NO. 1438. WINGS OF THE TKNTHREDINOWEA—MacGILLIVRAY. 607 



side AB is much greater than the side BC, we will find that any stress 

 exerted at the point B will not be eqiiall}' distributed, but that a nnich 

 larger part of the stress would fall on the side BC than on the side AB. 



We may assume that that wing is the most perfect mechanical 

 device which approaches the closest to some type of truss. From our 

 previous studies of the wing topography of the Tenth redinoidea we 

 are justitied in concluding that if such a thing as a ti'uss exists in their 

 wings it must l)e of the type where one side is longer than the otiier, 

 for there is no point situated near the middle of the front margin of 

 the wing to Avhich veins converge. 



Before taking up a direct comparison of the wings of the Tenthre- 

 dinoidea with the types of tru.sses given above, we should not overlook 

 the fact that we \vA\e to do not with a simple but with a complex type. 

 The front wings nmst in reality be trussed on both sides, for the 

 hymenopterous wing has stress exerted upon it by the air upon both 

 front and hind margins. The primary stress is exerted at some point 

 on the front margin where it strikes the air, while the secondary 

 stress is exerted on the hind margin where the hind wings are hooked 



Fig. 15.— Type of truss. Fk;. It;.— Type i>k truss. 



to it. This secondary stress is due to the necessity for a s3'nchronous 

 motion and to the fact that the hind wing must be pulled along. The 

 force exerted on the front margin of the front wing would be a push 

 or a force causing I'etardation, while the force exerted on the hind 

 margin of the front wing and the front margin of the hind wing 

 would be a pull or a force causing acceleration. 



A clearer conception of the arrangement of the trusses in the 

 hymenopterous wing will be had if we study tirst in some detail the 

 topograph}" of a wing in which these structures are self-evident. For 

 this purpose a front wing of Bhnnocampa alternlpes has been selected, 

 tracings from a photograph have been made, and the trusses found in 

 these wings marked as triangles by means of dotted lines (tig. 17). 

 For convenience in following the course of these triangles on the figure 

 they" have been numbered, the same number being placed on each side 

 of the same triangle. For the sake of brevity they will be referred to 

 in the following descriptions by these numbers. 



From what has already been said, it would be expected that these 

 trusses should arrange themselves into three groups, the first strength- 

 Proc. N. M. vol. xxix— 05 40 



