BRITISH ASSOCIATION MEETING. 
541 
increase of tenipcratare In llio earth's cnisi . Tlio experiments hitherto made 
give, unfortunately, somewhat eonflieting results, and even in the same 
mine the rate of increase of temperature is by no means uniform. This is 
shown very elcarly in the results obtained by Mr. Astley. It is scarcely pro- 
bable, however, that the temperature in tlie mine-shaft influenced the residts, 
and we must therefore seek the cause of this irregularity in the varying con- 
ducting power of the different strata, arising from different density, and 
different degrees of moisture of the strata. As to the rate of increase, they 
appear to confirm previous experiments, in which it has l)een shown tliat the 
temperature increases directly as the depth. The rate is at first rather less 
than this, afterwards somewhat greater, and at last again less, but on the whole, 
the straight line on which the temperature increases as the depths nearly ex 
presses the mean of the experiments. The amount of increase indicated in 
these experiments is from 51 dogs, to Syf degs., as the depth increases from 5f 
yards to 231 yards, or an increase of 1 deg. in 99 feet. But if wc take the results 
which are more reliable, namely those betweenthe depths of and G85 yards, we 
have an increase of temperature from 57t degs. to 75idegs., or 17? degs. Fahren- 
heit. That is a mean increase of 1 deg. in 70"8 feet. This rate of increase is not 
widely different from that discovered by other authorities. Walferdin and Arago 
found an increase of 1 deg. in 59 feet in the artesian well at Grenclle. At the salt- 
works at Kheme, where an artesian well penetrates to a depth of 7G0 yards, or 
rather more than the Dukinfield mine, tlie increase is 1 deg. in 5 1'7 feet. MM. de 
la Rive and Marcet found an increase of 1 deg. in 71 feet. In one respect the 
observations in the Dukinfield mine are peculiarly interesting. As they give 
the temperature in various descriptions ot rock, they appear to prove what has 
hitherto been partially suspected, namely, that the conducting powers of tlie 
rocks exercise a consideraole influence on the temperature of the strata. If 
wc add to this the influence of the percolation of water, we shall probably have 
a sufficient explanation of the irregularities observed in the experiments. 
From the above observations we have evidence of the existence in the earth of 
a central heat, the temperature, so far as can be ascertained, increasing in the 
simple ratio of the depth. We do not, however, presume to offer an opinion as 
to whether this increase continues to infinitely greater depths than we havc 
yet penetrated, as observations upon this point are still imperfect. But, as- 
summg as an hypothesis, that the law which prevails to a depth of 700 yards 
continues to oi)erate at still greater depths, we arrive at the conclusion that 
at a deptli of less than two and a half miles the temperature of boiling water 
would be reached, and at a depth of 10 miles a temperature of 3,000 degs. Fahren- 
heit, which we may assume to be sufheient to melt the most refractory rocks 
of which the earth's crust is composed. If, tlierefore, no other circumstance 
modified the conditions of liquefaction, all within a thin crust of this thickness 
would be in a fluid state. This, however, is not the case. At these depths 
the fusing point is modified by the pressure and conductivity of the rocks. 
We know that in volcanic districts, where the great subterranean laboratory of 
nature is partially opened for our inspection, the molten mass, relieved from 
fsressiire, pours forth from volcanic craters currents of lava which form a peeu- 
iar class of rocks. Besides this, it has been ascertained from Mr. Hopkins's 
experiments on soft substances, such as spermaceti, wax, and sulphur, that 
the temperature of fusion increases about 1'3 Farenheit for ever 5001b. pres- 
sure per square ineli, that is, in other words, that the temperature of fusion 
under pressure is increased in that ratio. If we assume this to be the law for 
the materials of the earth's crust, and correct our previous calculations in 
accordance M'ith it, we shall find that we have to go to a depth of ()5 miles, 
instead of merely 10, before the point of fusion of the rocks is reached. It 
must, therefore, be observed that Mr. Hopkins's later experiments with tin 
