4 14 PHILOSOPHICAL TRANSACTIONS. [ANNO 1783. 



only ; yet it did not. He added more spirit of nitre, but without effect. At 

 JO h 3 rn 35 s he took out the apparatus, and raised the bulbous end to make the 

 quicksilver run, but found it was frozen, so that it did not alter its figure in the 

 least. He then placed it in the mixture, where it continued till 10 h ll m , when 

 he made another trial as before, but without perceiving any alteration : however, 

 to be more certain of its being frozen, he proposed to take out the thermometer; 

 but all the strength in his fingers could not move it in the least, so that he and 

 the officers, who stood by, were convinced it was frozen fast. 



The thermometers used on this occasion were those marked a and p. 



A 2d experiment was made the next day, Dec. \6, with the same instruments 

 and artificial mixtures. Mr. H. was rather unfortunate in making too small a 

 quantity of the freezing mixture at the beginning, which obliged him to make 

 repeated additions to it : by this means the operation was not only retarded, but 

 sometimes it even undid what had been done before ; for in pouring in the nitrous 

 acid part of it unavoidably came in contact with the bulbs of the instruments 

 before it was mixed with snow. In this case it never failed making the thermo- 

 meters rise suddenly much higher than where they stood before the spirit was 

 added -, and at length it only descended to 206°, which is not half so low as on 

 the preceding day, though the temperature of the air was 10° colder, viz. 34°: 

 yet it is remarkable, that though the thermometer was so much higher, the ap- 

 paratus was sunk more than twice as low as the day before ; for after having 

 been lontj stationary at 40°, it sunk to Q5°. He then made a fresh mixture, but 

 it had no effect any way during three quarters of an hour he attended to it after- 

 wards. During this idle interval he made the 3d experiment. Finding no 

 alteration, he went down to breakfast, and on returning, was surprized to find 

 the quicksilver in the apparatus thermometer had subsided into the bulb, and the 

 standard thermometer had been very low, and was rising briskly. The spirit 

 thermometer also showed the mixture had a less degree of cold than before. To 

 be certain that the quicksilver in the apparatus thermometer was in the bulb, he 

 took the apparatus out of the mixture, and examined it minutely for half a mi- 

 nute, till he was quite certain of it ; and also that the quicksilver in the cylin- 

 der was frozen, and it is remarkable it did not liquify in all that time. 



The 3d experiment was made during the continuance of that which imme- 

 diately precedes it, and was the effect of chance ; for the first freezing mixture, 

 which had been used in the 2d experiment, standing in the glass close by him, 

 he took down the thermometer g, and charged its cylinder with quicksilver, as 

 in the other examples, and suspended it in the old mixture, with the mercurial 

 thermometer b, and a spirit thermometer; the mixture seemed to have lost 

 much of its coldness, as appeared by the thermometers. It seemed very extra- 

 ordinary, that the apparatus, after having been so long stationary at 43°, should 



