2o6 SECTIONAL ADDRESSES 



a sample for analysis ; but this will not matter provided the breathing 

 remains constant, as indicated by the series of observations. This diffi- 

 culty can be overcome by using two analysis apparatuses ; in one the 

 analysis is made, while in the other air is gradually being collected in the 

 burette during the five minutes by allowing the mercury to fall slowly ; 

 this is analysed, while a further sample is being collected in the first 

 apparatus, and so on ; alternatively the air could be collected in sampling 

 tubes and analysed later. 



Finally, I should like to thank Mr. T. W. Adams, who has been 

 associated with me in the calculations on which these results are based, 

 and Mr. W. J. Martin of the London School of Hygiene for advice on 

 the mathematics required, and to express my indebtedness to the Parsons 

 Fund at Guy's Hospital and the Medical Research Council for grants in 

 aid of the work, and to the Physiological Society and Guy's Hospital Reports 

 for the illustrations. This study is not yet complete, though I think the 

 main conclusions will stand. We hope in due time to publish all the 

 evidence on which these conclusions are based. 



