Pulmonary G-ases in the Respiration of Man. 227 



three-way iron stopcock. A scale is fixed at the back of the open limb, 

 the divisions corresponding exactly to those engraved on the other 

 limb. 



The flask C, open at both ends, is full of water, and communicates 

 by india-rubber tubing with the cylinder D, which contains the air for 

 .analysis. 



The process of analysis is as follows : The eudiometer is entirely 

 filled with mercury. Next, the hydrogen being placed under a 

 pressure of 1 or 2 inches of water by depressing the receiver, 

 the gas is driven through the three-way cock, while at the same 

 time it is aspired by a compressed india-rubber syringe, which is 

 suddenly released ; this washes out all the atmospheric air from the 

 tube connecting the supply of hydrogen with the eudiometer. The 

 three-way cock being turned so as to connect the hydrogen-receiver 

 with the eudiometer, mercury is let out at the bend of the instru- 

 ment, when hydrogen is drawn into it, and the volume of the gas 

 recorded. Air from the flask D is admitted into the eudiometer 

 through the india-rubber tube, pressure being exerted by the water 

 held in the flask D, while aspiration is produced by letting out 

 mercury from the eudiometer. The gases are brought under atmo- 

 .spheric pressure by adding or taking out mercury. 



The gases are mixed in the eudiometer by means of an india- 

 rubber syringe fitting to the open end of the instrument, where the 

 funnel is shown in the drawing; the syringe is compressed and 

 released repeatedly, thus driving the mercury up and down. The 

 little movable spirit level is convenient, though not indispensable, 

 towards the adjusting of the mercury in the two limbs. I find that 

 .such an instrument has been proposed, or used, for similar purposes 

 by Gr. Lunge (' Chem. Soc. Jl.,' 1892), but the readings are not so 

 reliable as those obtained from two identical scales, one for each limb 

 of the U-tuke. 



The gases are finally exploded in the usual way. 



The process is described in full in my paper read June, 1891, to 

 the Royal Society. 



