Henn: South American Pceciliid Fishes. 103 



Birth. 



In the two species, which Philippi had under observation, the young 

 were born singly and at intervals of a few minutes. Delivery usually 

 took place before eight o'clock in the morning, but in a few instances 

 it took place about noon. During this period the female, when not 

 tormented by males, remained just below the surface of the water. 

 The young are not expelled in a definite position; either the head or the 

 tail may appear first, and occasionally the young is expelled doubled 

 upon itself. Delivery is much slower when the young appear tail 

 first. In this case the adult may swim about for awhile with the half- 

 protruding young. The position of the young in the ovary, as shown 

 by cross-sections, is likewise undetermined. 



Contrary to Ryder's statement that fright seemed to hasten par- 

 turition, Philippi found that excitement tended to greatly retard or 

 postpone delivery. For better observation he placed females in the 

 act of delivery in smaller aquaria. This usually caused a delay of 

 four or five hours, and when parturition was resumed a number of 

 unripe eggs were also cast out. 



The cannibalistic habits of the fishes are pronounced, and when the 

 birth is at an end the adult will seize upon her own young. This takes 

 place even when the tank is well supplied with food. The males 

 likewise devour the young fishes, and if they are to be reared, they must 

 be separated from the adults. 



Seal (1911, p. 93) states that the young of Gambusia holbrooki and 

 Heterandria formosa are born one at a time. The ejection of each 

 fish is so rapid that they appear as though shot out with some force. 

 " This, however, might be due to the bursting of the follicle and the 

 uncoiling of the fish as it is released from restraint. When they first 

 appear they are still in a somewhat curved form, but they quickly 

 straighten out and swim into hiding. . . . The intervals between the 

 extrusions vary from several minutes to as many hours." Seal also 

 noted the presence of two or more generations in a single season; 

 fishes born early in May were themselves mature and producing young 

 by the middle of August. 



Number of Young in Some Viviparous Species. 



The number of young at birth in a given species is evidently quite 

 variable. The appended table gives the number of embryos found in 



