Before accepting as a demonstrated conclusion this notion of a constant increase of temperature from the surface to the centre of our globe, it may be well to re-examine the facts which are relied upon as proving it.

That there is a general increase of temperature so far as we are able to go downwards in the earth's crust, there can, as we have seen, be no doubt whatever. Yet it may be well to bear in mind how very limited is the range of our observation on the subject. The deepest mines extend to little more than half-a-mile from the surface, and the deepest borings to little more than three-quarters of a mile, while the distance from the earth's surface to its centre is nearly 4,000 miles. We may well pause before we extend conclusions, derived from such very limited observations, to such enormous depths.

But when we examine critically these observations themselves, we shall find equal grounds for caution in generalising from them. There is the greatest and most startling divergence in the results of the observations which have been made at different points at the earth's surface. Even when every allowance is made for errors of observation, these discrepancies still remain. In some places the increase of temperature as we go downwards is so rapid that it amounts to 1° Fahrenheit for every 20 feet in depth, while in other cases, in order to obtain the same increase in temperature of 1° Fahrenheit, we have to descend as much as 100 feet.

Now if, as is so often assumed, this increase of temperature as we go downwards be due to our approach to incandescent masses forming the interior portions of the globe, it is difficult to understand why greater uniformity is not exhibited in the rate of increase in different areas. No difference in the conducting powers of the various rock-materials is sufficient to account for the fact that in some places the rate of increase in temperature in going downwards is no less than five times as great as it is in others.

VARIATIONS IN UNDERGROUND TEMPERATURES.

Again, there are some remarkable facts concerning the variation in the rate of increase in temperature with depth which seem equally irreconcilable with the theory that the heat in question is directly derived from a great, central, incandescent mass. M. Walferdin, by a series of careful observations in two shafts at Creuzot, proved that down to the depth of 1,800 feet the increase of temperature amounted to 1° Fahrenheit for every 55 feet of descent, but below the depth named, the rate of increase was as much as 1° Fahrenheit for every 44 feet. On the other hand, in the great boring of Grenelle at Paris, the increase in temperature down to the depth of 740 feet amounted to 1° Fahrenheit for every 50 feet of descent, but from 740 feet down to 1,600 feet, the rate of increase diminished to 1° for 75 feet of descent. The same remarkable fact was strikingly shown in the case of the deepest boring in the world—that of Sperenberg, near Berlin, which attained the great depth of 4,052 feet. In this case, the rate of increase in temperature for the first 1,900 feet, was 1° Fahrenheit for every 55 feet of descent, and for the next 2,000, it diminished to 1° Fahrenheit for every 62 feet of descent. In the deep well of Buda-Pesth there was actually found a decline in temperature below the depth of 3,000 feet.

Perhaps the most interesting fact in connection with this question which has been discovered of late years, is that in districts which have recently been the seat of volcanic agencies, the rate of increase in temperature, as we go downwards in the earth's crust, is abnormally high. Thus at Monte Massi in Tuscany, the temperature was found to increase at the rate of 1° Fahrenheit for every 24 feet of descent. In Hungary several deep wells and borings have been made, which prove that a very rapid increase of temperature occurs. The deep boring at Buda-Pesth penetrates to a depth of 3,160 feet, and a temperature of 178° Fahrenheit has been observed near the bottom. The rate of increase of temperature in this boring was about 1° for every 23 feet of descent. In the mines opened in the great Comstock lode, in the western territories of the United States, an abnormally high temperature has been met with amounting in some cases to 157° Fahrenheit. Although this is the richest mineral-vein in the world, having yielded since 1859, when it was first discovered, 60,000,000l. worth of gold and silver, this rapid increase in temperature in going downwards threatens in the end to entirely baffle the enterprise of the miner. The rate of increase in temperature in the case of the Comstock mines has been estimated at 1° Fahrenheit for every 46 feet of descent, between 1,000 and 2,000 feet from the surface, but as much as 1° Fahrenheit for every 25 feet, at depths below 2,000 feet.

The facts which we have stated, with others of a similar kind, have led geologists to look with grave feelings of doubt upon the old hypothesis which regarded the increase of temperature found in making excavations into the earth's crust as a proof that we are approaching a great incandescent nucleus. They have thus been led to enquire whether there are any conceivable sources of high temperatures at moderate depths—temperatures which would be quite competent to produce locally all the phenomena of volcanic action.

There are not wanting other facts which seem to point to the same conclusion: namely, that volcanic action is not due to the existence of a universal reservoir of incandescent material occupying the central portion of our globe, but to the local development of high temperatures at moderate depths from the surface.

DEPTHS AT WHICH EARTHQUAKES ORIGINATE.