PETASIDAE. 101 



On the whole, it is best to discard the separation between Trachynemidae 

 and Aglauridae, and to unite the component genera of the two 

 into one family, for which the older name Trachynemidae may be retained. 

 This family will then be characterized as : — Trachomedusae with eight radial 

 canals ; without blind centripetal canals ; with well-developed manubrium ; 

 with numerous tentacles arranged in a single series, but of either one or two 

 kinds ; gonads either pendent or linear. 



The collection contains fifteen genera and eighteen species of Trachome- 

 dusae, of which two genera, three species and one variety are believed to be new. 



Petasidae Haeckel, 1879. 

 sens. em. Browne (:04). 



Trachomedusae with four or six radial canals on which the gonads are 

 borne ; with either enclosed sense capsules or free sensory clubs. 



We may well follow Browne in separating the Petasidae into two sub- 

 families, Petachninae and Olindiinae, according as the otocysts are free 

 clubs or enclosed capsules. The present collection contains members of the 

 Olindiinae only. Indeed, so far as I can learn, no one since Haeckel has 

 observed any members of the Petachninae, 



Although Goto (: 03) believes, from his studies on the development of 

 Olindioides, that the sense organs of Olindioides are ectodermic, and that 

 the Olindiinae therefore belong to the Leptomedusae, recent students 

 (Browne, : 04, Mayer, : 04, Maas, : 05) agree in classing them among the 

 Trachomedusae. Certainly in all respects, except possibly the origin of the 

 otocysts, they are typical of this order ; and very recently doubt has been 

 thrown on Goto's conclusion, that the sense organs are ectodermic, by 

 Murbach, who makes the positive statement (: 07, p. 2, footnote) that in 

 Gonionemus murbachii, at least, they are of endodermic tissue. I therefore 

 retain the Olindiinae among the Trachomedusae, where I believe they un- 

 doubtedly belong. 



Mayer (: 04) makes the presence of sucking discs one of the distinguishing 

 features of this family, thus excluding Gossea and Aglauropsis. Both of 

 these genera, however, are in other respects closely allied to Gonionemus ; 

 Aglauropsis even being believed by Maas (: 05) to be a synonym of the 

 latter, so that it is better to consider this character, with Browne (: 04), as 

 of generic value only. I regret that my account of rudimentary sucking 

 discs in Gonionemus pelagicus should have led Maas (: 05) to believe that 



