of the Arctic unil thf Attlarcttc Fannaa. 319 



America through tlic researches of the 'Albatross,' but tliere 

 still remains too much to be worked uj) to allow us to sutjgest 

 reasons why a subsurface-water connexion between the 

 similar forms of north and south has not been discovered 

 on the western shores of tropical America. Possibly such a 

 connexion may some day be established for one or other 

 group of animals. If, personally, I doul)t this, it is for two 

 reasons. In the Panama province there is certainly no coral- 

 reef formation, but there is a surface-water fauna of tropical 

 character, and in times not very long gone by there was really 

 coral-formation. The reasons for a more or less marked 

 suppression of the subsurft\ce-watcr fauna by the tropical 

 surface-fauna may hold good in this case also to a greater or 

 less degree. 



Secondly, corresponding to the remarkably equable climate, 

 we find, on tlie west coast of America from the temperate 

 southern to the temperate northern zone, a fauna of nearly 

 homogeneous character, interrupted only in the narrow pro- 

 vince of Panama. In general character it may be described 

 as a cool-water fauna, but it has undergone quite remarkable 

 local differentiation. This fauna springs, apparently, from 

 the southern hemisphere ; and thus, probably in ancient days, 

 possibly before the time of the separation of the faunas, 

 this fauna of southern origin gradually conquered its present 

 region — that is to say, it crowded out more or less the members 

 of the universal fauna. 



Thirdly, Agassiz expressly mentions the poverty of the 

 depths examined by him on the * Albatross.' He accounts 

 for this by the fact that the currents on the tropical shores 

 of West America waft in a comparatively limited quantity of 

 pelagic organisms, which would afford food for the inhabitants 

 of the deep sea. As the animal world of the deep sea is 

 wholly, and that on the slopes of the continents partly, 

 dependent economically on the pelagic fauna, we have here a 

 reason of great imj)ortance, and one which would apply in 

 the main to West Africa also. In the same way, it must 

 be taken into account that the extreme narrowness of the 

 continental slope affords anywhere the opportunity for an 

 interruption of faunal continuity. 



I cannot omit to refer here to a fact which may be brought 



carried out by the great Africun rivers." Probably the bottom of the 

 comiuentjil .slope shows these river-depoaits in a still hi^rlier deirree, so 

 that we uiay have a similar state of atlairs as, for instance, i>n the east coast 

 of temperate ^outh America, where the mouths of tlie liio Negro and 

 Kio La Plata form barriers between the South Bramlinn and the Pata- 

 gouiau littoral and contiiieutal faunas. 



22* 



