DEPARTMENT OF EXPERIMENTAL EVOLUTION. 129 



testes shorter than noraial. In the majority of the male intergrades some 

 such abnormahty exists in the reproductive system. These details are given 

 to show how complete a series the sex intergrades really form. 



"Sex here reveals itself not as a fixed and definite state, but as a purely 

 relative thing. With this material no arbitrary classification into males and 

 females is justifiable or possible, not only because of the confusing admixtures 

 of male and female secondary characters, but also because the same individual, 

 even the same sex-gland, may develop eggs and sperm at the same time or 

 sperm at one time and eggs at another time. 



"By selecting as mothers female intergrades with several male characters 

 the production of sex intergrades has been continued for 19 generations. 

 There has been no apparent reduction in vigor or change in the character of 

 the offspring produced, and (contrary to my earlier expectations) it is now 

 hoped that the material may be maintained as long as desired. 



"Several normal females from the sex-intergrade stock produced only 

 normal females, and from these only normal females were produced throughout 

 succeeding generations. The origin of the sex-intergrade strain may be re- 

 ferred to as a mutation. The origin of all-female-producing strains from within 

 this sex-intergrade stock is a return mutation. Several of these return muta- 

 tions have been observed and probably others occurred but escaped observa- 

 tion, since some of the mutants were not used in propagation of the stock. 



"On another point the occurrence of these sexual forms (in the sex-inter- 

 grade strain) throws important light on the problem of sex in Cladocera, 

 indicating that the capacity for sexual reproduction is not lost after long- 

 continued parthenogenetic reproduction (130 generations). Furthermore, 

 the origin of subsidiary lines within the sex-intergrade strain which produced 

 normal females exclusively (except that in the thirteenth and fourteenth 

 generations of one of these all-female-producing strains several normal males 

 were produced) indicates rather clearly that the sex-intergrade strain did not 

 occur in stock which had lost or had undergone any retrogression in sexual 

 capacity." 



The similarity of the general results as to sex obtained by Doctors 

 Riddle and Banta is obvious. Combined with the observations of 

 Goldschmidt on gipsy moths they lead to the conclusion, which Dr. 

 Banta draws, that probably siixiilar but less readily recognizable inter- 

 mediate sexual states may be of somewhat common occurrence and 

 that sex in general is a much less fixed and precise state than is com- 

 monly supposed. 



In the case of the sex intergrades of Siinocephalus it is not main- 

 tained that they were brought about by external conditions. They 

 suddenly appeared and were continued because of some change in the 

 germ-plasm. But, as hinted at in Year Book No. 14, page 133, there is 

 evidence that the exclusively female-producing series may be inter- 

 rupted and males made to appear with special changes in the environ- 

 ment. On this point Dr. Banta reports: 



"This evidence came in part from the fact that the very few occurrences 

 of sexual forms among the stock began at times of poor experimental condi- 

 tions — poor food or otherwise inferior breeding conditions. The sex-inter- 

 grade strain originated after the continuation for several generations of 

 conditions unfavorable to Simocephalus. One strain of the 'long spine' 

 Daphnia, after three or four generations of depression, produced a relatively 



