394 bulletin: museum of comparative zoology. 



tentacles. The Bache specimens add nothing, except by way of 

 verification to the earher accounts. The most characteristic features, 

 which by themselves are sufficient to separate C. percgrina from all 

 other species, are the absence of peripheral canals and of peduncle; 

 minor features of diagnostic value are the square marginal lappets, 

 the squarish outline and comparatively great length of the gastric 

 pockets, the large nematocyst pads beneath the bases of the tentacles, 

 and the large number of otocysts (5-8 per lappet). 



The Bache captures are of interest geographically (p. 431), all 

 previous records for C. peregrina being from the Pacific. 



Solmaridae Haeckel. 

 {Sensu. Maas, 1904b; Bigelow, 1909a; Mayer, 1910). 

 Narcomedusae without gastric pockets. 

 Pegaxtha Haeckel, 1879. 



Solmaridae with peripheral canal-system (at least in one generation), 

 and with otoporpae; with gonads localized at the edge of the stomach, 

 as numerous as, and in the radii of, the marginal lappets. 



As here defined, Pegantha corresponds to the genera Pegantha -|- 

 Polyxenia + Polycolpa of VanhofFen (1908a, 1912a, 1912b). Accord- 

 ing to him these three genera can be separated by the number of tenta- 

 cles, Pegantha having 8-13, Polyxenia 16-18, Polycolpa 24-30. But 

 specimens of Pegantha triloba having recently been found with four- 

 teen and fifteen tentacles (Vanhoffen 1912a, 1912b), there is no longer 

 any justification for separating Polyxenia from Pegantha, at least on 

 this ground. And the supposed gap, in tentacle-number, between 

 Polyxenia and Polycolpa proves to be equally non-existent, for 

 Pegantha laevis Bigelow with 19-22 tentacles when mature (1909a, 

 p. 97) is exactly intermediate between them. The number of tenta- 

 cles is, at most, only a specific character here; and even so, must be 

 used with caution, being variable in e\ery peganthid yet described. 



But it does not follow that all Solmaridae with peripheral canals 

 and otoporpae should be united in one genus, for they fall into two 

 groups, according as the sexual products are localized in pockets at the 

 margin of the stomach, as numerous as, and in the radii of, the mar- 

 ginal lappets (Pegantha), or form a continuous ring in the lower 

 gastric wall, as Haeckel describes them for his genus Polycolpa. And 

 a typical Polycolpa, i. e., solmarid with ring-like gonad, and with 



