224. Convection. In the experiment described in last Article it is necessary to apply the source of heat to the surface of the water, and the reason of this is very obvious;

for when the heat is applied to the surface, the heated particles, being rendered lighter, remain where they are, and the heat can only reach the differential thermometer by means of conduction. But if the heat be applied to the bottom of the liquid the heated particles will ascend, and their place will

be taken by colder particles carried down from above, and the process will continue until the whole liquid is heated. This process, by which heat is conveyed to the various particles of a liquid, is termed Convection.

By introducing a little colouring matter, the direction of the fluid currents may be rendered visible.

Thus (Fig. 63) if we heat a vessel containing water, and drop into it a few fragments of cochineal, we shall find that the ascending currents go up by the centre, while the descending ones come down by the sides, their course being denoted by the arrow-heads in the figure.

225. Freezing of a Lake. In nature we have several illustrations of convection on the large scale. Let us begin with the case of a lake which is cooled at its surface.

The particles so cooled become specifically heavier and descend, and are replaced by lighter particles from beneath, so that in a short time the whole body of water has been subjected to the cooling agency. This process will go on until the temperature of 4° C. has been reached, which is the point of maximum density of water.

After this the process of convection goes on no longer, and if the surface particles be further cooled they become lighter, not heavier; they do not, therefore, descend, but remain at the top.

Had water contracted down to o°, and had ice been heavier than water, the ice would have fallen down to the bottom as it was formed, and the whole lake would soon have become one mass of ice. It would in such a case probably have remained frozen all the year round.

But as it is, the body of the water of the lake never attains a lower temperature than the point of maximum density, a temperature which is not destructive to life; while the coating of ice is confined to the surface, and becomes thickened only by the slow process of conduction.

226. Convection Currents in the Sun.-It may here be remarked that convection depends on two things. First of all we must have the force of gravity—an up and down, for it is in consequence of this force that a body specifically

lighter ascends, and were there no gravity there would be no convection. In the next place, the body must expand through heat, for if it hardly expands at all convection will be very feeble, and on this account the convection of mercury is much less than that of water.

Let us illustrate this with reference to the atmosphere of our luminary, where we have every reason to suppose there must be very strong convection currents.

In the first place, there are naturally great changes of temperature occurring in those regions; secondly, gas is a substance which expands greatly through heat; and thirdly, the force of gravity is there very great. We are therefore led to expect in the atmosphere of our luminary storms of terrific violence; and we find that such is really the case, for Lockyer has observed in the sun storms which were travelling at the rate of more than one hundred miles in a second.

227. Trade Winds, &c.-In our own earth we have notable examples of convection currents, for we have the vertical sun shining full upon the equatorial regions of the earth, in consequence of which there is a rising of the rarefied air, and a mounting of it into the upper regions of the atmosphere. The place of the ascending air is supplied by colder air from the poles on both sides, so that we have an under-current sweeping from the poles to the equator, and an upper current of heated air travelling above from the equator to the poles. The under-currents form the trade winds, the upper-currents the anti-trades.

Now, owing to the rotation of the earth from West to East, the under-current coming from the North Pole, or region of less rotation, into the equatorial regions, or those of greater rotation, will have, from the first law of motion, a tendency to lag behind or to fall to the west at the same time that it advances southward; the under-current from the north will thus become in reality a north-east wind. like manner the under-current in the southern hemisphere will be a south-east wind. The reverse will take place with the upper-currents or anti-trades, for these will travel from

In

a region of greater to one of less rotation; they will therefore be pushed forward in the direction in which the earth

rotates.

Hence the return trade which goes north will also go east, that is to say it will be a south-west wind; and the return trade going south will also go east, that is to say it will be a north-west wind. We have thus in the northern hemisphere the trades blowing from the north-east, and the return trades blowing from the south-west; while in the southern hemisphere we have the trades blowing from the south-east, and the return trades blowing from the north-west.

The land and sea breezes are probably due to similar causes. During day the land gets much more heated than the sea, and hence there will be an upper-current from land to sea, and an under-current from sea to land, the latter constituting the sea breeze. After sunset, however, the land cools more rapidly than the sea, and we then have an undercurrent from land to sea, constituting the land breeze.

LESSON XXV.--SPECIFIC AND LATENT HEAT.

228. In discussing the laws which regulate the distribution of heat, a very important element is the quantity of heat which a body absorbs when its temperature is raised, and also the quantity which is absorbed when a body changes its state.

The quantity of heat necessary to raise a body one degree in temperature is called its specific heat. Thus we define the specific heat of any substance to mean the quantity of heat necessary to raise one kilogramme of the substance 1o C., the unit being the amount of heat necessary to raise one kilogramme of ice-cold water 1o C.

Suppose, for instance, that a kilogramme of any substance required as much heat to raise it from 100° C. to 101° C. as would raise of a kilogramme of ice-cold water one degree, then we should say that the specific heat of the substance in question at the temperature of 100° was 0'4.

229. One of the simplest methods of measuring specific

heat is the method by mixture, which is best understood by a numerical example.

Suppose, for instance, that 3 kilogrammes of mercury at 100° C. have been mixed with one kilogramme of ice-cold water, and that the temperature of the mixture is 9o C.; find the specific heat of the mercury.

Let x denote the unknown specific heat. Then, since the mercury has been reduced from 100° to 9', we have thus a loss of 91° in 3 kilogrammes of mercury, which will be represented by 3 × 91. Also, the gain of heat by the water will be IX 9 (the specific heat of water being unity). Now, as the operation is supposed to be conducted so as not to lose any heat, it is evident that the loss of heat by the mercury will be equal to the gain by the water, and hence 3 x X 91 033 nearly, from which we see

=

9.'.x=

9 273

=

that the specific heat of mercury is only of that of water. Other methods of estimating specific heat have been devised. In one of these we reckon the quantity of melting ice converted into water in consequence of the hot substance parting with its heat. Another is the method by cooling, for when two substances are exposed to the same cooling influence it is clear that the one with the smallest specific heat will cool fastest, so that the velocity of cooling will afford a means of estimating the specific heat.

230. Specific Heat of Solids.-Dulong and Petit were the first to prove that the specific heat of solids is greater at a high than at a low temperature. They obtained the following results :