352 ■ PROCEEDINGS OF THE NATIONAL MUSEUM. voi,. xxxii. 



latter name, however, is more often applied to members of the steno- 

 pelmatid genus Stenopelmatus. 



There has been recognized three distinct species of A/iahrus, ranging 

 from the plains east of the Rocky Mountains to the Pacific, and some 

 orthopterists believe there are really many more species. In my study 

 of the genus I have examined several hmidred specimens from localities 

 ranging from Kansas to California, north to Washington and British 

 Columbia, and from altitudes varying from the plains of Kansas to the 

 treeless heights of mountain peaks. Much time has been spent in 

 search for stable characters for the separation of the various described 

 species. So great is the range of variation in both structure and colora- 

 tion and so apparentl}'^ distinct are the extremes that the multiplicity 

 of forms seemed at first certain, but such seems now scarcel}' probable 

 of proof. With but a few specimens from different localities before 

 him a worker will find the recognition of two or more forms easy, but 

 the examination of a long series usually merges the supposedly distinct 

 species into one. In 181)7 Scudder" separates the species apparently 

 to his satisfaction. There he considers the form occurring in Wash- 

 ington and recorded as purpiirascens to be not that species, but smiplex. 

 But two years later, in his Catalogue of the Orthoptera of the United 

 States, he includes Washington in the habitat of purpurascens, thus 

 showing a reversal of opinion. An examination of the Scudder col- 

 lection shows no satisfactory division of this genus into species, a fact 

 expressive of the difficulty of the task. In the course of my investi- 

 gation and studies I found some promising characters, but upon trial 

 their value as stable meiuis of separating the mass of material into 

 species proved unreliable. 



Professor Gillette suggested, and for a time strenuously maintained, 

 that the relative lengths of the posterior and intermediate tibite were 

 of specific value for the separation of simplex iixvdi purpia'ascens^ in 

 simplex the posterior tibia? being less than two times as long as the 

 intermediate ones, while in purpurascens the hind tibijB are twice or 

 more than twice as long as the middle ones. But when determined by 

 this chai'acter both simplex and purpurascens were found in nearly 

 every State and Territor}^ where either species is known to occur, and 

 material, obviously of one catch, yielded both species. The subgenital 

 lobes of the females also presented promising possibilities, but upon 

 trial were also found wanting. Their extending to or beyond the apex 

 of the subgenital plate seemed very suggestive of a differentiating 

 character, but the examination of a large number of specimens resulted 

 in the rejection of this as a specific character. Color is also useless, 

 as individuals of all shades are said to be found among living swarms. 

 Putnam says that scarcel}'^ an}^ two of the man}^ specimens seen in Mid- 

 dle Park, Colorado, were colored alike. Length and shape of ovipos- 



» Psyche, VIII, p. 95. 



