June-Oct., 1921 Bulletin of the Brooklyn Entomological Society 89 



reticulata (Burmeister) Horv. (cf. fig. la). The metanotum is 

 always concealed at middle, appearing on each side as a small 

 triangular area, which I propose to term the metanotal triangle. 

 These triangles are widely different in shape in the different 

 species, as the figures show, and thus they furnish taxonomic 

 characters of great value. In the nymph (fig. ic) the metanotum 

 is not concealed at middle. The segment immediately following, 

 which has often been treated as the metanotum, is in reality the 

 first dorsal segment of the abdomen, which thus has seven seg- 

 ments, exclusive of the genital segments. 



M. atrata is peculiar in having minute vestigial wings, and the 

 pronotum is correspondingly enlarged, approaching somewhat the 

 condition characteristic of the winged forms. It is produced pos- 

 teriorly, so that it covers not only both mesonotum and meta- 

 notum, but also the median portion of the first abdominal seg- 

 ment. 



The structure of the tarsi has likewise been variously inter- 

 preted. In all the American and exotic species which I have been 

 able to examine there is agreement in the number of tarsal seg- 

 ments in the adult stage; all have the front tarsi with one seg- 

 ment, the middle and hind with two, or, as it is briefly expressed, 

 the tarsal formula is i : 2 : 2. I do not think that the minute en- 

 largement or "node" which may be perceived (with difficulty) 

 at the base of the first tarsal segment can be considered as a true 

 segment, because this node appears to be nothing more than a 

 slight basal swelling of the metatarsus when the leg is cleared and 

 examined under very high magnification. When it is counted, 

 as by Champion in the " Biologia," the tarsal formula becomes 

 2:3:3, and the species remain uniform in this particular. The 

 subgenus Kirkaldya Torre-Bueno was founded on the supposition 

 that americana differed in tarsal formula from the other species, 

 and since there are no other characters to support it, it must be 

 definitely suppressed. Van Duzee in his Catalogue of 1917 main- 

 tains this subgenus, although de la Torre-Bueno, in his later paper 

 on the Veliinae of the Atlantic States (Bull. Brooklyn Ent. 

 See, XI: 52-61, 1916), makes no mention of it; however, in the 

 table of species there given the tarsi are still made to vary, and 



