310 Comparative Evolution of Biomes 



Thus only the species able to exist under climatic or biotic condi- 

 tions of various kinds occurring on a connection are able to cross it. 

 Simpson ( 1947 ) has given the apt name "filter bridge" to terrestrial 

 connections, which probably fits all land connections sufficiently 

 narrow and intermittent to be considered as bridges. 



Only those members of a biome in actual geographic contact 

 with bridging areas can cross them. Thus of the many northern 

 species of the rodent genus Citelliis, only the most northerly and 

 holarctic species C. nndulatus appears to have made a recent cross- 

 ing of the Bering bridge between Siberia and Alaska (Fig. 135) 

 (Burt and Grossenheider, 1952; Rausch, 1953). In each of many 

 North American genera in divers groups only one species now 

 extends into Alaska and is in a position to cross the Bering bridge 

 should the climate change slightly. In the large caddisfly genera 

 Hydropsyche and Triaenodes only H. riola and T. tarda extend into 

 Alaska, and of the many nearctic pines only Piniis contorta extends 

 into the vicinity of this region (Munns, 1938; Flook, 1959). 



A remarkable parallel exists between these present-day examples 

 and indications of what happened in the past. In a surprising num- 

 ber of instances only one species of a moderately large genus has 

 dispersed over a land bridge and become established in another 

 regional association of its parent biome. This is suggested when- 

 ever all the species in each regional unit are monophyletic. Exam- 

 ples in the caddisflies include the North American genus Pycno- 

 psyche in the eastern deciduous forest, the entire American branch 

 of Agapetus, and the Australian branch of Agapetus. 



Virtual proof of past dispersals of only one species per group 

 is demonstrated in the caddisfly genus Wormaldia. One subgenus 

 consists of a number of Asiatic species and one species, W. mohri, 

 in the Great Smoky Mountains of eastern North America. The 

 American species is closely related to W. kisoensis, the only known 

 Japanese species, although it is distinct from it in many features. 

 When the phylogcny and geographic distribution for the subgenus 

 are correlated, it is evident that the evolution of the subgenus cen- 

 tered around China and that late in this evolution a northern spe- 

 cies evolved which dispersed into North America and to Japan, 

 the two migrant populations giving rise to W. mohri and W. kisoen- 

 sis respectively (Fig. 136) (Ross, IQDGa). In another family of 

 caddisflies, the genus Ilimalopsyche illustrates the initial evolution 

 of a single northeastern Asiatic branch and the subsequent dispersal 

 of this one distinctive line into North America. The single American 

 species of Himalopsyche lives in Californian streams. 



