404 AxxALS OF THE Carxegik Museu.m. 



since it enters the Indian Creek, but above the region where the 

 Youghiogheny cuts through Laurel Hill and Chestnut Ridge it is not 

 found any more. (2) This species is also found outside of the Ohio 

 drainage, namely: (a) in Erie county, in the lake-drainage in Con- 

 neaut Creek at Albion, and in Elk Creek at Miles Grove; (//) in the 

 Potomac drainage, in Wills Creek near Hyndman, Bedford county. 

 This latter locality, although not properly belonging to western Penn- 

 sylvania, as defined above, is m.entioned here, since it will give oc- 

 casion to a very interesting zoogeographical discussion, for which, 

 however, the investigations have not been finished. 



Remarks on the Geographical Distribution of the Craw- 

 fishes OF Western Pennsylvania. 



Only a few points shall be mentioned here, in order to call atten- 

 tion to some remarkable facts of the distribution, which will be dis- 

 cussed more fully in a subsequent paper. 



^Ve may lea\e alone C. bartoni, which is found everywhere in the 

 state, and its variety robustus, for which only slender material is at 

 hand. C. carolinus, motiongalensis and dio^^enes are restricted to the 

 southwestern section of the state. The distribution of C. carolinus 

 plainly indicates immigration from the south along the high plateau 

 and longitudinal valleys included between Chestnut Ridge and Alle- 

 gheny Mountains. In the valley between Chestnut Ridge and Laurel 

 Hill, it managed to reach the cross divide between Indian Creek and 

 the Loyalhanna River in Westmoreland county, without crossing it, 

 while in Sumerset county it has crossed the transversal divide between 

 Casselman River and Conemaugh system. C. inonongalensis very 

 likely is its parallel form on the low plateau (Cretaceous peneplain?) 

 northwest of the foot of the Chestnut Ridge : it has not yet been 

 traced beyond the limits of the state, but reports are at hand that it 

 is also found in West Virginia.' Northward, the Ohio-AUegheny- 

 Kiskiminetas River seems to form the boundary of its distribution '' 

 this is highly interesting, in so far as a large river marks in this case 

 a barrier to the further dispersal of an aquatic creature, but this is 

 easily understood, if we take account of the peculiar habit of this spe- 

 cies of living in springs on the hill-tops. Apparently, this species also 

 came from the south, and migrated on the hills parallel with the gen- 



' Sucli reports have been obtained from unprofessional people, and cannot be given, 

 unless verified. 



