CHAPTER XIII 



AGE AND AREA, AND SIZE AND SPACE, 



IN THE COMPOSITAE 



By James Small, D.Sc, F.L.S. 



Age and Area. In a previous contribution to the study of the 

 geographical distribution of the Compositae (103) many of the 

 conchisions were based upon the Age and Area hypothesis as 

 far as the phenomena could be determined roughly by simple 

 inspection of a series of maps which included all the genera. It 

 Avas mentioned (103, p. 190) that although this hypothesis was 

 still restricted to "age within a given country, its proved exten- 

 sion to absolute age and total area seems to be only a question 

 of time and application." This extension of the original hypo- 

 thesis, which Avas suggested in 1916 by the writer (103, p. 208), 

 has now been adopted by Dr J. C. Willis, and the present con- 

 tribution consists of a critical analysis of the statistics for Age 

 and Area in the Compositae in the light both of that extension 

 and of previous phyletic conclusions. These previous suggestions 

 were simimarised as "the basis of future discussions" (103, 

 p. 313) in a family tree which is reproduced upon p. 125. The 

 statistical data are given in Table I, and were obtained by the 

 following methods. (Table I. pp. 120-124.) 



In order to avoid the unbalanced effects of the inclusion of 

 new genera which have been discovered or resuscitated fre- 

 quenth'^ as the result of special studies of only one or a few 

 tribes, the data have been prepared only for the genera included 

 by Bentham in the Genera Plantarum. The area covered by each 

 genus has been determined approximately in millions of square 

 miles. For this purpose Mikania and Eujjatoriiitn have again 

 (cf. 103, pp. 133 and 204) been taken as one genus, and so have 

 Aster and Erigeron as two genera which are "so very closely 

 allied that the transitional species are comparatively numerous 

 and the genera in these cases are distinguished only by the so- 

 called indefinable characters of the taxonomist" (103, p. 307). 

 All genera occupying less than 1,000,000 square miles have been 

 included in Class 1 : while 59 other classes have been taken for 

 the other genera, the total area of the land surface of the world 

 being approximatelj^ 60 million square miles. This method is. 



