BIGELOW: MEDUSAE AND SIPHONOPHORAE. 371 



three tj'pe-localities : an opportunity afforded by the Eastern Pacific 

 collection (1909a), the present specimens, and by a series of S. atra, 

 including A. Agassiz's original specimens, and others collected off 

 Vancouver Island, in 1906. 



The Bache specimens, listed below, confirm my earlier conclusion 

 (1909a, p. 202) that none of the characters used by Maas (1897) to 

 separate S. divisa from S. pterophylla, i. e. presence of an apical pro- 

 jection, swollen tentacular bases, and a coronal furrow, can be relied 

 on. The first two occur equally in S. -pterophylla, as indeed Mayer 

 (1910) points out, while it is only in occasional specimens from the 

 type-locality of S. divisa that the coronal furrow occurs (Bigelow, 

 1909a, p. 202); this feature was lacking in most of the Albatross 

 examples, and in all recorded by Vanhoffen (1913a, p. 14) from Callao, 

 Peru. 



Mayer's figures of S. pterophylla (1910, pi. 29, fig. 3-14) suggest 

 that its gonads, and lip, are less complex than those of its Pacific 

 relative. But actual comparison of specimens shows that no sharp 

 line can be drawn between the two in this respect, for the primary 

 sexual folds of Atlantic specimens, 10-12 mm. high, being often bi- 

 or tri-fid with occasional pinnate sublobes (Mayer, 1910, pi. 29, fig. 3, 

 4), agree with those of Pacific specimens of a corresponding size, 

 though less complex than those of larger Pacific specimens, (Bigelow, 

 1909a, pi. 43, fig. 6). And this is also true of the crenulations of the 

 lip. In the number of primary sexual folds, too, (7-9), the two agree. 

 And the number of rudimentary tentacles per quadrant (16-21) is about 

 the same in the Bache as in the Pacific specimens of 20-25 mm. In 

 short, the large S. pterophylla agree very closely with § grown S. divisa 

 in all these characters. And further growth, on the part of S. ptero- 

 phylla, would no doubt lead to what may be called the S. divisa stage. 



According to Mayer, color separates the two species. And it is 

 certainly true that in all specimens of S. pterophylla studied in life 

 (Mayer, 1910, Bigelow, 1917) the stomach and tentacular bulbs were 

 deep brown, whereas in the Pacific specimens I have seen alive (1909a), 

 the gonads were orange to brownish red, tentacular bulbs pale yellow. 

 But color is proverbially inconstant among Medusae. In short, S. 

 divisa and S. pjtcrophylla belong to one species, of which they represent 

 afmost two color-phases. 



According to Vanhoffen (1913a) this is likewise true of S. atra, and 

 he explains the simplicity of the gonads and of the lip in A. Agassiz's 

 (1865) figures as evidence of immaturity merel}^ This assumption 

 is not unreasonable, for it is never possible to postulate how far com- 



