SUBTERRANEAN AMPHIPOD STYGONECTES 169 



physically isolated from adjacent valleys by ridges of insoluble 

 elastics such as shales, sandstones, and conglomerates. Ranges in 

 the Edwards Plateau are apparently restricted to isolated and rela- 

 tively deep systems of subterranean drainage, and extrinsic barriers 

 to species dispersal are in the form of extensive faults and stratigraphic 

 changes. 



5. With few exceptions, species ^vith the widest ranges are found in 

 the tenuis group, and in a few instances, ranges cover parts of several 

 contiguous physiographic provinces. The majority of species with 

 wide ranges are largely interstitial and noncavernicolous and are able 

 to disperse rather freely through ground waters that are situated 

 close to the surface. 



6. By and large, rivers do not appear to impose dispersal barriers or 

 to preclude gene flow between populations of the same species 

 situated on opposite sides. At least eight species display distributional 

 patterns, which are explained only if one allows for subfluvial dispersal. 

 In addition, and especially in limestone terrains, there is good pre- 

 sumptive evidence that in a few instances dispersal can take place 

 under siuface drainage divides by way of subterranean water courses 

 developed in different patterns than those on the sm'face. 



7. Several rather widely ranging species show morphological 

 variation, which, while in some cases geographically inconsistent, 

 implies that isolation of certain populations either is taking place at 

 present or has taken place in the recent past. 



8. The distribution of two species in coastal plain areas near 

 brackish water indicates that at least some members of Stygonedes 

 are probably still ecologically and physiologically close to assumed 

 brackish or marine ancestors and that colonization of fresh-water 

 habitats by some species might have taken place rather recently. 



9. A number of examples of sympatry (often more specifically, 

 syntopy) between two species of the same species group have been 

 noted and explained in terms of multiple invasion of the particular 

 range in question. In none of these situations is a hypothetical 

 model of sympatric speciation required to explain these ecological 

 associations. 



10. The possibility of a breakdown in ecological isolathig mech- 

 anisms leading to hybridization between two otherwise phenotypically 

 and presumably genetically distinct species has been observed in a 

 population sample from a spring in Maryland. 



11. The extreme variation noted between the mean size of sexually 

 mature individuals in different populations of S. inzzinii is attributed 

 to differences in the size of respective habitats rather than to differences 

 in respective genetic S3"stems. 



