AFEICAN DIPLOPODA OF THE FAMILY GOMPHODESMIDyE. 



By O. F. Cook, 



Custodian of Myriapoda. 



The GomphodesiTiida' constitute a clearly defined and homogeneous 

 group. They are apparently the dominant Meroeheta of tropical East 

 Africa, with reference both to individuals and species, the Oxydesmidie 

 being their only rivals. The Gomphodesmida^ are clearly a more si)ecial- 

 i/ed group than the Oxydesmida' and, indeed, present several charac- 

 ters unicpie in the order. Thtj first six legs of males, for instance, are 

 provided at the end of the last joint with a fleshy pad or sole, present 

 in all known species. In several genera the number of olfaetory cones 

 of the last antennal joint is 10, while 4 is the normal and constant num- 

 ber for all other known Merocheta. In some forms segment 15 bears 

 iu males a subtriaugular process from between the anterior pair of 

 legs; such a modification of the sternum of a single segnu'nt of the 

 posterior part of the body is elsewhere unknown in the l)ii)lopo(la. 

 Several other ecjually peculiar but not entirely unparalleled secondary 

 sexual characters are described under the various genera. 



In habit the Gomphodesmida' are also strikingly distinct from all 

 African Merocheta, the salient features being a robust and compact 

 body, a strongly convex, unsculptiiriMl dorsum, lateral carina* with 

 prominent, thickened, even, and entire lateial margins, the posterior 

 segments greatly shortened, especially segment 18, with the last seg- 

 ment short, broadly triangular, narrowly truncatt^ at apex and without 

 prominent setiferous tubercles. The presence of pores on segments 11 

 and 14 will also serve as a means of family diagnosis, with the single 

 exception of the genus Marptodesmus. 



An agreement in pore formula has led previous writers to refer the 

 members of this family to the genus Eurydesmus Saussure,' a very 

 imi)erfectly known genus supposed to come from South Amerit;a. 

 Probably related Brazilian forms belong to a series which has been 



' Essai Xlyr, Mex.— Mem. Soc. Phys. et Hist. Nat. Geneve, 1860, XV, p. 77. The 

 type is E. angulaiue Saussure, idem, p. 78, supposed to be from Brazil. 



Proceedings U. S." National Museum, Vol. XXI— No. 1 1 70. 



(>77 



