134 Bulletin of the Brooklyn Entomological Society Vol. XV 



solution. Indeed, Weisman postulated a temporary opening of 

 the stigmata to the atmospheric air. But after it had been shown 

 by Calvert and Dewitz that the filling proceeds while the larva is 

 submerged this idea was abandoned. 



It is one of the interesting coincidences that we so often find 

 in scientific research, that two separate investigators should 

 simultaneously arrive at similar conclusions on this question. 

 These two are Tillyard in Australia working with dragonfly 

 larvae, and Franckenberg in Germany working with Corethra 

 larvae. Both investigators ascertained that the tracheae will fill 

 _with air even when the larvae Jbe placed in deoxygenated water. 

 Both therefore concluded that the primary filling must come 

 from within and is not a diffusion process. 



Let me describe this process of filling. In Corethra, the so- 

 called " phantom " larvae, there are four air chambers connected 

 by tracheae. Filling begins at the rear behind the posterior cham- 

 bers ; these fill first, then the air can be seen advancing in the 

 lumina of the connecting tracheae, and finally the anterior cham- 

 bers fill. In the Odonata the process is reversed. The trachea 

 lying next the cephalic heart (found only in late embryos and 

 temporary in nature) fill with air, this air column is seen to move 

 in a posterior direction along the main trunks to the rectal gills, 

 which then begin to operate at once, the gaseous air appearin 

 to be the stimulus requisite for operation. In each case this fill- 

 ing occupies just a few moments, from three to ten seconds, 

 rarely as much as half a minute. To be sure of the details Till- 

 yard in an ingenious experiment was able to slow up the filling 

 so that it covered a period of several minutes. 



It must be noted that this sudden air supply cannot come from 

 the outside of the larva, as endosmosi.s is not rapid enough to 

 account for the rapidity of the filling. Furthermore, this change 

 will take place in water free from oxygen, hence in conditions 

 which bar the possibility of endosmosis altogether. Both authors, 

 Tillyard and Franckenberg, assume that the blood of the embryo 

 must hold air to a point of supersaturation, and that it goes out 

 of solution with such force that the serous fluid filling the tra- 

 cheae is forced through the tracheal walls into the body cavity. 

 Now, what is this gas ? Tillyard states that it is carbon dioxide, 



