16 U.S. NATIONAL MUSEUM BULLETIN 2 92 



Indeed, it is probable that instead of having one-half or less than one- 

 half the numbers of species reported for Dominica, both islands 

 actually have or have had richer faunas inasmuch as they offer a 

 greater variety of habitats. To what extent some of the habitats on 

 these French islands have been ravaged by man and perhaps the 

 mongoose is not known to us; nevertheless, such species as Aratus 

 pisonii, which is largely limited in its ecological distribution to man- 

 grove swamps, occiu-s on Guadeloupe but is absent on Dominica, 

 where no mangi'oves are present. Too, the larger area of low-lying 

 land on Guadeloupe has provided for a much more extensive estuarine 

 development than is possible on Dominica. 



In assessing the significance of the numbers of species reported 

 for the islands, one should take into account the fact that some of 

 the records cited herein are old ones based on collections made prior 

 to the deforestation associated with agricultural and other develop- 

 ments. Many of the dry arroyos that are so abundant on many of the 

 islands were, in years past, shaded ravines with permanent streams. 

 With water in them during only a few weeks or months of the year, 

 many of the freshwater decapods have disappeared, apparently 

 completely from some of the smaller islands. It is entirely possible, 

 if not highly probable, for example, that some of the species herein 

 reported to occur on Saint Croix and Saint Thomas, where fresh 

 water is at such a premium, no longer exist on those islands. 



We are aware that from a faunistic standpoint the decapods of at 

 least one of the geographic regions recognized (Mexico and Central 

 America) are probably composed of two elements, and it is regrettable 

 that the crustaceans of this area are too little kno^\^l to enable us to 

 distinguish between the northern and southern components of this 

 fauna. In order to avoid ambiguity resulting from our lack of knowl- 

 edge, we are considering the fauna of this region as a unit, designating 

 the area as the Central American-Mexican region, occasionally 

 referring to it as the Central American-Mexican continental mass. 



Except for Trinidad, the freshwater fauna is either so depauperate, 

 so little known, or pan-Antillean in nature on most of the islands off 

 the northern coast of South America that they seem to warrant no 

 special attention in this discussion, and they are referred to infre- 

 quently, some not at all. Trinidad, although occasionally included 

 herein as a member of the Lesser Antilles, should perhaps faunis- 

 ticaUy be considered more properly a part of South America. 



As might be expected, many (24) species present in the West Indies 

 also occur in coastal areas of all three neighbormg continental masses, 

 but the ranges of a majority of them are decidedly more restricted, 

 with 33 species and subspecies endemic to one or more of the islands. 



