Figure by means of the Pendulum. 419 



at every station. This requires an accurate determination of 

 the length of the pendulum as well as the time of oscillation, 

 and also necessitates the measure of the vibratory movement 

 of the stand on which the pendulum swings. The English 

 content themselves with relative measures from one station to 

 another and simply compare the forces of gravity, by counting 

 the number of oscillations made by the "same pendulum in the 

 two places. They determine absolute gravity only at a base 

 station. It is evident that the latter method is immeasurably 

 superior in point of economy, and we may say that it can be 

 made to yield results fully equal in accuracy to those of its 

 rival school. Moreover the differential method has the great 

 advantage of eliminating all those sources of error that are 

 practically the same for each station. 



General Results of Pendulum Work. 



Knowing now which method best suits our purpose a still 

 further question arises. How accurately shall we do the work ? 

 Certainly not as accurately as we can. That would be bad 

 economy from every point of view. Nothing is gained by 

 measuring the force of gravity to its 1/lOOOOOtk part to deter- 

 mine local deflections when these variations themselves are 

 several times as great. J^o fact is more certainly demonstrated 

 than that certain places on the earth's surface present varia- 

 tions of the force of gravity quite exceeding anything to be 

 expected either from the amonnt or density of the adjacent 

 matter. These anomalies in many cases baffle all attempts to 

 classify them, but one general result seems to be that moun- 

 tains are light and islands are heavy. Some of the first, if not 

 the very first pendulum observations made, gave the strange 

 result that the Andes in Equador are not much heavier than 

 water. Foster's celebrated series on Green Mountain gave a 

 result indicating that this volcanic formation is about twice as 

 heavy as cork. The sacred mountain in Japan has given a 

 similar result, namely, that the mountain is lighter than would 

 be indicated by its volume and density. Haleakala in the 

 Hawaiian Islands seems to be of the same mean density as the 

 rocks on its surface. One observer has even gone as far as to 

 say that the Alleghany Mountains in Pennsylvania weigh less 

 than nothing, meaning by this, that if gravity at the summit 

 be corrected for elevation, the result is not more than gravity 

 at the base, showing the downward attraction of the mountain 

 to be practically insignificant. If we glance at a few island 

 stations, gravity is found to be mostly in excess of what it 

 ought to be. The most striking examples are Fernando, St. 

 Helena, Ascension, Minecoy, Isle of France, Bonin, Maui and 



