VOL. LXXX.] PHILOSOPHICAL TRANSACTIONS. 677 



was .813 at 6o°. This was furnished by Dr. George Fordyce, p. r. s., who suc- 

 ceeded in bringing it to that strength chiefly by adding the alkaH very hot. Care 

 must be taken that none of the caustic alkaH comes over in the distillation. 

 Some alcohol was also sent, for trial, by Mr. Lewis, an eminent distiller in 

 Holborn, whose specific gravity, at the same temperature, was .814. 



It was with spirit rectified from malt-spirits that Dr. Dollfuss's series of ex- 

 periments was made; but he tried several comparative experiments with such as 

 had been rectified from rum and brandy, and found no other difference than 

 might fairly be ascribed to unavoidable errors. On examining the results of 

 Dr. Dollfuss's experiments it was perceived, that though the numbers agreed to- 

 gether tolerably well on the whole, yet in some places there was that degree of 

 irregularity in the first differences, as made it advisable to repeat several of the 

 experiments; and Dr. Dollfus leaving England about that time, the busi- 

 ness of this repetition was intrusted to Mr. Gilpin, clerk of the r. s. This 

 gentleman had already taken a part in the business, by assisting Dr. DoUfuss in 

 the former experiments, particularly in the very nice part of weighing the mix- 

 tures; and his great skill, accuracy, and patience, in conducting experiments, 

 as well as in computations, had on other occasions been proved to many 

 members of the society. One experiment leading on to another, Mr. Gilpin 

 was at length induced to go through the whole series anew; and as the de- 

 ductions in this report will be taken chiefly from this last set of experiments, it 

 is proper here to describe minutely the method observed by Mr. Gilpin in his 

 operation. This naturally resolves itself into 2 parts, the way of making the 

 mixtures, and the way of ascertaining their specific gravity. 



1 . The mixtures were made by weight, as the only accurate method of fixing 

 the proportions. In fluids of such very unequal expansions by heat as water and 

 alcohol, if measures had been employed, increasing or decreasing in regular pro- 

 portions to each other, the proportions of the masses would have been sensibly 

 irregular; now the latter was the object in view, namely, to determine the real 

 quantity of spirit in any given mixture, abstracting the consideration of its 

 temperature. Besides, if the proportions had been taken by measure, a different 

 mixture should have been made at every different degree of heat. But the prin- 

 cipal consideration was, that with a very nice balance, such as was employed on 

 this occasion, quantities can be determined to much greater exactness by weight, 

 than by any practicable way of measurement. The proportions were therefore 

 always taken by weight. A phial being provided of such a size as that it should 

 be nearly full with the mixture, was made perfectly clean and dry, and being 

 counterpoised, as much of the pure spirit as appeared necessary was poured into 

 it. The weight of this spirit was then ascertained, and the weight of distilled 

 water, required to make a mixture of the intended proportions, was calculated. 

 This quantity of water was then added, with all the necessary care, the last 



