5O ENTOMOLOGICAL NEWS. [Feb., '14 



sketches applied. This correspondence, which will be referred 

 to again, clearly indicates that in recent literature two or more 

 species of dragon flies have been confused under the name of 

 pallidus. 



I have no doubt of the existence in the genus Gomphus, as 

 generally used, of several subgroups, along the lines indicated 

 by de Selys and Professor Needham. As soon as possible 

 it will be convenient to use these subgroups as genera. But 

 before this can be done an exhaustive study of the approxi- 

 mately seventy species involved will be necessary. At pres- 

 ent no one can use these group names intelligently. These 

 groups have been defined by Professor Needham, so far as 

 imagoes go, in terms not used or emphasized by de Selys, 

 whose groups were based largely on thoracic pattern, though 

 the resultant groupings, in the two cases, have much in com- 

 mon. For example de Selys' Group 5 includes pallidus (and 

 villosipes), lividus, spicatus, ntinuius and exilis. Arigomphus, 

 as used by Needham, includes \pallidus, villosipes, spicatus and 

 other species not known to de Selys in 1858, the date of the 

 Monographic. Lividus and exilis are placed in another group 

 by Needham, who has not discussed minutus. 



As stated above the groups require accurate definition. So 

 far as de Selys goes, stpicatus and exills, at least, should not be 

 associated with pallidus; and in Needham's arrangement it is 

 certainly a mistake to separate exilis and spicatus, for example. 

 Arigomphus is defined (Aquatic Insects Adirondacks, p. 447-8) 

 as having two cells between the base of veins Ai and A2 at 

 their origin. Five males and one female of villosipes, selected 

 at random, all have a single cell. Three males of cornutus, 

 which is an Arigomphus, have two wings with one cell, and 

 four wings with two. This character is tabulated below for 

 the material discussed in this paper. To the shape of the apex 

 of abdominal segment 8 some importance may attach, but the 

 character is difficult of accurate definition (see Fig. 8, and ex- 

 planation). As to the hind femora in the two sexes, I have 

 examined thirty species of which I have both sexes, and the 

 femora are different in the sexes in all of them. In the males 



