THE CYPRINODONTS. 11 



family. In females it forms a single large chamber behind and above the 

 ovaries and intestine ; in males of those species in wliicli the anal fin is 

 much carried forward, the stays from the vertebrae and the supports of the 

 anal divide the air bladder into separate chambers. 



The ovaries vary somewhat according to the modes of reproduction ; 

 the simpler forms are the oviparous. Anableps represents the most modified. 

 The embryology of this genus as interpreted by Valenciennes and Wyman 

 is liable to some corrections; these are noted below under the generic 

 description. The development of another of the viviparous forms, Gam- 

 busia, has been well worked out by Kyder. The breeding female of some 

 Funduli is provided with a tube upon the anal fin, which, though less de- 

 veloped, recalls that of the female Khodeus of the Cyprinidae ; it varies 

 with the season. An urogenital tube on the anal fin is a marked feature 

 of the male in Anablepinae and Jenynsiinae, Plates VII., VIII. The absence 

 of this tube distinguishes other viviparous forms. In these latter the clasper 

 varies much in length and structure ; it may be barbed, or furnished with 

 hooks, or provided with fleshy pads, or may end in a simple point without 

 either barbs, hooks, or flaps. 



The kidneys and testes resemble those of most fishes. Their secretions 

 are not thrown together in a common receptacle, as was supposed to be the 

 case by Valenciennes in Anableps. The structure of the separate recep- 

 tacles is sketched on Plate VII., and described below. 



A minor sexual character is that of the small spines appearing on the fins 

 of males in several genera in breeding time ; and another is seen in the 

 growths of the fins, anal and dorsal in particular, at the same time. 



As in the organs of nutrition, excretion, and generation, there are con- 

 siderable differences in the sensory apparatus. The peculiar eye of Ana- 

 bleps, half of which looks up and half down, has been well known since the 

 time of Artedi. This genus has a tubular nostril ; also possessed, less de- 

 veloped, by one or more of the other genera. 



The affinities of the Cyprinodonts to Esocida3, Cyprinida?, and others, 

 were pointed out by Professor Agassiz in his Poissons Fossiles. The family 

 was differentiated quite early, as is shown by the Tertiary fossil types, 

 Lebias and others ; its derivation is to be traced through extinct forms. 

 A search for direct connecting links with other recent families is not a 

 promising field of inquiry. 



