JOHN H. NORTHROP 493 



acetate by a small amount of ammonia or to the hydrolysis of protein 

 by pepsin is therefore identical in form but differs as to the signifi- 

 cance of the term A—x. 



In all the foregoing experiments the rate of digestion has been fol- 

 lowed by means of changes in the conductivity of the solution. Since 

 this value does not accurately represent the course of digestion, the 

 objection might be raised that the agreement between observed and 

 calculated values is due to compensating errors in the derivation of the 

 equation and in the deviation of the conductivity changes from the 

 actual progress of digestion. This could not be the case in the tests of 

 the equation in the differential form since in this case the results are 

 comparative and any deviation of the conductivity changes from the 

 true rate would cancel out. It is possible, however, that the agreement 

 of the equation in the integral form might be due to some such com- 

 pensation of errors. In order to show that this is not the case a series 

 of experiments was made in which the course of digestion was followed 

 by means of the increase in amino nitrogen. This value was deter- 

 mined by Van Slyke's^" method as already described, ^^ and, as far as 

 is known, accurately represents the progress of digestion. In these 

 experiments the quantity of egg albumin was kept constant (0.5 per 

 cent egg albumin) and the concentration of pepsin varied. The 

 results are summarized in Table VIII. The figures given under X 

 are the increase in amino nitrogen in cc. per 660 cc, of solution. They 

 are the average of three determinations and have an experimental 

 error of about 10 per cent. This is sufficient to account for the var- 

 iations in the constant of equation (6). The errors in x are greatly 

 magnified in this constant as it depends on the difference between 

 two experimental values. 2- In every case sufficient time had elapsed 

 before the first observation so that x at the time this determination 

 was made was already large compared to the concentration of pepsin. 

 The change in value of the constants for the first minutes of the re- 



20 Van Slyke, D. D., J. Biol Chem., 1912, xii, 275. 



21 Northrop, J. H., /. Gen. Physiol, 1918-19, i, 607. 



2^ The errors in x are reduced in the constant of Schiitz's rule. The varia- 

 tions in this constant are therefore outside the limits of experimental error 

 while those of Arrhenius' constant are within the limits of experimental error. 



