534 OSTERHOUT— DYNAMICAL THEORY OF ANTAGONISM. 



tained one hundred disks. By this means the material in the dif- 

 ferent tumblers was made as similar as possible. The disks in each 

 tumbler were then packed together (like a roll of coins) to form a 

 cylinder whose resistance was measured. Throughout the experi- 

 ments the different lots were kept side by side and treated as nearly 

 alike as possible, except that they were placed in dift'erent solutions. 



Great care was taken that all the solutions (both pure and 

 mixed) should have the same conductivity as sea water. 



The distilled water used in making the solutions was prepared 

 with especial care, the first and last part of the distillate being dis- 

 carded. It was distilled from a glass still (which had been used 

 for some months) using plugs of cotton in place of cork or rubber 

 stoppers. 



The salts used were the purest obtainable, being for the most part 

 Kahlbaum's (in some cases Merck's blue label reagents were 

 employed). 



The mixtures were made in the following proportions : 



The results* are shown graphically in Fig. i (cf. Table III.). 

 As will be seen on inspection of the figure, the resistance rises at 

 first (except in pure NaCl) and subsequently falls. 



Evidently two processes are involved, one of which produces a 

 rise, the other a fall in resistance. While these processes might be 

 looked upon as independent, everything points to the fact that they 

 are casually connected, and it seems natural to assume that they 

 represent two chemical reactions, one of which is dependent on the 

 other. 



* The results are expressed as per cent, of the net resistance in sea water 

 at the start of the experiment. They might also be expressed as per cent, of 

 the control but this has no especial advantage for the present investigation. 



