4 PROCEEDINGS OF THE NATIONAL, MUSEUM. vol. 60. 



3-jointed (fig. 54) ; thoracic segments structurally much as Nema- 

 tinae; legs 4-jointed (fig. 52) ; abdominal segments (uropods normal 

 on urites 2-7, inclusive, and postpedes on urite 10 as in Nematinae) 

 4 annulate, A, B, C, and D (figs. 57-61), A, B, and C are large, promi- 

 nent, distinct, and ornamented with spines or hairs, A to a somewhat 

 lesser degree, while D is membraneous, inconspicuous, narrow, not 

 prominent, unornamented, and often infolded or depressed to apparent 

 absence; spiracle in alar region below B; spiracular area not large 

 and distinct; alar area, below C in alar region, consisting of a pair 

 of large dorsad-ventrad elongated folds, the anterior fold more dis- 

 tinct. 



Distributicrn. — This subfamily is represented in both divisions of 

 the Holarctic Region, but most of the species occurring in the Nearctic 

 Region are confined to the eastern part of the continent. The best- 

 known species of the subfamily, the bristly rose slug, Gladius 

 isomerus Norton, which is very abundant and injurious in the eastern 

 part of the Nearctic Region, and has heretofore been confused with a 

 European species. 



Generic Characters. — Heretofore with the exception of the synopsis 

 given by MacGillivray ^ the genera assigned to this group have been 

 separated by the modification of the flagellum, especially that of the 

 male. MacGillivray in his recently published synopsis introduces the 

 character of the relative length of the anterior basitarsis and separates 

 the genus Priophor-us from the other two genera by the absence of an 

 appendage to the radiellan cell. Although this last-mentioned char- 

 acter applies for certain few species of the genus Priophorus some of 

 the species described in this paper and assigned to Priophorus (and 

 which agree with the definition here used) have a distinct appendage 

 to the radiellan cell. The relative length and distinctness of this ap- 

 pendage is subject to individual variation and is sometimes different 

 in the two wings of the same specimens. Because of this variability 

 this character can not safely be used in separating genera and even 

 for species it should be used with great care. 



The secondary sexual characters, as found in the antennae and 

 heretofore used as generic, cannot be considered of such value, since 

 they group together species which on biology, larvae, and habitus are 

 not closely allied. Ti'ichio campus iiY-egularis (Dyar) has the an- 

 tennae in both sexes simple, and if those antennal characters are 

 used it would be placed in the genus Priophorus., but the larvae, lar- 

 val-habits, ovipositor, and leg characters show that it belongs to 

 Tricliio campus., where in the female it is so closely allied to grega- 

 rius (Dyar) as to offer difficulty in separating it, without examina- 

 tion of the ovipositor. Priophorus uhni (Linnaeus) has therefore 

 been placed in the genus Trichiocampus because of the character of 



•Bull. 22, Conn. Geol. and Nat Hist. Surv., 1916 (1917), p. 108. 



