CYDNIDAE OF THE WESTERN HEMISPHERE — FROESCHNER 379 



The separation of the genera m each of these two groups must be 

 based on an entirely different set of characters. In Group A the 

 modifications of the terminal lobe of the peritrcme furnish abundant 

 generic separations. One section of the group exhibits a short, 

 expanded lobe of various shapes and textures (figs. 90, 95-100); a 

 second section has the terminal modification markedly transversely 

 elongate and with or without a recurved apical part (figs. 89, 91-94). 

 In each section there appear some additional modifications that aid 

 in further separation of the genera, so that only very few of the 

 features shown by other body parts are required for delimiting the 

 genera within Group A. 



Group B is characterized by the lack of terminal modifications of 

 the peritreme (figs. 102-112); therefore, characters derived from other 

 parts of the body must be used for separating the included genera. 

 Several usable and apparently significant features may be used to 

 separate most of these genera, but at the end of the series there 

 accumulates a very heterogeneous mass of species for which no satis- 

 factory separation was found. An admittedly very weak feature is 

 presented to separate this unwieldy mass into two groups for which 

 generic names are already available. These two genera, Dallasiellus 

 Berg and Tominotus Mulsant and Rey, each hiclude species that 

 appear to be closer to certain of those in the other genus than to some 

 of the more remote members of the same genus, and this condition 

 led the author to hunt for additional breaks, but a more satisfactory 

 one was not found. 



The author has been deliberately conservative in accepting genera 

 in both groups and believes that numerous genera of but one or a 

 few species emphasize the dilTerence between species rather than their 

 relationships. Consequently, in this paper, genera are defined by 

 groups of characters possessed in common rather than by single 

 differences. The results of such an approach may be very unsatis- 

 factory to those who hold the opposite view, so an effort was made to 

 compromise the two viewpoints by retaining some of the lesser differ- 

 ences to establish subgenera. Thus the relationships as well as the 

 more conspicuous structural modifications may be recognized. 



The following tabulation of the Cydninae occurring m the Western 

 Hemisphere indicates the author's current conclusions. 



la. Anterior part of peritreme terminated by a differentiated lobe, loop, or 

 band (figs. 89-101); osteolar opening usually visible ventrally at base of 

 terminal process. 

 2a. Terminal process of peritreme elongate, transverse length more than 

 three times width (figs. 91-94). 

 3a. Terminal process fused with cuticula, forming a flat, polished band 

 extending almost or quite to lateral margin of evaporatorium and 

 separated therefrom by a distinct, impressed line (figs. 92-94). 



Rhytidoporus 



