ISOLATION AND SPECIFIC CHANGE 



EDMUND W. SINNOTT 



Connecticut Agricultural College 



Those regions of the earth which are so isolated biologically that 

 the dispersal of plants or animals between them and other areas is 

 difficult or impossible are characterized, as is well known, by large 

 numbers of species and genera which are peculiar to them or are 

 "endemic." In general, the more definitely isolated the region the 

 higher is its proportion of local forms. Why isolation should be 

 associated so universally with the presence of these endemic types 

 is a problem which has excited speculation. It is evident that once a 

 local race is established, isolation will operate effectively to maintain 

 it, both by preventing its dispersal abroad and by excluding invaders 

 which might supplant it. The difficult problem has been to account 

 for the actual origin of the endemic types themselves in the first place. 

 In an attempt to throw light on this problem a study has been made 

 of the floras of a number of islands which are isolated to a greater or 

 less degree from adjacent land masses and have developed a large 

 body of local species and genera of vascular plants.^ 



Certain elements of the endemic flora in all these islands are doubt- 

 less not of local origin but are "relicts," remnants of types once much 

 more widely spread, which owe their preservation to freedom from 

 the keener competition of the mainland. These types do not con- 

 cern our problem. As to just how abundant they are we do not know, 

 but those forms which stand well apart and have no near relatives in 

 the islands or elsewhere are probably to be looked upon as relicts. 

 From the close similarity of most of the endemic species and genera 

 with others on near-by islands or on the adjacent mainland, however, 

 and from their frequent occurrence as groups of related forms, it is 

 evident that the bulk of the endemic element in these floras is actually 

 of local origin. 



Several hypotheses have been put forward to account for the 

 origin of these endemic forms. Some investigators have pointed to 

 natural selection as the primary factor, believing that new types are 

 produced by this agency to fit exactly the peculiar conditions in each 

 region, rather than a wide environmental range. Others, less con- 



' These islands are New Zealand, Ceylon, Hawaii, the Galapagos, Juan Fernan- 

 dez, St. Helena, Sokotra and Mauritius. 



444 



