Page:Dictionary of National Biography volume 54.djvu/281

This page has been validated.
Stewart
274
Stewart

tion of the Eye to the Rays which emanate from Bodies’—were contributed in 1855 to the Philosophical Society of Victoria (Transactions, i. 92, 95). On his return he gave up business, and in February 1856 joined the staff of Kew observatory as assistant observer to John Welsh [q. v.] In October 1856 he became assistant to his former teacher, Forbes, at Edinburgh. Stewart at this time also worked at pure mathematics with Professor Philip Kelland, and in 1856 wrote a paper on a theorem in the theory of numbers (Trans. Royal Society of Edinburgh, xxi. 407), his only contribution to mathematics, for which he then showed distinct aptitude (Tait). In 1857 he published an interesting paper on the relation between the density and composition of sulphuric acid solutions, deducing therefrom the existence of definite compounds of the acid and water (Proc. Royal Society of Edinburgh, iii. 482; a preliminary abstract appeared in 1855, Brit. Assoc. Report, pt. ii. p. 70). Mr. Spencer Umfreville Pickering and others have since employed Stewart's method of research.

It was under Forbes's influence that Stewart undertook the researches on radiant heat which form his most important contribution to physical science, and for which in 1868 he was awarded the Rumford medal by the Royal Society. Stewart extended the ‘theory of exchanges’ due to Pierre Prevost (1751–1839) of Geneva, and proved, in opposition to the view of Jean Baptiste Joseph Fourier (1768–1830), that radiation is not a surface phenomenon; that it depends on the thickness of the radiating body, and in general that at any given temperature ‘the absorption of a plate equals its radiation, and that for every description of heat’ (Trans. Royal Soc. of Edinburgh, xxi. 1 sqq., read 15 March 1858); and that thus ‘the streams of radiant heat crossing any point of an enclosure of uniform temperature are not altered by the interposition of a body, whether opaque or transparent.’ It is remarkable, since Forbes had proved the similarity of radiant heat to light, that Stewart did not at once extend his results to optics. He also found out later that, cæteris paribus, the internal radiation in different substances varies as the square of the refractive index (Brit. Assoc. Report, 1861, i. 107), correcting an erroneous statement made by himself previously. Meanwhile Gustav Robert Kirchhoff (1824–1887) arrived independently at results which included those of Stewart, and led to the explanation of the dark lines in the solar spectrum as due to the absorption by layers of the vapours of various elements, and to the foundation by himself and Robert Wihelm Bunsen of spectrum analysis—one of the greatest discoveries of the century (Berichte der preussischen Akad. der Wissenschaften, 11 Dec. 1859).

Stewart had been pushing on in the same direction, but more slowly. In 1860 he showed by experiments on tourmaline, of which the experimental arrangement was suggested by Professor (afterwards Sir George Gabriel) Stokes, that his law held good for polarised rays of light (Proc. Royal Soc. x. 503, read 22 May 1860). In the same year he also showed that red glass, when raised to a sufficiently high temperature to emit light on its own account, gives out greenish light, and similarly that a piece of platinum foil blackened appears, when so heated, brighter in the blackened part than elsewhere (ib. x. 385, read 7 Feb. 1860). In May 1861 (ib. x. 193) he wrote a paper on the theory of internal radiation in uniaxal crystals, which was developed in the same year by Stokes (ib. p. 537). By this time, however, Kirchhoff had practically exhausted the subject for the time being.

On 1 July 1859 Stewart had been appointed director of the Kew observatory in succession to John Welsh, and henceforward he devoted himself mainly to meteorology, and especially to the phenomena of terrestrial magnetism. In 1861 he was appointed additional examiner in mathematics at Edinburgh University for five years, and there made the acquaintance of his future collaborator, Professor Peter Guthrie Tait. In August and September 1859 there had been a great magnetic disturbance, accompanied by auroral displays and by marked changes in sun-spots; the analysis of the photographic records of the magnetic storm at Kew directed Stewart's attention to the subject. General Edward Sabine [q. v.] had previously shown a connection between the occurrence of sun-spots and magnetic disturbances. Stewart now put forward the view that auroræ, magnetic storms, and earth-currents are due to variations in a primary electric current in the sun (Phil. Trans. 1861, p. 423). In 1862 he was elected F.R.S., and in the same year he suggested that the ‘red prominences’ are really solar auroræ (Phil. Mag. [4] xxiv. 302). In 1863 he made a careful investigation of the increase of pressure of a given volume of air between 32° F. and 212° F., his result for this important constant agreeing closely with that of Victor Regnault (1810–1878). In 1866 he redetermined the density of mercury at 32° F. with great accuracy. As it could be shown that the law of radiation of Stewart and Kirchhoff does not hold for a moving body in an enclosure of constant temperature, he made, in conjunction with Professor Tait between 1865 and 1873,