Page:Proceedings of the Royal Society of London Vol 1.djvu/36

Rh hope, he apprehends, must be relinquished of ever effecting the decomposition of the muriatic acid in the way of simple elective attraction; its basis being probably some unknown body, which nothing but the application of complicated aﬂinities will perhaps ever enable us to discriminate.

On doable Images caused by atmospheric-a1 Refraction. By William Hyde Wollaston, M.D. F.R.S. Read March 6, 1800. [Phil. Trans. 1800, p. 239.]

The remarkable instances of double and triple images of the same object produced by aerial refraction near the horizon, lately communicated to the Society by Mr. Huddart. Prof. Vince, and Mr. Dalby, have given rise to the present paper, in which the author attempts to explain these phaenomena on theoretical principles, and to illustrate his conclusions by artiﬁcial experiments.

Admitting the inference given by Professor Vince, that these appearances arise from certain unusual variations of increasing density in the lower strata of the atmosphere, our author undertakes, lst, to investigate the successive variations of increasing or decreasing density to which ﬂuids in general are liable, and the laws of the refractions occasioned by them; 2dly, to illustrate and conﬁrm the truth of this theory by experiments with ﬂuids of known densities; and lastly, to ascertain, by trial upon the air itself, the causes and extent of those variations of its refractive density on which the inversions of objects and other circumstances observed in the above phaenomena seem to depend.

Under the ﬁrst head we ﬁnd the demonstrations of three propositions, deduced from the general laws of refraction. The ﬁrst imports, that if the density of any medium varies by parallel, indeﬁnitely thin strata, a ray of light moving through it in the direction of the strata, will be made to deviate during its passage; and the deviation will ever be proportionate to the increment of density where it passes. From the second it appears, that when two ﬂuids of unequal densities are brought into contact, and unite by mutual penetration, if the densities at different heights be expressed by ordinates to a perpendicular line drawn across the ﬂuids, the curve drawn through the terminations of these ordinates will have a point of contrary ﬂexure. And in the third proposition it is shown, that if parallel rays pass through a medium, varying according to the preceding proposition, those rays above the point of contrary ﬂexure, where the line will be concave, Will be made to diverge, while those below the same point, where the curve will be convex, will converge after their passage through it. The converging rays, it follows hence, will at a certain distance, proportionate to the quantity of convergency, meet in a focus, beyond which they will diverge again, and thus produce effects perfectly similar to those caused by a medium of uniform density, having a surface similar to the above-mentioned curve of densities, whether convex or concave, according to the nature of that curvature. Hence may be inferred