The specific oxygen capacity of Mood 7 



It is not very easy to discern the processes by which conviction 

 grows in the mind ; probably the mere inspection of the figures given 

 is sufficient to convince the reader that so far as the relation of the 

 respiratory oxygen to the iron of haemoglobin is concerned, these 

 quantities are related in the proportions of two atoms of oxygen to 

 one of iron. To me, who had the privilege of seeing Peters work 

 from week to week, conviction came in a slightly different way; it 

 developed as it were like the image on a photographic plate. As one 

 experimental difficulty after another was overcome, as one source of 

 error after another was weeded out, as the worker himself developed 

 in skill and in capacity, just so surely did the results which he ob- 

 tained approach the theoretical figure with greater certainty till at 

 the end when all the difficulties had been overcome and when Peters 

 himself had attained to the rank of a first-rate exponent of the 

 technique, I arrived at a stage of conviction in which I never 

 doubted, when he undertook an experiment that the result would be 

 between 385 and 405. Perhaps there could be no surer proof that 

 all thought of the wide differences between different kinds of haemo- 

 globin, alleged to exist by Bohr and others, had passed out of our 

 horizon, than the fact of our almost laughable concern at the end of 

 the work as to why the average figure was .391 and not 401. We, in 

 the laboratory, thought perhaps that Peters did not perform sufficient 

 experiments to obtain a true average, or that some trace of 

 methaemoglobin was always present or, most probably, that some 

 trifling error had crept into standardisation of the apparatus used. 



The probability of the last source of error at least seemed 

 sufficiently great to warrant the initiation of a fresh research on 

 the subject, which was undertaken by Burn. Moreover on quite 

 general grounds it seemed desirable to undertake something of the 

 sort, for independence of the fallacies of a single experimental 

 procedure is of the essence of all sound experimental work. 



The problem was to find a method of calibrating the differential 

 blood-gas apparatus (12 ' in such a way that the possible errors involved 

 in the method which Peters had used would be of a different character 

 from those involved in the new method. 



The details of the differential apparatus will be found in the 

 Appendix ; the form used by Peters is shown in Fig. 1. All we need 

 say here is that the oxygen is liberated from the haemoglobin in one 

 of its two bottles. The pressures in these become unequal, and the 

 difference of pressure is indicated by the movement of the fluid in the 



