INTRODUCTION 7 



measured by the degree of resemblance between 

 parents and offspring, this treatment of the subject 

 will no longer suffice. Thus Brooks (1906) says, 

 C" So far as the word is used inductively in biology, 

 heredity is the resemblance of child to parent, of 

 offspring to ancestor, while the difference between 

 parent and child is called variation." The study 

 of alternative inheritance, which appears to be the 

 most usual form of heredity, has made it necessary 

 to revise such a definition of heredity, as well as our 

 outlook with regard to its incidence. It has now 

 become a commonplace of observation that the 

 differences between organisms, as well as their 

 resemblances, are often inherited, i If a tall is crossed 

 with a dwarf variety, we know that usually the second 

 generation will inherit tallness and dwarfness — 

 the parental differentiating characters — in a definite 

 proportion, and that certain of the tall individuals 

 will go on transmitting dwarfness. We may even 

 cross two white varieties of plants or two albino 

 animals, externally alike, and obtain coloured off- 

 spring. Yet we know that the colour in this case is 

 not the result of variation. One of the necessary 

 elements in its production has been inherited from 

 each parent, though neither possesses both. In 

 such instances invisible (probably nuclear) differences 

 have been inherited which, when combined, produce 

 a striking externalised difference. Hence it is neces- 

 sary, in speaking of inheritance, to recognise that 

 both similarities and differences may be inherited, 

 the one quite as truly as the other. Some of the 

 differences, particularly the quantitative ones, which 

 appear in offspring may, then, be the result of varia- 

 tion, germinal or otherwise; but many of them will 

 be the result of inheritance. 



