224 



BARBARA LEE LUND. 



placed the survival times of the sisters which died first in one 

 column, and the more resistant sisters in another (though as a 

 rule the survival times of sisters was the same) , took an average 

 of the two sets of sister cells of thirty individuals for each period 

 and then found the differences between these averages, which I 

 called the average difference between starved sisters, given in 

 Table VI. E. This difference was very small, being .02, .03, 

 and .02 minutes respectively for 15, 19 and 25 hours. This is 

 practically the same as saying that there is no difference, since 

 these numbers are within the limits, of error in determining the 



2.4 



it OUTS 



FIG. 3. Paramecium. Survival time of starved and yeast fed Paramecia in 

 KCN. (See Table VI.) The abscissa represents time in hours between division 

 and death of Paramecia. The ordinate represents survival time in minutes. 

 Curve B represents the average survival time of yeast-fed Paramecia. Curve D 

 represents the average survival time of the starved sisters of B. Curve E repre- 

 sents the difference between the average survival times of the fed and starved 

 sisters. That is, it is the difference between the curves B and D. 



death point. It is important to notice that the average difference 

 in survival time of thirty pairs immediately after division, and 

 the average difference after the sisters had been starving 25 

 hours, are exactly the same (viz., .02 minutes), even though the 

 actual survival time of sisters immediately after division and 

 after starving for 25 hours is very different. This close corre- 

 spondence is probably due to the fact that in the case of Para- 



