INTIIUDITTION. 7 



tuiiied are l)ut tutir, it' we except a coiipU- ut hcctles, Otidrlu/niliilr.sfossilis, 

 toiuid ut Fossil, Wyoming, and lli/lastr.s sqmdhlciis, iVoni the I'leistocene 

 beds of Scarboro, ( )ntario. These four locaHties are Florissant in central 

 Colorado, the crest of the Roan nioiiiitains near the head of East Salt creek 

 in western Colorado, the huttes bordering the AVhite ri\cr near the Colorado- 

 Utah l)iiundai-v, and (ireen river city, Wyoming. All of these localities, 

 except the Roan mountains, were described in more or less* detail in my 

 Tertiary Insects of North America. The Roan mountain beds are appar- 

 ently merely an extension of those found on the White river, 50 miles dis. 

 tant, but here contined to the very crest of the range. Fossil insects are 

 found at several points, but only in one spot have they been obtained in 

 any remarkable number; here, however, in extreme abinidanee. As this 

 spot was T) miles di.stant from our camp and (lui-tiiiie and supplies were lim- 

 ited, no great number of specimens were brought away, but enough was 

 seen to warrant; the belief that a prodigious nundjer of specimens might be 

 obtained there. 



The detailed study of the fossil Rhvncdiophora has made very clear and 

 specific one point which im])ressed me in general while working in the 

 tield, and that is the wide diHerence between ihe character of the fossils 

 obtained at Florissant and those obtained at anv of the other localities 

 (perhaps excepting Flko, Nevada, of which little is known) in tiie Rockv 

 mountain region. The Hymenoptera which al)ound at Florissant almost 

 disa()i)ear in the other localities, while the Coleoptera, which hold a third 

 place at Florissant, form the larger proportion of the mass in the other 

 deposits. To test the opinion formed l)y the cursory examination of speci- 

 mens in the field, I have counted the specimens obtained in each of the 

 ditl'erent localities visited during a single sunmier, and find the opinion 

 amplv confirmed. 



The first .set of columns in the accompanying table shows the total 

 number of specimens (regardle.ss of s])ecie.s) obtained during this season's 

 work, separated by onlers, (!) in all localities; (2) at Flori.ssant alone; and 

 (3) in the other localities, ex(duding Flori.ssant; and the second set of 

 colunms the same figures reiluced to percentages. Nothing couM well be 

 more striking than the contrasts in the 1 1 xinenoptera and Coleoptera. 



