Miscellaneous. 7 7 



of descent with modification. They indicate, in fact, that several 

 states of symbiotic equilibrium have been successively established 

 between the phylum of the parasites and that of their hosts. Still 

 more, in the particular case of the Bopy rinse, we can, by a careful 

 study of the embryogeny, determine the order in which these various 

 states of equilibrium have been produced, follow step by step the 

 modifications caused in the organism by a parasitism gradually 

 becoming more and more complete, and thus give a truly natural 

 classification of these animals. 



The first larva of the Bopyrinse is very uniform throughout the 

 group. By the long duration of its pelagic existence it teaches us 

 that the ancestors of the Bopyrinse were for a long time free forms. 

 By its general organization it shows us that this ancestral form 

 must have approached the -iHgidae, and more especially Eurydice, 

 The differential peculiarities which these first larvse present are 

 furnished chiefly by the sixth pair of thoracic feet, and are in 

 relation with the emergence of the embryo from the host which 

 harboured the parent, and not, as has been supposed, with its en- 

 trance into a new host ; from this it results that the modifications 

 are numerous, especially in the group in which the parasitism is 

 most decided, that is to say the Entoniscidse. 



The second free larval form has been called by us the Cryptoniscian 

 embryo or Cryptoniscus-stage, because the males of the Cryptonis- 

 cidee represent in a more complete fashion this transitory phase in 

 the development of the other Bopyrinae. It is under this form that 

 the fixation of the Bopyrian upon its host is efiected at the com- 

 mencement of its parasitic life. In several Eutoniscians {Portunion 

 Mienadis and P. Kossmanni), and in Phryxus Paguri, we have 

 ascertained the presence of several Cryptoniscian embryos, attached 

 to adult females provided with males. In some of them we have 

 even observed spermatozoids apparently mature and normal. We 

 may inquire whether, when the place upon the host is thus preoccu- 

 pied, the Cryptoniscian larvae do not, at least temporarily, play the 

 part of complemental males. The attached larva speedily under- 

 goes a series of transformations which, in the female Cryptoniscidee, 

 are accomplished in very different fashion from that which occurs 

 in the other Bopyrinae. 



Further, while in the Crj^ptoniscidse the male stops in its deve- 

 lopment at the second larval form, in the other Bopyrinse it con- 

 tinues its evolution, and acqiiires a more or less Idotheiform aspect. 

 We notice also that there exists an astonishing superposition of 

 parasites and a triple parallelism between the genera Cryptoniscus, 

 Zeuxo, and Bancdia of the family Cryptoniscidoe, and the genera 

 Peltogaster, Lernceodiscus, and SaccuUna of the group Ehizocephala, 

 and the genera Pagurus, Porcellana, and Cancer of the infested 

 Decapoda. 



Lastly, the singular coexistence of parasitic Cirripedes in all the 

 tj'pes of Decapoda infested by Bopyrinse, and the existence of forms 

 such as Phryxus resupinatiis, which, although no longer belonging 

 to the group Cryptoniscidse are stiU nevertheless indirect parasites 

 of the Decapoda, lead us to the hypothesis that the Bopyrinse were 



