NO. 1438. WINGS OF THE TENTHREDINOIDEA—MacGILLTVRAY. H25 



they have shown by the adoption of these modifications greater ability 

 to conform to envdromental modifications. 



The cell K^ group finds its greatest modifications in the position of 

 the radial cross-vein, the position of the medio-cubital cross-vein 

 between 8c+K+M and cu])itus, and in the loss of the anal cells. The 

 cell K., group finds its greatest modifications in the swinging of the base 

 of media toward the a])ex of the wing, the atrophy of the base of the 

 radial sector, and the loss of the second anal cell. It has been shown 

 that the trend of modifications in these wings is toward the arrange- 

 ment of the veins in such a way as to form supporting trusses in the 

 stigmatal area. The cell R^ group has accomplished this by means of 

 the medio-cul)ital cross-vein alone, while the cell R^ group has 

 employed not oidy the cross-vein, but combined it with the transverse 

 j)art of media. The fact that this latter type is the one preserved 

 throughout the higher Hymenoptera would seem to indicate that it is 

 the one that has been most successful in meeting the requirements of 

 natural selection, and consequently must he the most efhcient type. 

 The loss of the l)ase of the radial sector, which is peculiar to the cell 

 R,^ group, and likewise to the higher Hymenoptera, would also seem 

 to point in tliis same direction. On the other hand, the cell R^ group 

 has exceeded the cell R,., group in the loss of the anal cells, which is 

 likewise peculiar to this group and the higher Hymenoptera; but even 

 this condition is approximated by the cell Rj group in the genus 

 (h'l/ssns (fig. 1)7), where the second anal cell is apparently wanting. 

 So far as structural modifications are concerned, the weight of the 

 evidence shows that the modifications found in the cell R5 group have 

 dei)arted farthest from the primitive type, and we are therefore justi- 

 fied in giving it the precedence here. 



Another fact that should not be overlooked, although it does not 

 refer to structural predominance, is the number of descendants. The 

 cell R, group contains five families, all of which are limited as to 

 luunher of genera and species. The cell R^ group contains two fam- 

 ilies, one containing a single species and the other many times as 

 many genera and species as is found in all the remainder of the Ten- 

 thredinoidea together. The predominance of the cell R^ group would 

 seem to contradict our conclusions from structural superiority and 

 therefore of efficiency of type, namely, that the predominance of indi- 

 viduals is a direct confirmation of the superiority and efficiency of the 

 cell R^ type. This is only an apparent contradiction, for, if structural 

 superiority and pi'edominance of descendants are compared in other 

 groups of animals, it is found that in those groups where there is a 

 marked^ structural supei'iority there are a limited number of genera 

 and species, while in those groups where there is a marked predomi- 

 nance of descendants, they are as a rule only mediocre so far as struc- 

 tural superiority is concerned. 



