HISTOEY OF PHYSICAL ASTKONOMY.

157

I

teen different stations included between 10°.38 / north latitude and 62°.56'south latitude. All the details of these experiments appear to have beenconducted with the most scrupulous regard to accuracy, and the resultsdeserve to be classed among the most valuable data we possess fordetermining the ellipticity of the earth by means of the pendulum. Mr.Baily obtained f° r ^ ie value of the ellipticity as the final result

deducible from Captain Foster’s experiments. He also arrived at anellipticity equal to by combining together the results of all the

experiments hitherto made with invariable pendulums. These evaluationsagree very nearly with each other, and also with that due to ColonelSabine, whose experiments were more extensive than those of any othervoyager, previous to Captain Foster.

The experiments with the pendulum generally agree in assigning to theearth an ellipticity somewhat greater than that derived from the com-parison of arcs of the meridian. In the one case the ellipticity may besaid to be ; in the other case only The perturbations of the

moon again indicate an ellipticity equal to The agreement of the

second and third of these values is sufficiently satisfactory. It is difficult,however, to account for the discordance of the first and second; and alsofor the anomalies which present themselves in some instances when onlytwo arcs of the meridian or two experiments with the pendulum form thegrounds of comparison. The results of recent geodetic operations concurin shewing that the meridional anomalies are mainly attributable to theeffects of local disturbance, and that, when the amount of this disturbancecan be in any case independently ascertained, and then eliminated fromthe measurements, the comparison of the latter may be satisfied by anellipticity a little greater than *. With reference to the anomalieswhich occur in pendulum experiments, it appears that the results generallyindicate the intensity of gravity to be greater on islands than on con-tinents. This is partly explained by the greater density of islands whichare frequently of volcanic origin; but, on the other hand, it is not to bedenied that the density of the surrounding ocean is considerably less thanthe mean density of the earth, and therefore ought to produce an effect of anopposite kind. Much doubtless will depend on the depth of the ocean atthe place where the experiments are made, and the nature of the stratumon which it rests ; for it must be borne in mind that, although Clairaut ’stheorem allows the density of the terrestrial spheroid to vary from onestratum to another, according to any law whatever, it supposes thedensity to be uniform throughout each stratum. The influence exercisedon experiments with the pendulum by the geological character of thesubstratum was first remarked by Colonel Sabine.

When we consider how difficult it is to form a just estimate of theirregularities of the earth’s surface, and the variations in the density ofthe exterior stratum, we have ample reason to be surprised that theanomalies we have been referring to are not greater than they really are.“If the form of the earth’s meridian,” says Mr. Airy, “were traced onpaper, the nicest eye would be unable to distinguish it from a circle. The

* Some idea of the effect of local disturbance may be formed when we state that, at thestation of Banog, within fifty miles of the Himalayah Mountains, Colonel Everest foundthe difference between the computed and observed azimuths to amount to as much as20".156. It is true that the measurement of azimuth angles is the most delicate part ofgeodetic operations; but, in the present instance, the discrepancy manifestly exceedsthe probable error of observation.