Y. Henderson (Henderson and Morris) has devised a somewhat simpler 

 form of the Haldane apparatus which has been improved in certain details 

 by Bailey x at the N. Y. Post-Graduate Hospital. The degree of accuracy 

 necessary for ordinary routine analyses for the determination of the basal 

 metabolism in the hospital is easily attainable with this apparatus. 



Krogh has recently described an apparatus which is accurate to 0.001 

 per cent. He finds that the sources of error which prevent the oxygen 

 analyses from being highly accurate in the Haldane apparatus are inti- 

 mately connected with the presence of water and dirt in the gas burette. 

 Water must of course be present to insure the saturation of the gas with 

 water vapor and dirt accumulates rapidly from the contact of mercury 

 with the rubber tubing and with oxygen. Krogh gets rid of these inter- 

 fering factors by employing three separate burettes (Fig. 20, 1, 2, 3) of 

 which one (1) is employed exclusively for moving the air to and from the 

 absorption pipettes, while the second (2) is of a suitable size for meas- 

 uring the air before and after the absorption of CO 2 , and the third (3) 

 for measuring it after absorption of O 2 . The water vapor necessary for 

 saturating the sample air, when it has become partially dried in the ab- 

 sorption pipettes will be supplied by the first burette and the variations in 

 the amount of water present has no influence upon the accuracy of the 

 measurements. The two other burettes (2) and (3) contain just enough 

 water to insure that the samples remain saturated. 



A second improvement introduced by Krogh in this apparatus is that 

 the mercury is raised and lowered in the burettes not by raising and low- 

 ering a mercury reservoir but by means of air pressure, an arrangement 

 which obviates the use of rubber connections between the burettes and the 

 reservoirs and besides facilitates the manipulation considerably (Krogh 



Still another apparatus employing the open circuit method is deserving 

 of mention. This is the apparatus of Ham'oit and Richet. By means of 

 air valves the inspired air and the expired air are separated, both being 

 measured by meters. In addition the expired air is measured again after 

 absorption of the carbon dioxid in potash. The first meter gives the vol- 

 ume of the inspired air, the second of the unchanged expired air, and the 

 third the volume of the expired air minus the volume of carbon dioxid. 

 The volume of inspired air less the final volume of expired air gives the 

 amount of oxygen consumed. The method as carried out by Hanriot and 

 Richet does not seem to be particularly accurate; but Krogh expresses 

 the opinion that the method has great possibilities if used with modern 

 gas meters of sufficient size and placed in a water bath where the volumes 

 measured would be subject to the same fluctuations. Krogh notes that 



1 This construction of the apparatus is made by E. Machlett & Son, 153 East 84th 

 Street, New York City. 



