Appendix I. On methods 299 



In the first place it measures the difference of the total oxygen in each sample of 

 blood and not merely the oxygen in the haemoglobin. 



Secondly it assumes that the two samples of blood are of the total oxygen 

 capacity. The reasoning which underlies the method is as follows. Let the oxygen 

 in the arterial blood be A A and that in the venous blood A y . It is required to 

 measure A A A v . 



Let the amount of oxygen necessary to saturate the arterial blood be B A , and the 

 venous blood B r , and the total oxygen capacity C. 



A,= C-B A , 

 A,.=-- C-B,., 



A A -A r = C-B A - 

 = B,.-B A . 



The question then arises, Is the method applicable to cases where the oxygen 

 capacity of the arterial and venous bloods is different, such for instance as in the 

 case of blood flowing through an active gland ? 



There is one special case in which the method may be applied ; fortunately it is 

 the most common, but care must be taken to see that it obtains. It is the case in 

 which A A and G are almost equal, that is to say in which so little oxygen is taken 

 up on the arterial side of the apparatus that a small error in its quantity may be 

 neglected as compared with the difference between A A and A r . 



Thirdly we must consider the accuracy of measurement of the blood. The guiding- 

 principle is the same. So long as B A is almost nothing it is not material that the 

 artei'ial blood should be very accurately measured. The practical point that has to 

 be considered is the accuracy of the measurement of the venous blood. 



Moreover if any considerable quantity of oxygen is taken up by the arterial blood, 

 one has to allow for any difference which may exist between the constants of the 

 two sides of the apparatus. 



For all these reasons it is best when working with arterial blood which is un- 

 saturated to place the arterial and venous bloods each in a different apparatus, 

 and analyse each separately with saturated defibrinated blood in the other bottle 

 of the apparatus, and, having obtained the value of either A or B for each blood, to 

 subtract the one from the other by arithmetic. 



A useful check on the measurement of the blood is a determination of C for each 

 sample of blood after B has been determined, 



These complications, as Verzar found, make the estimation of the oxygen used by 

 organs during oxygen want very difficult ; for the reasons given above he made 

 separate estimations of arterial and venous bloods. 



Differential method with apparatus for 0*1 c.c. of blood. 



There is no difference in theory between this apparatus and that which we have 

 just described. The smaller apparatus was designed for work on human blood. 

 With it all the work described in this book on the human dissociation curves has 

 been performed, except where otherwise stated. 



