8 NEW YORK STATE MUSEUM 



This grouping separates p o p u 1 n e a and concolor from 

 the closely allied mutica and h o r n i i , and brings together some 

 that are not so closely allied, but it shows fairly well the degree of 

 specialization of the species, as will be shown later. 



If our species only are considered, a more natural grouping can 

 be made on antennal characters, the species having anhulated anten- 

 nae being more closely related to each other than to those with unicol- 

 orous ones ; but, when the foreign species are taken into account, 

 this character loses its value, as we find S. perforata and some 

 of the species that have been referred to Eutatrapha and Thyestes 

 with the antennae annulated, though the other characters would lead 

 us to look for unicolorous ones, they being very close to our t r i - 

 d e n t a t a and but distantly related to the species this character 

 would place them with. Using the antennae, the species would 

 divide as follows. 



Antennae annulated Antennae unicolorous 



obliqua Candida 



mutica calcarata 



hornii tridentata 



populnea cretata 



moesta discoidea 



tulari vestita 



concolor imitans 



fayi 



lateralis 



puncticollis 



The foreign species, carcharias, scalaris, perforata, 

 similis, quercus, Eut. varicornis, io-p u n c t a t a 

 and T. pubescens would come in the first group ; the others 

 in the second group. 



The development of the head usually follows that of the process on 

 the claws ; but there are exceptions to this rule, as is seen in the case 

 of scalaris of Europe, where the flat front of the head would 

 place it near calcarata instead of with populnea, where 

 it more naturally belongs. In fact, whatever character is taken, it 



