Graham's work is remarkable at once for its originality and for the simplicity of the methods employed obtaining most important results. He communicated papers to the Philosophical Society of Glasgow before the work of that society was recorded in _Transactions_, but his first published paper, "On the Absorption of Gases by Liquids," appeared in the _Annals of Philosophy_ for 1826. The subject with which his name is most prominently associated is the diffusion of gases. In his first paper on this subject (1829) he thus summarizes the knowledge experiment had afforded as to the laws which regulate the movement of gases. "Fruitful as the miscibility of gases has been in interesting speculations, the experimental information we possess on the subject amounts to little more than the well-established fact that gases of a different nature when brought into contact do not arrange themselves according to their density, but they spontaneously diffuse through each other so as to remain in an intimate state of mixture for any length of time." For the fissured jar of J. W. Döbereiner he substituted a glass tube closed by a plug of plaster of Paris, and with this simple appliance he developed the law now known by his name "that the diffusion rate of gases is inversely as the square root of their density." (See DIFFUSION.) He further studied the passage of gases by transpiration through fine tubes, and by effusion through a minute hole in a platinum disk, and was enabled to show that gas may enter a vacuum in three different ways: (1) by the molecular movement of diffusion, in virtue of which a gas penetrates through the pores of a disk of compressed graphite; (2) by effusion through an orifice of sensible dimensions in a platinum disk the relative times of the effusion of gases in mass being similar to those of the molecular diffusion, although a gas is usually carried by the former kind of impulse with a velocity many thousand times as great as is demonstrable by the latter; and (3) by the peculiar rate of passage due to transpiration through fine tubes, in which the ratios appear to be in direct relation with no other known property of the same gases--thus hydrogen has exactly double the transpiration rate of nitrogen, the relation of those gases as to density being as 1:14. He subsequently examined the passage of gases through septa or partitions of india-rubber, unglazed earthenware and plates of metals such as palladium, and proved that gases pass through these septa neither by diffusion nor effusion nor by transpiration, but in virtue of a selective absorption which the septa appear to exert on the gases in contact with them. By this means ("atmolysis") he was enabled partially to separate oxygen from air.