DISCOVERY BY EXAMINING RARE SUBSTANCES.

505

substance, and every new combination of substances, must possess new properties (otherwise we could not know them to be new), and produce new effects, so it follows that rare or peculiar substances are fertile sources of new discoveries.

Various important discoveries were made in thermoelectricity by Peltier, Matthiessen, and others, by the aid of the comparatively scarce substances, tellurium, selenium, and bismuth. Roscoe, by investigating compounds of vanadium, discovered that that element was closely allied to phosphorus, and determined its true atomic weight, and found that the weight given by Berzelius was not an accurate one. Arfvedson discovered lithia by analysing petalite and spodumene. It was by minutely examining the zinc ores of Frieberg that Reich and Richter discovered indium; and, by similarly examining the zinc ores of the Pyrenees, Boisbaudran discovered gallium.

After a new substance is discovered in one particular place or rare material, it is frequently found in a great many others. Thus the selenium of Fahlun was soon found in the curious and rare products of the Hungarian mines, and in the sublimates of Mount Stromboli. Soon after thallium was discovered in one substance, it was found in many others; and a similar result occurred with rubidium and cæsium.

i. By examination of the residues, &c., of manufacturing processes.—This method has on many occasions led to the discovery of new truths of science, and especially to the discovery of new elementary substances. By examining the solution of crude platinum in aqua-regia, obtained in his process of manufacturing that metal, Dr. Wollaston discovered palladium. Smithson Tennant, also, in the year 1802, tried to alloy with lead the powder left from native platinum after a solution of all the platinum

with aqua-regia; and Descotils, by further examining this powder, found that it contained a metal which imparted a red colour to the ammoniacal precipitate. Vauquelin, by treating the powder with alkali and heat, found a metallic oxide, which he considered to be the same as that discovered by Descotils; but Tennant, in 1804, finally showed that the powder really contained two metals, viz., osmium and iridium.' "Mitscherlich himself found, in the scoriæ of the mines of Sweden and Germany, artificial minerals having the same composition and the same crystalline form with natural minerals: as silicates of iron, lime, and magnesia, agreeing with peridote; bisilicate of iron, lime, and magnesia, agreeing with pyroxene; red oxide of copper; oxide of zinc; protoxide of iron (fer oxydulé); sulphurets of iron, zinc, lead, arseniuret of nickel; black mica. These were accidental results of fusion.' 2

Investigation of the concentrated residues of large manufacturing operations often yield new discoveries, because substances which exist only in very minute proportions in the crude or native materials of a manufacture frequently become concentrated to so great an extent by the processes employed that they become conspicuous. The concentrated residues of Courtois's manufacture of saltpetre so acted upon the vessels he employed that he was induced to investigate the circumstance, and thus discovered iodine. Balard, also, by analysing the concentrated motherliquor of sea-water, was led to the discovery of bromine. By analysing the residues of the vitriol works of Fahlun, Berzelius discovered selenium. The concentration of thallium in the process of burning sulphur and sulphides in

1 See Thomson, History of Chemistry, vol. ii. p. 234.

2 Whewell, Philosophy of the Inductive Sciences, vol. i. p. 510.

DISCOVERIES BY EXAMINING RESIDUES OF MANUFACTURES. 507

the manufacture of oil of vitriol enabled Crookes to discover that metal in the flue-dust deposited during that process.

Professor Bunsen, also, by concentrating a great bulk of the water of a mineral spring at Durckheim, was led to the discovery of two new alkali-metals, viz., cæsium and rubidium. The more soluble substances in the brine of salt-works become immensely concentrated by continual boiling and removal of the less soluble salts. During a long series of years of evaporation, in this way, and by further similar treatment of the brine of Stoke-Prior salt-works, and reduction of it to a very minute amount-less than one thousand-millionth part of its original bulk-I have been enabled, with the further aid of the spectroscope, to discover the presence of several alkaline metals (but no new ones) not previously known to be contained in it; and by washing with water large bulks of crude substances, such as chalk, sand, fire-clay, &c., and concentrating the liquid in a similar manner, I have found minute quantities of salts of lithium.

j. By examining the ashes of rare plants and animals.—Special opening for discovery lies in this direction, because we know that each different species of plant possesses special powers of assimilation, which enable it to select particular ingredients; this is particularly seen in the ability of sea-weeds to appropriate iodine and bromine. Similar remarks may be made with regard to the secretions and ashes of animals, which notably contain phosphorus ; it was the distillation of dried urine which led to the discovery of that element, and the analysis of the ashes of bones which further led to its abundant production.1

1 The earth of bones had been considered as a peculiar simple earth; but Gahn ascertained by analysis that it was a compound of phosphoric acid and lime.'—Thomson, History of Chemistry, vol. ii. p. 243.

CHAPTER LIV.

DISCOVERY BY DEVISING HYPOTHESES AND QUESTIONS, AND TESTING THEM.

As study, comparison, and inference often lead to the discovery of true explanations of phenomena, and also to the conception of new theories, hypotheses, and questions, so, on the other hand, the latter are often the cause of new experiments, observations, and discoveries. The method of discovery, by asking questions, and then attempting to solve them, is very similar to that of raising hypotheses and testing them; and the only difference is, that in the latter case we have already conceived an imaginary answer, but in the former we may have no preconceived idea of what the answer will be.

The method of discovery by conceiving new theories, hypotheses, and questions, and testing them, is a very common one. Many researches are commenced in order to settle a preconceived idea. Curiosity excites inquiry, and a favourite hypothesis or question is a powerful stimulant to research and to the making of new experiments. Hypotheses and questions are conceived in various ways, but chiefly by association of ideas and by inference, as already described in Chapter XXXVII. ; and they are tested either by comparing them with already known truths or with new ones, the latter being obtained in the usual way, viz., by means of new experiments, observations, comparison, or. inference. Dalton discovered his atomic theory by comparing known facts with an hypothesis he had inferred; and he did this at a time when circumstances were sufficiently ripe for the purpose, i.e., when knowledge of chemistry had sufficiently advanced.

DISCOVERY BY MEANS OF HYPOTHESES.

509

The hypotheses and questions chiefly referred to at the head of this chapter are those which form the startingpoint of a research, and to settle which a research is made; but we must remember that during the progress of every single investigation many hypotheses and questions are usually raised, and many discoveries are made whilst testing them, all of which are subsidiary to the settlement of the main idea which originated the inquiry.

There are many hypotheses and questions which, from their very nature, are of great importance; and if the experiments made to test them yield an answer, either for or against, the results must also be of great value; but in such cases this method of discovery is often very uncertain. Harvey was nineteen years verifying by experiment and observation his theory of the circulation of the blood. Newton was many years trying to verify his hypothesis of the law of action of gravity before he succeeded; Oersted's experience was similar with his conception of electro-magnetism; and Faraday's not unlike, with his idea of the relation of magnetism to light. Faraday also successfully sought to determine his hypothesis of the identity of frictional, voltaic, and animal electricity, in order to obtain the decision of a doubtful point which interfered with the extension of his views, and destroyed the strictness of reasoning;' but even after many years of trial he did not succeed in verifying his hypothesis of the relation of gravity to the physical forces. Many hypotheses in science have, however, been successfully verified.

In order to test the hypothesis of an exploded planet, 'the German astronomers agreed to examine the whole of the zone in which Ceres and Pallas move, in the hope of finding other planets-fragments, as Olbers conceived they might possibly be-of an original mass. In the course of this search Mr. Harding, of Lilienthal, on September 1,

1

4