IMIKKH \\< l. IN QTJINBA-PIGB. 



length measurement as indicated on MacDowell's figure 5, F. Two 

 observations have also been made of skull dimensions, one of basilar 

 skull length as indicated in MacDowell's figure 1, 0. M., and the other 

 of maximum lygomatic width [Z). The measurements here dealt with 

 found upon repeated measurement to be accurate within 0.1 mm. 

 The observations are combined in classes of 5 mm. range in tables 29 

 to :•] for the Beveral races and hybrids studied. We may consider 

 first the obeervations on skull length. 



THE CUTLERI HYBRIDS. 



1 1 ii adult cvthri females have skull lengths distributed as shown in 

 table *- )( .». The range extends over 10 classes; the mean is 51.55 mm., 

 and the standard deviation 13.50 mm. Twenty-eight adult females of 

 race B have a range in skull length of 15 classes; their mean is 58.14 

 nun., and the standard deviation 19.75 mm. F r hybrid females 

 between cuth ri males and females of races B and C available for study 

 number 24. with a range of 13 classes, a mean skull length of 57.70 mm., 

 and a standard deviation of 16.85 mm. These figures indicate (like 

 the weight observations) that the F x hybrids are practically as large 

 as the larger parent race and not more variable. The 33 F 2 hybrids 

 studied show a range of 18 classes, with a mean at 54.35 mm. and a 

 standard deviation of 17.20 mm. The Fj mean was about 3 mm. 

 greater than the intermediate between the races crossed, but the F 2 

 mean practically coincides with it. Judged by the standard deviation, 

 1 _ is not more variable than pure race B and is only slightly more 

 variable than F,. The means show in F x an increase in size over that 

 we should expect through inheritance, but a loss of this increase in F 2 . 



Observations on the male hybrids from the same cross are recorded 

 in the next four row r s of table 29. The mean of the F x hybrids is 

 again greater than that of either pure race and surpasses the inter- 

 mediate point by nearly 4.5 mm. The mean of the F 2 hybrids is close 

 to the intermediate, which it exceeds by about 0.6 mm. The vari- 

 ability standard deviation) of F 2 , however, is considerably greater than 

 that of 1' and even exceeds that of pure race B. Particularly note- 

 worthy is the occurrence of one very large F 2 individual, nearly as large 

 as the largesl 1 individual. Compare this with the occurrence of a 

 single very hh.iII 1' female, as small as the smallest pure cutleri female. 



The cull, ri hybrids with races B and C show similar results in regard 

 to the other measurements studied — zygomatic width and femur length. 

 t ablee 31 > and 3 1 and] >late6.) In every case F { exceeds the means of 

 both parent races, but I' approximates the intermediate between them, 

 which n exceeds by a fraction of a millimeter only. In no case is the 

 I mean as great as the race B mean— that of the larger parent race. 

 These facts, like the weight curves, indicate (1) that, so far as heredity 

 oncerned, an exact intermediate between the parent races would 



