150 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.40. 



forests are of relatively recent growth. Such indications can be sup- 

 ported by evidence drawn from other groups of biological facts. 1 



DESCRIPTIONS OF NEW GENERA AND SPECIES. 



The two new millipeds that are to be described from south Texas 

 belong to the order Anocheta, but are only remotely related to the 

 other*members of this group previously known from more northern 

 and eastern localities as species of the old cosmopolitan genus Spiro- 

 holus. Texas is not lacking in native species of Spiroholus or Arcto- 

 bolus, as the dominant eastern genus of this group has now been 

 named, but these belong to the temperate fauna and are not to be 

 considered in this paper. 



With a single exception of the genus OnycMus, described a few 

 years ago from the Colorado desert, the relatives of the new types are 

 found only in Mexico and Central America, whence three additional 

 new genera are recognized. Study of some further material of the 

 previously established genus OnycMus has also resulted in the 

 description of three more species from Arizona. 



ANALYTICAL KEY TO GENERA TREATED IN THIS PAPER. 



First segment not emarginate behind the antennae Genus Anelus, new, p. 160. 



First segment broadly emarginate on each side, behind the antennae 



Anal valves with prominent compressed, elevated margins. 



Genus Glosselus, new, p. 163. 



Anal valves with depressed beveled margins, meeting in a groove 



First segment not margined with raised ridges.. Genus Cyclothyrophorus Pocock, p. 160. 



First segment with a distinct raised ridge bordering the lateral corners and 



emarginations 



Striations of posterior segments few, produced obliquely upward into sharp 



gpineg Genus Centrelus, new, p. 154. 



Striations numerous, not produced into spines 



Coxge of third legs of male with a long, retrorse process Genus Eurelus, new, p. 151. 



Coxa? of third legs of male unarmed Genus Onychelus Cook, p. 157. 



The last four genera, characterized by the depressed margins of the 

 anal valves, undoubtedly constitute a natural group, but the affinities 

 of Anelus and Glosselus are still uncertain. They are not closely 

 related to the other American genera, unless it be with the widespread 

 tropical genus Trigoniulus Pocock. Nor can it be insisted that Anelus 

 and Glosselus have any very close affinity with each other, though 

 Glosselus seemed to be the nearest relative of Anelus available for com- 

 parison at the time the description was drawn. 



> Vegetation Affected by Agriculture in Central America, Bulletin 145, Bureau of Plant Industry, 

 U.S. Department of Agriculture, 1909. 



