208 / ARTHUR THOMSON [march 



ovum being relatively much larger than the sperm), but with further 

 growth the resemblance to the male gradually increased. But in the 

 experiments recorded by Mr. Tutt, it is noted that the influence of the 

 male parent seems to be less than that of the female. It seems likely 

 that the discrepancy is to be in part accounted for on the lines of Mr. 

 Vernon's conclusion as to sea-urchins, that the characteristics of the 

 hybrid offspring depend directly on the relative degrees of maturity of 

 the sex-cells of the two parents. Professor Ewart has not found any 

 reason to regard prepotency as correlated with sex. As he cautiously 

 says : — " When allowance is made for reversion, inbreeding, and various 

 other factors, it is extremely difficult to estimate how far the one sex 

 predominates over the other." 



Inbreeding. 



The diversity of opinion in regard to inbreeding, which some exalt 

 as essential to the success of a race, and others decry because of alleged 

 baneful influences, is probably due to the fact that it is beneficial only 

 up to a given limit, and that it is apt to fail prematurely because of 

 some taint in the stock. 



On the one hand, it is advantageous in fixing character or 

 developing prepotency, as Professor Ewart illustrates by Dalmatian dogs, 

 Basset hounds, and hornless Galloway cattle. As Galton maintains, the 

 mating of two extraordinary members of two stocks is likely to be 

 followed by a heavy filial regression, while this tends to be slight 

 between two equally-gifted but not extraordinary members of a high- 

 class inbred stock. Moreover, with the extension of an untainted 

 pedigree the risk of serious reversion is probably lessened, for there are 

 probably limits to the duration of latent characters. In other words, 

 the more thorough the inbreeding, within the limits of stability, the less 

 will be the normal quantitative filial regression, and the less will be 

 the risk of the more qualitative reversion. On the other hand, as to 

 disadvantages, there is on a 'priori grounds a necessary lessening of 

 variability, since that is in part conditioned by cross-breeding ; and there 

 is the actual fact that inbreeding often goes too far, and results in loss 

 of vitality and in degeneration. Professor Ewart instances the cases of 

 foxhounds, hogs, guinea-pigs, and race-horses. Sir Everett Millais noted 

 in regard to inbred dogs that distemper carries off about 60 to 70 per 

 cent, of those attacked, and that hereditary deformity and disease tend 

 to be induced. 



It may be of interest to recall the experiments which Eitzema Bos 

 made some years ago with rats (Mus deeumanus). From seven of one 

 family and an unrelated male which died after two crossings, he con- 

 tinued inbreeding for six years, about thirty generations. In 1887 the 

 average number of a litter was 7|-, in 1891 4:^, in 1892 3^ The 

 rate of mortality and the number of infertile pairings greatly increased 

 (Biol. Centralbl. 1894, xiv. pp. 75-81). 



