500 T, A. STEPHENSON 
verrucae and acrorhagi! cannot be used in limiting 
families, and this leads on to the question of vesicles. 
A certain number of forms develop, either all over their bodies 
or in certain parts only, various sorts of hollow vesicular out- 
growths of the coelenteron (see Text-figs. 2, a, and 18). These 
may be slightly or very highly specialized. It may be argued 
that they are only verrucae which have gone farther, but in 
most cases they have gone a good deal farther, and really 
TrExt-Fia. I. 
A. Small portion of the upper part of the body of Bunodactis 
alfordi, somewhat enlarged, to show the vertical rows of 
verrucae, three of them ending above in conical acrorhagi. 
B. Half a transverse section of an acrorhagus of B. alfordi. 
Mesogloea black, ectoderm and endoderm white, the black 
strokes in the former representing nematocysts, 
seem to constitute a definite and characteristic feature by 
which forms possessing them may be separated from those 
which do not. Since these forms also show an agreement among 
themselves in other ways, falling naturally into sets, we may 
fairly take ‘ presence of vesicles ’ as a family character for use 
among others. 
The presence or absence of a definite base seems a valid 
1 Jn this paper the term ‘acrorhagi’ is used to cover ‘marginal 
spherules ’ of any sort, whether simple or compound, whether nematocyst 
batteries or not. There seems to be too much variation in their structure 
for it to be possible to maintain a serviceable distinction of them into 
acrorhagi, pseudo-acrorhagi, &c, A sketch of typical acrorhagi from 
Bunodactis alfordi is given in Text-fig. 1, 
