590 - ON THE OPTICAL CHARACTER . [438 especially when diluted with white light reflected from an outer surface not forming part of the boundary of the thin plate*. . Michelson, who with his great authority supports the surface-colour theory, mentions several tests under four headings (p. 561). To my mind these tests are as well, if not better, borne by an interference theory. But reliance seems to be chiefly placed upon " the more rigorous optical test of the measurement of the phase-difference and amplitude-ratios " when polarized light is reflected. I agree that this is a cogent argument, and unless it can be met the balance of evidence derived from simple observation would perhaps incline to the surface-colour theory. It is, I think, the fact that many beetles" exhibit a less well-marked polarizing-angle than could be reconciled with the usual theory of thin plates constituted of non-absorbent material. An escape from the difficulty might perhaps be found in imagining a stratification composed of more than two materials, so that, for instance, the polarizing-angle for the first and second might differ considerably from that corresponding to the second and third. But such a structure seems rather improbable, and any combination of thin plates composed of two transparent materials only should give a definite polarizing-angle, abstraction being made from the minor deviations observed by Airy and Jamin. At this point it may be recalled that a well-marked polarizing-angle and a sudden change of relative phase through two right angles are more closely connected than is sometimes realized. The latter without the former would involve a physical discontinuity. Michelson considers that in practice the phase-change affords the more delicate criterionf, and that in most cases it is decisive in favour of surface-colour. A circumstance which may perhaps be regarded as telling upon the other side is afforded by the variety of colouring at different parts, but at the same angle (e.g. at perpendicular incidence) seen in certain beetlesDr Hodgkinson mentions Chrysocliroa fulminans. The " colours vary in an indescribable manner when attentively examined at different angles of incident light with the eye. alone; with the mirror (viz., at perpendicular incidence) the wing-oases are seen to be coloured successively from base to tip iridescent green, yellow, orange, and red, and these tints remain unaltered by change of position of the object." I have confirmed generally this observation, and other beetles show something similar. The explanation makes large demands upon the Surface-colour theory; but a moderate change -of structure is all that would be required by interference. . .A caution is perhaps required against regarding the two theories as mutually exclusive. Both Walter and Michelson admit exceptions, and * -Ed. Traris. loc. cit. - : . . t Some of Michelson's diagrams are rather confusing in that they suggest a phase-difference of 180° between'the two polarized components reflected perpendicularly, when evidently the distinction between the two components disappears. -.;-6,1807). If the mercury be wet, boiling may be dispensed with and negative pressures of two atmospheres are easily demonstrated.