32 



Scientific Proceedings (55). 



the actual counts. In other words it has been assumed that we 

 are dealing here with strictly integral variates. This assumption 

 seems justified for the present material, but not for the interracial 

 material of the previous paper. 



The constants deduced from Table I are given in Table II. 



TABLE II. 



Constants of Variation Deduced from Table I. 



Character. 



Mean. 



Standard Deviation. 



Coefficient of 

 Variation. 



Number of mammae 



ii.39±.io 

 8.88 ±.23 



1. 12 ±.07 



2.54±.i6 



9-84 =*= -63 

 28.63 ±1.95 



Number of young at birth. . . . 



Coefficient of correlation r = 0.195 =*= .086. 



From this table the following points are to be noted. 



1. There are approximately 2.5 more mammae in the dam, on 

 the average, than number of young in the litter in these swine. 

 This is a slightly larger "factor of safety" than is found inter- 

 racially. 



2. The intraracial correlation between these variables in swine 

 is not only absolutely low, but is relatively much lower than the 

 interracial correlation. Again it is not apparent here that natural 

 selection has operated in the expected manner. 



3. Intraracially, just as interracially, size of litter is relatively 

 a more variable character than number of mammae in the dam. 



4. There is, as would be expected, a very considerable reduction 

 in variability, in respect of both characters, in the single species 

 (intraracial) as compared with the composite group of 90 different 

 species (interracial). 



18 (835) 



The effect of animal extracts upon the volume of the spleen. 



By Isaac Ott, M.D., and John C. Scott, M.D. 



[Physiological Laboratory, Medico-Chirurgical College of 

 Philadelphia.] 



We studied the volume of the spleen with an oncometer 

 attached to a modified piston recorder. The animals employed 



