“All the solid bodies with which we are surrounded are constantly undergoing changes of bulk corresponding to the variations of temperature. * * The expansion and contraction of metals by heat and cold form subjects of serious and careful attention to chronometer makers, as will appear by the following statements:--The length of the pendulum vibrating seconds, in vacuo, in the latitude of London (51° 31′ 8″ north), at the level of the sea, and at the temperature of 62°, has been ascertained with the greatest precision to be 39·13929 inches: now, as the metal of which it is composed is constantly subject to variation of temperature, it cannot but happen that its _length_ is constantly varying; and when it is further stated that if the “bob” be let down ¹⁄₁₀₀th of an inch, the clock will lose 10 seconds in 24 hours; that the elongation of ¹⁄₁₀₀₀th of an inch will cause it to lose one second per day; and that a change of temperature equal to 30° Fah. will alter its length ¹⁄₅₀₀₀th part and occasion an error in the rate of going of 8 seconds per day, it will appear evident that some plan must be devised for obviating so serious an inconvenience.”[6] [6] “Noad’s Lectures on Chemistry,” p. 41. From these data it is readily seen that the variations in the rate of a pendulum as it is carried from the equator towards the north are sufficiently explained, without supposing that they arise from a peculiar spheroidal form of the Earth. Others have attributed the variable motions of the pendulum to increased density of the air on going northwards. That the condition of the air must have some influence in this respect will be seen from the following extract from experiments on pendulums by Dr. Derham, recorded in numbers 294 and 480 of the _Philosophical Transactions_:--“The arches of vibration _in vacuo_ were larger than in the open air, or in the receiver before it was exhausted; the enlargement or diminution of the arches of vibration were _constantly proportional_ to the _quantity of air_, or rarity, or density of it, which was left in the receiver of the air-pump. And as the _vibrations_ were _longer_ or _shorter_, _so_ the _times_ were accordingly, viz., two seconds in an hour when the vibrations were longest, and less and less as the air was re-admitted, and the vibrations shortened.” Thus there are two distinct and tangible causes which necessarily operate to produce the variable oscillations of a pendulum, without supposing any distortion in the supposed rotundity of the Earth. First, if the pendulum vibrates in the air, which is colder and therefore denser in the north than at the equator, it must be more or less resisted in its passage through it; and, secondly, if it vibrates _in vacuo_, the temperature being less, the length must be less, the arcs of vibration less, and the velocity greater. In going towards the equator, the temperature increases, the length becomes greater, the arcs increase, and the times of vibration diminish.