774 



THE AMERICAN NATURALIST 



[Vol. XL 



America. In Europe the varying coast Hnes, the insular condi- 

 tions, the archipelagic surfaces are to be more seriously studied 

 than they have been in connection with extinction, although it 

 must be stated at once that the main phenomena of extinction in 

 unstable Europe coincide with those in stable America. 



Wallace discussed the extinction of the large Pliocene Austra- 

 lian mammals chiefly from this standpoint (see also p. 785). He 

 {(Iror/. Dist. Mam., 1876, vol. 1, pp. 158-159) attributed the 

 Australian extinction chiefly to the possible glacial conditions and 

 to the increased competition and struggle for existence caused 

 by the progressively contracted land area due to subsidence. 



The substitution of insular for continental conditions by subsi- 

 dence has undoubtedly been a potent cause both of extermination 

 in certain localities and of the survival of very primitive forms 

 (Wallace), r. ij., ATonotremata and Marsupialia in Austraha. 

 While the ronfract ion of land areas may have resulted in general 



Insular Conditions. — On islands we oi)serve local dwarfing 

 and extincti..n rather than the -enerai extinction of a family or 

 order which is our ri al snliject. ( )f inland life, so thoroughly 

 stndie.l l.v Wallace, it niav he said at once that most of the causes 

 both of survixal au<l <'Xtinction which prevail on continents are 

 infrnsi,lrd on islands. Wallace ri-htly attributes the survival of 

 the M.Miotrenuita an.l Marsnpialia to the practi.'aily insular con- 

 dhiou of the Australian re-ion. Lycll, Walla.v, aiid others cite 

 many instances of profound and rapid modifications caused by 

 the introduction of new forms on i>!and>. 



Wc have to consi(l<>r temperature and nioistinv as brought about 

 by oeoh.o-i,. and physionTai)hic chanu'o, and also as affected by 

 biotic chaimv> or chann-o in the faun i uid flora. 



[nnrasin<i Cold 



Influence of Secular Cold. — The effects of secular lowering of 

 temperature must be analyzed with some care. At first sight 



