64 GEO. H. HORN, M. D. 



characteristic acumen. Later (Proc. Acad. 1873, p. 56) Mr. Crotch, 

 in a few words, suppresses many of the species as mere varieties 

 without giving any definite reasons for such action. The results of 

 the present study confirm the views of Dr. LeConte, basing the sepa- 

 ration of the species on structural characters not observed by either 

 of the preceding authors. 



Before giving the table of the species it is well to observe that, 

 contrary to the idea expressed by Crotch, the species are far less va- 

 riable in their markings than might have been expected. The ely- 

 tral sculpture is also remarkably constant, and forms two fairly-well 

 defined types, the first illustrated by tomentom and luteocincta, in 

 which the punctuation is so fine and dense as to appear alniost as an 

 opacity ; the second form by virgata and flavolimhata has the elytra 

 distinctly punctate and slightly asperate, the punctures closely placed, 

 but evidently distinct. 



In the following table the species are divided into two series, the 

 first without any trace of metallic surface lustre on any portion of 

 the body, while in the second there is more or less such lustre. There 

 is no trouble in applying this character, except in some forms of 

 luteocincta, in which the elytra are nearly black, but an examination 

 of the plaga of the occiput, or the thoracic spots, shows the surface 

 lustre very plainly. 



The occipital spot is of far greater constancy and importance than 

 would have been expected of so trivial a character. In about one- 

 half the species the occiput has merely a small spot in front of the 

 edge of the thorax, while in the other species a large transverse space 

 is piceous, extending at times nearly to the eyes. It is true that in 

 nearly all the species, when the head is unduly extended, a piceous 

 band is visible across the occiput, but the difl^erence between the two 

 sets of species can be easily appreciated by an examination of such 

 well-known forms as canadensis and flavolimhata. 



The following table has been prepared as an aid to the separation 

 of the species more fully described in the subsequent pages ; a cabinet 

 arrangement is suggested by the number preceding each species. 



Surface of body without any trace of metallic lustre in the markings, these 



being piceous or brownish 2. 



Surface with metallic lustre in the markings, if not on the elytra at least on the 



head and thorax 6. 



2. — Thorax more than twice as wide as long; epipleurse more or less piceous. 



2. brevicollis. 

 Thorax not twice as wide as long; epi pi eurse. always pale 3. 



