1902.] AND ANCIENT GEOGRAPHY. 339 



occurred at this time. The African stock of the Potamonince, 

 remains practically unchanged, the Madagassian forms alone 

 becoming separated from it; the Indian stock spreads over the 

 Malaysian islands to North Australia and Japan, 1 and further, 

 sends out a westward branch over western Asia, reaching Southern 

 Ezirope and Northern Africa. 



D. Distribution of ^Eglea and the Trichodactylin^e. 

 (Compare Fig. 2, p. 296, and Fig. 4, p. 311.) 



1. The remarkable resemblance of the range of AZglea to that of 

 Parastacus suggests identity of origin. This would mean that 

 sEglea, in the beginning of the Tertiary, inhabited Chili, and 

 migrated, at this time, into northern Argentina and southern Brazil. 

 Since there are no closer relations to this peculiar genus in any 

 other part of the world, sEglea apparently is indigenous to Chili, 

 i.e., to the northern extremity of the American part of Archinolis, 

 and subsequently extended only into the southern part of Archiplata. 

 Of course, the opposite direction of migration also is possible. 



2. As we have seen above (p. 312) the distribution of the Tricho- 

 dactylinoz offers no remarkable feature. It belongs to the Atlantic 

 slope of present South America (the Neotropical region of Wallace) 

 and seems to have formed under the recent conditions. Possibly, 

 this subfamily is a new addition to the freshwater fauna, and immi- 

 grated from the marine littoral of the Atlantic. Further investiga- 

 tion of this question, together with a closer study of the morpho- 

 logy and systematic relations of this group are very desirable. 2 



PART III. CLIMATIC AND BIOCOENOTIC-- BARRIERS TO THE 

 DISTRIBUTION OF CRAYFISHES AND CRABS. 



In the foregoing discussions we have repeatedly called attention 

 to some distributional facts which we were unable to explain. For 



1 This extension began possibly as early as the Upper Cretaceous and Lower 

 Tertiary. 



2 Everything here depends on the systematic position and affinity of this group. 

 If it should be a primitive one, and really belong to the PotamonidcE, it is 

 possible that it reached Brazil in Lower Cretaceous times, when it formed part of 

 Archhelenis. Its isolation in Upper Oetaceous Archiplata, which was not fully 

 destroyed when it became a part of Lower Tertiary Neonotis, would explain its 

 isolated morphological position. In the Tertiary this subfamily would then have 

 extended its range northward. 



3 As to the term " Biocoenotic barrier," compare Ortmann, Grundzuege der 

 marinen Thiergeographie, 1896, pp. 41 and 70. 



