150 F. B. SUMNER, M. E. MCDANIEL AND R. R. HUESTIS. 



Miss King gives the data for 425 hybrids (Fj, F L> and F 3 ) 

 between the Norway and the albino rat. The sex ratio found 

 was 119.07 or ii8.ii, depending upon whether the entire number 

 was included, or whether consideration was restricted to the mott 

 reliable series (277). In either case, the difference cannot be 

 regarded as significant, according to the standard here adopted. 

 In the first case, it is 14.5 8.3. 1 Miss King has also computed 

 the sex ratio, from the records given by von Guaita of hybrids 

 between albino mice and Japanese waltzing mice. The ratio is 

 113.17, but this is based on only 356 mice of known sex, and is 

 surely of doubtful value. 



Pearl and Pearl give the results of computations, based upon 

 vital statistics of the city of Buenos Aires. They have compared 

 the sex ratios resulting from Argentine X Italian and Argentine 

 X Spanish matings with those resulting from matings within each 

 " pure " race. The figures for the " hybrid " group were in each 

 case the larger, and the authors believed that the difference was 

 significant in three of the four cases taken singly. But when 

 these probable errors are computed according to the formula now 

 adopted by Dr. Pearl (see p. 127), the significance of these various 

 differences becomes doubtful in all cases except one. That be- 

 tween the sex ratio for the Italian-Argentine cross and for pure 

 Italian stock, is 4.95 1.05. In all of the other cases the probable 

 error is half or more times as great as the difference. Despite 

 the great numbers comprised in these statistics, there is thus 

 some doubt as to the reality of these differences. And their 

 biological meaning is further obscured by the fact that the " pure " 

 races differ among themselves in respect to their sex ratios quite 

 as significantly as they differ from the " hybrids " ;~ likewise by 

 the fact that tlie highest "pure'' ratio ( the Spanish X Spanish) 

 is very nearly identical with the lower of the two " hybrid " ratios. 

 It should be added that the authors themselves did not express 

 any great confidence that the higher ratios of the mixed matings 

 are due to hybridization per sc. We shall discuss later the sig- 

 nificant yearly differences which are shown in Table I. of the 



1 Assuming 104.6 + 2.77 as the normal ratio for the white rat. 



2 The same reservation must, though certainly with less cogency, be made 

 in the case of our own data, discussed above. 



