38 HELEN DEAN KING 



ference is nearly three times the probable error, and is, therefore, 

 large enough to signify that in this series the range of variability 

 in the body weight of the males was greater than that of the 

 females. 



In the B series the cofficients for the males were, as a rule, 

 larger than those for the females up to the 334 day period: 

 Beyond this age the coefficients for the females were the larger. 

 This latter relation is not a normal one, and it can" be attributed 

 in this instance to the fact that the number of records that were 

 available for use in calculating the coefficients for the older 

 males was very small: in this series only seventeen out of a 

 total of fifty-seven males lived to be one year old. 



At all age periods, except at 13 and at 304 days, the coefficients 

 for the males of the A series of inbreds were higher than the cor- 

 responding ones for the males of the B series, although the dif- 

 ference in some cases was less than the probable error. Between 

 the average coefficients for the two male groups there was a dif- 

 ference of 2.7 points, but the importance of this difference is 

 greatly lessened by the fact that the coefficients for the older 

 males in the B series were abnormally low. The evidence, on 

 the whole, would seem to indicate that there was little difference 

 in the range of variability in the bod^ weights of the males in 

 ' the two inbred series, since a comparison of the average coefficients 

 for the two male groups up to 334 days, taking all ages together, 

 shows a difference of only 1.1 points in favor of the individuals 

 in the A series. 



At ten of the sixteen age periods shown in table 15 the co- 

 efficients of variation for the females of the B series of inbreds 

 exceeded those for the females of the A series, but in many cases 

 the differences were less than the probable error and therefore 

 they can have no significance. The average coefficients for the 

 two groups differed by only 0.04 points, so it is evident that the 

 range of variability in body weights was practically the same 

 for the females of the two inbred series. 



A comparison of the coefficients for the males with those for 

 the females in the combined series (A, B) shows that in the 

 majority of cases the male coefficients were much the higher. 



