10 Chapter I 



We may now tabulate the results of three of Burn's experiments. 



The chemical method of calibration therefore gives to K quite 

 constantly a somewhat higher value, for the different apparatuses 

 which were tried, than does the physical method. The mean differ- 

 ence is 2'5 % 



Let us now look at Peters' results in the light of this newly 

 acquired information. Had the constants of his apparatus (and the 

 apparatuses alluded to in the above table were among those used 

 by Peters) been obtained by the chemical method, his figures for the 

 specific oxygen capacity would all have come out 2%5 / higher or 

 thereabouts, and would therefore have been as follows : 



Specific oxygen capacity as 

 given by Peters 



Specific oxygen capacity 



recalculated by chemical method 



of calibrating apparatus 



411 409 

 394 



Theoretical figure 401 



As a third method of calibrating the apparatus has been worked 

 out by Homiann which gives results identical with that of Burn (see 

 Journal of Physiology, vol. XLVII) it seems probable that Peters' 

 values as corrected by Burn are more trustworthy than his un- 

 corrected ones. 



The figures arrived at by the chemical method of calibration give 

 an average which is so near to 400'8 that there can be no doubt 

 that the oxygen and iron are united in haemoglobin in the ratio of 

 two atoms of the former to one of the latter. 



The propositions, therefore, (a) that the bloods of different animals 

 have fundamentally different oxygen capacities, and (6) that analyses 

 furnish any serious reason to doubt the validity of the oxygen to iron 

 being related as two atoms to one, have ceased to trouble my mind. 



