the Point of Mercurial Congelation . *355 
at once to 434 0 , and a few minutes afterwards (C) fell to 360°. 
Imagining that a new mixture would now bring it very low, I 
made another, but in the mean time the inftruments had rifen 
greatly, and after handing in the frefh mixture (C) funk to 
374 0 , and (B) to 438°. I fhould have mentioned, that thefe mix- 
tures were double in quantity to thofe ufed in the former experi- 
ments ; inftead of glafs tumblers, they were made in pint 
bafons. 
I obferved alfo, that the mixtures feemed to grow thin fooner 
than common ; for I always made them of the confidence of 
pap. I added fnow at times, to thicken it, but found it had • 
very little effect, but rather decreafed the cold. It is with great 
diffidence I offer it as my opinion, that the temperature of the 
air was too cold, and that the quickfilver being nearly in a hate 
of congelation before plunged into the mixture, was inhantly 
frozen on putting the inhruments into them ; and as the quick- 
filver in the tubes muh have been of the fame temperature 
with that in the bulbs of the thermometers before the experi- 
ment, I fhould imagine, that when the quickfilver in the bulb 
was frozen folid, it communicated an addition of cold to that in 
the tube, and froze it alfo, which prevented its fubfiding as 
ufual ; for in other cafes, the contra&ion of the quickfilver, 
when folid in the bulb, was the caufe of the quickfilver fub- 
fiding in the tube ; but then the latter was fluid, for the cir- 
cumambient air was warmer than the degree at which quickfil- 
ver freezes, and the increafed cold was applied only to the bulb. 
The obfervations made before the experiment began, as related 
in the beginning of thefe remarks, (hew the quickfilver 111 the 
thermometer was congealing, and that (A) and (C) were 
adtually frozen. 
When 
