66 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 89 



on the anterior end of the first valvifer, depresses the posterior end 

 of this valvifer by puUing upward on its anterior end (fig. 21 A, B). 

 The muscie, however, has no antagonist inserted on the first valvifer, 

 so that its contraction simply retracts the ventral valvula and pro- 

 tracts the dorsal valvula (B). After this, the principal movement of 

 the valvulae on each other evidently results from the motion of the 

 second valvifer on the first, and there can be no doubt that the mus- 

 cles effecting this action are those of the ninth tergum inserted on 

 the second valvifers (fig. 17 E, G, H, 6, 7). The first of these two 

 muscles (d) is attached on the anterior apophysis (g) of the second 

 valvifer; the second (7), having a horizontal position, is attached 

 on the posterior dorsal arm (i) of the same valvifer. The two mus- 

 cles thus oppose each other from opposite sides of the articular sur- 

 face (G, h) by which the second valvifer rocks on the first valvifer. 

 If the ovipositor were a rigid structure, this mechanism apparently 

 could only move the distal shaft of the ovipositor up and down on 

 the fulcral point (h), but as already noted, the demonstrable effect 

 of the rocking of the second valvifer on the first valvifer is an opposite 

 back-and-forth movement of the two valvulae of the same side on 

 each other. 



The two pairs of ventral muscles of the ovipositor (fig. 17 E. H. 

 c), 10) uniting the first valvifers and the second valvifers, respec- 

 tively, with the posterior intervalvula {p'lv), must be tensors in func- 

 tion, since their contraction could only produce a compression of the 

 basal parts of the ovipositor. The muscles of the seventh sternum 

 (fig. 19 A, 7), inserted on the anterior ends of the first valvifers 

 (fig. 17 A, E, F, H, i), are evidently retractors of the ovipositor. 



During oviposition, the egg passes through the entire length of the 

 ovipositor in the channel between the valvulae. It appears that the 

 movement of the valvulae on each other is the cause of the progress 

 of the egg, but it is not evident how the effect is produced, since the 

 inner surfaces of the valvulae are smooth and give no suggestion of 

 the means by which the movement of the &gg is controlled. 



The distal end of the ovipositor is somewhat enlarged and forms 

 a lanceheadlike termination of the shaft with the acute apex slightly 

 depressed (fig. 17 B). The tips of the valvulae are capable of opening 

 upon each other by reason of a hingelike membranous area (m) 

 at the base of the enlarged terminal part of each of the dorsal valvulae. 

 In handling narcotized specimens, the observer may be surprised to 

 see the ends of the valvulae suddenly gape widely apart (D), as they 

 sometimes do. There are no muscles in the terminal parts of the 

 ovipositor ; the opening of the distal extremities of the blades is pro- 



