162 
J. PLAYFAIR McMURRICH 
In A. bournei and A. brachiolata the conditions are different. 
Of the former species, Van Beneden (1891) observed a stage in 
which there were two couples of well developed marginal tentacles 
and a third couple was represented by two very much smaller 
tentacles, which were themselves decidedly unequal in develop- 
ment, one being only about half the size of the other (fig. 3). 
A median marginal tentacle was also present, being of about the 
same size as the smaller member of the third couple, and from this 
relation, combined with the fact that in younger larvae possessing 
only two couples of tentacles there was no trace of the median 
ventral tentacle. Van Beneden concludes that in this species the 
median tentacle develops synchronously with the members of 
the third couple. 
Fig. 3 Diagram of an oral view of A. bournei at the time of the formation of 
the median marginal tentacle (adapted from E. van Beneden). 
Fig. 4 Diagram of an oral view of A. brachiolata at the time of the formation 
of the median marginal tentacle. 
In A. brachiolata I found (1891) practically the same succession. 
In larvae with three couples of mesenteries there were only two 
couples of marginal tentacles, the third couple being represented 
however, by merely tuberculiform elevations of the disk. In 
older forms, in which the fifth couple of mesenteries was in process 
of development, the median marginal tentacle had appeared and 
had reached a stage of development in which it was practically 
the same size as the members of the third couple, which, on their 
part, were much smaller than the membrane of the second and 
first couples (fig. 4). It is evident, therefore, that the median ven- 
tral marginal tentacle of A. brachiolata, if it does not appear syn- 
