be evident, agrees as nearly as we can expect with that before obtained for the heating effects developed upon or by the black and white surfaces under examination; and which was shown to be nearly the same, whether the light was in its ordinary intensity, or at a high degree of concentration.

(41.) We have thus established that with considerable differences in the intensity of light acting, the heating effects on a black and a white surface maintain the same ratio very closely.

It has also been shown that on the same surface, with different intensities of light, the heating effect is proportional to the intensity of light.

At one intensity it is shown that the heating effects on the black and white surfaces are proportional to the quantities of light respectively absorbed by them.

Hence the heating effect is proportional to the light absorbed by the surface in respect to its colour, at all intensilies.

Hence also the light absorbed at different intensities is proportional to that impinging on the same surface.

These conclusions contain, perhaps, no information absolutely new; but in establishing experimentally what seems hitherto to have been only taken for granted on loose grounds, I conceive we may best prepare the way for investigating the nature of the heating power of light, and for examining whether it be analogous to any other phoenomena. One step appears to me to be gained in having, as I think, clearly shown the exact proportionality in the heating effect to the quantity of light acting, and shown to be actually absorbed by the surfaces. These experiments also confirm (if further proof be wanting) the conclusion that the sun's heating effect is of a simple nature.

(42.) It may not be altogether superfluous here to remark, the dependance of the results in the former portions of these inquiries (see (18) of the paper in the Annals for June), upon the con siderations laid down in the present paper (28). It will be thence evident that without knowing any thing of the relative powers of the surfaces for absorbing simple heat or radiating it again, if any such heat were intercepted by the glass, the effect on removing it would have been a diminution of ratio by the addition of equal quantities to its terms; supposing that the heat were instantaneously communicated from the front to the back of the bulb. If this were not the case, but a certain time were required for the effect to be produced, it would at the first moment be an addition of quantities in the ratio of the absorptive powers of the surfaces for simple heat: this, in the present case, would be a ratio of "greater inequality," and as appears from (38) nearly =

100

89

Again, with respect to the subsequent experiment (Annals,

June, (19), (20), it is equally obvious that the same distinction must be attended to; but if the lower bulb were only coated on the half of its surface exposed to the sun, the effect (if any were produced) would be greater, since here the ratio of "greater inequality" must operate. In this way I have repeated the experiment with a half coating of chalk, but with results so precisely the same as before, that no diminution was perceptible. In order to try the effect with a coating of still greater absorptive power, I repeated the experiment with a bulb half coated with white silk pasted on; the other being entirely painted with Indian ink. No diminution took place, as will be evident from the following results. The instrument was of a larger construction, and not graduated by Prof. Leslie's scale.

Large differential thermometer. Bulbs, Indian ink; and white silk on half. Graduation from white.

(43.) The question above alluded to (31, &c.) as to the existence and magnitude of a heating effect exterior to the cone of light formed by a lens, is one of the greatest curiosity and interest, especially as connected with what appears to be the analogous effect in the case of the prismatic spectrum. In a supplement to a paper on the latter subject, communicated some months since, and now before the Royal Society, I recorded a few imperfect experiments, in which it appeared to me that this phenomenon was clearly perceptible with a lens of about three inches aperture, and 7.5 focal length, by means of the differential thermometer; and I have since repeatedly observed the same thing, though from the smallness of the effects observed I am inclined to suppose that they could hardly have interfered in any sensible degree with the experiments described in the present paper. From the very small intensity of the effect in question, I have experienced great difficulty in applying both the test of its transmissibility through glass, and that of its relation to surfaces, so as to come to any decisive conclusion. I hope shortly to be able to bring forward some investigation of these points. Meanwhile, as connected with the subject of the present paper, I may be permitted to give the results of a few experiments, which clearly establish the existence, and convey an idea of the quantity of this effect; and which were made with a different instrument, and under different circumstances, from the few just alluded to.

(44.) Observations of the heat exterior to the cone of light formed by a lens. Aperture, 3.25 inches; focal length, 7.5. Bulb of photometer coated with Indian ink, in glass case.

The two last observations show how much of the effect is to be attributed to reflected light.

Lons

(45.) The experiment of Sir W. Herschel, from which a maximum heating effect further from the lens than the focus of greatest light is inferred, will be found in the Phil. Trans. 1800, No. 15, Ex. 23. It there appears that sealing-wax was scorched in the same time in the focus, and at half an inch further from the lens; whilst at half an inch nearer, no effect was produced in double the time. It can, perhaps, scarcely be inferred, that this effect is due to the same cause as that which operates outside of the luminous cone; since it is obvious, that beyond the focus the light again diverges, and we cannot with certainty distinguish the effects due to light under the peculiar modifications to which it may there be subjected, from those which may arise from some peculiar development of heat in the same position. The mere inspection of the adjoining diagram will illustrate the directions which the differently coloured rays, separated by the dispersive power of the lens, are made to assume; and with their different combinations it is highly probable that very different heating effects are produced. This is a topic of great interest, and one which, if more thoroughly examined, seems likely to lead to a more complete acquaintance than we at present possess with the nature of the....

violet

focus

rod

violet

heating effects developed both by the rays of light themselves, and at short distances from them.

ARTICLE II.

On the Corrosion of Copper Sheeting by Sea Water, and on Methods of preventing this Effect; and on their Application to Ships of War and other Ships. By Sir Humphry Davy, Bart. Pres. R. S.*

1. THE rapid decay of the copper sheeting of his Majesty's ships of war, and the uncertainty of the time of its duration, have long attracted the attention of those persons most concerned in the naval interests of the country. Having had my inquiries directed to this important object by the Commissioners of the Navy Board, and a Committee of the Royal Society having been appointed to consider of it, I entered into an experimental investigation of the causes of the action of sea water upon copper. In pursuing this investigation, I have ascertained many facts which I think not unworthy of the notice of the Royal Society, as they promise to illustrate some obscure parts of electro-chemical science; and likewise seem to offer important practical applications.

2. It has been generally supposed that sea water had little or no action on pure copper, and that the rapid decay of the copper on certain ships was owing to its impurity. On trying, however, the action of sea water upon two specimens of copper, sent by John Vivian, Esq. to Mr. Faraday for analysis, I found the specimen which appeared absolutely pure, was acted upon even more rapidly than the specimen which contained alloy : and, on pursuing the inquiry with specimens of various kinds of copper which had been collected by the Navy Board, and sent to the Royal Society, and some of which had been considered as remarkable for their durability, and others for their rapid decay, I found that they offered very inconsiderable differences only in their action upon sea water; and, consequently, that the changes they had undergone must have depended upon other causes than the absolute quality of the metal.

3. To enable persons to understand fully the train of these researches, it will be necessary for me to describe the nature of the chemical changes taking place in the constituents of sea water by the agency of copper.

When a piece of polished copper is suffered to remain in sea water, the first effects observed are, a yellow tarnish upon the copper, and a cloudiness in the water, which take place in two or three hours: the hue of the cloudiness is at first white; it gradually becomes green. In less than a day a bluish-green precipitate appears in the bottom of the vessel, which constantly

* From the Philosophical Transactions for 1824, Part I.

accumulates; at the same time that the surface of the copper corrodes, appearing red in the water, and grass-green where it is in contact with air. Gradually carbonate of soda forms upon this grass-green matter; and these changes continue till the water becomes much less saline.

The green precipitate, when examined by the action of solution of ammonia and other tests, appears principally to consist of an insoluble compound of copper, (which may be considered as a hydrated sub-muriate) and hydrate of magnesia.

According to the views which I developed fourteen years ago, of the nature of the compounds of chlorine, and which are now generally adopted, it is evident that soda and magnesia cannot appear in sea water by the action of a metal, unless in consequence of an absorption or transfer of oxygene. It was therefore necessary for these changes, either that water should be decomposed, or oxygene absorbed from the atmosphere. I found that no hydrogen was disengaged, and consequently no water decom. posed necessarily, the oxygene of the air must have been the agent concerned, which was made evident by many experi

ments.

Copper in sea water deprived of air by boiling or exhaustion, and exposed in an exhausted receiver or an atmosphere of hydrogene gas, underwent no change; and an absorption in atmospherical air was shown when copper and sea water were exposed to its agency in close vessels.

4. In the Bakerian Lecture for 1806, I have advanced the hypothesis, that chemical and electrical changes may be identical, or dependent upon the same property of matter: and I have farther explained and illustrated this hypothesis in an elementary work on chemistry, published in 1812. Upon this view, which has been adopted by M. Berzelius and some other philosophers, I have shown that chemical attractions may be exalted, modified, or destroyed, by changes in the electrical states of bodies; that substances will only combine when they are in different electrical states; and that, by bringing a body naturally positive artificially into a negative state, its usual powers of combination are altogether destroyed; and it was by an application of this principle that, in 1807, I separated the bases of the alkalies. from the oxygene with which they are combined, and preserved them for examination; and decomposed other bodies formerly supposed to be simple.

It was in reasoning upon this general hypothesis likewise, that I was led to the discovery which is the subject of this paper.

Copper is a metal only weakly positive in the electro-chemical scale; and, according to my ideas, it could only act upon sea water when in a positive state; and, consequently, if it could be rendered slightly negative, the corroding action of sea water