106 



CLAUDE W. MITCHELL 



uals under starvation, embryos were developed and might even 

 be born before feeding had begun. Such development and birth 

 must obviously depend upon the stored nutritive supply which 

 the young mother had derived from its parent. Thus the degree 

 of such development or the number of young produced under 

 conditions of low nutrition offers a practical measure of the initial 

 vitality or original nutritive supply, and this measure may be 

 applied to the successive individuals produced by an}-- one mother. 

 To test this, twelve individuals were placed in cultures in which 

 food was scarce although sufficient to maintain life. The young 

 born M^ere then isolated in similar cultures of low nutritive con- 

 tent. One individual, usually an early member of the family, 

 was put, as a control, into a culture well supplied with food. In 

 some cases where the food supply was very scarce actual starva- 

 tion occurred. But the general results as shown in table 6 and 

 chart 3 indicate that in the starved families, the central members 

 of a family receive more nutrition than do either earlier or later 

 members. It is true that under such conditions of low nutrition 

 which were deemed on the whole the most favorable for this 

 experiment, chance variations in feeding cannot wholly be pre- 



Reproduction of sisters. 



TABLE 6 



Individuals marked () heavily fed; all others starved or 

 on minimum nutrition 



