a well-known product of decomposition of chlorophyll. Another is a substance of well-marked properties, nearly resembling, but not identi- cal with, phyllocyanin. It has not, so far as my experience goes, been hitherto observed as a result of any process of decomposition to which chlorophyll has been subjected outside the animal body. I consider it as a body $ui generis, characterised by its fine purplish-blue colour and its brilliant metallic lustre. The existence of other products in addi- tion to these two is possible. On one occasion, indeed, a definite crystalline substance was obtained, which seemed to be peculiar, but that it was in any way connected with chlorophyll could not with certainty be maintained.
In March, 1894, Dr. G. Johnstone Stoney communicated to the Society a memoir by myself and Mr. P. L. Gray, entitled " Experi- mental Investigations on the Effective Temperature of the Sun," which was published in the Phil. Trans.,' A, vol. 185 (1894). In these investigations the method we adopted was as follows : A beam of sunlight was sent horizontally into the laboratory by means of a Stoney single-mirror heliostat. The mirror was an optical plane of unsilvered glass, and the beam was directed into one aperture (A) of a differential Boys' radio-micrometer. The other aperture (B) received the radiation from a strip of platinum, which could be raised to any desired temperature by an electric current supplied by a battery of accumulators. The temperature of this strip was at any moment determined by its linear expansion, the instrument being previously calibrated by melting on it minute fragments of AgCl and of pure gold, as in Joly's meldometer. In front of the aperture (B) of the radio-micrometer was placed a stop with a circular hole of 5 '57 mm., and the distance of this hole from the receiving surface of the thermo-couple was 60 "2 mm. This gave for the angle subtended by a diameter of the aperture at the receiving surface 5*301. Knowing then (i) the ratio which the angular diameter of this circular aperture bears to that of the sun, (ii) the temperature of the platinum strip at the moment that the radio-micrometer is balanced, (iii) the amount of the sun's radiation lost by reflection from the heliostat mirror and also by absorption in the earth's atmosphere, it is possible on any assumption with regard to the law connecting radiation with tem- perature, to determine the effective temperature of the sun. A series of very accordant observations were made in this way, the mean of which gave 6200 C. as the effective solar temperature.
