112 FURTHER EXTENSION [pt. ii, ch. xr 



definite proof, and he decided in favour of the latter. The theory 

 of natural selection of infinitesimal variations has, however, met 

 with great and increasing difficulties in explaining the general 

 occurrence of useless characters, or the manner in which natural 

 selection can take hold of the first beginnings of a change. It is 

 now generally recognised that the bulk of the morphological 

 characters by which the systematic arrangement of plants into 

 related groups is carried out have no physiological value to the 

 plant at all. 



At this point Age and Area comes in, showing that the dis- 

 persal of species is largely independent of their distinctive 

 morphological characters, for even in the youngest of them 

 (those most limited in area) no relation can be pointed out be- 

 tween these things, and yet the conditions under Avhich these 

 very confined species are living must approximate at any rate 

 to those under which they began. One must therefore conclude 

 that specific characters have evolved without any relation to 

 their possible significance in the struggle for existence. Area 

 occupied depends mainly upon age, and not upon morphological 

 characters (of course there are many exceptions); species spread 

 where they find suitable conditions, and the adaptation is not 

 on their side, but in the long run they choose the best environ- 

 ment. Prof, de Vries regards this as being the great proof which 

 the mutation theory still wanted for its complete acceptance. 



Finally, a brief chapter (xxii), Avhich does not lend itself to 

 a summary in ad^•ance, is given to shoAv the general bearings of 

 the subject-matter of the book upon the study of distribution. 

 Age and Area, and Size and Space, are both so valid, and can 

 be so successfully used to make predictions about geographical 

 distribution, and these predictions are so near to accuracy, that 

 it is clear that in general distribution has been mainly goA'erned, 

 positively by age, negatively by barriers (of course including 

 ecological barriers). This being so, it seems probable that a very 

 promising line of work for the present may be the study of 

 invasions of plants, of course taken in connection with ecological 

 investigation into the formation (or disappearance) of barriers. 

 Age, and geographical proximity, again, will have to be taken 

 into more serious account in dealing with taxonomic questions, 

 and there are other directions in which the changes in our methods 

 of viewing problems of distribution that seem necessary may 

 produce considerable effects. 



