236 PROCEEDINGS OF THE NATIONAL MUSEUM vol. 112 



portions of the two, they are separated by a considerable difference 

 in elevation. 



Aside from the variation in dorsal coloration, there is some local 

 differentiation in the pigmentation of other parts. Normally the legs 

 and underparts of Cherokia are yellow, but in the Smokies the pleural 

 regions and anal valves become nearly black; in the vicinity of High- 

 lands, North Carolina, the legs may be pink instead of yellow. In 

 xystodesmids generally, the legs assume the same color as the para- 

 nota and are brightest on the distal joints — -those which extend 

 beyond the paranota when the animal is walking, a coloration giving 

 the impression of much greater body width. 



Evaluation of characters: From the preceding accounts, one 

 can draw several inferences. To begin with , five characters have been 

 detected which seem to vary significantly with respect to the spatial 

 distribution of the genus. As a complicating factor, in only a few 

 cases do any of the variational phases coincide in any given region, 

 and at least one character — color pattern- — is subject to a great 

 amount of localized regional variation, sometimes three separate color 

 phases occurring together. What significance is to be assigned to 

 these structural variations in the definition of taxonomic categories? 



Of the five characters, one is represented by two quite different 

 forms, which are geographically exclusive in their distribution (fig. 2). 

 This character is the structure of the paranotal scapulorae. Most of 

 the specimens seen so far fall readily into one category or the other 

 and make a strong case for full specific value. There is, however, some 

 evidence suggesting that the geographic and morphologic gaps between 

 these two populations are more apparent than real, something to be 

 discussed under a subsequent heading (page 258). There is reason to 

 believe that intergradation between the two populations occurs in 

 extreme western Georgia, an area from which no material was avail- 

 able. Since other structural characters in the genus, such as male 

 gonopods, size, body form, and color pattern, transgress the boundary 

 between the paranotal types, the latter are considered as diagnostic 

 of a wide-ranging western subspecies of Cherokia georgiana, rather 

 than of a different species. 



Of the various names which have been based upon specimens of 

 Cherokia, all apply to the form characterized by submarginal scapu- 

 lorae. Of them, the oldest available name is Bollman's Fontaria 

 georgiana, proposed for specimens from Macon and Tallulah, Georgia. 

 To the best of my knowledge, there is no name available for the popu- 

 lation of Tennessee and Alabama with marginal scapulorae, and a new 

 one is herein proposed. 



The latter is quite uniform over its range with respect to body form, 

 but the distribution of the more eastern georgiana includes specimens 



