66 ENTOMOLOGICAL NEWS [Mar., '22 



have been examined. This article is an explanation of the 

 accompanying plate. 



If the reader will refer to the accompanying plate (IV), he 

 will note that the genus falls into three levels of differentia- 

 tion as indicated by the heavy horizontal lines. These are : 



Level I. Scinifasciata, angclina and foliata. By comparison 

 with any of the other penes figured it is obvious that these are 

 alike in that none of their parts are as much exaggerated or 

 specialized as are one or more of the parts of any of the spe- 

 cies figured in the two higher levels. By this same compara- 

 tive standard, scniifasciata is more generalized, hence older, 

 than foliata, which has the cornua slightly specialized and 

 than angclina, which has the lateral lobes lengthened and broad- 

 ened. As will be shown later, 3 all three are probably pre-Mio- 

 cene species. 



Level II. This includes all the species lying between the 

 two horizontal lines on Plate IV. These are all American and 

 are species that probably date from the Miocene or later. Some 

 of these groups are apparently at the height of their develop- 

 ment. 



Level III. This level includes the Eurasian species, less the 

 primitive angclitia. These are the most specialized of the 

 genus and are the postglacial remnants of a Eurasian fauna 

 that probably reached its climax in preglacial times. 



The individual species and minor groups of the genus will 

 be discussed as follows : 



Group 1. Semifasciata Burm. PI. IV, fig. 1. Maine to Florida, 

 west to Michigan and Texas. A spring and early summer species 

 found in woods swamps in the deciduous and southern pine forests. 1 



The primitiveness of this species seems to be confirmed by 

 its isolated position morphologically, by its non-Libellula wing 

 pattern, which has basal markings and color that recall CcJi- 

 thcinis and Pcrlthcmis, by its less rugged build, which is very 

 different from the husky proportions of many of the more spe- 

 cialized Libellulas, by its spring and early summer season which 



3 In the April number of the NEWS. 



* As far as possible the writer has tried to correlate the distribution 

 of the species of Libcllnla with the plant formations of the eastern 

 United States as worked out by Transeau. See "Forest Centers of 

 Eastern America," Amer. Nat., xxxix, pp. 875-889, 1905. 



