SEX-DIFFERENTIATION IN XIPHOPHORUS HELLERI. 47 



three years old they ceased producing young and, during the course 

 of several weeks, took on the secondary sex characters of males. 



When, a few days later, these specimens came into the writer's 

 possession, both were decidedly female as to shape and size of 

 body and decidedly male as to development of anal and caudal fins, 

 which were far advanced toward the male condition yet much 

 belated. Cytological preparations of the entire reproductive sys- 

 tem of both fishes disclosed that ripe sperm was in all parts of the 

 duct and gonad, but that the latter was juvenile in comparison to 

 the size and age of the fish. 



Suggestive as the case was, the data obtained were far too 

 meager to decide either for or against the reputed transformation. 

 That peculiarities do occur in the sex conditions of these teleosts 

 has been noted in this laboratory as well as by many fish fanciers. 

 The extent and meaning of these peculiarities became the aim of 

 this work. It was thought necessary to approach this problem 

 from three different angles : 



1. A detailed study of sex-differentiation in males and females. 



2. Isolation of a large number of females for sex observation. 



3. An attempt to control sex experimentally. 



Only the first part is considered in the present paper. The last 

 two are still in progress and will be the subject of a future publi- 

 cation. 



II. MATERIAL AND^METHODS. 



No attempt was made to study embryonic development during 

 gestation, except for a few later stages, which will be described 

 further on. Development during gestation seems to present some 

 interesting features which will be further investigated as soon as 

 the opportunity presents itself. My material includes fishes from 

 birth to adult stage. To secure all possible stages in sex-differenti- 

 ation, a large number of fishes were necessary ; not less than 400 

 specimens were studied, of which 300 were sectioned for cytologi- 

 cal study. 



As to fixatives, Allen's, Bouin's, Child's, Flemming's strong, 

 Smith's, and Zenker's fluids have been used, Bouin's giving con- 

 stant and satisfactory results. For general stains Heidenhain's 

 haemotoxylin and alum-cochineal counterstained with orange G 



