
robbed of much of its value by his curious error in supposing 
that Field and I have described in the latter a second (right) 
hydrocel. There is, however, some plausibility in the 
adoption of this comparision in the form which it assumes in 
MacBride’s hands—a comparison of the true hydrocel (as 
distinguished from the anterior enteroceel) with the collar- 
cavity of Enteropneusta. The situation of the hydroccel in 
young larve is certainly strongly suggestive of this; and the 
obvious objection that the hydrocel is unpaired in most larve 
is met by MacBride’s supposed discovery of a second hydroccel 
in Asterina; and this writer even goes so far as to suggest 
that in a pore leading directly from the hydroccel to the ex- 
terior which he has found in one larva of Asterina, we have 
the homologue of the collar-pore (20). As I have utterly 
failed to find any trace of this second hydroccel in any of 
the larvee I have examined, I may perhaps be forgiven if I 
refuse, at present, to accept the evidence of such an obviously 
secondary form as the larva of Asterina; and I would more- 
over point out that one of MacBride’s own figures (19, fig. 4) 
is out of accordance with his hypothesis, since the “ collar- 
cavity” should be posterior to the anterior body-cavity, and 
should not embrace its posterior end, though I must confess 
that I am at a loss to understand why he ever drew his figure 
in this form, as it is unnecessary for his hypothesis, and unsup- 
