Page:Encyclopædia Britannica, Ninth Edition, v. 5.djvu/480

Rh 468 CHEMISTRY LAWS OF COMBINATION. Chloride of silver, for example, in whatever manner it may be prepared, invariably consists of chlorine and silver in the proportions by weight of 35 36 parts of the former and 107-66 of the latter. But it is often the case that elements combine together in several proportions ; whenever this occurs the several proportions in which the one element unites with the other invariably bear a simple relation to one another. Thus, 1 part by weight of hydrogen unites with 8 parts by weight of oxygen, forming water, and with 16 or 8 x 2 parts of oxygen, forming peroxide of hydrogen. Again, in nitrous oxide we have a compound of 8 parts by weight of oxygen and 14 of nitrogen ; in nitric oxide a compound of 16 or 8 x 2 parts of oxygen and 14 of nitrogen ; in nitrous anhydride a compound of 24 or 8 x 3 parts of oxygen and 14 of nitrogen ; in nitric peroxide a compound of 32 or 8x4 parts of oxygen and 1 4 of nitrogen ; and lastly, in nitric anhydride a compound of 40 or 8 x 5 parts of oxygen and 1 4 of nitrogen. This law is known as the law of combination in multiple proportions. The proportions in which two elements combine with a third also represent the proportions in which, or in some simple multiple of which, they will themselves combine. For instance, 35 36 parts of chlorine and 7975 parts of bromine combine with 1U7 66 parts of silver; and when chlorine and bromine unite it is in the proportion of 35 36 parts of the former to 79 75 parts of the latter. Iodine unites with silver in the proportion of 126 53 parts to 107 66 parts of the latter, but it combines with chlorine in two proportions, viz., in the proportion of 126 53 parts either to 35 36 or to three times 35-36 parts of chlorine. This is known as the la ,v of combination in reciprocal proportions. In explanation of these three laws deduced entirely from experimental observations, chemists have adopted the atomic or molecular theory which was first introduced into the science by Dalton at the commencement of this century. According to this theory the exceedingly small masses or molecules of which it is supposed matter consists are com posite, being made up of indivisible particles or atoms (see the article ATOM, vol. iii. p. 36). The molecules of the elements are assumed to consist of similar atoms, whereas those of compounds are congeries of dissimilar atoms; and the molecules which constitute a given kind of matter, it is supposed, are alike in weight and general properties, but differ from those of which all other kinds of matter are composed, so that every molecule belongs to one of a definite number of species. The study of the alterations which take place in the composition of molecules under the influence of various forces, and which result from their action upon one another, is the work of the chemist, whilst it is the province of the physicist to study the influences of those forces upon matter which affect entire molecules without in any way altering their composition. The chemist has no means of ascertaining, nor does he attempt to ascertain, the absolute weights of the atoms or of the molecules of the various elements and their com pounds ; he concerns himself merely with their relative weights, hydrogen being adopted as the standard of refer ence since it is the lightest of all known elements. The relative weight of the atoms of the various elements referred to that of hydrogen regarded as 1 are given in the third column of the table on page 467. The determina tion of the exact atomic weight of an element is an opera tion of extreme difficulty, and one requiring the greatest analytical skill, so that as yet the atomic weights of only a limited number of elements have been ascertained with more than approximate accuracy. The most accurately determined atomic weights are those of hydrogen, oxygen, nitrogen, chlorine, bromine, iodine, lithium, potassium, sodium, silver, and thallium. Apparently the numbers obtained for these elements are practically perfect. The manner in which the relative weights of the atoms of the elements are determined will be evident from the following considerations. If, instead of comparing together the relative weights of the elements which enter into combination, the volumes which they occupy in the state of gas (at the same tempera ture and under the same pressure) before and after com bination are compared, it is found that gases always unite together in very simple proportions, viz., either in equal volumes, or in volumes which bear some simple relation such as 1 : 2, 1 : 3, 1 : 4, 2 : 3, &c. Moreover, whatever the number of volumes before combination, it always is reduced to two on combination. Thus, equal volumes of hydrogen and chlorine gases unite without condensation to form hydrochloric acid gas ; in the production of water 2 volumes of hydrogen and 1 of oxygen combine, but form only 2 volumes of water-gas or steam; and if ammonia gas be decomposed by heat or a series of electric sparks, 2 volumes of the gas yield 3 volumes of hydrogen and 1 of nitrogen. Now, according to the law of Boyle and Mario tte, the volume of a given mass of any gas varies inversely as the pressure, provided that the temperature remains the same ; for instance, the quantity of air which is contained in a vessel of the capacity of 1 pint under the pressure of 1 atmosphere, or 15 Ib upon the square inch, may be con tained in a vessel of half a pint capacity if the pressure be doubled. According to the law of Charles and Gay-Lussac, on the other hand, all gases expand equally by heat, provided the pressure remains constant, the rate of expansion being -^^ of the volume at C. for each rise of 1 C. in temperature ; or in other words, the volume of a gas varies directly as the absolute temperature. A gas which strictly conforms to these two laws is said to be a perfect gas, but none of the gases with which we are acquainted are perfect in this sense. Thus, Audrews a experiments show that carbonic anhydride, which under a pressure of 36 atmospheres at C. is reduced to the liquid state, condenses more than it should according to Boyle s law. Again, the density of chlorine gas referred to air, according to Stas s determination of the atomic weight of this element, should be 2 4501. The following table exhibits its density at various temperatures from 20 to 200 C., 1 and it is evident that it is higher than it should be at all temperatures below 200 C. : Temperature. 20 50 100 Density. 2-4807 2-4783 2-4685 Temperature. 150 200 Density. 2-4609 2-4502 From the few accurate observations which have been made on this subject it appears that, in general, the departure from the laws of Boyle and Charles is greater the more the temperature of the gas approaches to that at which it becomes liquid ; and chlorine affords an instructive illustra tion of this, since it is readily condensed to a liquid under the pressure of 4 atmospheres at 15 0- 5 C., or by cooling in a bath of solid carbonic anhydride and ether. The general resemblance in the behaviour of gases under the influence of pressure and heat is very great, however, although not in absolute accordance with the laws of Boyle and Charles ; by this we are led to the assumption that their physical constitution must be similar, and, therefore, to the acceptance of the proposition, originally stated by Avogadro in 1811, that equal volumes of different gases contain equal numbers of molecules. Obviously, therefore, if the relative weights of equal volumes of different gases are determined under the same conditions as to tempera- 1 Ludwig, Bcrkhte der deutschen chemischen Gesellsckaj t, 1868, 232.