The temperature of the terrestrial surface perpetually varies under the influence of local as well as general causes. Had this fact been known to the philosophers of antiquity, they would have taken advantage of it to liken the earth to an animal whose skin is more or less sensible of heat, not only according to the difference of the seasons, but according to the different hours of the day.
The diurnal thermometrical variations are those which penetrate the least profoundly into the interior of the soil. At a depth of about five feet they cease to be perceptible. The maxima and minima of the year, however, can be detected at a sufficiently considerable depth. The limit descends as low as 80 to 100 feet. Below 100 feet, the terrestrial stratum is found invariable,--that is, inaccessible to the thermometrical changes of the atmosphere. It is remarkable that the temperature of this stratum differs but little from the mean annual temperature of the air, which, in the latitude of London, is 49°.
The maxima and minima of the yearly heat are propagated very slowly in the earth, and their difference gradually becomes less and less. Thermometers buried 26 feet deep in the ground, mark, in our latitudes, the maximum of temperature only on the 10th of December, and the minimum on the 15th of June. But there are certain elements which we must take into account. Thus, the depth of the invariable thermometrical curve depends both on the latitude of the place, on the conductibility of the strata, and the difference between the highest and lowest temperature of the year. The less this difference, the more nearly does the invariable stratum approach the surface. Here we have the reason why, in the intertropical torrid zone, where the temperature scarcely varies above two or three degrees in the whole year, the invariable curve is not found more than forty centimetres beneath the surface.
In the temperate zone, lying between the torrid and the frigid zones, the same phenomena assume, apparently, a more complex character; the isogeothermal lines inflect as diversely as the isothermal, and the former are far from running parallel with the latter. And this is easily understood, even without any experiment, for there is no relation between the ever-varied composition of the terrestrial strata, and the much more uniform composition of the atmosphere.