Liquids do not possess great conductivity, but they, as well as gases, are influenced by _convection_, or the transport of heat from the bottom layers to the top (_conveho_, to carry up). We have already mentioned that the heated particles of water rise to the top because they expand, and so become lighter. This is convection of heat; and by it liquids and gases, though actually bad conductors, may become heated throughout to a uniform temperature. Of course the more easily expansible the body is the more rapidly will convection take place—so gases are more readily affected than liquids. Solids are not affected, because convection of heat depends upon molecular movement or mobility, and it is obvious that the particles of solid bodies are not mobile. Professor Balfour Stewart says with reference to this that “were there no gravity there would be no convection,” for the displacement of the light warm particles by the heavier cold ones is due to gravity. The instances of convection of heat in nature are numerous, and on a gigantic scale. The ocean currents, trade winds, lake freezing, etc., while the chimney draught already referred to, is another example; and in all these cases the particles of air or water are replaced by convection. In the case of the lake freezing the cold particles at the top sink, and the warmer ones ascend, until all the lake is at a temperature of 36·2°, or say 4° above freezing. At this temperature water assumes its maximum density, and then _expands_, as we have seen, instead of contracting. Ice is formed, and being thus lighter than water, floats; and so unites to cover in the water underneath, which is never frozen solid, because the cold of the atmosphere cannot reach it through the ice in time to solidify the whole mass. [Illustration: Fig. 84.—Radiant heat.] Radiant heat is the motion of heat transmitted to the ether, and through it in the form of waves. The sun’s heat is radiant heat, and radiation may be defined as “The communication of the motion of heat from the articles of a heated substance to the ether.” The fire gives out radiant heat, and so does heated metal, and it is transmitted by an unseen medium. It is quite certain that the heat of a suspended red-hot poker is not communicated to the air, because it will cool equally in a vacuum. Sir Humphrey Davy proved that radiant heat could traverse a vacuum, for by putting tin reflectors in an exhausted receiver he found that a hot substance in the focus of one reflector caused an increase in the heat of the other. If we put a red-hot or a hot substance in one reflector, and tinder in the other, the latter will take fire. The velocity of heat rays is equal to that of light, 186,000 miles in a second, and indeed, radiant heat is identical with light. Heat is reflected as is light, and is refracted in the same way as sound.