CH. IX] OBJECTIONS TO THE HYPOTHESIS 89 



•out in its only natural habitat, the Monterey peninsula of 

 California, probably on account of the secular drying of the 

 Californian climate. 



The comparative rarity of seriously broken areas of distribu- 

 tion among endemic forms, especially south of the influence of 

 the last glacial period, is much against any very large amount of 

 dying out. One would not expect a moribund species to retain 

 its area intact— though it is true that with the Cycads, often 

 supposed moribund, this is largely the case. 



It is very hard to suppose that a genus would choose certain 

 spots upon the globe where its species should die out in large 

 numbers, yet the facts of distribution require that this should 

 be so under this explanation. Why should the Seiiecios retire 

 to die, in large numbers, to Mexico, California, Bolivia, Peru, 

 South Africa, Australia, etc.? The larger the genus the greater 

 the number of local species, and the greater the number of 

 places in which they occur. 



As this is the principal argument brought forAvard by oppo- 

 nents of Age and Area, it will be well to bring up other points. 

 If all or most endemics are to be regarded as relics, then they 

 must evidently be, on the whole, older than the "wides," and 

 the reply to another objection (23) that greater distribution may 

 be due to youth, rather than age, may be given at the same time. 

 The great difliculty is to explain why, in most countries remote 

 from the influence of the last glacial period, the "dying-out" is 

 purely mechanical. Every family and genus behaves in the 

 same way, whether it has or has not wides, and whatever its 

 habit of growth, its origin (local or foreign), or its distribution 

 generally. The general type of distribution — in "wheels within 

 wheels" — is shown below (Chapter xv) in several maps, and not 

 even the most determined upholder of a general dying-out can 

 interpret these maps into a support for his position. There is no 

 doubt that a large number of species and genera in the north 

 temperate zone may be interpreted as dying out (cf. footnote, 

 p. 86), but they are insignificant in number beside the endemics 

 of more southern regions. North-temperate America has perhaps 

 400, but Ceylon alone has SOO endemics, and Brazil perhaps 

 12,000. The latter country has 240 endemic Eugenias alone. 



A still greater difliculty for the supporters of general dying- 

 out is to explain why there should be many more endemics at 

 the point of death (VR in Ceylon, for example) than there are 

 a little further removed from it (R), and more of these than of 



