334 W. D. FUNKHOUSER 



(Plate XXXV, 10, 11). This would suggest that perhaps the piece is a 

 true meron, and indeed it might be considered as such in the cicada. In the 

 cicada this meron of the posterior leg shows a strong spine and the entire 

 piece is much enlarged. The coxa in the Membracidae also shows spines 

 or protuberances in many species (Plate xxxv, 5), and the question arises 

 as to whether these might be homologous with the meral spines of the 

 cicada; but this interpretation would hardly be reasonable because of 

 the fact that the spines are chiefly on the interior rather than the lateral 

 margins and are never set off by sutures. In fact these spines, or teeth, 

 are rather irregular in position and show much variation. In no case has 

 any evidence been found that they are indicative of separate sclerites. The 

 coxa has not been used for systematic work in the Membracidae, and it is 

 doubtful whether it is of value for this purpose. Such distinctive struc- 

 tures as may be present, as spines or elbows, are usually on the mesal 

 angles rather than on the ventral or the lateral angles, where they might 

 be easily identified, and are, moreover, not at all constant in the forms that 

 have been studied. 



The trochanter is normally an elbow-shaped segment attached to the 

 ventro-mesal extremity of the coxa (Plate xxxv, 12-15). The proximal 

 half projects directly ventrad, while the distal half turns ventro-mesad. 

 The segment is freely movable in the Membracidae, and the articula- 

 tion with the coxa is comparatively weak. The coxa-trochanter joint, 

 however, is often strengthened by the overlapping hooks or projections, 

 which tend to overcome this weakness (Plate xxxv, 12). No special modifi- 

 cations are found in the trochanters of the first or the second pair of legs, 

 but in those of the hind legs most interesting developments may be found. 

 The commonest variation is that of general shape. In most species the 

 segment is practically cylindrical, bent in the middle but nearly equal in 

 diameter at each end (Plate xxxv, 13). This shape graduates to a roughly 

 spatulate outline (Plate xxxv, 14), in which the proximal end is much 

 narrowed and nearly cylindrical while the distal end is broadly flattened 

 and paddle-like. In certain species of the subfamily Membracinae the 

 segment is shortened and nearly straight, the internal angle being hardly 

 recognizable and the articulatory surfaces almost in a line with each other 

 (Plate xxxv, 15), giving the entire segment a spindle-shaped outline. 



By far the most interesting modification of the trochanter, however, 

 and one that is extremely valuable for systematic purposes, is the develop- 



