taken after this took place have been rejected. Occasional clouds. Occasional clouds, strong gusts of wind. No note of cloud, very little Wind blowing strong into No note of cloud till just at the end of these observations. A very little wind; occasional clouds. Halo with hazy clouds: moon seen through them with much-diminished brilliancy. Frequent passing clouds during the latter part of these observations. No cloud visible, but haziness suspected, as it existed both at sunset and at sunrise. We have then Q (quantity of heat coming from the moon's surface) *This formula is based on the assumption that the heat coming to the earth where e-apparent distance between the centres of the sun and In column 6 we have the deviation for full moon calculated from the observed mean deviation for each night. In column 7 the supplement of the apparent distance between the centres of the sun and moon. In column 8 the approximate mean altitude of the moon. In column 9 the number of times the telescope was put on or off the moon during the observations included in the mean result. In all these observations the deviations which have been measured are those due to the difference between the radiation from a circle of sky containing the moon's disk, and that from a similar circle of sky close to it not containing the moon's disk. The annexed diagram will show approximately the rate at which the moon's light increases and diminishes with its phases as deduced from formula (a); and the ringed dots with the accompanying Roman figures (for reference) give the quantity of the moon's heat as determined by observation on different nights. Although there is considerable discordance between some of the observed and calculated quantities of heat, the results suggest to us that the law of variation of the moon's heat will probably be found not to differ much from that of the moon's light. It therefore follows that not more than a small part of the moon's heat can come from the first of the three sources already mentioned. With the view of ascertaining what proportion of the sun's heat does not leave the moon's surface until after it has been absorbed, some readings of the galvanometer were taken on four different nights near the time of full moon, with a disk of thin plate glass in front of the face of each pile; and the deviation was about six or eight divisions. As the glass screens were examined with care for dew after removal on each night, and none was perceived except on one occasion, the probable percentage of the moon's heat which passes through plate glass is 8, or rather less. Few experiments appear to have been made on the absorptive power of glass for the sun's rays; but, from the best data that I have been able to obtain, I find that probably about 80 per cent. pass through glass. The greater part of the moon's heat which reaches the earth appears, therefore, to have been first absorbed by the lunar surface. from an element (S) of the moon's surface K. S. cos. cos, 0 and being respectively the inclinations of the lines to the sun and to the earth from the normal to that point of the moon's surface, and K a constant. It now appeared desirable to verify this result, as far as possible, by determining by direct experiment the proportion which exists between the heat which reaches the earth from the sun and from the moon. If we start with the assumption that the sun's heat is composed of two portions, and the luminous rays, whose amount = L, also that the moon's light consists of two corresponding portions, L', O', the luminous not being absorbed, and the non-luminous being entirely absorbed in their passage through glass, then L = 8, L+O L' L' L+O ='08; L+0 ] 80,000 its generally received value (800,000), we have (b) Owing to the extremely uncertain state of the weather, only one series of eighteen readings was obtained for the determination of the sun's heat. A beam of sunlight was thrown, by means of a plane mirror, alternately on and off a plate of polished metal with a hole •175 inch in diameter. At a short distance behind this the pile was placed. The deviation thus found was connected with that previously found for full moon by using the deviation produced by a vessel of hot water as a term of comparison. The relative amount of solar and lunar radiation thus found was which is quite as near that given by (b) as we could expect when we consider the roughness of the data. As a further confirmation of the correctness of the two rough approximations to the value of the ratio existing between the sun's and the moon's radiant heat already given, the subject was investigated from a purely theoretical point of view. It was assumed (1) That the quantity of heat leaving the moon at any instant may without much error be considered the same as that falling on it at that instant. (2) That the absorptive power of our atmosphere is the same for lunar and solar heat. (3) That, as was already assumed in obtaining formula (a), the moon is a smooth sphere not capable of reflecting light regularly. Then the heat which leaves the moon in all directions = quantity which falls on the moon = of the quantity which falls on the earth from the sun 1 13.55 K = K. {(-e). cose + sin e} sin e. de= 3п. 4 therefore (if we may be allowed the expression) sun-heat 13.55 × 3π 79,000 = moon-heat E = (quam proximè).. 1 (d) E In the above, the proportion between the areas of surface presented by the moon and earth to the sun is taken 13.55, and the angle subtended by the earth at the moon =1° 55'. The value of the readings of the galvanometer was determined by comparison with those obtained by using a vessel of hot water coated with shellac and lampblack varnish as a source of heat. The vessel was of tin, circular, and subtended the same angle at the small concave reflectors as the large mirror of the telescope. It was thus found that (the radiating power of the moon being supposed equal to that of the lampblack surface and the earth's atmosphere not to influence the result) a deviation of 90 for full moon appears to indicate an elevation of temperature through 500° Fahr.* In deducing this result allowance has been made for the imperfect absorption of the sun's rays by the lunar surface. In the present imperfect state of these observations it would be premature to discuss them at greater length; but as some months must elapse before any more complete series can be obtained, and the present results are sufficient to show conclusively that the moon's heat is capable of being detected with certainty by the thermopile, I have thought it best to send this account to the Royal Society; and I shall be most happy to receive suggestions as to improvements in the method of working, and as to the direction in which it may be most desirable to carry on future experiments. GEOLOGICAL SOCIETY. [Continued from p. 243.] February 10th, 1869.-Prof. T. H. Huxley, LL.D., F.R.S., The following papers were read :— *This may seem a very large rise of temperature; but it is quite in accordance with the views of Sir John Herschel on the subject (Outlines of Astronomy, section 432 and preceding sections), where he says that, in consequence of the long period of rotation of the moon on its axis, and still more the absence of an atmosphere, "The climate of the moon must be most extraordinary, the alternation being that of unmitigated and burning sunshine, fiercer than that of an equatorial noon, and the keenest severity of frost, far exceeding that of our polar winters, for an equal time." And again, ". . . . the surface of the full moon exposed to us must necessarily be very much heated, possibly to a degree much exceeding that of boiling water." |