LUTZ, LIST OF GREATER AXTILLEAX SPIDERS 127 



lected especially iu the vicinity of Baiios San A'incente. The difference 

 in elevation between tliis place and the plains is not enough to cause a 

 noticeable difference in temperature apart from that connected with soil 

 conditions and moisture. Figure 4 gives an idea of the character of 

 the region — -moist atmosphere^ lime-humus soil, rather ^"tropical'' vege" 

 tation. Leaving out of account the Coliors Araneus and Dendrij pliant es 

 for taxonomic reasons and not considering the minor ecological differ- 

 ences in each region, we found 10 genera on the ^^lains which we did not 

 find at BaxLOS, 24 at Bahos which we did not find on the plains and only 

 9 genera at both places. That is, 52.6 per cent, of the plains genera 

 were not found in the hills and 72.7 per cent, of the hills genera were 

 not found on the plains. In other words there is at least as much differ- 

 ence between these two nearby but ecologically different localities as 

 there is between Cuba and the other islands or between the Antilles as a 

 whole and the mainland. 



Furthermore, this must give us a pause in considering the relationships 

 of the various islands to each other and to the mainland, for 90 per cent, 

 of the genera we took on the plains are found in United States and only 

 77.3 per cent, of those from the hills; 40 per cent, of the former are 

 known elsewhere only in United States, as far as America is concerned, 

 and only 4.5 per cent, of the latter. The general impression is that the 

 West Indies are more closely related to Central and South America than 

 to United States and, going still further, that the Lesser Antilles are 

 especially related to South America and the Greater Antilles to Central 

 America. If this be true, in how far is it due to land bridges, ocean 

 currents or wind and in how far is it due merely to similarity, and hence 

 congeniality, of environment ? The first part of the question needs prior 

 consideration, and the second part cannot be satisfactorily answered until 

 we know more about animal ecology. 



Maixlaxd Affixities of the Axtilleax Fauxa 



As far as the data concerning the genera of spiders go, the Lesser 

 xintilles (St. Vincent) seems to have a closer aflSnity with South America 

 than does the other division of the Antilles, since 22.2 per cent, of the 

 genera of the Lesser Antilles are known elsewhere in this hemisphere 

 only in South America and the corresponding percentage for the Greater 

 Antilles is only 8.6. However, if we include all the genera whose dis- 

 tribution we can study we note that only 82.4 per cent, of the genera 

 kno-^Ti from the Lesser Antilles are known also in South America, while 

 the corresponding percentage for the Greater Antilles is 84.3 per cent. 

 The belief that work in the northern Lesser Antilles will still further 



