NUTRITION LABORiVTORY. I9I 



Study of the mechanical and electrical features of this ergometer has brought 

 out a number of theoretical points of great practical importance, which have 

 justified a recalibration of the apparatus on a much more extended and accu- 

 rate scale. Through the courtesy of Dr. C. F. Langworthy, of the U. S. 

 Department of Agriculture, an ergometer has been loaned to the Nutrition 

 Laboratory to be duplicated and for calibration. The apparatus, which is 

 practically a stationary bicycle, is calibrated by placing it inside the chair 

 calorimeter, rotating it from the outside by means of a flexible shaft, and 

 measuring the heat actually produced. An extended series of tests was made, 

 not only with this apparatus, but with a duplicate ergometer constructed in 

 the laboratory machine-shop. 



THE DETEIRM I NATION OF SUGAR IN DIABETIC URINE. 



The numerous clinical methods for determining sugar in diabetic urine 

 have proved inadequate for the accurate studies which are to accompany the 

 research on metabolism in diabetes now being carried out in the Nutrition 

 Laboratory. Accordingly, Dr. A. W. Peters has been occupied during the 

 past year in developing and modifying a number of methods for determining 

 sugar, with a view to selecting, or devising if necessary, the most accurate 

 and rapid method for the determination of sugar in pure-sugar solutions, 

 and of reducing power in pathological urines. An especially critical study 

 has been made of the estimate of sugar by the method of copper reduction, 

 particularly the conditions affecting the reduction process and the quantita- 

 tive determination of the copper involved. From the results obtained a pro- 

 cedure has been devised which is satisfactorily rapid and gives a high degree 

 of accuracy in the determination of sugar in both diabetic urines and sugar 

 solutions. Throughout the study, correct and delicate copper analysis was 

 regarded as the chief means of investigation, and in this connection some 

 results were obtained which are of critical importance for the accuracy of a 

 sugar determination as well as of some general interest in copper analysis. 



ANALYSES OE OUTDOOR AIR. 



The Nutrition Laboratory is fortunate in being supplied with probably the 

 most accurate type of gas-analysis apparatus in existence, namely, a modifi- 

 cation of an apparatus devised by Pettersson, according to plans of Dr. Klas 

 Sonden, of Stockholm. This has been in constant use for the past three 

 years in the laboratory, and in the hands of Miss Alice Johnson has given 

 extremely accurate results. One fundamental error, however, inherent in all 

 methods of gas analysis as yet developed, has shown itself to be likewise 

 inherent in this apparatus, although the other errors have been reduced to a 

 minimum. A year ago we were firmly of the opinion that there were con- 

 siderable fluctuations in the composition of outdoor air with particular refer- 

 ence to the oxygen content — fluctuations that were wholly inexplicable — but 

 during the past year's research, in which a control has been established by 

 the analysis from day to day of samples of a cylinder of compressed air 



