402 EMERGENCE OF DRAGONPLY-LARV^., 



system of larva; more tlian a day old (Expt.l), yet, in larvw only 

 an hour oi- two old, the whole system is filled with COj, except 

 the rectal capillaries, where there is a mixture of gases (Expt.2). 

 Also, though the Azol solution appeared to be unsuitable for 

 penetration of chitinous membranes, yet the fact that it ex- 

 tracted some of the gas from the rectal capillaries of a 2-3 days 

 old larva, and failed to do so in the case of a 6-12 hours old 

 larva, suggests that the proportion of O in the gills of the former 

 was considerably greater than that in the gills of the latter (Expt. 

 3). We are thus led, on all three counts, to the definite con- 

 clusion that, (i. ) the original gas which fills the, tracheal system of 

 the larva chiving the pvocei^s of hatching is COo, but that, (u.) the 

 estahlishment <>/' regnlav rectal respiration sets np a process of 

 diffusion hetween the rectal cajnllaries and the water in the 

 rectum, ivhereby the COo in the tracheal system is gradually with- 

 drawn, and replaced by air, or a mixture of nitrogen and oxygen 

 not differing much from the ordhmry composition of air. 



It is easy to show that these results are in agreement with the 

 physiological aspects of the problem. Firstly, the exertions of 

 the larva to burst open both the egg and the pronymphal sheath 

 involve a sudden and rapid increase of metabolic activity. Thus, 

 just at the period when large quantities of CO^ are being 

 liberated by the activities of the larva, gas replaces the liquid in 

 the trachea;. If this gas be CO., as our experiments suggest, we 

 have solved at once the problem of what the larva does with the 

 large amount of CO^ liberated by its new exertions, and why it 

 is that the gas comes into the tracheal system in the manner it 

 does. Secondly, if we grant that this gas is CO.,, the gradual 

 change to air, or a mixture of gases not difiering appreciably 

 from air, is not only to be expected as a result of rectal respira- 

 tion, but must inevitably take place, by the ordinary laws of 

 diffusion. Eor, in the tracheal system of the newly-hatched 

 lai'va, the pressure of COo will be 760 mm., or atmospheric 

 pressure. But in the aerated water of the rectum, the partial 

 pressure of O is approximately 150 mm., that of N approximately 

 610 nnn., and that of CO. less than 1 mm. It follows that the 



