48 MICHIGAN ACADEMY OF SCIENCE. 



verse bands of green or brown. In such case, all European species have 

 three bands, the American four bands and the Asian are multilineate. 

 There is no exception known to this rule. The American forms are 

 closely related to the European, and indeed in the case of the two more 

 common species in these countries, the difference in the number of 

 bands is almost the only distinguishable characteristic. Curiously 

 enough, although the European species are found in England, and, at 

 least, one American form extends as far north as Minnesota, the family 

 is wholly wanting in the Pacific states, whose fauna in many other 

 respects is more closely allied to that of the old world than that of any 

 other part of this continent. In addition to the true Viviparas we have 

 three genera peculiar to North America. One, Campeloma, is of almost 

 universal extent throughout the Interior Region. Another, Lioplax, has 

 only two species. One abundantly extended through the Mississippi 

 valley and the other confined in the Coosa river in Alabama. While 

 the third and most remarkable of all is its bizarre appearance, Tulotoma, 

 is confined wholly to the upper portion of that river. 



Passing now to the bivalves, we find in the enormous development of 

 the Unionidw by far the most striking feature of our fauna. No less 

 than 645 species of this family are catalogued by Lea in his last 

 synopsis. While the increase of our knowledge of the great amount of 

 variation exhibited by this family under local influences has already re- 

 sulted in the diminution of recognized species and the future will un- 

 doubtedly increase the result, the enormous and peculiar development 

 of this family, particularly in the southeastern portion of the United 

 States, forms with the Poh/gi/rw among the land shells, and the 

 PleurocerUkr among the fluviatile univalves, the distinguishing features 

 of our fauna. The distribution of these forms over the continent is 

 general. That is. there is no portion affording a suitable habitat that 

 is without some representative of the family, yet the limitations upon 

 the distribution of many of the various groups of species, and even upon 

 individual forms are well marked and often very remarkable, and offer 

 to the inquiring mind many problems for investigation. In the main, 

 there are substantially what may be called four sub-faunas represented. 

 The great lines of archean rocks now known as the Sierra Nevada 

 range on the west and Appalachian on the east have proved to be almost 

 impassable barriers to the dispersion of this family, and the existence 

 of the distinct faunas separated by these ranges is conclusive proof that 

 they antedate the origin or the immigration of the forms peculiar to 

 them. The California Region so peculiar in its land species is equally 

 well characterized by its unione fauna. The great genus Unio, which 

 is represented in the Eastern States by more than 200 species in Ala- 

 bama and 40 in Michigan, is wholly wanting west of the Rocky moun- 

 tains, this being the largest area destitute of unio life in the temper- 

 ate or tropical regions of the globe. The only Margantina is a European 

 species of circumpolar range. While the Anodontas belong to a peculiar 

 group entirely distinct from those found in the eastern states, and 

 so closely allied to the prevalent pal?earctic type, that by eminent 

 conchologist they have considered no more than a geographical race 

 of a well known European form. 



In the same way east of the Appalachian Range, is a group of distinct 

 species extending along the entire Atlantic coast from the extreme north 



