Page images

number of facts connected with the heat of combination of carbon which have hitherto appeared anomalous seem to find their explanation in this variability of the carbon atom. Thus the heats of combination in certain reactions are as follows::

[merged small][merged small][merged small][ocr errors][merged small][merged small][merged small]


[ocr errors]

C, H1= 1701

[blocks in formation]

On the assumption that the nature of the carbon atom does not vary, these facts are difficult of explanation. Regnault's results regarding the specific heats of liquid and gaseous carbon compounds are explained by adopting the hypothesis that the nature of the carbon atom varies in its different compounds. The great variations in the specific heats of the liquid carbon compounds observed by Regnault are due, not to the fact that these compounds are liquids, but to the immediate influence of the carbon atoms contained in these compounds. Hence all Regnault's endeavours to establish a relation between the varying specific heats of the liquids and their relative volumes must be fruitless. The unknown cause which Regnault admits, in the last words of his research, "on est obligé d'admettre qu'en outre des causes, telles que la dilatation, qui font certainement varier la capacité calorifique d'un même corps avec la température, il en existe d'autres, que nos moyens d'investigation ne sont pas parvenus jusqu'ici à définir," is to be found in the varying specific heat of the carbon atoms, which changes with the changes of temperature and of molecular constitution. Regnault, in his researches upon the specific heats of gaseous carbon compounds, gives merely the mean result for a certain temperature-interval, except in the case of carbon dioxide; from these mean results little can be deduced, if it be true that the specific heat is a function of the temperature, as in these cases the nature of this function is unknown. It is only from widely extended researches upon the specific heats of a number of gaseous and liquid carbon compounds that any deep insight can be obtained into the nature of carbon, and into the general laws regulating the specific heats of the compounds of this element. With this work I am now engaged; but some time must elapse ere my results are sufficiently advanced for publication; I have therefore summarized the points of interest, and showed the consequences which seem to follow from my researches, so far as these have been carried out.

(3) The question has yet to be answered, How, in reference to the atomic constitution of the elements carbon, boron, and silicon, can we on physical grounds harmonize all the foregoing results regarding the specific heats of these elements? Only

* Mém. de l'Acad de Paris, vol. xvvi.

by assuming that the atom of these elements is not a simple but a complex structure built up of simpler elements. This hypothesis, put forward by Kopp, Berthelot, and others to account for the seeming exception of carbon &c. to Dulong and Petit's law, yet remains as the most probable means of accounting for the changes in the specific heats of those bodies, although it is now shown that they obey the above-named law. Before these questions regarding the constitution of the atoms of these elements can be answered, careful researches upon the specific heats of their compounds must be undertaken. I propose to discuss the theoretical bearings of these questions in another instalment of this paper.

Hohenheim, October 1874.

XXXII. On the Action of Solids in liberating Gas from Solution. By CHARLES TOMLINSON, F.R.S.*


F Fact and true Inference were friends and fellow-travellers, or if the latter followed the former at an easy distance, we should always find them on smooth roads and under a cloudless sky. Whoever discovered the one, would be quite sure that the other was close at hand, neither impeded by the quicksands of doubt, nor hidden from view by the fog of hypothesis. Mediocrity would then be as efficient as genius, and we should be constrained to acknowledge the truth of Lord Bacon's assertion, that " our method of discovering the sciences is such as to leave little to the acuteness and strength of wit, and indeed rather to level wit and intellect" †.

The difference between a non-scientific and a scientific intellect is supposed, at least in modern times, to consist in this, that while the one mixes up facts with their inferential ocular spectra in such a way as almost to justify the sarcasm that "there are more false facts than false theories current," the other carefully isolates facts from their theoretical or hypothetical shackles, and proclaims from the housetops "we know that these are facts, we infer that this is their explanation."

And with good reason; for a fact stands out in objective prominence, and, as such, must present the same aspect to every sane observer-whereas the inference is eminently subjective, taking its tone and texture from the mind of him who draws it, so that it is exposed to the warping influences of prejudice, of self-esteem which prefers its own theories to nature's truth, of * Communicated by the Author.

"Nostra vero inveniendi scientias ea est ratio, ut non multum ingeniorum acumini et robori relinquatur; sed quæ ingenia et intellectus fere exæquet."-Novum Organon, lib. i. sect. Ixi.

dyspepsia and other accidents. Hence two observers may draw entirely different inferences from the same fact, and both be equally wrong when the case comes to be referred to nature's high court of appeal.

When a scientific man says, "In the year 18- I established this fact," he is within philosophical bounds, supposing the fact to be such as can be repeated and verified by persons capable of performing an experiment with accuracy. But when he says, "In the year 18-I established this explanation," he becomes non-scientific, because he confuses that which admits of proof with that which can only be a matter of inference.

If, for example, I pour into a test-glass a quantity of sodawater and observe that numerous bubbles of gas cling to the side, and then pour soda-water into another test-glass that has just been made chemically clean, and find that no bubbles adhere to the side, I infer that chemical cleanliness in the latter case, and the absence of it in the former, are concerned in the phenomena. If a line be drawn within the clean glass by means of a glass rod smeared with grease or oil, such line becomes instantly covered with gas-bubbles. In such case it is fair to infer that a portion of the chemically clean surface is thus made unclean, and that the different behaviour of the gaseous solution depends on the distinction between chemically clean and chemically unclean.

But M. Gernez has recently informed us that in 1866 he established the conclusion that no gas is liberated below the surface of a supersaturated gaseous solution, such as Seltzer water, unless we introduce some kind of gaseous atmosphere, such as that which is retained on the surface of a solid body or in the capillary cavities of a porous body, and that it is into this atmosphere that the gas held in solution expands and


In 1867 I published in this Magazine † an account of a number of experiments in opposition to the view of M. Gernez, the object of which was to show that while a gaseous supersaturated solution, with its upper surface exposed to the air, gives off gas, either with effervescence or imperceptibly, the surface of the liquid confined by the sides of the vessel is subject to two modifications-(1) The state of chemical purity of their surface, and (2) the pressure exerted by them virtually on the liquid. (1) If the vessel be chemically clean, no gas is disengaged and no bubbles form on the sides, because the adhesion

* Comptes Rendus, 4 Janvier 1875, p. 44. A translation of this Note is given in the Phil. Mag. for February last.

"On the so-called 'Inactive' Condition of Solids," Phil. Mag. August and September, 1867.

between the sides and the gaseous solution is perfect; and therefore the sides may be regarded, pro ratá, as merely a continuation of the liquid itself, and no bubbles will form there any more than in the central parts of the liquid. (2) But suppose the sides to be chemically unclean, adhesion is diminished or destroyed, and therefore the surface of the liquid next to such sides is virtually as free as its upper surface; consequently bubbles will form there, just as they do on the upper surface; only in the latter case they do not appear as bubbles (except during effervescence), because there is no pressure; the sides do exert pressure, and therefore bubbles are formed. Now it does not matter whether there be air or not between the sides and the liquid, since it is no function of air to induce the liberation of gas or the formation of gas-bubbles. It is really want of adhesion. A glass rod or other solid introduced into the liquid does nothing more than form new sides, as it were, to the vessel; and its effect is merely that of the sides. If chemically clean, the rod will form no bubbles around it, because the adhesion between it and the gaseous solution is perfect. If dirty, the adhesion is imperfect, and the surface of the liquid in contact with it will be as free, or almost so, as the upper surface.

I must refer to my paper for the experimental details; but I may be permitted to state briefly, that the various solids made chemically clean, or partly so, and plunged into soda-water, displayed no gas-bubbles except on the parts that had not been made clean-that a flint-stone was immediately covered with bubbles, but, broken into two parts, not a single bubble appeared on the newly fractured surfaces-that a rat's-tail file, properly cleaned, liberated no bubbles; so that I could not agree with M. Gernez that a solid, whatever its polish, "is covered with roughnesses that form a sort of network of capillary conduits, into which the surrounding gases penetrate and condense," that "the gas-bubbles thus imprisoned become the centres to which those that are dissolved pass."

But M. Gernez says that if a small cavity be worked into the end of a glass rod, and this, full of air, be lowered into a gaseous solution, the gas escapes into it in bubbles. I submerged a cage of fine wire gauze in soda-water, but there was no escape of gas so long as it was chemically clean. When taken out, rolled between the slightly greased hands, and again lowered into the soda-water, the gas escaped from its side in bubbles with an audible noise.

Now it cannot be maintained that contact between the cage and the hands imparted to the cage a film of air instead of a film of oil-or that a slightly greasy rod introduced into a chemically clean glass containing soda-water, and rubbed against

the side below the surface, introduces air and not oil-or that a clean glass rod on which no bubbles appear, taken out and drawn with friction through the hand, acquires an air coating, which causes it to be covered with bubbles when returned to the soda-water. A man must be strangely fascinated with his own hypothesis to maintain such a proposition; and yet M. Gernez, after a lapse of eight years, attempts to explain my results (which have been confirmed by Professor Schrötter *) by seriously maintaining that the effect of washing with caustic potash solution, boiling distilled water, or alcohol is not to make the solid surfaces chemically clean, but to deprive them of the film of air which clings to them. I quote his language: "Si l'on a soin de dissoudre, par des lavages successifs à la potasse, à l'eau distillée bouillante et à l'alcool, la couche superficielle des vases de verre en certains points de laquelle se trouverait retenue une petite quantité d'air, on constate qu'il ne se forme plus une seule bulle gazeuse sur la paroi baignée par le liquide, pas plus qu'à l'intérieur de la solution sursaturée entre des limites de température et de pression très-étendues" (p. 44).

As to the action of porous bodies in liberating gas from solution, I have long since endeavoured to show that they act by reason of that strong adhesive force of which the absorptive power of charcoal for various gases affords striking examples. Charcoal, or other porous body, in a boiling liquid (which I have defined as a supersaturated solution of its own vapour †) powerfully absorbs the vapour of that liquid, but cannot retain it so long as the liquid is maintained at or near its boiling-point; hence it is constantly pouring forth streams of vapour, and by its presence so far facilitates the escape of vapour, that in distillations from 20 to 30 per cent. more liquid is condensed in consequence of the presence of a few bits of charcoal, while at the same time soubresauts are prevented.

It appears to me that porous bodies liberate gas from gaseous supersaturated solutions by virtue of that same force of adhesion by which gases become imprisoned in charcoal &c. A piece of cocoanut-shell charcoal, for example, in soda-water, condenses a certain amount of carbonic acid within its pores; and the adhesive force continues active even though a further supply of gas can find no entrance. In seeking to enter, it becomes disengaged from the water, and rises in a stream apparently from the surface of the charcoal. Under diminished pressure the gas escapes in large quantities from the charcoal itself, while the adhesive force is restored with the pressure.

* Pogg. Ann. vol. cxxxvii.

Proceedings of the Royal Society, January 21, 1869.

« PreviousContinue »