Chemistry and Physics. 419 



ide, furnished with platinum electrodes and supplied with about 

 three amperes of current by means of a Grove battery. Only 

 one of the gases is collected, the other being allowed to escape 

 through a mercury seal. If the hydrogen is to be collected, this 

 gas is led, first, through a tube of hard glass containing metallic 

 copper and heated to redness, thence through a flask containing a 

 strong potassium hydroxide solution, then through a second hot 

 tube containing copper, through a regulating tap and a tube con- 

 taining solid hydroxide, through a long tube containing phos- 

 phoric oxide, and finally through glass-wool. The globe in which 

 the gas is weighed is connected to one terminal of a four-way 

 tube, the other terminals leading to the pump, to the generator 

 and to a blow-off tube of the barometric length, respectively. 

 In making an experiment, gas from the generator was allowed to 

 flow through the purifying train, under the action of the pump, 

 for half an hour. Then the pump was put into communication 

 with the globe and with its set of tubes and a high vacuum pro- 

 duced in them. On making connection with the generator the 

 globe slowly filled with gas, the operation taking from two to 

 three hours. The gas was allowed to escape from the blow-off 

 tube for some minutes, under a pressure of half an inch of mer- 

 cury. The cistern was then lowered, leaving the end of the tube 

 free, and the flow of gas was continued for two minutes. Four 

 minutes were allowed after the tap to the generator was closed, 

 for equilibrium of pressure to be attained. Then the tap to the 

 globe was turned off, the barometers and thermometers read, and 

 the globe weighed. The finally corrected values obtained were 

 for hydrogen 0*158531 gram, and for oxygen 2*51777; so that 

 the density-ratio is 15*882. Combining this with Morley's ratio 

 of volumes 2*0002 : 1, the value 15*880 is obtained as the ratio of 

 the atomic masses. The following summary of the results of 

 different experimenters is given in the paper : 



Name. Date. Atomic masses. Densities. 



Dumas, 1842 15*96 



Regnault, 1845 15*96 



Rayleigh, 1888 15*884 



Cooke and Richards, 1888 15*S69 



Reiser,... 1888 15*949 



Rayleigh, 1889 15*89 



Noyes, 1890 15*S96 



Dittmar, 1890 15*866 



Morley, 1891 15*879 



Leduc, 1891 15*905 



Rayleigh, '.... 1892 15*882 



— Proc. Roy. Soc, Feb. 18, 1892; Nature, xlvi, 101, June, 1892. 



G. F. B. 



2. On the Properties ■ of liquid Oxygen and liquid Air. — In a 

 recent lecture at the Royal Institution, Dewae illustrated the 

 properties of liquid oxygen by some remarkable experiments. 



