INTRODUCTION 15 



The action of isolation in producing discontinuity is dealt 

 with in Chapter V and that of correlation in Chapter VI. 

 The efficacy of Natural Selection as the most generally ac- 

 cepted theory of the spread of new characters is examined 

 in Chapter VII. It is shown that the scope of this process 

 is questionable. In Chapter VIII we examine the other 

 theories of evolution, and in Chapter IX the nature of adap- 

 tation and the special difficulties of explaining its origin 

 are detailed. In a general summary (Chapter X) we 

 attempt to define the relationship between adaptation, varia- 

 tion and group-formation and to distinguish between their 

 presumed causes. 



We may conclude this chapter with some remarks on 

 procedure in evolutionary inquiry in so far as our methods 

 are involved. 



Many of the subjects mentioned above can be investigated 

 experimentally. The origin and mode of inheritance of 

 variation are almost exclusively to be treated in this way. 

 The validity of the selection hypothesis, as an explanation of 

 the spread of variants, has been likewise tested by experiments 

 in the field and in the laboratory, and the formation of new 

 habits, food preferences, reactions to the environment, etc., 

 have been similarly investigated. The behaviour of animals, 

 their interrelationships, seasonal occurrence and the incidence 

 of actual environmental pressure on animal populations are 

 most profitably studied by direct observations in the field. 

 For the study of the distribution of variants in nature, the 

 formation of groups and the incidence of correlation we fall 

 back on the methods of taxonomy and statistical analysis, 

 though the findings of genetics are of service here : of supreme 

 importance is the method of population-analysis, which is a 

 combination of statistics, field observation and taxonomy. 

 This has been much in vogue during the past thirty years. 

 It dates further back indeed, viz. to the pioneer work of 

 Coutagne, Gulick, Duncker and Heincke, and to other studies, 

 particularly of economically important animals (fishes). 

 More intensive and critical work supported by modern gene- 

 tical and statistical methods has been conducted by such 

 workers as Crampton, Schmidt, and Sumner. 



In this work we are approaching the subject of evolution 

 primarily as taxonomists. We believe that all theories of 



