THE PROBLEM AND METHODS OF IWESTIGATIOX 71 



that the decrease in ox^^gen consumption may apparently difTer 

 widely in different narcoses, and Alexander and Cserna have found 

 that not only is this true, but that the decrease in carbon-dioxide 

 production is not parallel to the decrease in oxidation in different 

 brain narcoses. In short, it is possible that the changes in the cell 

 which bring about narcosis may differ in character with different 

 narcotics and perhaps with different cellular conditions. Perhaps, 

 as so often in the history of biological theory, all the theories of 

 narcosis are more or less correct. 



But, however the narcotic substances act upon the cell, there 

 can be no doubt that within a given species or organism a general 

 relation exists between metabolic condition and susceptibiUty to a 

 given narcotic. Differences in metaboHc condition do not exist 

 independently of differences in condition of the colloid substratum, 

 and whether the narcotic aft'ects primarily the substratum or cer- 

 tain of the chemical reactions, the susceptibility of the organism 

 or part to its action must differ as the conditions which determine 

 or are associated with metabohc activity differ. 



Narcosis is only one stage in the action of the narcotic sub- 

 stances. When they are present in sufficiently high concentration 

 or act for a sufficiently long time, they bring about changes which 

 are not reversible and which finally end in death by making the 

 continuation of metabohsm impossible. The wide range of varia- 

 tion observed in some cases between narcotic and killing concen- 

 trations, both with different narcotics and with the same narcotic 

 at diff'erent stages of development (Vernon, '13), indicates that 

 the reversible changes involved in pure narcosis are different in 

 some way from those which result in death. With the killing con- 

 centrations the relation between susceptibility and metaboHc con- 

 dition is more distinct and uniform than with the lower, purely 

 narcotic concentrations, where incidental factors may sometimes 

 mask or reverse the fundamental relation (see pp. 75-76). With 

 the cyanides, however, where narcotic and killing concentrations 

 do not differ very greatly, this relation appears more distinctly 

 and uniformly than with any other agents thus far used. 



It cannot of course be maintained that the susceptibility to 

 cyanides or other narcotics of an organism or part at a given moment 



