24 McCMURRICH. [VoL. III. 
and the tubercles project more above the general surface. They 
are simply batteries of nematocysts, the presence of these bodies 
at once distinguishing them from the verrucze of Az/actinia, etc. 
R. Hertwig (82) has stated his objections to the presence or 
absence of elevations of the column wall being considered of 
systematic importance except in so far as genera and species 
are concerned. That these objections must be sustained is 
certain; but it also seems probable that more attention ought 
to be paid to the structure of the elevations than has hitherto 
been done. It is possible to distinguish in them at least three 
varieties: (1) simple elevations of the mesogloea unaccompanied 
with any extensive modification of the ectodermal epithelium, 
e.g. Bunodes gemmacea according to Jourdan (80), and ‘Bunodes 
(Cylista) minuta, R. Hertwig (82) ; (2) evaginations of the entire 
column wall furnished with numerous nematocysts, as in the 
acrorhagi of Awthea, etc., and the tubercles of Bunodes tentatus ; 
and (3) verrucz in which the ectoderm consists of peculiar elon- 
gated cells, quite different from those of the column wall in 
general. Jourdan, in his description of the tubercles of Bunodes 
gemmacea, apparently considers them to be produced by the 
dipping down of the ectoderm between them into the mesogloea 
to form what he terms “verrues glandulaires.” R. Hertwig has, 
however, shown by his observations on Lunodes (Cylista) minuta 
(82) that Jourdan’s interpretation of the appearance presented 
by his sections was erroneous, and has maintained that these 
apparent ectodermal enclosures within the mesogloea were pro- 
duced by the contraction of the column wall of the specimen 
examined, whereby deep pleatings were produced which “may 
look like detached epithelial islands in transverse section.”’ This 
being the case, the elevations of Bunxodes gemmacea belong to the 
first of the three groups enumerated above, and are markedly 
different from those of B. ¢enzatus. 
The sphincter muscle (Pl. III., Fig. 7) is very strong, forming 
a large projection into the body cavity. In transverse section 
it has a more or less circular outline, and is attached to the 
column wall by a pedicle of mesogloea which has a fenestrated 
appearance, containing numerous cavities lined with muscle 
cells. From this pedicle the mesogloea processes arise in a very 
irregular manner and anastomose in all directions, so that the 
central portion of the muscle thickening appears to have a 
