128 



weight of solutions of different salts and other substances, as compared 

 with the loss of weight in water. 



The force which holds together the particles of a vaporizable 

 liquid is gradually overcome, if that liquid be exposed to air, by 

 another force which separates, expands, and diffuses those particles 

 in the form of vapour ; and this separation takes place, even at a 

 common temperature, so rapidly, provided the surface be sufficiently 

 extensive, that an easy opportunity is afforded of determining the 

 loss of weight by a common balance. 



A subject for investigation, possessing much interest, thus presents 

 itself, and, in its pursuit, some new and unexpected results are 

 encountered. 



The method which I have pursued has been to expose to the 

 atmosphere, for a definite period, solutions of different salts, and also 

 pure water under like conditions of quantity and area, temperature, 

 atmospheric moisture, and atmospheric pressure. 



Different salts and other soluble substances are thus found to 

 possess, when in solution, different powers of retarding or accelerating 

 evaporation, and hence, from its amount, as compared with that which 

 takes place in pure water, we can estimate the comparative value of 

 those powers. 



The powers themselves being established as facts, the next point 

 is to endeavour to discover the cause or causes on which they depend, 

 and a wide field of inquiry is thus opened. 



The following are the instruments which have been employed : 



1 . A balance, for one of the scales of which is substituted a flat 

 metal plate, six inches square, on which the vessels to be weighed 

 can be conveniently supported. This balance will turn sensibly at a 

 grain, even with a weight of 4 Ibs. on either side. 



2. A number of copper pans tinned within, all of the same size, 

 being precisely 5 inches square inside, with perpendicular sides 

 f ths of an inch in height, also a number of earthenware pans of the 

 same dimensions, The area of 25 square inches has been chosen, 

 partly because this size is convenient for manipulation, and partly 

 because the results obtained can be easily represented in decimals. 

 This facility of decimal calculation would be of importance should 

 such pans come into general use as hygrometers, for which purpose 

 they are well adapted. 



