j- 1 5 Mr. c .wallo’s Account of 
ter, the operation muft be continued one or two minutes 
longer ; after which the wire H will be found to be kept very 
tight by the ice. Now the bottle with the ether is left upon a 
table or other place, and to the outfide of the glafs tube the 
hand mull be applied for a moment, in order to loften the fur- 
face of the ice, which adheres very firmly to the glafs, and 
then pulling the wire H out of the tube, a folid and hard piece 
of ice will come out, fattened to its lpiral extremity. 
Inftead of the wire H fometimes I put a fmall thermo- 
meter into this tube fo as to have its bulb immerfed in the 
water. With this thermometer I have obferved a very re- 
markable phenomenon, which feems to be not explicable 
j n the prefent ttate of knowledge concerning heat and cold. 
This is, that water will freeze in the winter with a lefs de- 
gree of cold than it will in the fummer, or when the wea- 
ther is hotter : for inftance, in the winter the water in the 
tube AB will freeze when the thermometer is about 50° ; but 
fummer, or even when the tempeiature of the atmo- 
sphere is about 6o°, the quickfilver in the thermomer muft be 
brought ten or fifteen, or even more, degrees below the freezing 
point, before the water which furrounds the faid thermometer 
will be converted into ice, even fuperficially ; hence it appears, 
that in the fummer time a greater -quantity of ether and longer 
time is required to tree z a given quantity of watei than in the 
winter, not only becaufe then a greater degree of heat is to be 
overcome, but principally becaufe in- the fummer a much greater 
degree of cold muft be actually produced before the water that 
is kept in it will afl'ume a folid for ni. When the temperature 
of the atmofphere has been about 4b 0 , I have froze a quantity 
of water with an equal weight of good ether, but at prefent. 
