Weight afcribed to Heat. 169 
Had any difference of weight really exifted, had it been no 
more thant ove millionth part of the weight of either of the 
fluids, I fhould certainly have difcoyered it; and, had it 
amounted to fo much as +5254 part of that weight, I fhould 
have been able to have meafured it: fo fenfible, and fo very 
accurate, is the balance which I ufed in thefe experiments. 
I was now much confirmed in my fufpicions, that the ap- 
parent angmentation of the weight of the water upon its 
being frozen, in the experiments before related, arofe from fome 
accidental caufe; but I was not able to conceive what that 
caufe could poffibly be,—unlefs it were either a greater quantity 
of moifture attached to the external furface of the bottle 
which contained the water, than to the furface of that con- 
taining the fpirits of wine,—or fome vertical current or cur- 
rents of air, caufed by the bottles, or one of them, not being 
“exactly of the temperature of the furrounding atmofphere. 
Though I had forefeen, and, as I thought, guarded fuffi- 
ciently aguinft thefe accidents, by making ufe of bottles of 
the fame fize and form, and which were blown of the fame 
kind of glafs, and at the fame time, and by fuffering the bot- 
tles in the experiments to remain for fo confiderable a length 
of time expofed to the different degrees of heat and of cold, 
which alternately they were made to acquire; yet, as I did 
not know the relative conducting powers of ice, and of {pirit 
of wine, with re{pect to heat; or, in other words, the degrees 
of facility or difficulty with which they acquire the tempera- 
ture of the medium in which they are expofed, or the time 
taken up in that operation; and, confequently, was not ab/fo- 
lutely certain as to the equality of the temperatures of the con- 
tents of the bottles at the time when their weights were com- 
pared, I determined now to repeat the experiments, with fuch 
variations as fhould put the matter in queftion out of all 
doubt. 
I was the more anxious to affure myfelf of the real tempe- 
ratures of the bottles and of their contents, as any difference 
in their temperatures might vitiate the experiment, not only 
by caufing unequal currents in the air, but alfo by caufing, 
at the fame time, a greater or lefs quantity of moifture to | 
remain attached to the glafs. | 
Vou. V. Z Re 
