66 A CHEMICAL SIGN OF LIFE 
during the experiment. With this contingency in view, 
certain possible objections are here considered. 
The method employed for carbon dioxide determina- 
tion is so delicate that a change in the reaction of the 
sea-water, brought about possibly by the addition of 
the narcotic, might be sufficient materially to alter the 
values obtained. This, indeed, is the reason why we 
have never been able to investigate the effects of potas- 
sium cyanide, since the slight trace of alkalinity thus 
introduced seriously modifies the results. This objec- 
tion, however, we have been able to refute by direct 
experimental means. 
If the solution of the narcotic differs in reaction 
from sea-water sufficiently to influence the determina- 
tion, a similar effect should be observed in the case of a 
nerve which has been killed. Two freshly isolated 
nerves of approximately the same weight were killed 
simultaneously by means of steam and left for twenty 
minutes, one in a 2 per cent solution of chloral hydrate 
and the other in sea-water. A measurement of the 
adventitious carbon dioxide production from the two 
nerves so treated gave no evidence of any difference. 
The diminution of carbon dioxide from nerves subjected 
to the action of narcotics cannot, therefore, be referred 
to any change in the reaction of the sea-water produced 
by the narcotic. 
Another possibility is involved in the fact that certain 
narcotics produce phenomena other than those of nar- 
cosis. This is probably the reason why the metabolism 
change is never exactly the same in the case of two 
nerves in which typical narcosis has been induced by 
different means. One of these effects must be a change 
