ATOPETHOLID MILLIPEDS — HOFFMAN AND ORCUTT 99 



Since 1940 a steady flow of publications has swollen the ranks of 

 the family. Most of these papers have been concerned solely with the 

 description of new species and genera, and need not be summarized 

 in detail. It may be added, however, that no consistency has been 

 obtained as regards the systematic position of many genera. In a 

 contribution appearing in 1941, Chamberlin (1941a) included the 

 genera Hiltonius and Messicobolus in the Atopetholidae although both 

 belong elsewhere. In the previous year, the new genus Arinolus had 

 been compared in the generic diagnosis only with Tyloholus, Hiltonius, 

 and Spirobolus, of v/hich all belong to the Spirobolidae. In 1943, he 

 (1943a) treated five genera as atopetholids — Hiltonius, Tarascolus, 

 Toltecolus, Messicobolus, and Aztecolv^—smd of these only the second 

 and third actually belong to the family. 



In 1949, a short paper by Dr. Chamberhn reviewing the genera of 

 both families restated the miain differences between the two and 

 corrected the previous erroneous allocations. Yet, even in this work, 

 with three new atopethohd generic names proposed, the established 

 genera Arinolus and Piedolus were completely omitted. 



The recently published "Checldist of the MilHpeds of North 

 America" (Chamberlin and Hoffman, 1958) lists all the known genera 

 and species occurring within the United States, but, being largely a 

 compilation from the literature, it perpetuates a number of errors, 

 which we discuss fully in the following systematic account. We 

 earnestly hope that with the pubhcation of the present work, the 

 misunderetood and abused Atopethohdae will at last become an 

 intelligible gi-oup upon which future systematic work can be based 

 with a considerable degree of confidence. This synopsis claims only 

 to be the rough foundation upon which a handsome systematic edifice 

 may someday be erected. 



Taxonomic Characters 



In preparing descriptions, we have devoted attention to aU of the 

 structural features of atopethohds hitherto utilized for diagnostic 

 purposes, as well as to numerous others that were entirely neglected 

 in the general preoccupation with those characters easiest to observe 

 and mention. Many characters in this last category have been found 

 worthless for taxonomic purposes. An attempt to study variation, 

 whenever adequate series permitted, has been moderately successful. 

 We have also determined that it is possible in many cases to identify 

 female specimens at least to genus, although our knowledge of the 

 family is still too meager to warrant the description of new species 

 from female specimens. When sufficient material has accumulated to 

 permit careful revision of genera and actual comparison of known 

 females is possible, it seems quite hkely that determinations can be 



