ofi THE REPOET OF THE No. 36 



Various subdivisions and combinations have been proposed by modern writers. 

 Konow in 1890, recognized a single family divided into three subfamilies and 

 these in turn divided into eleven tribes. Dalla Torre in his Catalogue, recognizes 

 one family and eighteen subfamilies. A still different grouping is that of Ashmead, 

 published in 1898, where eighteen families are recognized, but different in extent 

 from those of Dalla Torre. The most recent grouping is that of Eohwer, who 

 would divide the group into four superfamilies. These superfamilies comprise 

 fourteen families, and one of these families, the Tenthredinida;, includes eleven 

 subfamilies. In the following paper the classification proposed by the writer in 

 1906, based on a phylogenetic study of the wings, has been followed, in which 

 a single superfamily and nine families are adopted. Representatives of the families 

 Blasticotomidse and Megalodontidae are not found in America, and they are not 

 considered in any reference in the following pages to the group as a whole. These 

 families would be grouped as follows : — 



The generalized Tenthredinoidea. 



The vein E2 conservers Xyelidce 



The vein E, losers Pamphiliidce 



The specialized Tenthredinoidea. 



The cell E4 group. 



Primitive cell Eg conservers .... BlasticotomidcB 



Primitive cell E, losers Tenthredinidce 



The cell E5 group. 



The radial cross-vein and M, conservers. 



The Medio-cubital cross-vein conservers Xiphydriidce 



The medio-cubital cross-vein modifiers. 



The M and m-cu equalizers. 



The Cell O-j-c^ losers Siricidce. 



The coll C-f-Sci conservers Megalodontidce. 



The M and yn-cu modifiers Oephidce. 



The radial cross-vein and M, losers Oryssidce. 



The great difference in the number and rank of the groups is due in the 

 main to the difference in opinion as to what position shall be assigned the com- 

 ponents of the restricted family Tenthredinidse. This family contains a large com- 

 plex of genera that are readily segregated into a number of definite groups that 

 are easily differentiated by means of structural characters, but they all have a 

 similar or common habitus. The writer believes that these groups are best handled 

 by considering them simply as subfamilies. In the work referred to above, twenty- 

 four subfamilies are listed. Of these, the following are found in the United States 

 and Canada : 



Contraction of the anal cell conservers Lophyrince 



Empliytinrp 

 SelandriincB 

 Dolerince 

 Phyllotomince 



