160 
J. PLAYFAIR McMURRTCH 
From their general appearance these four species may be divided 
into two groups, one consisting of A. albida and A. sibogse, and 
the other including the two remaining forms, the most noticeable 
difference being the greater proportionate length of the tentacles 
in the members of the first group. In the species that form it 
these organs are many times the diameter of the disk in length, 
while in A. brachiolata and A. bournei they are about the same 
length as the diameter of the disk or but little longer. A difference 
of more importance is, however, revealed by an examination of 
the succession shown in the development of the tentacles. 
It is a well known characteristic of the Cerianthege that their 
mesenteries appear in couples, the members of each couple lying 
one on each side of the median sagittal plane of the body. Conse- 
quently the intermesenterial chambers will be also arranged in 
the same manner, except that there will be a single chamber at 
the ventral edge of the sagittal plane and another at its dorsal 
edge. The tentacles, both marginal and labial, are outgrowths 
of the roofs of the intermesenterial chambers, and will therefore 
also be arranged in couples, except that there will be a median ven- 
tral tentacle corresponding to the median ventral chamber. 
No tentacle develops in the roof of the dorsal median chamber, and 
hence the number of tentacles of the marginal series at least, will 
be odd in all later stages of development. It has been found, 
however, that the median ventral tentacle in both the marginal 
and labial series, makes its appearance only after a number of 
couples have formed — indeed, it may altogether fail to develop 
in the labial series — and the number of couples that precede it 
in development appears to be definite for a given species. 
The marginal tentacles need alone be considered here. Boveri 
(1890) found that in A. albida there was no trace of a median 
marginal tentacle in larvae which already possessed three couples of 
well developed tentacles of the marginal series as well as the two 
tentacles of the fourth couple in process of development, and Van 
Beneden (1898) found that in larvae of the same species in which 
there were three couples of well developed tentacles and a fourth 
couple whose members were somewhat smaller, there was present 
a still smaller marginal tentacle (fig. 1, Tm). Hence it may be 
