Chemical and Physical Researches. By THOMAS GRAHAM, D.C.L., F.R.S. Collected and printed, for private circulation only, by JAMES YOUNG, F.R.S. and Dr. ANGUS SMITH, F.R.S. Edinburgh: 1877.

It would have been difficult to offer more graceful evidence of friendship, and nothing would have touched and gratified Mr. Graham more, than the publication of this volume by his valued friends Mr. Young and Dr. Angus Smith.

Mr. Graham's work was peculiar, as his researches were mainly devoted to the elucidation of questions which occupy an intermediate position between chemistry and physics. Dr. Smith therefore wisely determined not to place the papers in strict chronological order, but to arrange them under three divisions, headed respectively "Gases," "Salts and Solutions," and "Unclassified Papers." The plan is a most fortunate one, for it well shows the gradual development of Mr. Graham's thoughts in each division of his work, and at the same time it exhibits its strength and coherence.

His earliest paper "On the Absorption of Gases in Liquids," published in 1826, is remarkable for ingenuity and close reasoning. In it he considers that " gases may owe their absorption in liquids to their capability of being liquefied, and that when gases appear to be absorbed by liquids they are simply reduced to that liquid inelastic form which otherwise, by cold or pressure, they might be made to assume; their detention in the absorbing liquid is owing to that mutual affinity between liquids which is so common. Faraday had shown, in 1823, that in the physical states of gas, liquid and solid, there was nothing of absolute permanency, and that any body may assume consecutively all these forms." Hence Mr. Graham concluded that those bodies which at the temperature of the atmosphere we experience to be gases, may be considered, without impropriety, as volatilised liquids, and he pointed out that it was not necessary that gaseous bodies -whose absorption in liquids he was explaining-should be presented to the liquids in a liquefied state, for the mere absorption of such gases by liquids occasioned their liquefaction. He gives as an instance the liquefaction which must accompany the absorption of steam, at 600° F., by sulphuric acid heated to the same temperature, while in order to liquefy the gaseous body in the ordinary way it would be necessary to cool it down through a range of nearly 400°. The absorption is, in fact, dependent upon "the affinity which occasions the miscibility of two liquids." In his last paper in the "Philosophical Transactions," published more than forty years afterwards, he refers to the liquefaction of gases in colloids in much the same terms, for he alludes to the general assumption of liquidity by gases when

absorbed by actual liquids or by soft colloids," and he shows that those gases penetrate rubber most readily which are easily liquefied by pressure, the rates of passage of carbonic acid and hydrogen being 1181 and 473 respectively; that gases undergo liquefaction when absorbed by liquids and such colloid substances as india-rubber; and finally, that the complete suspension of the gaseous function during the transit through india-rubber cannot be kept too much in view.

Probably the subject with which Mr. Graham's name will always be specially connected is the diffusion of gases. The first paper on it was published in the "Quarterly Journal of Science" for 1829. He found that the lighter a gas is the more quickly it diffuses away from an open cylinder, and the experiments led him to believe that the diffusiveness of gases is inversely as some function of their density, apparently the square root of their density. In this long series of researches the molecular aggregation of the septa through which the gases penetrated was always carefully considered, and in dealing with gaseous penetration he clearly saw that orifices of excessive minuteness might be quite impassable by gases of low diffusive power; that in other cases pores of graphite could only be permeable by molecules; and lastly, as will be subsequently shown, that there might be an inter-molecular porosity due entirely to dilatation at a high temperature.

In a paper published in 1831 he established the following law of the diffusion of gases :-The diffusion or spontaneous intermixture of two gases in contact is effected by an interchange in position of indefinitely minute volumes of the gases, which volumes are not necessarily of equal magnitude, being in the case of each gas inversely proportional to the square root of the density of that gas. He speaks of diffusion being effected by a force of the highest intensity, and urges that sensible masses are not affected by diffusion, but only molecules. Diffusion afforded, in Graham's hands, the most conclusive proofs of molecular movement, and as an instance of the rapidity of such molecular mobility it should be remembered that Maxwell has estimated that the initial velocity of a molecule of hydrogen is nearly 1860 metres per second.

The continuation of the researches led him to study the passage of gases through a minute orifice in a thin disc of platinum, a mode of passage, termed effusion, which left no doubt of the truth of a general law, that different gases pass through minute orifices in times which are as the square roots of their respective densities, or with velocities which are inversely as the square roots of their respective densities. The research afforded an experimental verification of the truth of the mechanical law, that the velocity with which a gas rushes into a vacuum, through a minute aperture, is the same as that which a heavy body would acquire in falling from the height of an atmosphere composed of

Mr. Graham

the gas in question of uniform density throughout. carefully points out that the phenomena of diffusion and effusion are essentially different, although the rates of passage are alike; diffusion only permits molecules to pass, but masses pass by effusion.

He found that if the orifice in the platinum disc became tubular, the passage of carbonic acid and nitrous acid became quicker in relation to air than when the rates of passage through a thin punctured disc were examined, and from facts such as this a new kind of passage was detected, which he called transpiration. By employing capillary tubes the effusion and transpiration rates of passage were found to differ widely, "for if the length of the tube is progressively increased, and the passage of all gases becomes greatly slower, the velocities of passage of the different gases are found to diverge greatly from their effusion rates." The velocities at last, however, attain a particular ratio with a given length of tube and resistance, and preserve the same relation to each other with greater lengths of tube and resistances; the most simple result being probably that of hydrogen, which has exactly double the transpiration rate of nitrogen, the relation of these gases as to density being as I 14. Mr. Graham viewed transpirability as being a kind of elasticity, depending upon the absolute quantity of heat which different gases contain under the same volume, and it is therefore more immediately connected with specific heat than with any other property of gases. One very important branch of Mr. Graham's labours appear to have commenced in 1829, with a notice of the "Singular Inflation of a Bladder," which was two-thirds filled with coal-gas, and on being introduced into carbonic acid for some hours became distended. Mr. Graham pointed out that, although Dutrochet would probably view in the experiment the discovery of endosmose acting upon aëriform matter as he had observed it to act upon bodies in a liquid state, the penetration of the carbonic acid was really preceded by an actual absorption in the water with which the pores of the bladder were filled. He continued the research at intervals and published a series of papers a few years before his death, in which he showed that the penetration of gases through colloid septa, such as india-rubber, was the result of an actual occlusion or absorption. A comparison of the relative rates of the penetration of oxygen and nitrogen through india-rubber led to a most remarkable experiment. Oxygen penetrates 24 times as fast as nitrogen; therefore by dialysing air Mr. Graham actually increased the quantity of oxygen from 20.8 to 41 per cent, just as he had effected a partial separation of oxygen from air by the slightly greater diffusion velocity of nitrogen. The phenomena of diffusion and penetration of india-rubber are, however, widely different, for the rate of passage through the colloid depends on the facility with which the gas is capable of being liquefied within its pores.

MM.

Deville and Troost discovered that certain gases penetrated tubes of iron and platinum, and this fact led Mr. Graham to enquire whether the penetration was not preceded by an absorption of the gas. His results proved that palladium, platinum, and iron did absorb hydrogen. Palladium, for instance, absorbs nearly 1000 times its volume of the gas, and most metals appear to have the power of selecting one or more gases in virtue of their colloidal character; for crystalline metals, such as osmium and iridium, do not occlude gas at all. The absorption of hydrogen by palladium led him to conclude that the gas was condensed into the metallic state; but, although much evidence was adduced, it may be doubted whether the metallic character of hydrogen was established.

Under the second classification of the papers, the research on "Arseniates, Phosphates, and Modifications of Phosphoric Acid," occupies a prominent place. It commenced with the explanatory hypothesis that phosphoric acid is disposed to unite with 3 atoms of base, and he then traced the method by which the metaphosphates and pyrophosphates were produced. It is impossible to overrate the importance of this research, for on it the theory of the polybasity of acids mainly rested.

The diffusion of gases appears to have led to the study of the diffusion of liquids, and in 1849 the splendid monograph on the latter subject was published, the results of which showed that diffusion supplied the densities of a "new kind of molecules," for in liquid diffusion we appear to deal no longer with chemical equivalents or Daltonian atoms, but with masses even more simply related as to weight. By continuing the investigation he was enabled to divide various soluble substances into crystalloids and colloids, the former having a rapid diffusion rate and the latter being marked by low diffusibility. He pointed out that, although chemically inert, in the ordinary sense, colloids possess a compensating activity of their own, arising out of their physical properties. The colloid is, in fact, the dynamic state of matter, the crystalloid being the statical condition. The colloid possesses energy, and it may be looked upon as the primary source of the force appearing in the phenomena of vitality.

His paper on "Speculative Ideas concerning the Constitution of Matter" will always be viewed with special interest, as it contained several remarkable expressions of belief, such as the suggestion that the various kinds of matter recognised as different elementary substances may possess one and the same ultimate or atomic molecule existing in different conditions of movement. With the atom at rest, the uniformity of matter would be perfect: but it always possesses motion due to a primordial impulse, and, as differences in the amount of this motion occasion differences of volume, matter only differs by being lighter or denser matter. The gaseous molecule is composed of a group of the preceding inferior atoms, following similar laws, and is thus

a reproduction of the inferior atom on a higher scale. Chemical combination consists in equal volumes of the different forms of matter coalescing and forming a new atomic molecule, and is therefore directly an affair of weight; and the combining weights differ because the densities, atomic and molecular, differ. And he points out that liquefaction or solidification may not therefore involve the suppression of the atomic or molecular movement, but only the restriction of its range.

All Mr. Graham's results were obtained by the simplest possible means, and it is not a little remarkable that throughout the forty-six papers which the volume contains but few plates and woodcuts are necessary for their complete illustration. All his apparatus which could be found was exhibited at the Loan Collection of Apparatus held during the past year at South Kensington Museum, where it excited much interest from its simplicity as compared with the complicated appliances there gathered together. Some remarks which concluded a description of it may not be out of place here, for it taught that although in certain researches, or for accurate observation and measurement, delicate and complicated instruments may be necessary, the simplest appliances in the hands of a man of genius may yield the most important results. Thus with a glass tube and a plug of plaster-of-paris Mr. Graham discovered and verified the law of diffusion of gases. With a tobacco-pipe he proved indisputably that air is a mechanical mixture of its constituent gases. With a tambourine and a basin of water he divided bodies into crystalloids and colloids, and obtained rock crystal and red oxide of iron soluble in water. By the expansion of a palladium wire he did much to prove that hydrogen is a white metal. And, finally, with a child's india-rubber balloon filled with carbonic acid he separated oxygen from air, and established points the importance of which from a physiological point of view it is impossible to overrate.

Dr. Smith has devoted 29 pages of the volume to a very accurate and valuable analysis of the papers, which he has in many cases collected from scientific periodicals now almost inaccessible even in this country. He has also added an excellent preface entitled "Graham and Other Atomists," which it may be well to quote at some length.

"Atoms and eternal motion are among the first-known scientific ideas. We find them discussed with full keenness by the earliest Greeks of whom we have received definite accounts." Dr. Smith then states the views which have been held from the time of Leucippus, " to whom the action of the atom as one substance, taking various forms by combinations unlimited, was enough to account for all the phenomena of the world." Graham took a similar view, and advanced the idea (of atomic motion with unity of material) to its utmost limit. The Greek told us all was motion; Graham considered that the diversity in motion was only the basis of the diversity of the material, or, in other words, that an atom constituted an element of a special kind ac