SUBTERRANEAN AMPHIPOD STYGONECTES 167 



As presently delineated, the range of S. ozarkensis does not pose a 

 serious problem of interpretation. This species is apparently re- 

 stricted to ground-water habitats (primarily caves) developed in the 

 nearly flatbedded Mississippian limestones of the Boone formation 

 (= Osagean series of southwestern Missouri), which slopes gently to 

 the west and houses extensive underground drainage systems. The 

 range of S. ozarkensis is transected, however, by the north-south 

 drainage divide which separates the Arkansas River basin on the west 

 from the White River basin on the east. Similarity in the phenotypes 

 of animals in populations that occvipy habitats on opposite sides of 

 this divide strongly suggests that dispersal from one drainage basin 

 to the other can occur through subterranean routes. This suggestion 

 is further supported by distributional data on four other species of 

 of aquatic cavernicoles, which also inhabit the Ozark region and whose 

 ranges cross this divide. Distributional data on two species of cave 

 fish, TypIiHchthys suhterraneus and Amblyopsis rosae, have been given 

 by Woods and Inger (1957); data on the distribution of the Grotto 

 Salamander, Typhlotriton spelaeus, have recentlj' been summarized by 

 Brandon (1965); and information on the distribution of a troglobitic 

 craA\'fish, Cambarus sefosus, have been presented by Bedinger and 

 Hobbs (1965). All four of these species, except possibly T. spelaeus, 

 are known only from subterranean waters, and their movements are 

 assumed to be restricted to caves and solution channels. 



The ciu-ious, archlike range of S. clantoni would appear to represent 

 a greatly delimited corridor of dispersal, the exact nature of which is 

 not fully understood. Data on morphological variation have been 

 compiled on several populations of S. clantoni (see table 6 and 

 p. 105), and while these data are admittedly inconclusive because of 

 few and small samples, they still tend to indicate patterns of in- 

 complete isolation within the range of this species. 



Migration of preciu-sor clantoni into eastern Kansas by way of the 

 Arkansas and Neosha Rivers Avould partially account for the present 

 distribution of this species, but it does not necessarily explain the 

 range extension eastward into the northwestern Ozarks (Salem 

 Plateau). The present range of /S. clantoni is dissected by a number 

 of major rivers, and from east to west, this range is traversed at near 

 right angles by several major stratigraphic changes. Most of the 

 topography of this area, however, is gentle, and as Fenneman (1938) 

 has pointed out, local relief in much of eastern Kansas is generally less 

 than 250 feet. If subterranean dispersal is still possible along the 

 range of this species, it must be restricted to shallow ground waters 

 situated near to the surface. But to what extent shallow ground 

 waters occur in this area is unknown to me, and there appears to be 

 very little detailed literature on the subject. Although more than 



