xxiv] THE DIFFERENTIATION THEORY 305 



Naias marina, giving birth to new species in various parts of 

 its range. 



There is another case among water plants which, though 

 Guppy does not allude to it, seems to the present writer to be 

 readily interpreted on the differentiation theory. The case in 

 question is that of the family Aponogetonaceae, with its one 

 genus Aponogeton, the Arrowgrass, often cultivated in England 1 . 

 Africa and Madagascar appear to be the headquarters of the 

 genus; the species in this region consist almost entirely of 

 plants with forked inflorescences, while the Indo-Australian 

 species have simple inflorescences. The species can be placed, 

 according to their geographical position, in a series extending 

 from west to east which also represents their affinities. The 

 African species lead on to the Madagascan ; these show affinity 

 with the Indian, while the North Australian are the most 

 remote. It seems that we must interpret the genus Aponogeton 

 as having reached a more advanced stage of differentiation than 

 such genera as Naias and Limnanthemum. Aponogeton no longer 

 contains any species whose range is approximately coterminous 

 with that of the genus, but the original area has become 

 "divided up into a number of smaller areas each with its own 

 group of species 2 ." However it must not be overlooked that 

 this case might be interpreted in other ways by those who hold 

 different views on plant evolution. 



Willis 3 has in recent years put forward a remarkable hypo- 

 thesis which is in many ways easily related to Guppy's theory 

 namely, the " Law of Age and Area," according to which the 

 relative size of the geographical territory occupied by each 

 species within a genus (or genus within a family) is, in general, 

 proportional to the age of that species. According to this 

 hypothesis, the most widely distributed genera and species 

 instead of being the best adapted, as is maintained by orthodox 

 Darwinians are in reality the most primitive, while those occu- 

 pying limited areas are relatively modern. It is impossible here 



1 Krause, K. and Engler, A. (1906). 2 Guppy, H. B. (1906). 



3 Willis, J. C. (i9i4 2 ), and a number of earlier and later papers. 



