HEREDITY AND VARIATION 



parents that produce the gametes are not themselves as a rule exactly 

 alike, and the gametes produced by them will therefore differ. The 

 dissimilar can breed together within certain limits. Meml races 



or varieties commonly interbreed freely ; sometimes -pecies and even 

 genera are interfertile. The artificially produced generic hybrids of 

 certain Orchids are a well-known feature of horticultural shows; 

 and other cases are known of fertile hybrids arising from crosses 

 between species of distinct genera, such as the " Raphanobrassica " 

 derived from a cross between Radish and Cabba^ The breeding 

 together of members of distinct races, species, etc., is called hybridisa- 

 tion, and the offspring are known as hybrids. Since the experimental 

 study of hybrids has thrown much light on the mechanism of inherit- 

 ance, it will be treated in some detail ; but before this can be done, 

 the nature of variation must be briefly considered. 



The characteristics of organisms have been classed under two heads : 

 those which are heritable and those which are not. The latter category 

 includes such features as can be related directly to the impress of 

 external circumstances upon the parent ; the former comprise those 

 features which cannot be so related. Mutilations can be quoted as 

 examples of characters not transmitted ; likewise the immediate 

 accommodations of the growing parts to the impact of gravity, light, 

 etc., such as have been described in Chapter IX. However effective 

 these may appear to be in determining the mature form of the parent, 

 there is no reliable evidence that such modifications are transmitted 

 to the offspring. But there are other larger or smaller deviations from 

 type, appearing suddenly and individually, which have not been 

 referred directly to known causes and are found to be [heritable. 

 Individuals showing variation of this type are called mutants and the 

 process of their formation mutation. It may be emphasised again 

 that modification leaves no permanent impress upon the organism 

 so as to affect its gametes. On the other hand, it has now been esta- 

 blished with certainty that mutation is the chief source of the varia- 

 tions upon which Natural Selection can work, and that it has therefore 

 played an important part in Evolution. 



Mendelian Segregation. 



It has long been known that offspring produced by the crossing of 

 closely related forms, whether of animals or plants, do not alw. 

 come true to type. But it remained for Mendel to discover, in the 

 latter half of the nineteenth century, the laws, since verified by many 



