158 Increase in Number of Species 



by long-term changes in the ecology of the region or by coloniza- 

 tion of an area which is physically isolated from the parent range 

 but which can be reached on rare occasions by stragglers. The im- 

 portance of geographic isolation in species fission has been recog- 

 nized by many authors (Wagner, 1868; Jordan, 1905; Rensch, 1929) 

 and was accorded an extended treatment by Mayr (1942) and 

 Stebbins (1950). 



DIVISION BY RANGE MOVEMENTS 



Species fission by range movements is a simple extrapolation of two 

 dynamic aspects of ranges. Frequently in large ranges populations 

 from different areas are intersterile; this intersterility is a force which 

 tends to tear the species apart but which is counteracted by inter- 

 breeding through adjacent parts of the range. Ranges move with 

 time and in doing so may be broken into two or more fragments 

 by ecologically inimical areas encountered in the range movement. 

 This fragmentation then breaks the continuity of gene flow, and 

 each part of the range is free to evolve even greater genetic indi- 

 viduality. 



The extremely common operation of this process is manifest by 

 the large number of known disjunct ranges and by closely related 

 species occupying different ranges which can be explained satis- 

 factorily only on the basis of the fragmentation of a previously 

 continuous range. 



Examples of such disjunct ranges are common. The range of the 

 boreal forest race of the smooth green snake Opheodrys vernalis 

 vernalis (Fig. 64), has a large eastern segment and a small western 

 segment restricted to the Black Hills of South Dakota. Presumably, 

 late in the last glacial advance this snake had an unbroken and more 

 southerly range; as the ice dissipated, the range moved northward, 

 and ultimately a segment was left isolated in the Black Hills forests 

 surrounded by relatively xeric areas (Smith, 1957). Range fission 

 must be invoked to explain the disjunct range of many plants such 

 as the purple skunk cabbage Syinplocorpus foetichis which has iso- 

 lated populations in Asia and eastern North America ( Fig. 65 ) and 

 to explain the disjunct ranges of many animals each having isolated 

 populations in Florida and other areas (Neill, 1957). 



To match these patterns of disjunct ranges, there are a remarkable 

 number of instances in which exactly the same sort of division of 

 species ranges has resulted in the evolution of two or more species 

 from each parent species. Many pairs of sister species of the caddis- 

 flies have one member in western North America and the other in 



