S Mr. kihwan’s Experiments , &c. on the fpecifc Gravities 
zdly. That if bodies, fpecifically heavier than water, be 
weighed in air and in water, they lofe in water part of the 
weight which they were found to have in air ; and that the 
weight fo loft is juft the fame as that of an equal bulk of wa- 
ter, and confequently that their fpecific gravity is equal to their 
weight in air, or abfolute weighty divided by their lofs of weight 
in water* 
^dly. That if a folid, fpecifically heavier than a liquid, be 
weighed fir ft in air, and then in that liquid, the weight it lofes 
is equal to the weight of an equal volume of that liquid ; and 
coiifequently if fuch folid be weighed firft in air, then in water, 
and afterwards in any other liquid, the fpecific gravity of this 
liquid will be as the weight loft in it by fuch folid, divided bv 
the lofs of weight of the fame folid in water. This method of 
finding the fpecific gravity of liquids I have found much more 
exadt than that by the areometer, or the comparifon of weights 
of equal meafures of fuch liquids and water, both of which are . 
fubjedt to feveral inaccuracies. 
4thly. That where the fpecific gravity of bodies is already 
known, the weight of an equal bulk of water may alfo be 
found, it being as the quotient of their abfolute weight divided 
by their fpecific gravity. This I (hall call their lofs of weight 
in water. 
Hence, where the fpecific gravity and abfolute weight of the 
ingredients of any compound are known, the fpecific gravity of 
fuch compound may eafily be calculated as it ought to be inter- 
mediate betwixt that of the lighter and that of the heavier, ac- 
cording to their feveral proportions : this I call the mathema- 
tical fpecific gravity. But, in fact, the fpecific gravity of com- 
pounds, found by actual experiment, feldom agrees with that 
found by calculation, but is often greater without ai\y diminu- 
2 tio n 
