segregated out at certain faces of the lead crystals. By the addition of a small quantity of arsenic or antimony the pink colour was replaced by a dull purple; and a clear pink tint was only obtained when all the oxidizable metals had been removed. I come now to the discussion of the state in which the silver exists to cause a pink or reddish reflection of light. Silver does not oxidize under the conditions of exposure to acetic-acid vapour and oxygen of the air. Moreover oxide of silver and silver carbonate are themselves decomposed and reduced to a metallic state by a heat below that attained in the stacks of fermenting tan. The silver must consequently be in the metallic state. As confirming this statement I made the following experiments :— Silver carbonate was triturated with white lead and water and then dried. Upon increasing the temperature, a delicate pink tint became visible upon the reduction of the oxide of silver. If a small quantity of silver carbonate be precipitated along with lead carbonate, the colour upon drying and heating is more uniform, and it may be obtained exactly resembling the tint seen on white-lead corrosions. The colour of the photographs obtained by means of silversalts is also evidence in favour of the metallic state of the silver; and I may also adduce the fact that a ray of light, when reflected ten times from a polished silver surface, is distinctly of a reddish colour. Collegiate Laboratory, Sheffield, XLVIII. On the Colour of the Lake of Geneva. By L. SORET*. MY DEAR TYNDALL, Geneva, March 31, 1869. N thanking you for your letter and the pamphlet you sent me, I take the opportunity of communicating to you an observation which may interest you. IN Whilst dealing with a different subject, I was led to consider whether the blue colour and the absorption of certain rays of light by water are due to the liquid itself or to the therein suspended solid particles. This question has been often discussed by others, as well as by yourself in your work The Glaciers of the Alps.' Your memoirs on the polarization of the blue light of the sky have suggested to me the idea that if the blue colour of the water be due to suspended solid particles, phenomena of polarization will be produced analogous to those observed by you on the light of the sky. The water of the Lake of Geneva, owing to the well-known * Communicated by Professor Tyndall, F.R.S. beauty of its colour, is very favourable to investigations of this kind. For this purpose I had a very simple apparatus constructed. It is a kind of telescope; a flat plate of glass with parallel surfaces, fitted hermetically at its one end, serves as object-glass. The instrument, therefore, can be immersed in water without the latter being able to enter it. The eyepiece consists of a Nicol's prism. It is easy to understand that, by immersing the telescope in the water, the eye receives the blue rays of light emanating from the water, and that this light can be analyzed by turning the Nicol. By proceeding in this manner, I found that the water of our lake really exhibits phenomena of polarization comparable to those observed on the light of the sky; only their observation is more difficult, and up to this time I have not been able to study them as well as I could have wished. Supposing the surface of the water N N perfectly plane, which is the case in time of perfect calm, a beam of solar light incident with the direction S I will be deviated to IR after the refraction. Now from a boat the telescope can always be placed in the vertical plane passing through the sun. If the telescope be inclined in such a manner that its axis becomes perpendicular to I R, then the light received by the eye is emitted perpendicularly to the direction of the solar rays in water. This arrangement is analogous to that where the maximum of polarization of the light of the sky appears-that is, when one is looking at a right angle from the sun. In this manner I made a series of observations on the Lake of Geneva in a place where the water was sufficiently deep not to allow the ground to be seen; and I was able to perceive a marked polarization. The plane of polarization was coincident with the plane of incidence. If the telescope, always in the same vertical plane passing through the sun, be inclined in any other direction, then the polarization is the less marked the more the direction of the telescope differs from the perpendicular to IR. If the direction of the telescope be vertical, so that one is looking downwards, no trace of polarization appears. If the telescope be placed in a vertical plane perpendicular to the former one (that is, at 90° to the sun), the polarization is the more distinct the more the direction of the telescope becomes horizontal, and the plane of polarization passes through the direction of the telescope and the sun. I have not been able as yet to prove that the maximum of polarization corresponds exactly to the position of the telescope where its axis is at a right angle to the direction of the solar rays. It cannot escape your attention that the phenomena are more complicated in the case of water than in the case of the firmament. In the first place, it is evident that, if the surface of the water is agitated, the solar rays cannot remain parallel after the refraction. The phenomenon, therefore, is the less distinct the more agitated the water is. This is exactly what I observed. When I tried my apparatus for the first time, the water was agitated, and I remarked no appreciable polarization-although this may be attributed as well to the fact that, for some days before, we had strong northerly winds and the water was not perfectly transparent. On two other days, when the water was only slightly agitated, an appreciable polarization could be observed. On a very fine and calm day at last, the polarization was as marked as that of the sky, which, however, was not very blue at the time. In the second place, the solar rays entering the water are already partially polarized by refraction; but when the telescope is placed in the vertical plane passing through the sun (that is, in the position most favourable for observation), it is easy to see that the rays already polarized must be extinguished (exactly as in the experiment where you produce a blue cloud with a pencil of light already polarized). Finally, the direct solar light is not the only light that enters the water. There are scattered rays of diffused light, which, impinging upon the water from all directions, produce after refraction a blue non-polarized light, or, properly speaking, an infinite number of rays polarized with different planes of polarization. I have satisfied myself that, when the sky is covered, no appreciable polarization is observed. As I know of no prior description of this phenomenon, and as peculiar circumstances prevent for some time my pursuing this investigation, I communicate to you these results, although I regret that I was not able to finish the series of observations, and to repeat them with the use of artificial light. Yours &c., L. SORET. I hope my friend Soret will compare the action of the water of the Lake of Geneva upon light with that of other waters. By intensifying his illuminating beam he may be able to operate upon small masses. His method of experiment holds out a promise of a definite solution of a much discussed and still open question. An elaborate memoir, "Sur la Polarisation Atmosphérique," published in 1864 by Dr. Rubenson of Upsala, has just reached me. I promise myself much instruction from the perusal of this essay. Royal Institution, IN J. TYNDALL. XLIX. Fundamental Principles of Molecular Physics. [Continued from p. 287.] III. N his paper on the "Fundamental Principles of Molecular Physics" Professor Norton undertakes not only to answer my objections against his theory, but also to show, as far as he can, that some of my own views on the same subject are questionable, and others inadmissible. Having in my last article examined briefly his system of defence, I come now to a rapid review of his means of attack. The reader, if he has watched attentively the progress of our controversy, will have already noticed the striking ability displayed by my learned opponent in framing arguments out of objections. The last example of such tactics I have reserved for this part of my reply as a natural introduction to what I shall say concerning his other arguments. Bulk of atoms.-As Professor Norton had assumed the atom of "gross matter" to be indivisible and spherical in form, I took the liberty to object that atoms "indivisible" cannot be either extended or spherical in form; for "if they were extended and indivisible, they would be so many pieces of continuous matter, which we have already proved to be impossible." To this the learned Professor gives no less than six distinct replies, which I am now going to examine. The first is as follows: "Professor Bayma assumes that every point of matter acts instantaneously upon every other point at all distances, however great or small, with a force having the same character at all distances, and inversely proportional to the square of the distance. This may be probable, but it is not self-evident; and in fact no reason can be assigned why one material point having no extent should act upon another with a force decreasing with the distance according to any law whatever. The law of inverse squares is a consequence of wavepropagation, or of radiation along definite lines, received on a molecule of definite size, and cannot be predicated of a force that acts instantaneously between two mathematical points. To suppose such a law is an arbitrary assumption." I beg leave to make some remarks upon the few expressions which I have italicized. 1st. The word assumes should be changed into proves. (See Molecular Mechanics, pp. 31, 32, and 53-65.) 2nd. It is not self-evident: of course; and therefore it was made evident by the help of special proofs. 3rd. No reason can be assigned: and yet many were assigned, and others are still assignable. 4th. Wave-propagation is a propagation of motion, and has nothing to do with elementary action, which cannot be propagated (Molecular Mechanics, pp. 63-65). 5th. On a molecule of definite size. Continuous or not? If continuous, then the reply confirms my objection: if not, then the action is received on single material points, contrary to the assertion of my learned critic. 6th. Cannot. Why not? 7th. Mathematical points: mathematical does not here exclude physical. 8th. An arbitrary assumption: here the learned Professor gives himself the innocent pleasure of applying to me, by way of retaliation, what I ventured to say and to prove of some of his fundamental principles. Fortunately however those who have read my Molecular Mechanics' know that I have done enough not to deserve the compliment. I wish my learned opponent had done as much. But, even setting aside all these imperfections, the reader will undoubtedly see that this first answer of the learned Professor is not calculated to meet my objection. Accordingly I consider all further discussion of it as unnecessary. His second answer is the following: "If matter consists of material points, as supposed by Professor Bayma, it is no more difficult to conceive of an atom of continuous matter than of the space coextensive with it." This second answer I cannot well understand. Surely, the learned Professor does not mean that, if matter (as I have not only supposed, but proved) consists of separate material points, then continuous matter can be more easily conceived. Yet what else is the natural sense of his conditional proposition? How |