coi Ohfervtttkiu and Enp/rinuntJ on 



pieviouny iried, by adding the powdered f.iUs gradually to the liquid till the thermometer 

 ccafed to fmk; obferving to produce the full eficclof one fait before a fecond is added, and 

 likewife of the fccond before a third is added. Neither foda, phofphorated foda, nor Glau- 

 ber's fait, fliould be mixed with nitrous ammoniac, or the powder compofcd of fal am- 

 nioniac and nitre, unlefs at a low temperature, that is, below o; but pounded ai-d kept apart. 



In the experiments alluded to in the table, the precaution of frelh-cryflallizing the falls 

 was not obferved, becaufc \ chofe to give the ordinary elTccts only. 1 therefore then ufcd 

 falts in their common ftate, taking care, however, to choofc fuch as had not in the leaJl 

 eflorefced. 



Since it is always ufcful, and generally abfoIutSly neceflary, to know how much room 

 in a veflel the feveral materials take up fcparately; and when mixed, it will be right to ob- 

 ferve, that fnow or ice in powder, at near o", occupy in meafure nearly two thirds more 

 than their weight ; that is, one ounce weight of water will, when in the form of fnow, or 

 ice ground to powder, nearly fill a vcflel which holds three ounces wine meafure; powdered 

 falts, nearly double their weight; ftrong nitrous acid, about three-fourths its weight; and a 

 mixture made of falts and diluted nitrous acid ; meafures rather lefs than two-thirds of tlia 

 weight of the ingredients. Witiiout a pre\ious knowledge of this, it is impofl'ible tu adjuft 

 the fize of the vcflels to the mixtures which are to be made ; becaufe, in moft nice experi- 

 ments of this kind, the height to which a vefTcl will be filled is indifpenfably necefTary to 

 be known beforehand. 



The long continuance of the late froft having afforded me opportunities of repeating 

 thefe experiments in various ways, I fliall mention briefly the refult of fuch as appear t« 

 mc to be material. 



I have found, that ice may be ground fo fine as to be equal to froren vapour ; and the 

 harder it is frozen, the finer it is ground, but with more labour. 



That quickfilver may be frozen by cooling the nitrous acid only, faving the trouble and 

 inconvenience of cooling the fnow likewife ; either by adding fnow at -|- 32 degrees, to 

 nitrous acid at — 29 degrees ; or fnow at -J- 25 degrees, to nitrous acid at — 20 degrees ; 

 or fnow at -j-20 degrees, to nitrous acid at — 12 degrees. Moft winters offer an oppor- 

 tunity of doing it in this way ; the nitrous acid may be cooled in a mixture of fnow and 

 nitrous acid. 



That it may likewife be frozen by mixing expedltloudy together fnow and nitrous acid, 

 when the temperature of each is + 7 degrees. 



Or by mixing ground ice and nitrous acid at -j- 10 degrees. 



Hence it follows, that the cold of this climate offers occafionally opportunities of freezing 

 quickfilver, without prcvioufly cooling by art the materials to be mixed ; for I have once 

 feen the thermometer at -|- 6 degrees — and others, I believe, have feen it lower. 



I expeded an opportunity would have offered this winter, but the lowed; point I faw my 

 thermometer at this feafon was only -f- 10 degrees. At this temperature I-mixed nitrous 

 acid (cooled out-of-doors to the temperature of the air) and fnow on January 23d laft ; 

 but the cold produced was not quite fufficient to freeze the quickfilver, although very near 

 It, as indicated by a thermometer. From what I have obferred fince thefe latter experi- 

 ments were made, I think it may be reafonably expedled that powdered ice and nitrous 

 acid at -f 14 degrees, or fnow at + 10 degrees, will fucceed, if mixed expcdiiioufly. 



Strong 



