from a somewhat different point of view, he expressed it in a different way. Still, the general principle involved in the two statements is the same, and we may, therefore, designate this third law as the law of GayLussac. Pass now to the inference which, after what we have learned, the general truth just stated suggests. As we have seen, the ratio of the gas or vapor densities of any substances always stands in the direct ratio of their molecular weights. Hence, it follows that the definite proportions of which we have been speaking are always the proportions of the molecular weights of the substances involved in the chemical process in question, or else some simple multiples of these proportions. In the case we have cited, the ratio of 2:18 is the ratio of the molecular weight of hydrogen gas to the molecular weight of water. In other cases we should find that the definite proportion observed in the chemical process would be the ratio of the molecular weight of one substance to twice or thrice the molecular weight of another, and sometimes of twice the molecular weight of one substance to thrice the molecular weight of another; but the proportions are seldom more complex than these. Finally, there are three important deductions which immediately flow from the principles we have discussed: In the first place, it will be obvious how very greatly these chemical facts confirm the molecular theory. Thus far we have based this theory on physical phenomena alone, and we have deduced the molecular weights of substances from the densities of these substances when in the condition of gas or vapor. Now, we find these same values reappearing in purely chemical phenomena, and, if there are such things as molecules, we should naturally expect that in a chemical process the action would take place between the molecules of the substances involved, and, if so, the definite proportions observed must be some multiples of the relative molecular weights, as we find that they are. In the second place, it can easily be seen that the definite proportions observed in chemical processes may give the means of correcting the molecular weights deduced from determinations of gas or vapor density. Such determinations can rarely be made with accuracy, and there are known to be causes independent of the molecular weight which influence the density of aëriform substances to a limited extent. The definite proportions, on the other hand, can usually be determined with great accuracy, and are invariable. It is true that in a new problem we may not be able to tell whether the proportion is the ratio of the weights of single molecules, or of several molecules; but it gives us an exact ratio, and, by comparing this with the approximate ratio of the weights of single molecules obtained from the gas or vapor densities, we can at once interpret the result, and deduce in each case the correct value of the molecular weight sought. Or, in other words, the gas or vapor densities give us an approximate value of the ratio between the weights of single molecules. The definite proportions give us the exact value of the ratio between the weight of single or multiple molecules, as the case may be. By comparing the two we can see at a glance for which of the possible multiples the definite proportions stand, and we can then very easily deduce an accurate value of the simple ratio at first only approximately known. In the third place, the definite proportions observed in chemical processes enable us to determine with cer IMPORTANT DEDUCTIONS. 97 tain limitations the molecular weights of non-volatile substances, to which the vapor-density methods are obviously inapplicable. Thus, in the process described on page 90 neither the oxide of copper used nor the metallic copper formed, is a substance whose vapor-density can be determined. But, in the proportion already given if two microcriths is the weight of a molecule of hydrogen gas, then 79.3 must be the weight of a molecule of oxide of copper, and 63.3 the weight of a molecule of copper, or else these two values are multiples of the molecular weights; and with this limitation we can thus determine the molecular weights of all similarly nonvolatile substances. Moreover, in most instances, principles or analogies of chemistry, of which we shall gain some knowledge as we proceed, enable us to decide whether we are dealing with multiple molecules or not. There are, however, many cases in which these guides are insufficient, and then our knowledge is uncertain to just that extent. But we have now pushed this discussion as far as can be profitable at this time. Indeed, I fear that you have found it abstruse and dull. But in chemistry, as in other sciences, we must apprehend the fundamental conceptions before we can advance in our study, and you will not regret the tedium it may have involved, if you gain a clear conception of the three great laws on which the whole superstructure of chemistry rests THE LAW OF CONSERVATION OF MASS, LECTURE V. CHEMICAL COMPOSITION-ANALYSIS AND SYNTHESIS-THE ATOMIC THEORY. In my previous lectures I have endeavored to give you a clear idea of the meaning which our modern science attaches to the word molecule. I must next attempt to convey, as far as I am able, the corresponding conception which the chemist expresses by the word atom. The terms molecule and atom are constantly confounded; indeed, have been frequently used as synonymous; but the new chemistry gives to these words wholly different meanings. We have already defined a molecule as the smallest mass into which a substance is capable of being subdivided without changing its chemical nature; but this definition, though precise, does not suggest the whole conception; for the molecule may be regarded from two very different points of view, according as we consider its physical or its chemical relations. To the physicist, the molecules are the points of application of those forces which determine or modify the physical condition of bodies, and he defines molecules as the small particles of matter which, under the influence of these forces, act as units. Or, limiting his regards to those phenomena from which our knowledge of molecular masses is chiefly derived, he may prefer to CHEMICAL DEFINITION OF MOLECULES. 99 define molecules as those small particles of bodies which are not subdivided when the state of aggregation is changed by heat, and which move as units under the influence of this agent. To the chemist, on the other hand, the molecules. determine those differences which distinguish substances. Sugar, for example, has the qualities which we associate with that name, because it is an aggregate of molecules which have those qualities. Divide up a lump of sugar as much as you please. The smallest mass that you can recognize still has the qualities of sugar; and so it must be, if you continue the division down to the molecule. The molecule of sugar is simply a very small piece of sugar. Dissolve the sugar in water, and we obtain a far greater degree of subdivision than is possible by mechanical means; a subdivision which, we suppose, extends as far as the molecules. The particles are distributed through a great mass of liquid, and become invisible; still, the qualities of the sugar are preserved; and, on evaporating the water, we recover the sugar in its solid condition; and, according to the chemist, the qualities are preserved, because the molecules of sugar have remained all the while unchanged. Consider, in the second place, a lump of salt. You do not alter its familiar qualities, however greatly you may subdivide it, and the molecules of salt must have all the saline properties which we associate with this substance. Dissolve the salt in water, and you simply divide the mass into molecules. Convert the salt into vapor, as you readily can, and again you isolate the molecules as before. But, through all these changes, the salt remains salt; it does not lose its savor, because the individuality of the molecules is preserved. So is |