164 On the different apparent Magnitudes of the same Objects. seemed to me a great deal smaller than when I looked at it directly. Since that time I have made repeated experiments of this matter, and always found the same circumstances." In another place he says:

"I shall recount still something more extraordinary on this variation of the magnitude of the visual angle, or of the image of objects,"

"Last winter I was in the country. In the night it froze hard, and there fell a little snow. On going out of my chamber in the morning, all objects appeared to me sensibly smaller than they had done the evening before.

"Since I made this discovery, and have been guarded against the rule of comparison, I plainly perceive that a very illuminated object seems smaller, and an object feebly supplied with light appears larger. The reason of this is evident. A strong light puts the whole globe of the eye on contracting itself, and a feeble one leaves it relaxed and dilated."

This author is very correct in his observations; but his explanation is founded on a false theory. The true reason is this: A strong light contracts the pupil of the eye, in which state it forms a small picture of an object upon the retina; but in a weak one the pupil is dilated, and all objects then appear larger. This property of vision will, I think, appear evident, from the following experiments.

It is difficult to enlarge the pupil of the eye to any particular dimension, but it may be contracted at pleasure, by means of perforations made either in a thin plate or a slip of paper.

Now, if an object be viewed through an aperture, about

of an inch in diameter, it will appear much smaller than to the naked eye, in consequence of the aperture of the crystalline lens being contracted; but if the perforation be removed from before the eye, the object will instantly appear increased in magnitude: and as no change can take place in any part of the eye instantaneously, it is therefore evident, that the apparent magnitudes of all objects are increased by an increase in the aperture of the crystalline lens, and consequently by an enlarged pupil.

The same thing may also be proved by looking at an object though perforations of different dimensions; for it will appear smaller through a perforation of of an inch, than through one that is four times as large; and an object viewed through a perforation as large as the pupil, appears of the same magnitude as to the naked eye.

Whence it is manifest, that all terrestrial objects appear larger to the naked eye in the mornings and evenings, when the pupil is large, than at noon when the pupil is less; and for the same reason they appear larger in winter than in

summer.

I am, sir, your humble servant,

E. WALKER.

Lynn,

March 18, 1808.

XXXII. On the Identity of Silex and Oxygen. By Mr. HUME, of Long-Acre, London.

SIR,

To Mr. Tilloch.

To inculcate any science with success, there is nothing so essential as a simple and perspicuous display of its first principles; and if there be any department in philosophy to which this observation is more peculiarly applicable, it is certainly the study of chemistry, than which there is, probably, none more useful to man.

The present period is, of all others, the most opportune for an improvement in chemical theory, as, from the very. brilliant discoveries of Professor Davy, it is obvious that a most material revolution is now dawning upon the modern system of chemistry, and, possibly, an entirely new structure must eventually prevail. I now allude to the word oxygen particularly, which, in its present limited sense, stands as a solecism in language, and a mère absurdity in the nomenclature of the day, since it has been lately proved to be at once the principle of acidity and likewise that of alkalescence.

In all modern authors, the classification of simple and

elementary bodies is, I presume, too diffuse, in respect to the number of subdivisions; and many of the titles employed might, with propriety, be expunged. But, though there are many other imperfections in chemical arrangement that require reform, I mean on this occasion to confine the following observations to one article only; and shall endeavour to prove that, in this instance at least, we should revise the list of simple substances, as far as regards SILEX; which is still continued, I think, with great impropriety, to rank as a species of earth.

To include in any one genus both silex and the other earths, as they are now called, seems extremely improper and palpably erroneous; nor can this classification be supported by any reasonable argument whatever. The definitions given from time to time, to distinguish an earth from any other elementary body, have never been sufficiently explicit, for they do not precisely exclude the alkalis: they make a useless division under the name of alkaline earths; and, as salifiable bases, an earth, a metal, an alkali, and silex, may be said to range as four distinct species of the same genus.

All earths are declared to be salifiable bases, and this I take to be the most essential clause in every definition; for, generally speaking, the earths have a ready affinity for every acid, even from the weakest, particularly the carbonic acid, to the most powerful that exists. Indeed, as far as concerns the combination with carbonic acid, with which the earths form nearly insoluble compounds, this peculiar property alone might have served to distinguish an earth from an alkali. Here, however, the principal force of the definition fails, the exception to silex is decisive; there is no carbonate of silex; no nitrate, no sulphate, nor, in short, any other salt, in which the acid is saturated by this simple element: neither art nor nature ever produced a perfect neutro-saline compound, in which silex could fairly be considered as a real and independent base.

To constitute a true salt, we know, there must not be less than one acid and one base, reciprocally saturating each other; and when the number of either exceeds, and the

salt

salt is not a binary compound, we may then fairly suspect an imperfection; for, one of the elements at least is frequently in the state of mere suspension, and not in chemical union with either of the other ingredients.

There are, indeed, numberless examples of such salts, and, I am ready to allow, some in which silex is found, whether as a mere contingency or otherwise; but it never exists as a perfect base, that is, possessing the capability of saturating the whole, or any part of the acid in such compositions.

If silex be, what I have long considered it, not only dissimilar to every elementary ponderable material besides, especially in generic characters, but, also, so vastly superior in its importance and bulk, as to leave no room for comparison; surely it ought then to be instantly removed, and no longer suffered to remain in the list of earths, but should be placed in the most prominent station in the arrangement of elements. Such is its consequence, that nothing in nature is so predominant or so universally disseminated; no compound solid substance of any magnitude is exempt from it, but contains always some, if not a very large portion of this insinuating, and as I conjecture, most essential of all terrestrial matter.

All organised bodies either contain silex, or, what I shall consider as a modification, oxygen. If there be any exceptions to this conclusion, they are so few and of such trivial import, that when they do occur they should be rated as anomalies; and it may happen, that the apparent absence of silex or oxygen is rather to be attributed to our want of means, and the imperfection of science to discover it.

In a geological view of this subject, where can we turn our eyes or employ our thoughts, without meeting this grand and multifarious cement-this bond of aggregation, that fixes the solidity of all tangible nature? The very outlines of our planet are traced out with it; and all primitive matter, from the most stupendous mountain or ragged precipice to the deepest cavern, even to the centre of gravitation, we are warranted to say, is replete with silex. If we contemplate the nature, volume, and importance of this, and then recolL 4 lect

lect the insignificance of zircon, glucine, and indeed the whole of the species of earths, none of which exists without an association of silex-all comparison vanishes, there is no estimate; these are as the mere spots to the brightest of luminaries, and therefore, in all systematic classification, should be separately arranged.

Where then ought silex to be placed in the arrangement of simple elements?-should it link with any other ponderable body as a species of the same genus, or preserve a station to itself? Were I asked for an answer to such a question, I would say--that seeing nothing to which it has the slightest resemblance but oxygen-gas, of which I conceive it to be the true base, here I would not only assign its proper rank, but give it also a precedence to all other elementary matters that had resisted decomposition,

It is hardly necessary for me now to add, that I do not consider oxygen in the state of gas to be a simple body; for whatever is susceptible of spontaneous change should always be deemed a compound, of at least two elementary substances. If one instance of this can be adduced, we may naturally infer that others will be found; and, fortunately for my present purpose, a most appropriate example has lately occurred, which confirms this conclusion: I allude to the experiments of Messrs. Allen and Pepys, upon carbon and carbonic acid, which appear to have been conducted with uncommon precision and genius, From these gentlemen we learn, that oxygen-gas is subject to spontaneous change, or, as they very properly express it, a deterioration; and that this will happen, though the gas be of the purest kind, that obtained from oxy-muriate of potash; and even when secured in glass vessels with glass-stoppers.

Having assumed silex and the base, of oxygen gas to bẹ synonymous and simple bodies, I shall now proceed, as far as my humble pretensions and knowledge of this subject will permit, to substantiate this position, by offering a few only of the numberless facts, which seem to confirm this identity. It is a task, I confess, I have imposed upon myself; for, having nearly three years ago permitted my opinion to be *Philosophical Transactions 1207.

1

pub