TRANSACTIONS OF THE SECTIONS. 23) 
To arrive at a general conclusion as to the accuracy attainable with this micro- 
meter, agreeably to the sixty-five observations here given, we shall find that for the 
mean power 152°7, and mean distance 7'67, the probable error of one observation, 
based upon five measures, will be 0’°086. 
Also, if we suppose the accuracy to increase in direct proportion with the power, 
the extreme difference to be expected amongst five such observations, taken with a 
magnifying power of 200, will be 0'"21. 
On the Distribution of Heat in the Interior of the Earth. 
By Dr. F. A. Strsestr6M, of Stockholm. 
Both the plutonic and volcanic phenomena are generally ascribed to causes residing 
within the earth’s solid crust, whilst the interior fluid mass is taken into con- 
sideration only with regard to what Humboldt calls its “ reaction.” Without dis- 
puting this way of explaining the phenomena—though I really think it liable to 
several grave objections—I wish to call attention to a cause of dilatation (and con- 
traction) that certainly exists, or at least has existed within the fluid mass itself, and 
which, I think, must be considered as an important item in this question. 
It seems impossible originally not to suppose different temperatures in different 
parts of the fluid earthy mass. Considering the great absolute temperature, as well 
as the immense bulk of the earth and other circumstances, differences as great even 
as 100° C., nay more, can in no way be regarded as improbable. The natural con- 
sequence thereof was the formation of currents, by means of which differently heated 
parts were brought together, and the temperature of the mass was made more and 
more uniform. However, on various grounds I conclude that even now the tem- 
perature cannot be one and the same through the whole fluid nucleus of the earth, 
but that currents of the said description still exist in the interior of our planet. 
This assumed, let v, v’ be the volumes, and ¢, #’ the temperatures of two differently 
heated fluid parts, which are mixed together, and which, for more simplicity, may 
be regarded as having the same mass and as being of the same chemical nature. 
Let, further, w be the volume and T the temperature of each after the mixture. As 
both the dilatation and the specific heat change with the temperature, let A, e be 
the mean dilatation and the mean specific heat between the temperatures ¢ and T, 
and A’, e’ between T and ¢ (¢ being > #’). Hence it follows that 
v =w(t+A¢—T)) 
=w(e—A'(T— t’)) 
(t—T)e=(T-?'e’, 
and consequently that "\(Ad~A'e) 
(ea w (t—?) (Ae’'—A’e 
v+y=2w+ ata F 
Now v-+»' being the original volume and 2w the volume after the mixture, it will be 
seen that there must needs be a change of volume, unless Ae'=A'e, which at least is 
not the case with the substances, for which Dulong has determined the variations of 
dilatation and specific heat. 
If we take as an example the values found by Dulong for iron between 0°— 200°, 
it will be seen that if only Paneaes of the earth’s volume were subject to the above- 
named process of mixture—one part being considered 200° warmer than the other— 
the result would be a change of volume certainly not less than five or six times the 
whole bulk of Vesuvius. This may in some way be illustrative of the quantity of 
action that might be supposed. As to the absolute intensity and the mode of work-” 
ing, I think no other force could be imagined more suited to the purpose, 
An Account of some Experiments on Radiant Heat, involving an Extension 
of Prévost’s Theory of Exchanges. By B. Srewarr. 
These experiments were performed with the aid of the thermomulti plier, the source 
of heat being for the most part bodies heated to 212°. Four groups of experiments 
were considered. Group the first contains those experiments in which the quantities 
