with a white film of alumina, which may be detached in flakes. Clearly the condition of the aluminium has been modified by its union with the mercury. It has already been stated that water can be frozen by cold produced by the solution of finely divided fusible-metal in mercury. It is not a matter of indifference whether the powders of the mixed constituents of the alloy are employed, or whether the alloy is previously prepared by fusion, and then powdered, which shows that the act of fusion has effected some change in the molecular arrangement of the metals. The explanation of the depolymerisation of the metals when they are united with each other is somewhat complicated. First, Mazetto* has shown that there is a similar lowering of temperature, though to a far less extent, when molten tin is mixed with molten lead, so that the lowering of the temperature is by no means confined to the solution of metals in mercury. The next step we owe to Professor W. Spring, of Liége, whose results in building up alloys by compressing the powders of their constituent metals have already been referred to. Spring finds that by determining the amount of heat given out by alloys of lead and tin on cooling from a molten state, that more heat is actually given out than might be expected from the results of calculation ;† the difference is so great that it could not be due to errors of observation, for in actual numbers it amounts to many hundreds of calories for a weight of 100 grammes. He concludes that when molten tin is added to molten lead, the atomic constitution of the molecules is simplified—that is, depolymerisation takes place. Let it be assumed that each molecule of molten lead contains an arbitrary number of atoms, say five, and that the molecule of molten tin also contains five atoms. Then, if one molecule of lead be added to three molecules of tin, so as to form the alloy PbSn,, five groups of PbSn, will be the result, but each molecule of the alloy will contain four atoms instead of five. This molecular change requires heat to effect the re-arrangement in the molten admixture of metals, and as this heat is absorbed cold is produced, and it will therefore be evident that both theory and experiment tend to the view that molecular change may be produced by alloying metals. Debray has given a case of an alloy in which a simple elevation of temperature induces allotropic change in the constituent metals. It is prepared as follows:-95 parts of zinc are alloyed by fusion with 5 parts of rhodium, and the alloy is treated with hydrochloric *Rendiconti del R. Instituto Lombardo [2], vol. xviii., No. 3. acid, which dissolves away the bulk of the zinc, leaving a rich rhodium-zinc alloy, containing about 80 per cent. of rhodium. When this alloy is heated in vacuo to a temperature of 400° C., a slight explosion takes place, but no gas is evolved, and the alloy is then insoluble in aqua regia, which dissolved it readily before the elevation of temperature caused it to change its state. We are thus presented with another undoubted case of isomerism in alloys, the unstable, soluble modification of the alloy being capable of passing into the insoluble form by a comparatively slight elevation of temperature. Influence of Varying Quantities of Metals on Each Other. There is, undoubtedly, a firm experimental basis for the view to which Matthiessen was guided,nearly thirty years ago, by a study of the electrical resistance of solid alloys, that when metals are united to form alloys, in many cases one metal, and sometimes both metals, assume the allotropic state. He showed, for instance, that silver has a conductivity represented by 100, and that the addition of a small quantity of gold to the silver is attended with a rapid fall in the conducting power. The conductivity of pure copper may be represented by the number 98, the addition of a small portion of tin greatly diminishes the conductivity, as is proved by the curves given on p. 68. He pointed out that the amount of tin is too small to admit of the possibility of a chemical compound being formed, and from this fact and other evidence he concludes that the passage to an allotropic state can alone explain the result. In this connection the influence of small quantities of one element on large masses of another may be referred to. Submarine telegraphy will present us with the first case. The commercial suc cess of a submarine cable is measured by the speed with which messages can be sent through it, and upon this point we have the testimony of Preece, who tells us that a cable made with the copper of to-day, when the necessity for using pure copper is recognised, will carry twice the number of messages that a similar cable of less pure copper would in 1858, when the influence of impurities in increasing the electrical resistance of copper was not understood. A paper by Sir William Thomson* shows how important the purity of copper is, and how remarkable is the action of the impurity. It is safe to say that the presence of 0.1 per cent. of bismuth in the copper would, by reducing its conductivity, be fatal to the commercial success of the cable. The influence of small quantities of foreign matter is more marked in the case of iron, but this question may be more conveniently studied in the following chapter. Proc. Roy. Soc., vol. x. (1860), p. 301. With regard to gold, the addition of 0.2 per cent. by weight of bismuth would, from the point of view of coinage, convert the gold into a useless material, which would crumble under the pressure exerted through the die. Instances of a similar nature might be multiplied indefinitely; it will, however, be sufficent to quote a statement of Sir Hussey Vivian, who says that 1000 part of antimony will convert the best selected copper into the worst conceivable. of In order to explain facts such as these, it is necessary to ascertain what relations may subsist between the atoms of a mass of metal and the atoms of the added impurity. First, as regards the cohesion of a metal, this property may be investigated by the aid of heat, or by submitting the metal to mechanical stress; and, in a research to which the author* devoted much time, tenacity was selected as the property to be tested, with a view to ascertain the effect of the added matter upon a metal or alloy. Gold was chosen as the subject of the experiment for the following reasons:-First, it is a metal which it is possible to purify in a very high degree, it is not liable to oxidation, and the accuracy the results is not affected by the presence of occluded gases. The purest gold has a tenacity of 7 tons to the square inch, and it elongates about 30 per cent. before breaking. Standard gold, which contains over 91 per cent. of gold, the alloying metal being copper, has a tensile strength of 18 tons to the square inch, and it stretches 34 per cent. before breaking; in fact, when an eminent engineer saw the results of these tests, he expressed an opinion as to the possibility of making a very good gun of standard gold, if the cost of the material were no object. When, however, a small quantity of certain metals, 0.01, 0.1, or 0.2 per cent. is added to the gold, the cohesion of the metal is reduced in a very remarkable way, as Hatchett showed to be the case in 1803. The author has tried the effect of adding to pure gold various metals and metalloids, introducing in each case 0.2 per cent. Some of these elements reduced the tenacity and extensibility of gold to a very low point, while others increased one or both of these properties. Since 1826, when Gmelin called attention to the relations between the atomic weights of elements which have similar properties, chemists have been actively engaged in establishing analogies between the properties of the elements and in arranging them systematically, and the result has been (mainly through the labours of Newlands, Mendeléeff, and Lothar Meyer) the promulgation of the Periodic law. This law states that the properties of * Proc. Roy. Soc., vol. xliii. (1888), p. 425; and Phil. Trans., vol. clxxix. (1888), A., p. 339. the elements are a periodic function of their atomic weights. Lothar Meyer has gone further, and has shown that a remarkable relation exists between the atomic volumes of the elements. Now, however tiny the atoms may be, they must possess volume, and the volume of each element will be peculiar to itself. The space occupied by one atom cannot yet be measured absolutely, but relative measurements may be obtained "by taking such quantities of the elements as are proportional to their atomic weights, and comparing the space occupied by these quantities." The relative atomic volumes of the elements are found by dividing the atomic weights of the elements by their specific gravities. The atomic weight of gold is 196.2; = 10.2 the atomic volume, 196.2 19.3 or, expressed in the metric system, 196.2 grammes of gold would occupy a space of 10.2 cubic centimetres. Lead, on the other hand, would have the large atomic volume of 18.1 and potassium that of 45.1. The question now arises-Does the power to produce fragility, which we have seen certain elements to possess, correspond to any other of their properties by which they may be classified? The facts represented in the Periodic law were, in 1869, graphically represented by Lothar Meyer in his well-known curve of the elements. By adopting atomic weights and atomic volumes as co-ordinates he showed that the elements can be arranged in a curve representing a series of loops, the highest points of which are occupied by cæsium, rubidium, potassium, sodium, and lithium, whilst the metals which are most useful for industrial purposes occupy the lower portions of the several loops. An examination of the results the author has obtained shows that not a single metal or metalloid which occupies a position at the base of either of the loops of Lothar Meyer's curve diminishes the tenacity of gold. On the other hand, the fact is clearly brought out that metals which render gold fragile all occupy high positions on the curve. This would appear to show that there is some relation between the influence exerted by the metallic and other impurities and either their atomic weights or their atomic volumes. It seems hardly probable that it is due to atomic weight, because copper, with an atomic weight of 63.2, has nearly the same influence on the tenacity of pure gold as rhodium, with an atomic weight of 104, or as aluminium, the atomic weight of which is 27.0. It will be evident from the following table, which embodies the results of the author's experiments, that metals which diminish the tenacity and extensibility of gold have high atomic volumes, while those which increase those properties have either the same atomic volume as gold, or a lower one. Further, silver has the same atomic volume as gold, 10.2, and its presence in small quantities has very little influence, one way or the other, on the tenacity or extensibility of gold. These results are shown graphically in the diagram (Fig 31). The tenacity of pure gold is 7 tons per square inch. INFLUENCE OF IMPURITIES ON GOLD. |