358 EARTH that the distance between these two small spheres (about 6 ft.) is equal to the distance between the two large ones. When the rod bearing the two small spheres is in as nearly perfect equilibrium as possible, the bar bearing the globes of lead is rotated on its vertical axis until these globes are brought nearly into perfect contact with the small balls on oppo- site sides, as at / and g. Their attraction on these balls being thus called into play, they tend to draw the light rod from its position of rest. The amount of the torsion thus pro- duced in the supporting wire is observed through a telescope placed some distance off, so as to avoid disturbing influences. ^ Then the bar is turned round in a contrary direction until the large balls are again nearly in con- tact with the small ones, as at h and &, so that the fine rod is swayed in a contrary di- rection from its position of rest, and the tor- sion thus caused is observed as before. The mean of the two results indicates the ac- tual amount of torsion which the attraction of the two large globes is capable of produ- cing. The experiments of Cavendish gave the attractive force exerted by two leaden spheres, each 174 Ibs. in weight, as equivalent to ^^nr of a grain weight, and he thence computed the density of the earth to be 5 '48 times that of water. Reich of Freiberg, in two series of experiments, made the density 5*438 and 5'582. The late Francis Baily made more than 2,000 experiments by this method, and deduced from them a density equal to 5*660. It is worthy of notice how closely the results obtained by Michell's method (or, as it is called, the Caven- dish experiment) agree together. The infer- ence appears to be that this method is more trustworthy than any of those before described. It is also remarkable that Newton had stated in his Principia that probably the mean den- sity of the earth is five or six times that of water ( Verisimile est quod copia materice totivs in terra, quasi quintuple vel sextuplo sit quam si tota ex aqua constaref). Sir John. Herschel ("Outlines of Astronomy," llth ed., p. 559) thus sums up the evidence hitherto obtained respecting the earth's density and mass : " The final result of the whole inquiry will stand as below, the densities concluded being arranged in order of magnitude : Density Scbehallien experiment by Maskelyne, calculated by Playfair * 4.713 Carlint, from pendulum on Mont Cenis (corrected by Giulio). . v '4.950 Col. James, from attraction on Arthur's Seat 5-316 Belch, repetition of Cavendish experiment .... . . 5-438 Cavendish, result 5-480 corrected by Mr. Baily's re- cotnputation 5-448 Baily's repetition of Cavendish experiment ...... .. 5-660 Airy, from pendulum in Harton coal pit 6-565 General mean 5.44! Mean of greatest and least 5-689 Calculating on 5 as a result sufficiently ap- proximative and convenient for memory ; ta- king the mean diameter of the earth considered as a sphere at 7,912-41 m., and the weight of a cubic foot of water at 62-3211 Ibs., we find for its solid content in cubic miles 259,373 mil- lions, and for its weight in tons of 2,240 Ibs. avoirdupois each5,842trillions(=5,842 x 10 18 )." The low specific gravity of the earth, com- pared with that which might be expected from the enormous pressure to which her interior parts are subjected and the compressible nature of their materials, has led some men of science to the conclusion that the temperature of the interior is sufficiently high to exert an im- portant counteracting influence. The princi- pal motions of the earth are her rotation on her axis and her revolution around the sun. The main proofs of both these motions are found in the results of astronomical observation. The revolution of the earth around the sun is in particular placed beyond all possibility of ques- tion by the phenomenon called the aberration of light, which affects every star in the heavens, and is at once explained on the hypothesis of the earth's motion ; while it remains absolutely without explanation, and even without the possibility of explanation, if the earth be re- garded as at rest. And when the revolution of the earth is once admitted, her rotation must be admitted at the same time; for it would ob- viously be absurd to regard the sun's apparent annual motion around the heavens as due to a real motion of the earth around the sun, while at the same time the sun's apparent diurnal motion around the heavens was regarded as due to a real motion of the sun around the earth. But there are certain proofs of the earth's rotation which may be regarded as in a sense terrestrial, since they have no reference to the celestial bodies. Among these we may mention Foucault's experiments with the gyro- scope and pendulum. The gyroscope as applied by him to this problem is a heavy disk or ring rapidly rotating in its medial plane, and so sus- pended as to be free to turn in any direction. Such a disk or ring tends to preserve the plane of rotation unchanged in position, and if the earth were not rotating would remain unchang- ed under the closest scrutiny. But as the rota- tion of the earth tends to change the position of the medial plane of the rotating body, the ten- dency before alluded to is called into action and causes the medial plane apparently to shift, while in reality it is only maintaining its po- sition against the effects of the earth's rotation. The result is, that when carefully examined through a telescope the rotating disk or ring is seen to shift steadily in a direction opposed to that of the earth's rotation. Foucault's pendulum experiment depends on similar prin- ciples: the "plane of swing" of a pendulum tends to resist any motion by which it would be caused to take up a position intersecting its former one. Now, if a pendulum be set swinging in a north and south plane, a tangent to the arc of swing at its lowest point passes when produced through a point on the pro- duced axis of the earth ; and if the pendulum
