Figure of the Earth. 157 



of the place and its elevation above the mean level of the sea, 

 are not the only, nor even the chief, circumstances, that affect 

 the length of the pendulum ; and consequently, that the measure- 

 ment in any one place, even supposing it to be correctly made, 

 and reduced to the level of the sea, by an amount which should 

 not be arbitrarily assumed, does not determine the pendulum of the 

 latitude, because the nature and density of the substances, that 

 compose the upper crust of the earth at the place of observation, 

 have a most important bearing, and which cannot be neglected. 

 The clock and the experimental pendulum were found to be liable 

 to variations of not less than ten seconds per day in the same la- 

 titude, according to the nature of the materials upon which they 

 rested ; and, as all the observations were necessarily made upon 

 the land, it is inferred that an equal variation in an opposite di- 

 rection, might be considered as likely to occur if the experiments 

 could be performed at different points on the surface of the 

 ocean : the whole difference, then, that might arise from the ac- 

 tion of the different substances that are found on the surface of 

 the globe may, in the same latitude, amount to no less than twenty 

 seconds ; and the difference in the length of the pendulum at sta- 

 tions differing in local circumstances, but still under the same 

 parallel, might be equal to 0.01 of an inch ; or nearly one-tenth 

 of the whole difference of the intensities of gravity at the pole 

 and the equator, or to -o-xnro^h part of the absolute attraction of the 

 earth. It would thus appear, that before the mean force of gra- 

 vity, in any parallel of latitude, can be inferred with certainty, 

 numerous observations, indeed an almost indefinite number, ought 

 to be made in or near that parallel, to produce by their combina- 

 tion, a near result. 



With respect also to the allowance to be applied to the length 

 of a pendulum measured at an height above the sea, to reduce it 

 to what it would have been if measured at the level, it is shewn 

 that the correction which has been recently proposed, for the 

 error arising from the figure of the surface, by which the regular 

 decrease of gravity in proportion to the squares of the distances 

 from the centre is affected, may be safely neglected ; but that a 

 far greater uncertainty than from external conformation, and for 

 which it would be far more difficult to assign a specific correction, 

 is involved by the variable density of the materials, on which the 

 pendulum is raised above the surface of the sea. From these 

 considerations, Captain Sabine concludes that the pendulum of a 

 particular latitude cannot become a standard of reference, because 

 its length is not practically determinable ; that the pendulum of 

 a particular city, London for example, (whereby it is implied that 

 a length measured in one part of the city should be recoverable 

 by a measurement made in some other part of the city,) is open 

 to the same objections, though in a le.ss degree ; but that the more 



