BIGELOW: MEDUSAE AND SIPHONOPHORAE. 385 



Both radial and circular canals are narrow and smooth walled, as 

 in most Leptomedusae. 



Gonads. The gonads occupy the major portion of the radial canals, 

 leaving only their proximal and distal ends bare. Each consists of a 

 single, broad, thin lamella, a simple dilation of the canal, thrown into a 

 complicated series of alternating transverse folds (Plate 3, fig. 3). 



The type is a sexually mature female, large eggs being plainly 

 visible. The other specimen is apparently a male, and its gonads are 

 smaller and simpler, corresponding to its smaller size. 



Manubrmm. The manubrium is voluminous, hanging to, or per- 

 haps below, the opening of the bell, somewhat cruciform in cross- 

 section, the radial canals traceable to the center over its aboral 

 surface. It is squarish in outline, without any clear morphological 

 division into basic, gastric, and labial portions. In both specimens 

 it is more or less twisted, and longitudinally wrinkled and folded in 

 the interradii. But since these folds follow no definite plan, but vary 

 from quadrant to quadrant, they may safely be classed as contraction 

 phenomena. Along the perradii the walls of the manubrium are 

 smooth (Plate 3, fig. 3). The mouth is wide-open, as broad, in both 

 specimens as the greatest diameter of the manubrium, and surrounded 

 by a simple hp. But although the latter has no distinct lobes, or radial 

 prolongations, its margin bears a series of small folds, and crenula- 

 tions (Plate 3, fig. 3). 



Marginal organs. There are four large radial tentacles with much 

 swollen bases (Plate 3, fig. 2). And though all but the bases of these 

 have been destroyed in all cases, enough remains to show that they 

 are hollow. Between these the margin bears a considerable number 

 of very small papilliform tentacles, without filaments (Plate 3, fig. 4). 

 And fortunately many of them are in such good condition that I can 

 safely assert that their present rudimentary condition is normal, not 

 the result of mutilation. They, too, are hollow, as is clearly visible, 

 when torn or broken. They are of various sizes, the presence of 

 several very small ones in the type showing that interpolation of new 

 ones takes place right up to maturity. But the larger ones are ex- 

 tremely uniform in length, and there is nothing to suggest that they 

 ever develop into large tentacles. 



In the only quadrant wholly intact (type) there are twenty-three 

 small tentacles, suggesting a total of 80-90. In the smaller specimen 

 there are about twenty in one quadrant. 



A careful search with the compomid microscope has failed to reveal 

 any otocysts, and as portions of the margins of both specimens are 



