vapour issuing from it is made to pass through a coil of pipes surrounded by cold water at C. Thus vapour is no sooner formed at A than it rushes along the tube, and is immediately condensed and its place supplied by new vapour; and the rapidity of the process will only be limited by the supply of heat at A and the supply of cold water at C, for if there be a small supply of heat at A there will be a small formation of vapour, and if there be a deficient supply of cold water at C the vapour will not rapidly condense. In fact, the vapour in this arrangement acts as a vehicle by which the heat

is drained away from A to C. D is the pipe by which a constant supply of cold water enters the condensing vessel, and at E the heated water is carried off, which being lighter naturally rises to the top. Thus the outside of the coil, or worm as it is called, is kept constantly surrounded by cold

water.

The object of distillation is generally to free the liquid from certain impurities with which it is associated. These are left behind in A, and the condensed vapour of the liquid is caught in a vessel at B.

202. Ebullition.-When a liquid, such as water, is heated in the open air, it continues for some time to increase in temperature, and the evaporation becomes more and more

rapid. At length bubbles of vapour break out and reach the surface, and the process of boiling or ebullition has begun. When this takes place the temperature ceases to rise, and remains stationary until all the water has boiled away, the only difference being that if the supply of heat be very great the process is very rapid, and if the supply of heat be small the process is very slow. The point at which ebullition commences is called the boiling-point.

203. This point, in the first instance, depends upon the nature of the liquid. The following table contains the boiling-points under the ordinary pressure of the atmosphere of some of the most important liquids :

Kopp appears to have traced a definite relation between the chemical composition of certain substances and the temperatures at which they boil.

204. In the next place, the boiling-point of the same liquid depends upon the pressure, a liquid boiling at a lower temperature if the pressure be smaller.

This may be easily shown by means of a few simple experiments. First, exhaust a vessel containing ether under the receiver of an air-pump, and it will be found to boil at

the ordinary temperature of the air. Secondly, fill a Florence flask (Fig. 60) half full of water, and boil it in the open air until the upper part of the flask be filled with the vapour of water; let it now be corked tightly and inverted. When it has ceased to boil, pour some cold water upon the flask, and ebullition will again commence. The reason is, that the cold water, by condensing the vapour which fills the upper part of the flask, withdraws the pressure, and thus enables the water to boil at a lower temperature.

It follows from all this that, inasmuch as the atmospheric pressure at the top of a mountain is smaller than at the bottom, so the boiling point of water at the top of a mountain is lower than at its bottom.

Thus at the top of Mont Blanc water boils at 85° C. instead of 100°, a temperature which is too low for culinary purposes. Food cannot, therefore, at such altitudes be cooked in open vessels of water, but people are there compelled to heat water in a close vessel under the pressure of its own vapour, in order to obtain a temperature sufficiently high for cooking purposes.

It has been previously stated (Art. 89) that the height of a mountain can be estimated by means of the barometer, and we may now add that it can also be ascertained by observing the boiling-point of water by means of a thermometer. For, knowing as we do the relation between the boiling-point of water and the atmospheric pressure (Art. 212), an observation of the former enables us to obtain the latter. In fact, we read the boiling-point of water in order to deduce from it the atmospheric pressure, and there is therefore no advantage in using a boiling-point thermometer, except that it is a more portable instruinent than an ordinary barometer.

205. In the next place, the nature of the vessel affects the boiling-point of the fluid which it contains. Thus it has been found that the boiling-point of water is somewhat higher in a glass vessel than in a metal one. If, however, iron filings be dropped into the glass vessel, or, according to Tomlinson, anything capable of acting as a nucleus, the temperature of the boiling-point is lowered, and ebullition is promoted.

206. It would also appear that the air dissolved in the water has some effect upon its boiling-point, and M. Donny, by depriving water as far as possible of the air which it contained, and by enclosing it in a peculiarly-shaped vessel, was able to raise the temperature to 135° C. without ebullition. 207. If the liquid is not pure, but contains substances in solution, this will likewise affect the boiling-point. Thus the general effect of salt dissolved in water is to raise its boiling-point.

208. Spheroidal state.—If a drop of water be thrown upon certain surfaces at a very high temperature, it does not adhere to the surface, but moves about and evaporates without boiling. This peculiarity of liquids has given rise to some very curious experiments. For instance, M. Boutigny poured liquid sulphurous acid upon a platinum capsule heated to a white heat; yet this very volatile liquid did not boil, and its rate of evaporation was very slow. Faraday, again, poured upon a red-hot platinum capsule a mixture of ether and solid carbonic acid, which evaporated very slowly, and

nevertheless solidified some mercury brought into contact with it.

Want of contact appears to be the explanation of this behaviour, and M. Boutigny has found that in certain cases the light of a taper may be seen between the liquid and the surface.

It is no doubt this want of contact that prevents the heat from reaching the liquid sufficiently fast to cause it to boil, and the difference between contact and non-contact is well exemplified in Faraday's experiment: for when the intensely volatile mixture lay on the red-hot platinum capsule without boiling, there was clearly a want of contact between the two; but when the mercury was thrown into this mixture, it became immediately frozen, because it was brought into intimate contact with it.

209. We have seen that the transition from the solid to the liquid state is gradual in the case of many substances, and that there is an intermediate condition of viscosity in which the substance partakes of the character of both states. The experiments of Cagniard de la Tour, and more especially those of Andrews, lead us to believe that there is an intermediate state between the liquid and the gaseous conditions of matter. Thus, Dr. Andrews finds that if we heat liquid carbonic acid under great pressure in a closed tube, when we reach the temperature of 31° C. or thereabouts, the surface of demarcation between the liquid and the gas becomes fainter and fainter, loses its curvature, and at last disappears.

210. Sublimation.-Generally speaking, the order of things is, that, when the temperature is increased, the solid passes into a liquid, and finally into a gas, but sometimes the solid passes at once into a gas without assuming the intermediate state of liquidity. This is called sublimation, and we have instances of it in arsenic acid and solid carbonic acid, which pass at once into the gaseous state. Snow also slowly evaporates, and thus assumes the gaseous form even at temperatures much below its melting-point.

211. Change of Composition in Evaporation and Condensation. Sometimes if we heat a mixture of two liquids,