374 NEOTROPICAL PSELAPHIDAE 



subspecies on the average to one species, this gives us 24,300,000 species popu- 

 lations. In our speculative argument this first step may be characterized as 

 the Subspeciation Factor. (PI. XXI). 



2. It is obvious that the number of species is subject to great variation 

 over such a large area. Area always covered by water (lakes, river systems, 

 water of swamps) would be wholly uninhabited. Grasslands have a much lower 

 yield in species. The periodically inundated forests (Haviland, 1926) and 

 swamps would have a lower yield than moist higher lying forests. Palm 

 swamps where the floor is thigh-deep mud. Mangrove areas, and all coastal 

 zones periodically invaded by tides, and brackish water communities, would 

 have very few species. Dissected rocky terrain, or areas with little or no 

 humus, would be almost barren of species. Montane zones above the tree- 

 line would be similarly almost barren, and the impoverishment above the 

 rain forest proper would be notable and progressive. 



If this is so then the fixed subspeciation rate may not be employed as 

 the process does not take into account the distinctive communities involved. 

 Unfortunately exact information is lacking for most pselaphid records, for 

 example, exact altitude in feet, whether the collection was from the plains of 

 southern Brazil or the brazilian rain forest. Grassland or pampas of Argentina, 

 plains of Paraguay or Uruguay, these and their many gradations may be 

 reflected in the sum total of subspecies selection pressure and, without in- 

 formation, a purely artificial but logical reduction must be made. 



Let us assume one-tenth of the area has maximum density, that is one- 

 tenth the area has one-tenth of 24,300,000 or 2,430,000; one-tenth of the area 

 has nine-tenths maximum density, and so on: 



One tenth 2,430,000 



One tenth 2,187,000 



One tenth 1,944,000 



One tenth 1,701,000 



One tenth 1,458,000 



One tenth 1,215,000 



One tenth 972,000 



One tenth 729,000 



One tenth 486,000 



One tenth 243,000 



13,365.000 



This revised total does not take into account the probability that some 

 portions could not support any pselaphids. This should allow for at least some 

 reduction and at a conservative figure would reduce the total to 13,000,000 

 species. This second step may be characterized as the Community Factor. 

 (PI. XXI). 



3. Species are populations of individuals and these latter must eat. Since, 

 with few exceptions, predaceous species feed upon animals smaller than them- 

 selves, they are ecologically fixed by their size within one of the horizontal 

 strata of the Eltonian pyramid (Elton, 1927; Park, Allee and Shelford, 1939; 



