94 Annals Entomological Society of America [Vol. VI, 



Such variation as this is not uncommon in the Mem- 

 bracidse. For this reason it would seem that taxonomic 

 characters based on the shape, size and number of cells should 

 not be attached with the greatest importance unless it can be 

 clearly proven that these irregularities do not occur in the 

 forms in question. 



It may be mentioned in passing, that Thelia bimaculata 

 shows also a great variation in the length of the pronotal horn. 

 An attempt has been made to compare this variation with that 

 of the wing but the results were negative, and the variation of 

 the two structures seems to be entirely independent. 



THE HIND WING 



As has been stated (p. 81) the hind wing in the Membracidae 

 has more nearly kept pace with the fore wing in specialization 

 than is usually the case in Hemiptera. An interpretation, 

 therefore, of the fore wing leaves little to be determined so far 

 as homologies are concerned and in fact the venation, after the 

 tracheation has been worked out, is almost self-evident. In 

 the hind wing the reduction has gone further than in the fore 

 wing as is shown in the nymphal tracheation (Fig. 25). The 

 tracheae for costa and subcosta have disappeared. The wing, 

 however, shows a thickening or ridge along the cephalic margin 

 which is probably due to subcosta or perhaps in some cases to 

 costa-plus-subcosta, although the preceding tracheae are not 

 distinguishable and the vein itself not prominent. Radius 

 behaves much as in the fore wing. The trachea is apparently 

 two-branched but high magnification shows that Ri is present, 

 running close to R2+3 (Fig. 26). Occasionally also, Ri upon 

 reaching the margin of the wing turns backward to meet the 

 costal thickening so that in some cases the point at which 

 costa-plus-subcosta unites with R2+3 represents what remains 

 of Ri (Fig. 27). The most important point of difference 

 between the two wings is found in the cross vein r-m connecting 

 R4+5 with Mi+2. This difference has been noted by Redten- 

 bacher in his explanation of the hind wing of Centrotiis cornutus^'" 

 the only Membracid which he figures and which happens • to 

 show this character. In the fore wing these longitudinal 

 veins (R4+5 and M1+2) usually anastomose. In the hind wing 



35. Redtenbacher, Josef. Vergleichende Studien uber das Flugelgeader der 

 Insecten, Ann. k. k. Naturh. Hofmus. I, 1886, p. 187. 



