BARRIERS 57 



in possession is increased by the same means, and the out- 

 come is then largely determined by other factors. To avoid 

 repetition, the bearing of occupation upon invasion will be 

 considered under succession. 



BARRIERS. 



DeCandolIe (1820:45) seems to have been the first to use 

 the term barrier and to distinguish the various kinds, though 

 Hedenberg (J754:67) clearly saw that stations of one kind 

 were insurmountable obstacles to plants belonging to a very 

 different type. DeCandolle points out that the natural 

 barriers to continuous invasion ("transport de proche en 

 proche") are: (1) seas, which decrease invasion almost in 

 inverse proportion to their extent ; (2) deserts; (3) mountain 

 ranges, which are less absolute on account of passes, valleys, 

 etc.; (4) vegetation, marshes being barriers to dry land 

 plants, forests to those that fear the shade, etc. Henfi-ey 

 (J852 : 56) discusses the effect of geological and climatic bar- 

 riers, and recognizes that man and animals often place a 

 limit upon the spread of plants. A. DeCandolle (1855 :70, 207) 

 treats of climatic barriers at considerable length, while 

 Datwin (1859) discusses topographical barriers in a more 

 general way. Gfisebach (t872:4), in discussing the effect of 

 barriers upon the constitution of vegetation, lays down the 

 fundamental rule that: "La loi supreme servant de base h 

 I'^tablissement persistant de flores naturelles. doit done §tre 

 reconnue dans les barriferes qui en ont entrav^ ou compl^te- 

 ment emp§ch6 le melange." Pound and Clements (J898:I8) 

 distinguish the physiographical barriers in invasion as 

 obstructive, i. e., barriers, and conductive, agencies for 

 migration. 



Any feature of the topography, whether physical or 

 biological, that restricts or prevents invasion, is a barrier. 

 Such features are usually permanent and produce permanent 

 barriers, though the latter may often be temporary, existing 

 for a few years only, or even for a single season. In this 

 last case, however, they are as a rule recurrent. Barriers 

 •Bay furthermore be distinguished as complete or incomplete 

 with respect to the thoroughness with which they hmit 



