Dr. Beddoes’s Account , &c. 283 
thing in this way. Accordingly, April 28, 1786, the ther- 
mometer {landing at 47 0 , he made a folution of a powder, 
confiding of equal parts of fal ammoniac and nitre, in a bafon, 
by means of which he cooled fome water, contained in a 
glafs tumbler, to 22 0 . To this he added fome of the fame 
powder, and immerfed two very fmali phials in it ; one con- 
taining boiled, the other unboiled water ; when he foon found 
the water in the phials to be frozen, the unboiled freezing 
firfl. 
H aving obferved that Glaubf.r’s fait, when it retains its 
water of cryflallization, produces cold during its folution, he 
thought 0/ adding this to his other powers, and July 18, 
1 786, reduced the thermometer 46 degrees. In this experiment 
the following proportions were ufed : the temperature of the 
air being 65°, to water four ounces, at 63°, were added, 
Of fal ammoniac ^ xi thermometer funk to 32 0 , that is, 31 0 
Of nitre ^ x - - - 24 0 , that is, 8° 
Of Glauber’s falts g ij - - 17’, that is, 7 0 
* 46° 
In this way he froze water on a day fo hot that the ther- 
mometer in the fhade flood at 70°. By firfl cooling the falts 
and water in one mixture, and then making another of thefe 
cooled materials, he funk the thermometer 64 degrees. 
Augufl 28. The temperature of the air being 65°, half an 
ounce of redlified fpirit of wine was diluted with three ounces 
and an half of water, and immerfed in the fame frigorific 
mixture. When cooled to 24% it began to freeze. A quan- 
tity of the neutral falts, likewife cooled in the mixture, were 
put into the diluted fpirit, when the thermometer fell to - 4% 
fo that the liquor was cooled 69 degrees. 
7 Spirit 
