32 THE AMERICAN NATURALIST [Vol. LV 



"Weight at inoculation is of course completely deter- 

 mined by the weight at 33 days and the rate of gain and 

 age at inoculation. The correlations between weight at 

 inoculation and these three factors are accordingly high. 

 They are rather different in the two groups of experi- 

 ments, but this is to be expected. There was much less 











































































\ 

































\ \ 



































\ 























4- 









































































































































































IRT 





















,..:\ 





































variation in age in lots 7-9 than in lots 10-14 and varia- 

 tion in age accordingly plays a less important part in de- 

 termining the variations in weight than in the latter case. 

 The rather high correlation between age and rate of gain 

 is due to the fact that nearly all of the guinea pigs suf- 

 fered a loss in weight following shipment at a little over 

 33 days of age. Thus the older they were at inoculation, 

 the greater the time in which they had to recover from 

 the effects of shipping. The rate of gain before 33 days 

 is generally an indication of the probable later rate df 

 growth. In the present case, however, the slight nega- 

 tive correlation^ between the rates of gain before and 

 after 33 days seems to indicate that shipping delayed 



