CH. ii] timiriazeff's eudiometer. 47 



there transferred to the vessel^ of water in which E 

 stands, and the gas in G is poured up into the funnel F 

 at the lower end of E. The funnel is connected with the 

 graduated part of the tube by a capillary passage ^ so 

 that the gas transferred from the carrier remains in the 

 funnel, whence it is drawn into the eudiometer by pulling 

 out the glass rod R working like the piston of a syi'inge. 

 The gas having been measured by means of the gradua- 

 tions on E, the piston is pushed in and the bubble forced 

 down into the funnel F. 



The pipette P is dilated into a bulb at B and ends 

 below in a bent capillary tube which can be inserted into 

 the funnel of the eudiometer, when by drawing out the 

 pistons of the syringe at the upper end of P, the gas is 

 dra^^^l into the pipette. For estimation of oxygen the 

 pipette contains freshly made solution of potassium pp'o- 

 gallate^; a second pipette filled with potash solution 

 serves to estimate the COo. After two or three minutes, 

 the gas is returned to the funnel of the eudiometer, drawn 

 in by the eudiometer syringe, and once more measured. 

 The difference between the first and second readings of 

 the eudiometer gives the amount of ox3^gen absorbed by 

 the pyrogallate, or of CO2 by the KOH, as the case may 

 be : the whole operation is so quickly done that no 

 corrections for barometric or thermometric changes are 

 necessary. 



1 The vessel should be of a dark colour so that the glass funnel at the 

 lower end of E may be perfectly visible. 



2 This arrangement, which is the only essential point of difference 

 between our apparatus and Timiriazeff's, is due to Mr F. F. Blackman. 



2 0-4 gram pyrogallic acid in 20c.c. of KOH (30 — 50 p. c. solution). 



