710 PHILOSOPHICAL TRANSACTIONS. [aNNO 1780. 



seems probable from a number of common appearances. We often find, after a 

 night of frost, tlie slates and other tiiinner parts about a house whitened with 

 hoar-frost, when the walls and more solid parts of the building remain quite free. 

 In like manner the smaller branches and twigs of trees often acquire this frozen 

 ornament, when the main branches and trunk remain naked for a long time; 

 and, in general, any thin or detached body, capable of being easily cooled, at- 

 taches hoar-frost the soonest. 



In favour of this general position, the following remarkable case lately oc- 

 curred. Between the public library and the buildings of the new court, there is 

 a long rail composed of bars of cast iron, but divided into 2 parts by 2 massy 

 stone pillars, which support the iron gate-way that leads into the garden. The 

 bars are about 6 feet high, and an inch square, and fastened with lead into a 

 stone parapet below in the usual way. A few bars much larger are set in among 

 the rest at regular distances, to give the rail more stability. On Sunday morning, 

 Feb. 13, when there was a slight frost accompanied with a fog, it was enter- 

 taining to observe how the hoar-frost had settled during the night on these bars. 

 Very little was to be seen on the fiat sides, but a great deal on the angles, by 

 which means from the top downward every bar was garnished with 4 fringes, 

 which made the whole rail look very gay and ornamental. Running the eye along 

 the foot of the bars near the parapet, it was observed, that the fringe of hoar- 

 frost on the corners stopptd short about 12 inches from the bottom, and that so 

 much of every bar was entirely free. Two bars next the house and 2 next the 

 library were likewise perfectly clear of it from top to bottom. One bar next the 

 pillar of the gate was quite free, and the 2d had contracted but little. The same 

 thing precisely may be said of the 2 bars contiguous to the other pillar. And 

 it was also observed, that the few thicker and stronger bars were less fringed at 

 the corners, and were quite free much farther above the parapet than the others. 



It is manifest, that during the night the air surrounding the bars must have 

 been constantly endeavouring to make them as cold as itself: while they, on the 

 other hand, resisted this change by drawing heat from every neighbouring source 

 which offered it, namely, from the parapet, from the pillars in the middle, and 

 from the pillars at both ends immediately adjoining to the library, and to the 

 house in the new court: for these bodies, from their great bulk, must have been 

 but very little cooled in the course of the night. Wherever the air seems to 

 have got the better in this struggle, as at the angles of the bars, wliich evidently 

 must be the parts the soonest cooled, there we find that the hoar-frost was de- 

 posited, but no where else. 



Several other instances were found quite of the same kind with that of the 

 rail. Among the rest, a figure of a unicorn in stone, which stands within the 

 college, had resisted the attacks of the air all to the tip of his horn, which ac- 



