We saw that, in the use which he made of the laws of Mitscherlich, and of Dulong and Petit, Berzelius recognized the importance of the aid given by physical methods towards solving the atomic problems of chemistry; but among those who have most thoroughly availed themselves of such aids Graham must always hold a foremost place. Graham devoted the energies of his life to tracking the movements of atoms and molecules. He proved that gases pass through walls of solid materials, as they pass through spaces already occupied by other gases; and by measuring the rapidities of these movements he showed how it was possible to determine the rate of motion of a particle of gas so minute that a group of a hundred millions of them would be invisible to the unassisted vision. Graham followed the molecules as in their journeyings they came into contact with animal and vegetable membranes; he found that these membranes presented an insuperable barrier to the passage of some molecules, while others passed easily through. He thus arrived at a division of matter into colloidal and crystalloidal. He showed what important applications of this division might be made in practical chemistry, he discussed some of the bearings of this division on the general theory of the molecular constitution of matter, and thus he opened the way which leads into a new territory rich in promise to him who is able to follow the footsteps of its discoverer. Other investigators have followed on the general lines laid down by Graham; connections, more or less precise, have been established between chemical and physical properties of various groups of compounds. It has been shown that the boiling points, melting points, expansibilities by heat, amounts of heat evolved during combustion, in some cases tinctorial powers of dye-stuffs, and other physical constants of groups of compounds, vary with variations in the nature, number and arrangements of the atoms in the molecules of these compounds. But although much good work has been done in this direction, our ignorance far exceeds our knowledge regarding the phenomena which lie on the borderlands between chemistry and physics. It is probably here that chemists look most for fresh discoveries of importance. As each branch of natural science becomes more subdivided, and as the quantity of facts to be stored in the mind becomes daily more crushing, the student finds an ever-increasing difficulty in passing beyond the range of his own subject, and in gaining a broad view of the relative importance of the facts and the theories which to him appear so essential.