274 



SCIENCE 



[N. S. Vol. XLIV. No. 1130 



Yellowstone and Snake rivers mingle during 

 the wet season and afford a means by which 

 fresh-water animals have crossed from one 

 drainage to the other. The dispersal of all 

 fresh-water forms has been normally by means 

 of the changes in river systems, the fauna fol- 

 lowing up the river as the head waters of the 

 latter work their way into new territory. Fre- 

 quently, ancient changes in streams, incident 

 tb piracy or beheading, etc., may be known 

 long afterward by the peculiarities of the 

 fauna inhabiting the present river system, in- 

 dicating many times that the present system is 

 made up of several ancient systems. A case 

 in point is the Tennessee Eiver system which 

 has been shown by C. C. Adams, 5 from a study 

 of the distribution of the molluscan genus Io, 

 to be made up of several smaller systems once 

 separated by divides. Ortmann's studies on 

 the Unionidag and the crayfishes also bring out 

 the value of distributional areas by river sys- 

 tems. The peculiar physical changes in the 

 Ohio River previous to and following the gla- 

 cial period, will doubtless be reflected in the 

 fauna, both recent and extinct, when detailed 

 studies are made bearing on this subject. 



The point which the writer wishes to bring out 

 and emphasize is that while it is true that there 

 are the three primary divisions as indicated 

 by Simpson and so strongly advocated by Ger- 

 main, there are also in addition many smaller 

 divisions which form precise faunal areas just 

 as true and natural as the three major areas. 

 The true relation of the different fresh-water 

 faunas can only be determined by dividing the 

 continent into areas separated by natural water 

 partings, as has been done by Dr. Dall and the 

 writer. That too many divisions have been 

 made by the writer in his Lymnsea monograph 

 may be true and is to be expected in a first at- 

 tempt, but the method is the only satisfactory 

 one for the study of fluviatile animals, a state- 

 ment in which I am sure all American stu- 

 dents will agree. Studies from this standpoint, 

 however, have not yet been made in sufficient 

 number and detail to work out a comprehen- 

 sive scheme of subdivision. It was with a view 



s National Academy Sciences, Memoir XII., No. 

 2, 1915. 



to stimulating such studies that the map in 

 question was published. 



Germain (page 397) criticizes the author for 

 his statement (page 84) that " It is not be- 

 lieved by the writer that the supposed land 

 connection with Europe via Greenland con- 

 tributed to any extent in the formation of the 

 present Lymnseid fauna," and states that it is 

 dangerous to base a general conclusion on a 

 particular case. The statement was not based 

 on the Lymnagidse alone, but on the whole 

 Basommatophorous group, the exotic species 

 of which, from the data at present known, 

 seems to have reached America by way of 

 Alaska rather than by the Greenland connec- 

 tion. The absence of such striking forms as 

 Lymnwa stagnalis, GaTba palustris and Aplexa 

 hypnorum from the Greenland fauna and their 

 presence in the Alaska fauna is tangible evi- 

 dence, to say the least. It is of course pos- 

 sible that this condition is due to a lack of 

 sufficient detailed field work in northeastern 

 America, but until this has brought to light 

 the missing data the deductions must remain 

 as based on present information. The invasion 

 from Siberia was evidently contemporaneous 

 with that of the larger mammals which oc- 

 curred in the late tertiaries. The northeastern 

 land connection is thought to have been used 

 by several mollusks (Margaritana and some 

 helices) and it would be strange indeed if some 

 fresh-water mollusks of other groups did not 

 also take advantage of the land bridge. How- 

 ever, in this as in other things the deductions 

 must be based on the available facts and not 

 on theories. The discovery of the European 

 land snail Tachea hortensis in Pleistocene de- 

 posits 6 goes a long way toward establishing 

 the existence of a northeastern land connec- 

 tion in late Tertiary time. 



The critical study of the fresh-water faunas 

 of many states and the ecological work of sev- 

 eral universities is providing a mass of data 

 which will ultimately afford the material for 

 a satisfactory division of North America into 

 natural faunal areas. It is quite possible, how- 

 ever, that it will be difficult to establish a sys- 

 tem that will include both fresh-water and ter- 



e C. W. Johnson, Nautilus, XX., p. 73, 1906. 



