18 



PROCEEDINGS OF THE NATIONAL MUSEUM 



these species seem to have had a common ancestor; in fact, it is en- 

 tirely probable that dadylifera is little more than an extreme develop- 

 ment of the trend in the ^e/a-group toward increase in the size of 

 process B and reduction in the length and arch of the telopodite blade. 

 It has also largel}^ lost the small but acute coxal spines found in pela. 

 The subspecies of pela constitute a distinct Rassenkreis pattern, 



Figure 4. — Distribution of the subspecies of Dixioria pela in western North Carolina, 

 northeastern Tennessee, and southwestern Virginia. The area is included by the head- 

 waters of the Holston River on the upper left, of the New River on the lower right, and 

 the Nolichucky River on the lower left. Dixioria pela pela is represented by solid circles, 

 D. p. acuminata by a solid inverted triangle, D. p. brooksi by a solid upright triangle, 

 D. p. fowleri by open squares, D. p. coronata by solid squares, and D. p. wrighti by open 

 circles. Intermediate samples are indicated by an X, and are discussed in the text under 

 one form or the other of the two involved. 



forming nearly a closed circle, and involving modification from a 

 simple gonopod to a more complicated one (or vice versa) in three 

 major succeeding forms. That systematically different forms rather 

 than a single geographically variable one are involved is borne out by 

 the fact that each of the groups is homogeneous ^vithin itself, even 

 though its range may extend almost a hundred miles. Intergradation 



