Royal Institution. 311 



graduated arc which it crossed at the two places, would show how 

 much the direction of gravity at one place is inclined to the direction 

 of gravity at the other place.] Now from our knowledge of the 

 form and dimensions of the earth, we know that the direction of 

 gravity changes very nearly one second of angle for every 100 feet 

 of horizontal distance. Suppose then that two stations were taken 

 on Schehallien, one on the north side and the other on the south 

 side, and suppose that their distance was 4000 feet ; then, if the 

 direction of gravity had not been influenced by the mountain, the 

 inclination of the directions of gravity at these two places would 

 have been about 40 seconds. But suppose, on applying the zenith 

 sector in the way just described, the inclination was found to be 

 really 52 seconds. The difference, or 12 seconds, could only be 

 explained by the attraction of the mountain, which, combined with 

 what may be called the natural direction of gravity, produced 

 directions inclined to these natural directions. In order to infer 

 from this the density of the earth, a calculation was made (founded 

 upon a very accurate measure of the mountain) of what would have 

 been the disturbing effect of the mountain if the mountain had been 

 as dense as the interior of the earth. It was found that the dis- 

 turbance would have been about 27 seconds. But the disturbance 

 was really found to be only 12 seconds. Consequently the propor- 

 tion of the density of the mountain to the earth's density was that 

 of 12 to 27, or 4 to 9 nearly. And from this, and the ascertained 

 density of the mountain, it followed that the mean specific gravity 

 of the earth would be about five times that of water. The only ob- 

 jection to this admirable experiment is, that the form of the country 

 near the mountain is very irregular, and it is difficult to say how 

 much of the 12 seconds is or is not really due to Schehallien. 



The second class is what may be called a cabinet experiment, 

 possessing the advantage of being extremely manageable, and the 

 disadvantage of being exceedingly delicate, and liable to derange- 

 ment by forces so trifling that they could with difficulty be avoided. 

 Two small balls upon a light horizontal rod were suspended by a 

 wire, or two wires, forming a torsion balance, and two large leaden 

 balls were brought near to attract the small balls from the quiescent 

 position. We could make a calculation of how far the great balls 

 would attract the little ones, if they were as dense as the general 

 mass of the earth ; and comparing this with the distance to which 

 the leaden balls really do attract them, we find the proportion of the 

 density of the earth to the density of lead. The peculiar difficulty 

 and doubt of the results in this experiment depend on the liability 

 to disturbances from other causes than the attraction of the leaden 

 balls, especially the currents of air produced by the approach of 

 bodies of a different temperature ; and after all the cautions of Ca- 

 vendish, Reich and Baily, in their successive attempts, it seems not 

 impossible that the phenomena observed may have been produced 

 in part by the temperature of the great balls as well as their attrac- 

 tion. 



These considerations induced the Lecturer, in 1S2G, to contem- 



