THE CANADIAN ENTOMOLOGIST. 33 



having the outer submarginal partly divided, while in the second the right 

 wing shows a portion of the cross-nervure, which is totally wanting in 

 the left. A rudimentary, or incomplete, cross-nervure in the outer 

 submarginal cell occurs in several specimens, and in one $ the third sub- 

 marginal nervure is continued half-way across the cell below. Another 

 male has the third submarginal cell divided into two cells by a cross-nervure, 

 which nearly coincides with the second recurrent. The left wing of one 

 specimen has the third submarginal nervure forked at the anterior end, so 

 as to form a minute triangular areolet, which, on the opposite wing, is 

 almost square, and gives from the lower outer corner a branch partly 

 across the cell. The outer submarginal cell is also in one instance partly 

 divided longitudinally by a branch from the centre of the third submarginal 

 nervure. The consideration of variations such as these specimens afford 

 will indicate one of the difficulties which may attend the determination of 

 a species (especially in the case of single insects) from descriptions, and 

 the possibility of its being placed in a wrong genus and confounded with 

 some species resembling it in color and markings. In a species whose 

 wing-venation is evidently so unstable as that of the present insect, the 

 specimens with additional complete or rudimentary cells appear to be 

 reversions toward an earlier type, in which the wing-cells were more 

 numerous. Another point in regard to the wings of this species is that the 

 outer cells of the under wings of the male (in all my specimens) are closed, 

 as in several of our species of Strongylogaster. This fact is not mentioned 

 in the descriptions before quoted, and seems to me sufficient reason to 

 question the propriety of placing the species in Tenthredo, from the 

 members of which genus it also differs in general appearance, and to 

 suggest the advisability of including it for the present in Strongylogaster. 

 From the first tribe of this genus (as divided by Cresson) it seems to 

 differ chiefly in having the lanceolate cell with a short, straight cross-line, 

 instead of an oblique one. That its true position in the family is some- 

 what uncertain is evident from the fact that it was originally described as 

 a species of Pachyprotasis, a genus much further removed from Tenthredo 

 than is Strongylogaster. My specimens were all taken in the same 

 locality — a swampy meadow margin, luxuriant in ferns, herbaceous plants 

 and shrubs. The majority of them were taken during June and July. 



