125 



was tapped quickly against the curtain. At once they moved backward and 

 downward, and the female struck the water with her abdomen. Then they rose 

 again, hovered a moment a few inches in front of the curtain, and repeated the 

 performance. After some time they separated and alighted among some cat-tailg, 

 growing near. Oviposition was not interrupted by copulation. Part of this 

 curtain of algae was collected. Portions of it were literally piled up with the- 

 dragonflies' eggs. Doubtless some of the eggs were washed from the abdomea 

 into the water, but the majority were placed on the algae. Eggs had been 

 placed at the top of the curtain, but this had become thoroughly dry. Females, 

 which I saw ovipositing were placing the eggs two or three inches above the 

 water where the curtain was very damp. The hatching of the egg, and possibly 

 the first moult of the nymph, takes place on this curtain. 



24. Though the subject of Odonate copulation has been considered 

 by many authors with "presque toujours une description detail^e et souvent. 

 poetique," I have been unable to find any statement concerning the filling of the 

 seminal vesicle of the male dragonfly, other than that this takes place before copu- 

 lation. In the case of Calopteryx, Argia and Enallagma, where I have been able to 

 make positive observations, the male fills the seminal vesicle at once after he has 

 captured the female. It seems probable that during the wild flight of mating 

 Aeschnas and some of the gomphines (I have noticed especially Dromogomphus 

 spoliatus) the seminal vesicle is being filled, and, this accomplished, the pair come 

 to rest in tree-top, on the ground, or where not, and copulation takes place. 

 The Anisoptera, which I have observed, do not copulate while flying, if they are 

 undisturbed. 



