AN ARRANGEMENT OF THE GEOPHILID^, A FAMILY OF 



CHILOPODA. 



By O. F. Cook. 



That the genera included in this family present structural char- 

 acters of great diversity has been known since the publication of Mei- 

 nert's investigations. That author attempted no subdivision of the 

 family into groups higher tlian genera, a course to be explained by the 

 fact that the number of genera recognized by him was very small, and 

 by the further consideration that some of the more important structures 

 were misunderstood. Thus the labrum of Orya is given as " bipartitum,''^ 

 while m reality it is entire, the bipartite appearance resulting from the 

 fact that the part in question is arched when in place, and usually 

 becomes wrinkled in the middle when depressed by a cover glass. The 

 labrum of Orphmvus is said by Meinert to be free; in reality it is com- 

 pletely coalesced and closely homologous to that of Orya. The labrum 

 of the primitive Chilopoda was, in all probability, tripartite, and the 

 coalescence of the parts with each other and with the frontal lamina 

 are to be-viewed as deviations from the ancestral form. Relationships 

 can not, however, be inferred merely from such a fact as coalescence; 

 Orya and SchendyJa have the labrum entire and completely coalesced, 

 and yet represent two very distinct lines of development. 



The present method of describing the mandibles has been another 

 source of confusion. As in other Chilopoda the mandibles of Geophi- 

 lidse may be supposed to have had originally both i^ectinate and dentate 

 lamellae. The comijound pectinate lamellie of DicelIophih<s,^ Orya and 

 Himantarium are evidently the homologues of the laciniate processes 

 of the mandibles of Scolopendrid;v and Lithobiidtv, while the mandibles 

 of such genera as Geophilus and Schendyla have developed differently, 

 the laciniate processes being now represented by a row of simple 

 spines. Thus one of the simple spines of Geophilus is to be looked upon 

 as homologue of a whole <' pectinate lamella" in Himantarium, and the 

 mandibles of the two genera are structurally much wider apart than 



'A new genus partially equivalent to Mecislocepltalus of Meinert and recent 

 authors, but not of Newport. According to Meinert, the mandibles of Mecistocepha- 

 lus have only dentate lamellie, but the reason for this view is not apparent. 



ProceedinL's of the United States National Museum, Vol. XVIII— No. 1039. 



63 



