388 C. W. MITCHELL AND J. H. POWERS 



It was difficult to ascertain whether the tendency to remain in 

 the saccate form was a cumulative one or whether it was remaining 

 constant throughout the experiment. Perhaps the most obvious 

 sign of the decreasing potential was the fact that it was becoming 

 more difficult to produce male production in the saccate series. 



In connection with male production, in / I^, there occurred, too, 

 for the first time, an interesting phenomenon which points with 

 some probability toward the possibility of a transition to a lower 

 type than the ordinary saccate A. amphora. This was the pro- 

 duction of a rather unusual number of males, among which a 

 number appeared like transitional forms between the males of 

 k. amphora and A. brightwelh. By isolating mothers about to 

 produce males it was discovered that these unusual individuals 

 were the early members born in male families; their humps were 

 small or almost lacking, the body was much smaller and more 

 compact than normal, and the corona was much narrowed. All 

 these characters suggest the males of the smaller species. Transi- 

 tional males were found in all stages between this much reduced 

 type and the customary type of A. amphora. 



One more series of saccates was bred, which we name / 5. It 

 was started in the usual way by resting eggs secured from mass 

 culture, which latter in its turn, bad been derived from the seventh 

 generation of 7 4- This last series of saccate rotifers was bred for 

 ten generations (table 5) . It showed every evidence of possessing 

 the full vigor of its predecessors and was likewise perfectly true 

 to type so long as bred under the customary and constant condi- 

 tions. A mass culture founded from the fifth generation failed 

 entirely to throw the humped type. It again "produced a small 

 number of the transition-like males. Last of all, in the seventh 

 and eighth generations, the series was itself again tested, by forced 

 and alternate feeding, to see whether the potentiality of the 

 humped type was still present. Mutation was indeed induced 

 showing that the long hereditary transmission, sexual as well as 

 asexual, of the simple saccate form had not produced a race 

 sufficiently stable to resist the cause for change which lies in an 

 unwonted food stimulus. 



It is natural to ask whether any tendency toward such race 

 formation had manifested itself throughout the entire succession 



