HYDROMEDUSAE, SIPHONOPHORES, AND CTENOPHORES. 307 



the west coast of Mexico, described by me as O. polynema Haeckel, 

 but which Mayer (1910) thinks is a new species; the other O. 

 polynema of Maas (1906) from Amboina, characterized by very 

 numerous tentacles. The last of these, it is true, much resembles 

 Philidium globosum Mayer (1910, p. 272, pi. 34, fig. 4) in general 

 appearance; but, apart from the difference in the number of canals, 

 is separated from it by having nearly four times as many tentacles 

 (upward of 100) and otocysts, the only species which approaches 

 it in this respect, P. gr eg aria (of which I have studied a large series) 

 having at most only about half as many tentacles (up to about 60). 

 And of course the absence of peduncle separates it from the many- 

 tentacled species of Eirene or Phortis. For this same reason it can 

 not be united with Haeckel's O. polynema; and as it must be called 

 something for the present, even if later united with some Philidium, 

 it may be named aphrodite. 



OCTOCANNA APHRODITE, new name. 



Plate 42, figs. 1, 2. 



Octocanna polynema Maas, 1906, p. 95, pi. 3, fig. 10 (not Octocanna poly- 

 nema Haeckel, 1879, Maas, 1905, Browne, 1905, Bigelow 1909a). 



Octocanna aphrodite — material examined. 



The three specimens from Amboina described by Maas (1906) 

 all had 8 canals ; it is remarkable that the two from the Philippines 

 should show variation one in one direction, the other in the other, 

 from this number. And it is fortunate that there are now enough 

 records to show that here as in Irenopsis the number is not fixed. 1 



Marginal organs. — The very numerous tentacles, with large basal 

 bulbs, are closely crowded on the bell margin. Here and there there 

 are to be seen younger onces in process of interpolation, at various 

 stages in development (pi. 42, fig. 2). There are no permanent rudi- 

 mentary knobs, neither are there any cirri. The bases of the tentacles 

 bear well-developed excretory papillae. 



The number of tentacles to the segment of the margin is variable, 

 corresponding to variation in the distance between radial canals. 



1 In Irenopsis hexanemalis in which the usual number of canals is 6, Browne (1905) has 

 recorded specimens with 4, 7, 8, 9, and 11 canals. 



74841°— 19— Bull. 100, pt. 5 3 



