﻿6 



Indiana University Studies 



A few more of the species collected by Schott were later de- 

 scribed by Eigenmann and Ogle (Proc. U.S. Nat. Mus., XXXIII, 

 1907, pp. 1-36). They are Prochilodus ~beam=magclalenae, 

 Astyanax orthodus; A. atratoensis; Gharax atratoensis. 



In 1913 I collected at Boca de Kaspadura near the divide 

 between the Atrato and San Jnan, at Manigru, Boca de Certegui, 

 Qnibdo, and Rio Sucio. 



During the Landon-Fisher Expedition of Indiana University 

 Mr. Charles Wilson collected along the same general route and 

 in the Truando, a western tributary of the lower Atrato. 



The species in the lowest course of the Atrato and its species 

 of Astroblepus and Pygidium living in the highlands have not 

 been collected. These will in part be identical with Magdalena 

 species. In spite of the deficiencies, 45 per cent of the species 

 of the Atrato are known to occur in the Magdalena. Eight more 

 species (about 10 per cent) have parallels in the Magdalena. 

 Ninety per cent of its genera are also found in the Magdalena. 

 The affinity or origin of these 90 per cent is certainly Magdelenan. 



Four more genera, Bunocephalus, Hemiancistrus, Ancistrus, 

 and Piabucina are found east of the Cordillera of Bogota and 

 will probably be found in the Magdalena. 



The genera not represented in the Magdalena are : 



1. Pristis, a marine genus. 



2. Lebiasina. otherwise found only on the Pacific slope. 

 (Probably immigrants from the San Juan and the south.) 



3. Phanagoniates, autochthonus or from the Tuyra. 



4. Xematobrycon, confined to the Atrato and the San Juan. 



5. Parastremma, Atrato. San Juan, and Patia. 



6. Pterobrycon, autochthonus. 



7. Microbrycon, probably the female of the preceding. 



8. Neoheterandria, Atrato. 



9. Thalassophryne, marine. 



It appears that either the Atrato and Magdalena received the 

 ancestors of their fishes from the same source or the one derived 

 its fauna from the other. 



The degree of affinity of the Atrato fauna to that of the Mag- 

 dalena is about the same as that of the Paraguay to that of the 

 Amazon. As far as known the per cent of Atrato species found 

 in the Magdalena is really less, but the extreme lowland fauna 

 and the extreme highland fauna of the Atrato will most probably 

 bring the per cent of identical species into the neighborhood of 

 50, if not to a higher per cent. 



