GENERAL LAWS OF ORIGIN AND MIGRATION. 263 



tanfc climatal changes must also occur. On the re-emer- 

 gence of the land such of these species as remained would 

 again extend themselves over their former areas of distri- 

 bution, in so far as the new climatal and other conditions 

 would permit. We would naturally suppose that the first 

 of the above processes would tend to the elimination of 

 varieties, the second, to their increase ; but, on the other 

 hand, the breaking up of a continental flora into that of 

 distinct islets, and the crowding together of many forms, 

 might be a process fertile in the production of some varie- 

 ties if fatal to others. 



Further, it is possible that these changes of subsidence 

 may have some connection with the introduction, as well 

 as with the extinction, even of specific types. It is cer- 

 tain, at least, in the case of land-plants, that such types 

 come in most plentifully immediately after elevation, 

 though they are most abundantly preserved in periods of 

 slow subsidence. I do not mean, however, that this con- 

 nection is one of cause and effect ; there are, indeed, in- 

 dications that it is not so. One of these is, that in some 

 cases the enlargement of the area of the land seems to be 

 as injurious to terrestrial species as its diminution. 



9. Another point on which I have already insisted, and 

 which has been found to apply to the Tertiary as weU as 

 to the Palaeozoic floras, is the appearance of new types 

 within the arctic and boreal areas, and their migration 

 southward. Periods in which the existence of northern 

 land coincided with a general warm temperature of the 

 northern hemisphere seem to have been those most fa- 

 vourable to the introduction of new forms of land-plants. 

 Hence, there has been throughout geological time a gen- 

 eral movement of new floras from the Palaearctic and 

 Nearctic regions to the southward. 



Applying the above considerations to the Brian and 

 Carboniferous floras of North America, we obtain some 

 data which may guide us in arriving at general conclu- 



