-og £)r. blagden’s Hijloty of 
44 obferved them in the night. I now made hafte to call fome 
44 of my friends, who came in time to fee evidently all this 
44 with me. They were the younger Dr. vogel, and Meft. 
44 weber, Wagner, and graumann. Toward eight o’clock 
44 the cylindrical piece began to foften in the lame manner as 
44 the former, and the other four foon followed. About eight 
“ they fell from the lurface of the glafs, and divided into many 
44 fluid (hilling globules, which were foon loft in the inter- 
44 ftices of the frozen mixture, and re-united in part at the 
44 bottom, being now exactly like common quickfilver.” 
In this experiment at Gottingen it is not a little furprifing, 
that the cold fhould have proved fufficient to effect fuch a con- 
gelation of the quickfilver. A mercurial thermometer would 
probably have flood lower than M. blumenbach’s of fpirits ; 
but how much cannot poffibly be determined, without knowing 
the ftrength of the fpirits, and after what method it was gra- 
duated. At all events, the difference could fcarcely have 
amounted to ten degrees, which would ftiil give the tempera- 
ture of the air at leaft 20° above the freezing point of quick- 
filver. Sal ammoniac with melting fnow produces 32 degrees 
of cold ; may we luppofe, that it has a power of increafing the 
cold nearly as much when it and the fnow are previoufly cooled 
below that point * ? But if fo, why was not the quickfilver 
congealed till after a period of feveral hours, lince other frigo- 
rific mixtures begin to a£l almoft immediately ? Befides, there 
was not here the appearance of adlion, which confifts in a folu- 
tion of the fnow, inftead of its freezing into a mafs. Or had 
the cold been greater where the mercury was placed than at M. 
blumejjbach’s thermometer ? The whole experiment remains 
* See Biihop yvatson’s Chemical Eflays, vcl. III. p. 138* 
