130 PHILOSOPHICAL TRANSACTIONS. [ANNO 1781. 



No aurora this evening; the air very still and serene till about 2 o'clock. Tues- 

 day morning, when the wind rose remarkably, and clouds formed in the north- 

 east. 



On Thursday, January '25, the difference of _ 



P , , , , Thurs. morning. Therm. Therm, 



temperature was found to be as here set down h. m. " in air. in snow. 



in the margin. 9 43 + 10 + 3 



, . 10 . . . . + 10 +3 



From 10 till 11 o'clock this forenoon the 10 30...'.".!+ 14." '. + i 



thermometer on the ballustrade in air, 6 inches n + 1+ + 8 



. . , 11 30 + 17 + ') 



above the snow, pointed to + 14, and when 12 + 2 o . . + 12 



tried on the snow to + 10. About noon this '2 30 + 22 + 20 



. , , r 1 1 • 1 u ' aftern. +2.5 + 2fj 



day some clouds were formed, which became 1 30. ..... + 27. . -+- °7 



quite general by 1 o'clock. 



During the last 2 times of observing, 3 experiments were made with a view 

 of discovering whether the snow without doors was gaining any thing from the 

 air; or if any of it was carried off in the way of evaporation. For this purpose, 

 a shallow dish, made of sheet brass, 4 inches in diameter, was exactly filled with 

 snow, and carefully weighed. In order to defend the outside of the dish from 

 the air, that no hoar-frost might attach itself to the metal, a circular hole was 

 cut in the lid of a paste-board box, so wide, as just to let in the dish to the very brim, 

 so that nothing communicated with the external air but the snow itself. The 

 apparatus, in this state, was set without doors for 3 hours each time, and then 

 brought in to the lobby of the observatory, where the dish was again weighed : 

 but in none of these trials did it ever appear that any weight was lost. On the 

 contrary, at the first weighing, which was on Monday night, 12 o'clock, it had 

 gained 5 grains. In the other 2 trials the increase of weight was scarcely 

 perceivable. 



The temperature of the air in the west room of the observatory remaining 

 very constantly for nearly 2 days at -+- 27°, a dish of snow, similar to the other 

 exposed there, was found to lose weight very sensibly, and for the most part 

 at the rate of 2 grains in an hour. Notwithstanding this, the snow thus wasting 

 or evaporating had no power of sinking the thermometer below -f 27, the tem- 

 perature of the surrounding air; though at one time it was fanned for 4 minutes 

 by a piece of paper fastened to the end of a long stick. Not to disturb the 

 uniform temperature of this room during these experiments, care was taken to 

 stay in it a very short time at every visit, and to keep the door and the window- 

 shutters close. 



On Christmas-day there was a frost, which in the morning made the thermo- 

 meter in air point to -(- 21 ; and during the preceding night there had been a 

 profuse deposition of hoar-frost. A pound of this was collected, and ils capacity 

 for heat compared with that of ice, and found equal as nearly as could be judged. 



