TRANSACTIONS OF THE SECTIONS. 49 



variations in its temperature, first observed in the descent of the lead which covers 

 the south side of the choir of Bristol Cathedral, and communicated by me to the 

 Royal Society in the year 1855, I have since confirmed by the following experi- 

 ment. I fixed a deal board, 9 feet long and 5 inches broad, to the south side of my 

 house, so as to form an inclined plane, and upon it placed a sheet of lead, turning 

 its edges down over the edges of the board, and taking care that it should not bind 

 upon it, but be free to move with no other obstruction than that which arose from 

 its friction. The inclination of the board was 18° 32', the thickness of the lead was 

 one-eighth of an inch, and its weight 28 lbs. The lower end of the board was 

 brought opposite to an upper window, and a "vernier" was constructed which 

 could be read from within, and by w r hich the position of the lead upon the board 

 could be determined to the 100th of an inch. I began to measure the descent of the 

 lead on the 1 6th of February, 1858, and recorded it every morning between seven 

 and eight o'clock, and every evening between six and seven until the 28th of June. 



The lead had descended between the 1 6th of February and the 30th of April (a 

 period of seventy-four days) 10'Gl inches, being an average daily descent of "1433 

 inch. On the 4th of May it was drawn up the board again to its first position. 

 Between that date and the 28th of June (a period of fifty-five days) it had descended 

 11*97 inches, being an average daily descent of *217o" inch. Its descent was far 

 from uniform, being on some days scarcely (if at all) perceptible, and on others 

 amounting to nearly half an inch. The average daily descents in successive months 

 from February to June were in 



inch. 



February '10000 



March. .' '13806 



April -16133 



May *2 1500 



June *21888. 



Every variation in the temperature of the lead contributed to its descent. The 

 extreme temperatures of the day and the night could not therefore determine its 

 daily motion ; for with the same extremes of day and night temperature, there may 

 be great differences in the number and amount of the intervening variations of tem- 

 perature. It is the effect of these daily variations of temperature, up and down, by 

 which the descent of the lead totalizes. Although, therefore, we are to look for the 

 influence of the extremes of day and night temperature upon the daily descent, we 

 are also to look for that of the variations between each two extremes. I accordingly 

 remarked that it was on days when a thermometer in the sun varied its height 

 rapidly and often, that the lead descended most. On the contrary, when the sky 

 was open and the heat advanced and receded uniformly, the descent was less, although 

 the difference of the extreme temperatures might be greater. It was least of all 

 when there was continuous rain. During the night the descent of the lead was 

 often imperceptible. I have explained the descent of the lead in a paper published 

 in the ' Proceedings of the Royal Society ' for April 1855. 



If we suppose the sheet of lead to become ice, and its dimensions to be increased 

 20,000 times, and if for the board on which it rests we substitute a mountain side lying 

 at a slope of 18*°, we shall have a glacier H mile broad, 200 feet deep and 34 miles 

 long, and which in no other respect than as it regards its length will be an exaggera- 

 tion. By converting the lead into ice, its physical properties will, however, have been 

 in some important particulars changed. It will have become twice as dilatable as lead 

 is, that is, it will dilate twice as much by a given variation of temperature when un- 

 opposed as lead does*. It will have become a more elastic substance than lead is, that 

 is, it will be capable of overcoming a greater force opposed to its dilatation under a 

 given change of temperature. It will have lost its ductility and have become friable, 

 that is, its parts will have become more liable to separation from one another and 

 its mass to disintegration. But, together with the last-mentioned quality, it will 

 have acquired the property (called regelation) of easily, and under a moderate press- 

 ure, returning from a state of disintegration to one of solidity ; which qualities of 



* See the experiments of Schumacher at Pultowa, as detailed in the paper of W. Struve, 

 " Sur la dilatation de la Glace," Mem. de I'Acaderaie de St. Petersbourg, ser. 6. torn. iv. 1848. 

 1860. 4 



