Now (by 1) the weight of the balloon filled with dry air was (7) The difference gives the weight of the empty balloon . The o2 cubic inch of residual air measured at 57°, and at 29.1 inches barometer, would become at 212° and 29′1 barometer 8.387 1187.20 8.39 1178.81 Cubic inch. 0*260 291 Grain. 0'08 Cubic inches, 27.676 0.260 at 27'416 The difference gives the space occupied by alcohol vapour at 212 and 29'1 barometer . Now 27.416 cubic inches of vapour, measured at 212° and 29′1 barometer, if they could exist uncondensed as vapour, at 60°, and at 30 inches barometer, would become 20'569 Grains. the 1189-20 But (by 2) the united weight of the balloon, the vapour, and the residual air was found to be Deduct the weight of the residual air (8) The difference gives the weight of the balloon and vapour alcohol} The difference gives the weight of 20:569 cubic inches of alcohol vapour at 60°. 10:31 100 cubic inches of alcohol vapour would therefore be 50 123 grains at 60° and 30 inches barometer. Now, 100 cubic inches of air are 30'938 grains, at 60° and 30 inches barometer, therefore 50 123 divided by 31 gives 1.620 as the density of the vapour of alcohol, from the foregoing experiment. Deville and Troost (Comptes Rendus, 1857, xlv. 821, and 1859, xlix. 239) have extended this method of Dumas to the determination of the density of vapours of bodies of very high boiling point. They employ light vessels of porcelain, instead temperature of sealing, and to add this difference to the capacity as calculated above. When the temperature at sealing is very elevated, this correction acquires some importance; but it is insignificant in most cases, more especially as the vapour densities obtained by experiment never coincide accurately with the theoretical results, and a fair approximation is generally all that is required to indicate the state of condensation of the constituents of the compound. Since the coefficient expansion of flint glass between 32° and 212° F. (0° and 100° C) is equal to 000228366, the increase in capacity of the balloon in the foregoing experiment between 57° and 212° F. is 0'046 cubic inch. 148.] EQUILIBRIUM OF TEMPERATURE. 315 of the glass balloon, and seal the exit tube by means of the oxyhydrogen jet. The constant temperature at which the density of the vapour is determined, is obtained by placing the porcelain vessel in the vapour of a body which boils at a much higher temperature than the substance subjected to experiment; the distillation of the body which furnishes the vapour-bath being conducted in vessels of iron. Boiling mercury, for example, gives a vapour of constant temperature of 662° (350° C.); the vapour of boiling sulphur is estimated at 824° (440° C.); that of cadmium at 1580° (860° C.) ; and that of zinc at 1904° (1040° C.). But since the determination of these high temperatures is liable to some uncertainty, a comparative experiment is made in a separate porcelain vessel, by employing a substance like iodine, which furnishes a very dense vapour, the density of which at measurable temperatures is exactly known. The experiments thus give the direct relation between the density of the vapour under trial, and that of iodine at the same temperature. By employing an iron bottle Dewar and Dittmar (Proc. Roy. Soc. 1873, xxi. 203) have determined approximately the density of potassium vapour. The interior of the bottle was first deoxidized by heating it to redness, whilst a current of hydrogen was passed into it; it was subsequently placed in a bath of melted zinc, and 200 grammes of pure mercury introduced. When of the mercury had distilled off, an iron test tube, containing 4 or 5 grammes of potassium, was dropped into the bottle, and, after the vapours of potassium had ceased to escape, the orifice of the bottle was closed. When cold, the bottle was opened under water, and the hydrogen produced by the action of the potassium on the water was pumped out by a Sprengel, and measured. this way it was determined that the vapour of potassium is not more than 45 times as dense as an equal volume of hydrogen, the theoretical density being 39. § II. ON THE EQUILIBRIUM OF TEMPERATURE. In (148) All bodies, when heated, return sooner or later to the temperature of surrounding objects; the tendency of heat being constantly to preserve or recover an equilibrium. This balance is restored either by the process termed conduction, that is, by transmission of heat from particle to particle; or by convection, or the motion amongst the particles of liquids or gases; or by radiation between bodies at a distance from each other. 316 EQUILIBRIUM OF TEMPERATURE. Conduction. [149. (149) If we place the end of a short strip of glass and of a strip of metal, of equal length, in the flame of a lamp, we shall soon be sensible that heat reaches the fingers more rapidly through the metal than through the glass; and shall have a convincing proof that these two substances differ greatly in their rate of conducting heat. Of all known substances, metals possess the greatest conductivity, but even they differ considerably when compared with each other. It may be taken as a rule, although it is liable to numerous exceptions, that the greater the density, the greater the conductivity. Despretz, many years since, and Langberg, as well as Wiedemann and Franz, have more recently published a series of experiments upon the relative conductivities possessed by different solids. In the experiments of the observers last named (Pogg. Annal. 1853, lxxxix. 497), bars of each substance similar in dimensions were exposed at one extremity to a constant source of heat, and the progress of the temperature along each bar was measured, at intervals of 2 inches, by means of a thermo-electric pair. They concluded that the conductivities for heat in metals follows the same order as their electrical conductivities. According to J. D. Forbes, the conductivity of wrought iron for heat diminishes considerably as the temperature rises, and a similar diminution in the conductivities of metals generally for electricity has been ascertained to exist as the temperature rises (276). Calvert and Johnson (Phil. Trans. 1858, 349) have investigated the conductivity of the metals by a still more direct method. Their plan of operating consisted in employing two vessels made of vulcanized caoutchouc, on account of its feeble conductivity. The bars of the metals under trial were each 6 centimetres long, and I centimetre square. Each bar in succession was passed through an opening in one of the sides of each vessel into which it projected one-sixth of its length, the intervening portion being covered with vulcanized caoutchouc. A given quantity of cold water sufficient to cover the bar was then introduced into one of these vessels, and the temperature accurately observed; into the other vessel a given quantity of water at about 90° C. was introduced, and the temperature was maintained steadily at this point for 15 minutes by the occasional injection of steam in sufficient quantity. At the end of this time, the temperature of the colder vessel was noted. A comparison of the rise of temperature experienced in this vessel when bars of different metals were employed in succession, furnished the relative conductivities, correction being made for the loss of heat by radiation and transfer from one vessel to the other during the experiment.* It is to be regretted that the authors did not test the accuracy of their method by repeating their experiments with bars of the same metals of a dif The preceding table gives some of the results obtained in this way, compared with those of Wiedemann and Franz. In the experiments of Calvert and Johnson, the platinum, aluminium, iron, and sodium employed were ordinary commercial samples; the other metals are believed to have been chemically pure. The purity of the metals is indeed a point of great importance, because the presence of small quantities of foreign metals or other substances greatly impairs the conductivity of the mass. It was found, for instance, that gold, when alloyed with 1 per cent. of silver, lost nearly 20 per cent. of its conductivity. Alloys of tin and lead, and lead and zinc, were ascertained to conduct in the ratio of the mean conductivity of the two metals, and these alloys were found by Matthiessen to conduct electricity in like manner, forming an exception to the generality of the alloys. Some alloys of good and bad conductors, with the inferior conductor in excess, give a conductivity no higher than that of the inferior metal; bronze, for example, and the alloys CuSn,, CuSng, conduct no better than tin. The presence of carbon ferent length-say of ten centimetres; they would no doubt have then obtained the same sequence; but the ratio of the quantities of heat conducted would prcbably have been different. Their numbers at present must simply be regarded as representing the order of conductivity, per unit of volume, not per unit of mass. 318 CONDUCTIVITY OF BODIES FOR HEAT. [149. diminishes the conductivity of iron. If that of silver be taken as 1000, malleable iron, steel, and cast iron will be represented It is principally owing to differences in conductivity that bodies at the same temperature excite when touched very different sensations of heat or of cold. A piece of metal feels much hotter or colder than a piece of wood heated to the same degree, because the metal, from its superior conductivity, according as it is above or below the temperature of the hand, imparts heat or receives it more quickly than the wood. This property of conduction is possessed by liquids in a very limited degree. On filling a test-tube with water, and holding it by the lower part, whilst the top of the tube is placed across the flame of a spirit lamp, the water at the top of the tube may be kept boiling for many minutes without occasioning the slightest inconvenience to the person who holds it. Gases are inferior even to liquids in conductivity; hence it is that porous bodies, such as wool, fur, and eider-down, which imprison large bodies of air within them, are so well adapted for winter clothing, by preventing the escape of the heat of the body outwards. For the same reason, chiefly, the employment of double doors and windows, which include a layer of air between them, is so useful in preventing the heat of our apartments from escaping outwards; or, as in the case of fire-proof boxes and icehouses, in preventing that of the outer atmosphere from penetrating. In a similar manner snow preserves the warmth of the earth during the rigour of winter. There seem, however, to be differences in the conductivity even of gases for heat. Magnus considers that the conductivity of hydrogen surpasses that of all other gases, and it is increased by increasing the density of the hydrogen employed. In his experiments he placed a thermometer at the lower part of a glass cylinder, which could be deprived of air, and filled successively with the various gases under trial; the upper part of the cylinder was then heated by means of boiling water. The temperature of the external air was uniformly at 15° C. during the course of the experiment, and care was taken to protect the apparatus from the disturbing influence of radiation; the temperature rose higher when hydrogen was employed than when any other gas was admitted. |