ultimate particles both of simple and compound bodies, the number of simple elementary particles which constitute one A Fig. 21. compound particle, and the number of less compound particles which enter into the formation of one more compound particle." How then did he determine the relative weights of the ultimate particles of simple bodies? Let us take the case of oxygen. The atom of oxygen, said Dalton, is 8 times heavier than the atom of hydrogen: let us call the weight of the atom of hydrogen, or the atomic weight of hydrogen, one; then the atomic weight of oxygen is asserted to be 8. Masses of hydrogen and oxygen combine in the ratio 1 : 8, to form water; but an atom of water is formed by the union of atoms of hydrogen and oxygen; if it is assumed that an atom of water is formed by the union of one atom of hydrogen with one atom of oxygen, then the atomic weight of oxygen must be 8. In saying that the atomic weight of oxygen was 8, Dalton implicitly made this assumption. But it might be assumed that an atom of water is composed of one atom of oxygen united with two atoms of hydrogen; as an atom of hydrogen is the unit in terms of which the weights of the atoms of other elements are stated, it follows from this assumption that the atomic weight of oxygen is 16; because 2 : 161: 8. Or it might be assumed that an atom of water is composed of three atoms of hydrogen united with one atom of oxygen; in this case the atomic weight of oxygen must be 24. Or it might be assumed that an atom of water is composed of two atoms of oxygen united with one atom of hydrogen; in this case the atomic weight of oxygen is 4. Before the atomic weight of oxygen could be determined 280 from the data of the composition of water, it was necessary to determine the number of atoms of oxygen and hydrogen which united to form an atom of water. Dalton's conception of the atom supplied no method whereby this could be done. To get over this difficulty if possible, Dalton framed several empirical rules regarding the compositions of the atoms of binary compounds. He classified compound atoms formed by the union of two elements into binary, ternary, quaternary, &c., atoms. Calling the two elements A and B, he said that a binary atom is formed by the union of one atom of A with one atom of B ; a ternary atom, by the union of, either one atom of A with two of B, or two atoms of A with one of B; a quaternary atom, by the union of, either one atom of A with three of B, or three of A with one of B. He then laid down the following rules: "I. When only one combination of two bodies [elements] can be obtained, it must be presumed to be a binary one, unless some cause appears to the contrary. II. When two combinations are observed, they must be presumed to be a binary and a ternary. III. When three combinations are obtained, we may expect one to be a binary, and the other two ternary. IV. When four combinations are observed, we should expect one binary, two ternary, and one quaternary, &c. &c." By applying these rules to water, which was the only compound of hydrogen and oxygen known at the time, Dalton concluded that the atom of water was a binary atom; but his own analyses had convinced him that hydrogen and oxygen combine nearly in the ratio 17 to produce water [we now know that the ratio is 1 : 8]; therefore he concluded that the atomic weight of oxygen was approximately 7. Dalton's rules for determining the compositions of compound atoms were not based on any general principle deduced from his fundamental conception of the atom; this conception could not indeed supply such a principle. If only one com pound of two specified elements was known, the simplest assumption to make was certainly that embodied in Dalton's first rule. But the simplest assumption is not always the best. 281 The law of combination by volumes of gaseous elements enunciated by Gay-Lussac in 1809 (s. Chap. VI. par. 87) may be stated as follows. The gaseous elements combine in the ratios of their combining volumes, or in ratios which bear a simple relation to these. 282 By combining volume is here meant the smallest volume of a gaseous element which combines with unit volume of hydrogen; and unit volume of hydrogen is defined to be the volume occupied by unit mass of hydrogen. All measurements of volumes are assumed to be made at the same temperature and pressure. Gay-Lussac interpreted his law, in the light of the Daltonian theory, to mean that the ratios of the masses of the combining volumes of gaseous elements are also the ratios of the masses of the atoms of these elements. Thus; 2 volumes of hydrogen combine with 1 volume of oxygen to form water; but a volume of oxygen weighs 16 times as much as an equal volume of hydrogen; therefore an atom of oxygen is 16 times heavier than an atom of hydrogen; but the atomic weight of hydrogen is 1, therefore the atomic weight of oxygen is 16. Again, 1 volume of chlorine combines with 1 volume of hydrogen to form hydrogen chloride; but chlorine is 35.5 times heavier than hydrogen, bulk for bulk; therefore, the atomic weight of chlorine is 35.5. If this interpretation of Gay-Lussac's law is admitted, the law supplies a means for determining the atomic weights of gaseous elements. But Gay-Lussac ventured on the further generalisation that equal volumes of gases (measured at the same temperature and pressure) contain equal numbers of atoms. Dalton shewed that this generalisation was inadmissible. Thus, consider the combination of hydrogen and chlorine. One volume of hydrogen combines with one volume of chlorine to form 2 vols. of hydrogen chloride; therefore, by GayLussac's generalisation, x atoms of hydrogen combine with x atoms of chlorine, to form 2x atoms of hydrogen chloride. To make the statement more definite, let us assume that x = 1; then, a single atom of hydrogen by combining with a single atom of chlorine has produced 2 atoms of hydrogen chloride. Hence, each atom of hydrogen chloride is composed of half an atom of hydrogen united with half an atom of chlorine. But, by definition, an atom of an element is not separated into parts when it interacts chemically with atoms of other elements or compounds. Hence either Gay-Lussac's generalisation is wrong, or the Daltonian definition of the elementary atom must be modified. In 1811 the Italian naturalist Avogadro modified the 283 Daltonian atomic theory by introducing the conception of two orders of small particles, the molecule and the atom. The molecule of an element or compound, said Avogadro, is the smallest mass of it which exhibits the characteristic properties of that element or compound. The molecule, he said, is formed of smaller parts; these are atoms. The atoms which form the molecule of an element are all of one kind; the atoms which form the molecule of a compound are of two, or more, different kinds. Avogadro's conception of the structure of matter applied to the case of water asserted that, if the separation of a quantity of water could be carried far enough, we should at last come to very minute particles each of which would exhibit the properties of water; but if these particles were separated into parts we should no longer have particles of water, but particles some of which would exhibit the properties of hydrogen and some the properties of oxygen. Similarly, if the separation into parts of a quantity of hydrogen could be carried far enough, the hypothesis asserted that we should at last come to very minute particles each of which would exhibit the characteristic properties of hydrogen; but if these particles were separated into parts we should no longer have particles of what we know as hydrogen, but particles more or less unlike hydrogen, yet each the same as all the others. In other words, the Avogadrean conception of the structure of elements and compounds asserts, (1) that a quantity of a compound, or of an element, consists of a vast multitude of minute particles each of which possesses the characteristic properties of the compound, or of the element; these particles are called molecules; (2) that each of these molecules itself consists of a fixed number of yet smaller particles; these smaller particles are called atoms; (3) that the properties of the atoms which form the molecule of a compound are very different from the properties of the molecule itself; (4) that the properties of the atoms which form the molecule of an element are also different from the properties of the molecule of that 284 element, but that inasmuch as the atoms which form the molecule of an element are all of one kind, and are only more minute portions of the same kind of matter as the molecule itself, there is not so marked a difference between the properties of these atoms and the properties of the molecule formed by their union, as there is between the properties of the atoms of the different elements which form a compound and the properties of the molecule of that compound. Avogadro modified the generalisation of Gay-Lussac, and gave it the following form : Equal volumes of gaseous elements and compounds, measured at the same temperature and pressure, contain equal numbers of molecules. 285 Let us apply this generalisation to the combination (1) of hydrogen and chlorine to form hydrogen chloride, (2) of hydrogen and bromine to form hydrogen bromide. In each case we shall suppose that a certain volume of hydrogen, which we shall call 1 volume, is caused to combine with the other element, and that the volume of the gaseous compound is measured, the temperature and pressure at which all measurements are made being the same. The data are these ;— 1 volume of hydrogen combines with 1 volume of chlorine to form 1 volume of hydrogen combines with 1 volume of bromine to form Let there be x molecules of hydrogen in the 1 vol. used, then the data translated into the language of the Avogadrean hypothesis read thus ; х x molecules of hydrogen combine with x molecules of chlorine and produce 2x molecules of hydrogen chloride. x molecules of hydrogen combine with x molecules of bromine and produce 2x molecules of hydrogen bromide. Now as every molecule of hydrogen chloride is composed of both hydrogen and chlorine, and as every molecule of hydrogen bromide is composed of both hydrogen and bromine, the necessary conclusion-if we grant Avogadro's hypothesis -is that one molecule of hydrogen chloride (or bromide) is composed of half a molecule of hydrogen and half a molecule of chlorine (or bromine); in other words, that each molecule of hydrogen, and each molecule of chlorine and bromine, has separated into at least two parts, and that these parts of |