34 



culty exists in accurately graduating this scale for a particular 

 bell, as the disturbing effects of varying temperature, &c, 

 affect the result. The laws laid down by Boyle, Gay Lussac, 

 and others that all gases expand and contract the same for the 

 same temperature, that expansion is in direct ratio to the abso- 

 lute temperature, and that the bulk is in inverse ratio to the 

 pressure, only hold good so long as the gas under consideration 

 is a perfect gas, and as soon as the vapour element is intro- 

 duced all this is modified. As it happens that in practice gas 

 and air are always more or less saturated with aqueous vapour 

 the normal state must be taken as that of saturation. To meet 

 this difficulty the Astronomer Royal in 1861 compiled a table 

 of expansions of gas in contact with water at various tempera- 

 tures, based upon factors supplied by J. Glaisher, F.R.S. It 

 is said to be calculated in accordance with the best modern 

 theories. It is the table used in official verifications of local 

 standards at the Standards Department, and is found in all 

 practical treatises on gas manufacture, but is of course not 

 used in the scientific comparisons and verifications conducted 

 at the department, as it is only calculated to ore-half percent., 

 and much closer observations are required. As an example of 

 the dilation I give an extract from the table : — 



A volume of gas at 32° Fahr. increases at 42° by 2\ per cent. 



a Q9° " " Pf>° " 7JL " 



a ^9° " " ^9° (t TO " 



or about 2\ per cent, for each ten degrees of Fahrenheit. As 

 the dilatation of pure or dessicated gases is in direct ratio to 

 the absolute temperature, which is found by adding 459 to the 

 Fahrenheit temperature, it will be found that the volume of a 

 pure gas at 32° will increase 2*036 per cent, for every ten 

 degrees of Fahrenheit, as against 2-J per cent, in the case of 

 saturated gases. 



In order to calculate such a table the amount of aqueous 

 vapour in the air must be ascertained by determining the tem- 

 perature of the dew point. This is done by means of M. Eegnalt's 

 hygrometer, as used by the British Standards Department 

 (Plate XII.). It consists essentially of two test tubes of glass, 

 each fitted with a silver thimble. Ether is poured in that 

 marked b, and the air or gas bubbled through it ; the tempera- 

 ture is thus reduced until dew is observed upon the silver 

 thimble, and the temperature is observed by the thermometers. 

 Of course the lower the temperature required to deposit dew 

 the less vapour exists in the atmosphere or gas being tested. 



The mode of verifying a cubic foot measure is very similar 

 to that employed in weighing the contents of a bushel measure, 

 and similar corrections are required. But in verifying the 



