148 U.S. NATIONAL MUSEUM BULLETIN 259 



origin for ;S'. indentatus. Even a third possible explanation is to sup- 

 pose a pre-Pleistocene invasion of fresh water by ancestral popula- 

 tions, followed by a short inland migration; then, after the last 

 interglacial period, a re-invasion of marginal coastal areas from the 

 non-inundated region to the west. 



In any event, it appears unlikely that S. indentatus is far removed 

 from a brackish water ancestry, although substantiating evidence 

 from salinity tolerance tests would be desirable and perhaps illum- 

 inating. It might be pointed out that experimental evidence from 

 such tests have proven invaluable in supporting earlier theories that 

 postulated a recent marine ancestry for European, subterranean 

 amphipods of the genus Niphargus (Jersche, 1963; Vandel, 1964). 



In conclusion, it is suggested that populations that invaded coastal 

 areas to the north and west of the present Chesapeake Bay and 

 subsequently gave rise to S. pizzinii might have been representatives 

 of the same widespread brackish water stock which later invaded 

 coastal areas along the lower margins of the Chesapeake Bay and 

 gave rise to S. indentatus. This suggestion seems feasible because, 

 first, it accounts for the morphological similarities of these two species, 

 and, second, it accounts for then- present distribution. 



Variation in the mean size of sexually mature animals from five 

 different populations of S. pizzinii is shown in table 4 (see p. 50). 

 Comparison of means by t-tests indicate that differences between 

 some of these means are highly significant (^<^.001). Initial in- 

 spection of these data might indicate geographic variation, but closer 

 examination tends to rule out this factor. The size mean of amphi- 

 pods in the population from the spring in Chester Co., Pa., is close to 

 the size mean of amphipods from Wetzels Spring in Washinton, D.C. 

 These two populations occur at opposite ends of the species range 

 and are further separated by the Susquehanna River. In comparison, 

 the size of the single female collected from a well on Mineshoe Island 

 in the Potomac River near Washington, D.C, is close to the size 

 mean of animals in the population from Refton Cave in Lancaster 

 Co., Pa. Refton Cave, however, is rouglily 75 miles from Mineshoe 

 Island and on the opposite side of the Susquehanna River. On the 

 other hand, this cave is less than 50 miles from the spring in Chester 

 County and is situated on the same side of the river. 



Both Pequea Silver Mine and Gables Woods are located near 

 Refton Cave in Lancaster County. The size mean of amphipods 

 from the silver mine population does not differ significantly from that 

 of the Refton Cave population, whereas the size mean of amphipods 

 in the population from Gables Woods (seep from a limestone outcrop) 

 is intermediate between those of the populations in Refton Cave and 

 Wetzels Spring. 



