180 PROFESSOR THOMAS GRAHAM's SCIENl.FIC WORK. 



the Government, and in tbe publication of various scientific nieft-Oirs, 

 several of tbeui possessing a higli degree of interest; but it was not till 

 1846 that he produced a research of any considerable magnitude. In 

 that year he presented to the Royal Society the first part of a paper 

 " On the motion of gases," the second part of which he supplied in 1849. 

 For this research Mr. Graham was awarded a second royal medal of 

 the society in 1850. The preliminary portion of the first part of the 

 paper related to an experimental demonstration of the law of the effu- 

 sion of gases, deduced from Torricelli's theorem on the efflux of liquids 

 — a demonstration that was achieved by Mr. Graham with much inge- 

 nuity, and without his encountering any formidable difficulty. But the 

 greater portion of the first part, and whole of the second part, of this 

 most laborious paper were devoted to an investigation of the velocities 

 of transpiration of different gases through capillary tubes, with a view 

 to discover some general law by which their observed transpiration rates 

 might be associated with one another. Again and again, with charac- 

 teristic pertinacity, Mr. Graham returned to the investigation ; but, 

 although much valuable information of an entirely novel character was 

 acquired — information having an important bearing on his subsequent 

 work — the problem itself remained, and yet remains, unsolved. Why, 

 for exam[)le, under an equal pressure, oxygen gas should pass through a 

 capillary tube at a slower rate than any other gas is a matter that still 

 awaits interpretation. 



Near the end of the same year, 1849, Mr, Graham communi(!ated, also 

 to the Koyal Society, a second less laborious, but in the nov<*lty an<l 

 interest of its results more successful, paper ''On the diffusion of 

 liquids,'* It was made the Bakerian lecture for 1850, and was supple- 

 mented by further observations communicated to the society in 1850 and 

 1851. In his investigation of this subject, Mr. Graham ai)plied to liquids 

 the exact method of imjuiry whi(;h he had applied to gases just twenty 

 years before, in that earliest of his papers on the snbjcctof gas-diffnsion 

 published in the Quarterly Journal of Science ; and be succeeded iu 

 placing the subject of liquid-diffusion on about the same footing as that 

 to which lie had raised the subject of gas diffusion prior to the discovery 

 of his numerical law. 



In 1854 Mr. Graham communicated another paper to the Eoyal 

 Society, " On osmotic force," a subject intimately connected with th.it 

 of his last i)revious communication. This paper was also made the 

 Bakeriau lecture for the year; but, altogether, the conclusions arrived 

 at were hardlj' in proportion to the very great labor expended on the 

 inquiry. In the next year, 1855, just five-and-twenty years after his ap- 

 pointment at the Andersonian University, Mr. Graham was made nmster 

 of the mint; and, as a consequence, resigned his professorship at Uni- 

 versity College. During the next five years he i)ublished no original 

 work. 



Thus, at the beginning of the year 18G1, Mi; Graham, then fifty-six 



