138 The Ohio Journal of Science [Vol. XVIII, No. 4, 



the results. A further explanation of this will be given later, as 

 the specific cases come up for discussion. Finally, the appear- 

 ance of the composites themselves are clearly indicative of the 

 nervous reaction or "temperament of the animal. 



The variations within each individual are beautifully 

 illustrated by the polygons themselves. Beginning with 

 Seven H, (Fig. 8), it will be seen that a marked difference is 

 shown between the feeding condition as compared with those of 

 before and after feeding. The composite of the former lies 

 considerably to the right of the other two, which explains the 

 fact that the mean has increased from 6.41 in the before feeding 

 condition to 10.42 in the feeding condition, or the length of 

 respirations increased, which indicates that feeding acted as a 

 stimulus to the animal. Comparing the standard deviations, it 

 will be seen that the feeding condition is the more variable. 

 The composite of after feeding is the most uniform, the mean 

 having gone down to 5.34 and the standard deviation to 2.20, 

 the latter figure is perhaps a little high because of a few" respira- 

 tions of extreme length. The results indicate that during 

 feeding, increased nervous activity or nervous reaction accom- 

 panied the feeding stimulus, and that before and after feeding 

 the animal was more quiet and not so nervous. 



The composite of Three H, (Fig. 7), shows the same results, 

 but not as marked as with Seven H. The mean increased 

 during feeding, but the standard deviation remained nearly the 

 same as in the before feeding condition, while after feeding it 

 increased slightly as compared to the other two conditions. The 

 stimulus of feeding upon this animal did not show as marked 

 effects as upon Seven H, merely increasing the mean or ampli- 

 tudes of respiration, and so moving the polygon a little to the 

 right. Not much difference is displayed between the before 

 and after feeding conditions, both in appearance of the com- 

 posites and their respective means and standard deviations. As 

 a whole all three composites are very similar, which leaves the 

 impression that Three H, is not easily disturbed and that her 

 nervous make-up is of a stolid nature. 



In the case of subject Two H, (Fig. 6), a more evident result 

 is evinced, namely, that the animal has a marked tendency to 

 be less active before feeding than after, which was not so 

 marked in Seven H and Three H. The mean 6.73, and the 

 standard deviation 3.09, in before feeding increased to 9.09 and 



