vi Introduction. 



to ascertain, as our knowledge progresses, what extent 

 of variation a given form is susceptible of, and in what 

 direction variation tends. 



When referring to works of descriptive Zoology and 

 Botany, I have frequently been disappointed at finding 

 that definitions of common species, of which I could bring 

 together large series of specimens, even from a single 

 locality, were far from covering the amount of variation, 

 and I have often wished to see descriptions drawn up 

 that would answer such a requirement ; I do not, of 

 course, allude to malformations or the like, but to charac- 

 ters such as, without being regarded as absolutely diag- 

 nostic, are nevertheless used in the definitions. The 

 reason for this shortcoming appears to be, either that 

 the author has not thought it worth his while to test 

 every one of the characters on as large a series as he 

 could get hold of, or that he has purposely omitted any 

 allusion to departures from the normal in order not 

 to obscure the conception of the species. As regards 

 numerical variation, descriptions, if giving consideration 

 to the matter at all, are usually restricted to an indica- 

 tion of the extremes. 



The imperfections of this work are no doubt numerous, 

 and will be pointed out in due course by others, but I 

 feel sure those who use it will find that every effort has 

 been made to render the descriptions as comprehensive 

 and accurate as the available material permitted. 



The phylogenetic speculations which appear in this 

 monograph, making it something more than purely 

 descriptive, need apology in these days of scepticism 

 concerning all theories of evolution. I beheve it is 

 possible to discover the tendencies of each group to 

 evolve in a particular direction, and then to decide what 

 characters have been modified or lost and what new 

 ones have been produced. The way in which a combina- 



