522 SCIENCE PROGRESS 



in these chapters we find some extremely interesting results, especially with 

 regard to latent factors. The investigations show in a striking way how differ- 

 ent may be the visible characters of the individual and the qualities which it 

 transmits ; they carry us far indeed from the old idea that heredity was the trans- 

 mission of a visible character from parent to offspring. We find that albinos 

 may carry a factor for black which shows itself * only when united with another 

 factor for colour. In flowers two whites carrying invisible colour factors may, 

 when crossed, produce a coloured flower : two white sweet peas, for example, 

 differing only in the shapes of their pollen grains, produced, when crossed, a 

 heterozygote with purple flowers like the original Lathyrus odora/us, and in 

 F 2 gave three kinds of flowers, purples, reds, and whites. To explain these 

 colours and their numerical proportions three pairs of allelomorphs are assumed, 

 two of which are factors that must occur together to produce any colour at all. 

 Various other factors produce subordinate types. Thus we arrive at the Men- 

 delian interpretation of reversion on crossing, and of variation, or certain cases 

 of variation, in cultivated organisms. Reversion is the union of factors occurring 

 separately in the types crossed, and variation is the omission or addition of a 

 factor. With regard to eye-colour in man and the colours of race-horses, com- 

 parisons are made between the Mendelian and biometric methods, much to the 

 advantage of the former. It is obvious that the units of eye-colour required 

 by the biometric method are very difficult to determine, while the Mendelian 

 units, on the other hand, have sometimes the disadvantage of being purely 

 theoretical. 



The most unconvincing chapters are those which deal with heredity and sex. 

 Prof. Bateson's treatment of this subject fails because he has not taken into 

 account what is known of sexual dimorphism in nature. He quotes the results 

 of crossing sheep horned in both sexes with another breed hornless in both 

 sexes, the hybrids being horned in the males, hornless in the females. The 

 author merely compares this result with other cases of sex-limited inheritance, 

 ignoring the fact that in the ancestral sheep the females were hornless or nearly 

 so. The discussion of colour-blindness in man is left in complete obscurity. 

 After a number of definite statements and a diagrammatic scheme of the heredity 

 in the original text, a half-page of corrections has been inserted in the volume 

 which cancels all the principal conclusions reached : colour-blind men, we are told, 

 do not have colour-blind sons, so that the descent in this case cannot at present 

 be explained on Mendelian principles. Several other cases of human abnor- 

 malities are discussed, and it is shown that many of them behave as dominants 

 and are transmitted in accordance with Mendelian laws. 



An impartial account is given of the few cases in which the evidence in- 

 dicates that no segregation takes place, as in the cross between negro and 

 European. One most remarkable case in which the absence of segregation was 

 observed by Mendel himself, and puzzled him completely, is that of Hieracium. 

 It has now been discovered that this plant and several others have the power 

 of producing seeds parthenogenetically, and as no reduction in the nuclei of 

 the gametes takes place there can naturally be no segregation. 



Although as mentioned at the beginning of this re/iew, Prof. Bateson does 

 not profess to discuss fully the relation of Mendehsm to evolution, he has a 

 short chapter on Biological conceptions in the li^ht of Mendelian discoveries. 

 He suggests that the something which is present in a dominant character may 

 be of the nature of a ferment, but that it is rot the ferment which is present 

 in the gamete, but the power to develop the ferment. He does not think, in 



