fome Thermo metrical Experiments. 517 
being fummer, betw een two- and three times the quantity of 
the fame ether mufthje to produce the fameeffect. 
There feems to ? he /fo^et l)in g /in t J^e ai^r which, belides heat, 
interferes with, tile treezing v of water,,- and perhaps of all fluids, 
though I cannot fay from? experience whether the above men- 
tioned difference between the freezing. of water in, winter and 
fummer, takes placer with -other fluids,, as milk and other 
animal fluids, oils, wines, &c* 
The proportion between the quantity, of the ether and of the 
water that maybe frozen by it, feems to vary according to the 
quantity of water ; for a larger quantity of water feems to 
require a propor.tionably lefs quantity of ether than a (mailer 
quantity of water, fuppofing that the water is contained in 
cylindrical glafs veflels ; for I have not tried whether a metal 
veflfel inftead of a glafs one, and whether fome other fhape 
belides the cylindrical, might not facilitate the congelation. In 
the beginning of the fpring I froze about a quarter of an ounce 
of water with nearly half an ounce weight of ether, the appa- 
ratus being larger, though fimilar to that deferibed above. 
Now as the price of ether, fufficiently good for the purpofe, 
is generally between eighteen pence and two (hillings ounce, 
it is plain, that with lefs than two (hillings a quarter of an 
ounce of ice, or ice cream, may be made in every climate, and 
at any time, which may afford great fatisfadtion to thofe per- 
fens, who living in places where no natural ice is to be had, 
never faw or tailed any fuch delicious refreihments. 
When afmalTpiece of ice, for inftance, of about ten grains 
in weight, is wanted, the neceffary apparatus is very fmall, 
and the expence ^of the ether not worth mentioning. I have 
a fmall box, which is four inches and a half long, two inches 
abroad, and one inch and a half deep, which contain s all the appa- 
ratus 
