11 



land digestion. To study this, the ordinary methods of counting the 

 blood corpuscles and measuring the coloring matter in the blood were 

 followed. Valuable help in the initiation of this work was obtained 

 from Dr. Wm. B. French and Dr. E. B. Behrends. The actual exami- 

 nation of blood, for the purposes mentioned, was conducted by Messrs. 

 B. J. Howard and C. P. Knight. The work on the examination of the 

 blood was not commenced at the beginning of the experiment, and so 

 ;it does not cover the whole time of the experimental work. 



DETERMINATION OF TEMPERATURE. 



; The temperature of the blood was taken sub lingua before and after 

 'dinner each day. This method is probably the least accurate of any in 

 1 common use. It is, how r ever, convenient and easy. Since the chief 

 object of the determination \vas to disclose any notable departures from 

 the normal the method was considered fairly reliable. 

 :> Standard clinical thermometers of maximum registration were used 

 for this purpose, each subject being supplied with a separate thermom- 

 eter. These thermometers w r ere all graduated through the courtesy of 

 the Bureau of Standards. 



The rate of pulse was also determined in connection with the deter- 

 mination of the temperature. This is, however, not a matter of so 

 very much importance because of the ease with which the rate of pulse 

 is varied by exercise and emotional influences. 



In general, an attempt was made to control as fully as possible all 

 the avenues which might lead to any useful information concerning 

 changes, even of a minute character, in the functional activities of the 

 body during the period of observation. As has already been intimated, 

 the final verification of any small changes of an organic nature, espe- 

 cially of incipient lesions which may take place, is denied in experi- 

 ment upon human beings, but, in so far as possible, any intimations of 

 such changes, which could have been secured by any of the ordinary 

 methods of study, were noted. 



In data of this kind, namely the determination of the temperature, 

 pulse beat, etc., where dependence is placed upon the subject himself, 

 there are doubtless errors of observation which are undetected. 

 Instructions, however, were given, and in so far as possible carried out, 

 to the effect that any variation of a marked character must be verified 

 by a second observer. This rule applied, not only to the variations in 

 the body weight from day to day, but also to the departure of the tem- 

 perature from the normal, and to the variations in the rate of pulsation 

 of the heart. Thus, whenever one individual in the class noted any 

 marked variation from the normal he called upon either one of the 

 superintendents or one of his fellows to verify the numbers which he had 

 observed. By taking this precaution many errors which otherwise 

 Id have crept into the reports were avoided. 



wou 



