370 APPENDIX B 



as can the pure recessives. Nothing of this kind ever occurs as 

 regards the sexual characters, for we cannot mate males 

 with males, or females with females. We must perpetually 

 mate together impure dominants, or at least cross dominants 

 with recessives. 



It is evident that the significance of the Mendelian phenomena 

 has been mistaken. They are anomalies rare anomalies of 

 sexual reproduction. Interesting and important as they un- 

 doubtedly are, they furnish no master key to all the problems 

 of heredity. 



Varieties are usually fertile when crossed, and their offspring 

 and descendants frequently exhibit Mendelian phenomena. 

 Species tend to be infertile when crossed, and Mendelian 

 phenomena are less common ; the hybrids appear, rather, to 

 blend the parental characters. Probably this apparent blending 

 is nothing other than a more or less complete return to the 

 ancestral type, which resembles either species more than they 

 resemble one another. The difference of behaviour between 

 crossed varieties and species may perhaps be explained by the 

 hypothesis that characters tend to behave as Mendelian allelo- 

 morphs only when they differ within certain limits. If the 

 difference is too small the characters blend ; but if it is too great, 

 neither character is inherited, or there may be a combination of 

 blending and loss. 



Apart from Mendelian or sexual latency (i. e. the latency of 

 the recessive character in the impure dominant) is that more 

 permanent kind of latency which occurs when ancestral charac- 

 ters lie dormant through an indefinite number of generations. 

 According to Darwin, " Besides the visible changes that it (the 

 germ) undergoes, we must believe that it is crowded with 

 invisible characters proper to both sexes, to both the right and 

 left sides of the body, and to a long line of male and female 

 ancestors separated by hundreds and even thousands of genera- 

 tions from the present time; and these characters, like those 

 written on paper with invisible ink, lie ready to be evolved 

 whenever the organism is disturbed by certain known or 

 unknown conditions." l If, however, Darwin's works be carefully 

 studied, it will be found that all, or almost all, the instances of 

 latency quoted by him are drawn from domesticated animals and 

 plants. 2 A similar lack of evidence that latent traits are common 

 in natural varieties may be noted in the works of other writers, 

 for example the Mendelian experimenters. If we cross such 

 varieties, we may render characters latent (usually in the 



1 Animals and Plants, vol. ii., pp. 35-6. 



3 Unfortunately my acquaintance with Botany is so small that I am not 

 iure that all the plants mentioned b} r Darwin are cultivated varieties. (See 

 Animals and Plants, vol. ii., pp. 32-5. 



