Scientific Intelligence — Meteorology. 387 



Anotlior theory had also been put forward, wliich attributed the mo- 

 tion of glaciers to the expansion of water in the act of freezing after 

 it had filled, not the minute pores of the ice, but internal cavities of 

 considerable dimensions. But, since the temperature of the glacier, 

 at considerable depths, must be sensibly constant, how were new cavi- 

 ties to bo formed when existing ones had been once filled up ? The 

 author, regarding both this theory and the preceding one as untenable, 

 was thus led to examine how far the apparent objections to De Saus- 

 sure's theory were really valid, by a series of experiments on the 

 descent down inclined planes. The experiments were made in the fol- 

 lowing manner : — A slab of sandstone, prepared to be laid down as a 

 part of a common flagstone pavement, was so arranged as to be easily 

 placed at any proposed inclination to the horizon. The surface of 

 the slab, so fai* from being polished, retained the grooved marks of 

 the instrument with which the quarryman had shaped it. A quan- 

 tity of ice was placed on the slab, within a frame nearly a foot square, 

 intended merely to keep the ice together, and not touching the slab, 

 with which the ice alone was in contact. The following were results 

 obtained in one set of experiments, the ice being loaded with a weight 

 ofabout IdOlb.r- 

 Inclination of the planes, 3°, 6^ 9'', 12°, 15". 

 Mean space for one hour, 0.31, 0.6:^, 0.90, 2.0, 2.5 inches. 



When the weight was increased, the rate of motion was also in- 

 creased. The least inclination at which sensible motion would 

 take place, was not determined ; but it was ascertained that it could 

 not exceed half a degree in the case of a smooth but unpolished 

 surface. With a polished surface of a marble slab, the motion 

 of the ice indicated a deviation from horizontality, with as much 

 sensibility as water itself. It will be observed, in the results 

 above given, that (1.) the motion was unaccelerated, and (2.) it in- 

 creased with the inclination, and (when the inclination was not 

 greater than nine or ten degrees) in nearly the same ratio ; and (3.) 

 the rate of movement was of the same order of magnitude as in actual 

 glacial motion, which may be stated generally, in cases yet observed, 

 never to exceed two feet a-day. The extreme small friction between 

 the plane and the ice, indicated by the small inclination necessary to 

 produce motion, was manifestly due to the circumstance of the lower 

 surface of the ice being in a state of gradual disintegration, which, 

 however, was extremely slow, as proved by the small quantity of 

 water proceeding from it. In the application, therefore, of these re- 

 sults to the case of actual glaciers, it was necessary to shew that the 

 temperature of their lower surfaces could not generally be lets than 



