410 bulletin: museum of comparative zoology. 



one merely curving into the other. The lateral margins of the ventral, 

 like the dorsal facet, are strongly toothed along their lower halves. 



The dorsolateral facet resembles A. trigona, the lateral ridge ex- 

 tending to the basal tooth of the hydroecium, instead of terminating 

 some distance above it, as is the case in A. leuckartii. But the ventro- 

 lateral facet is single, as in the latter, instead of being subdivided by 

 a transverse ridge, as it is in both A. trigona and A. haeckeli. And 

 corresponding to the continuity of ventral and apical facets, its 

 ventral margin is rounded, instead of angular. Its basal margin is 

 strongly toothed as is also the lateral ridge. The apical facet (Plate 5, 

 fig. 4), is essentially pentagonal, but it has only three distinct angles, 

 dorsal, and two lateral, there being no ventral angles, owing to the 

 continuity of apical with ventral facet. This outline, particularly 

 the fact that the apical facet is pointed at its dorsal end, is charac- 

 teristic, for in both A. trigona, A. haeckeli, and A. leuckartii it is hex- 

 agonal, its dorsal end bounded by a transverse margin instead of an 

 angle, owing to the different shape of the dorsal facet. 



As illustrated by the apical view (Plate 5, fig. 4), all the facets are 

 deeply concave, a fact which gives the nectophore a very characteristic 

 appearance. 



The hydroecium (Plate 5, fig. 1), though of the ordinary abyhd 

 type, is shallower than in other members of Abyla {scnsu strictu) 

 occupying only about two thirds of the length of the nectophore. But 

 the nectosac extends almost the whole length of the nectophore, the 

 pedicular canal joining it a considerable distance below its apex; 

 and consequently its dorsoradial canal follows a diphyid rather than 

 an abylid, course, just as in Ceratocymbo, (1911b, p. 232). 



The somatocyst is oval, much as in A. leuckartii, except that its 

 pedicular canal joins it some distance below its apex. 



No account of the appendages is possible, only a few much con- 

 tracted young siphons, with their tentilla, being visible, and the 

 inferior nectophore is lost. 



The features just mentioned, of which the triangular dorsal facet, 

 the continuity of ventral and apical facets and the dorsal angle of the 

 latter, are most diagnostic, suffice to separate Abyla dentata from all 

 abylids yet described. But it is not certain that the Bache specimen 

 is the first which has been studied, for Moser (1913a, p. 149) mentions 

 as close allies of the A. trigona -A. leuckartii group, two new ab^dids 

 (as yet unnamed) from the collections of the German South Polar 

 Expedition, one or other of which may prove to be identical with A. 

 dentata, when described. 



