334 



Aaron Franklm Shull 



examined with a microscope and found to be free from protozoa. 

 Rotifers were reared in various concentrations of this filtrate, one- 

 fourth, one-half, three-fourths, and undiluted, as well as in pure 

 spring water. The five lines were derived from sisters, and were 

 fed equal quantities of food from the same fresh cultures. The 

 flagellate food lived readily in the filtrate, of all concentrations, 

 and when the records were made two days later it was always 

 abundant. Starvation, therefore, could play no role in the re- 

 sults. Table X shows the results. 



A comparison of the totals shows that there was a gradual de- 

 crease, not only in the percentage of male-producers, but in their 

 absolute number, from the line bred in pure spring water to that 

 bred in the concentrated filtrate. 



The three series in dilute filtrate were discontinued at the 

 end of the twelve generations shown in Table X. The line in 

 spring water and that in the undiluted filtrate were bred further, 

 the additional generations in each line being shown in Table Xa. 



If the series of generations in the latter table be combined with 

 the corresponding series in Table X, of which they are continua- 

 tions, it is found that there were nineteen successive generations 

 in the filtrate without a single male-producer. In none of the 



TABLE Xa 

 Continuation oj the series oj generations bred in springwater and in undiluted filtrate shown in Table X. 



* Remainder of family not recorded. 



