lH) 



DENSITIES OF OXYGEN AND HYDROGEN 



Hence we have 



Mean barometric pressure 737-70 nim. 



Water pressure in l>(iilint;-point apparatus reduced to equiva- 

 lent column of mercury 23 nim. 



I'rcssure of steam in boiling-point apparatus 737-93 '""'• 



Pressure reduced to sea level in latitude 45 737-67 mm. 



Boiling point corresponding 99. 1 70 



Mean reading of maximum depressed zero — 0.300 



Correction for internal pressure -\- .014 



Correction for inequalities of bore — -coj 



Corrected reading for freezing point — 0.289 



Mean reading for boiling-point 98.880 



Correction for internal pressure + .098 



Correction for inequalities of bore — .002 



Corrected reading for boiling-point 98.976'' 



Observed interval 99-265'' 



True interval 99. 1 7°° 



Correction to observed interval — -095° 



The followintj tal>le shows the corrections foiiiid for 



a leading between 



1 



With this theiinoineter were compared two thermometers of Jena normal glass 

 numbered 826 and i;}5;^, each haviiiir a scale in fiftieths of a deiijree enclosed in its 

 tube. This pair was used for the actual determination of the temperature of the 

 oxygen ; their indications are given, as well as the indications of the hydrogtii air 

 thermoiueter computed from them. 



During the present series of determinations, the cistern of my ])arometer was 

 3.9 metres higher than the cfiitie of the globe while it was filling w itli o.vygen- 

 The temperature nf tlie tube connecting them was assumed to be the .same as the 

 temperature of the room. Tlie indications of the mano-barometer, reduced for the 

 e.\-i)ansion of the mercury and of the scale, were therefore still further corrected by 

 adding the term •^•9"',T,7^f^, where D is the weight of a litre of oxygen at the 

 temperature of the room and at the pressure shown in that experiment. 



( 



