Page:Madras Journal of Literature and Science, series 1, volume 6 (1837).djvu/58

50 Professor Forbes from a comparison of most of the recorded observations, has deduced a formula for finding the mean oscillation between 10 and 4  at any given latitude. The result of this for latitude 16° 30' is 2.35 millimetres or 0.092 of an inch. The mean of the observations in the table cannot be considered the experimental mean at Moulmein, for, three out of the four unrecorded months being rainy, the former would almost certainly exceed the latter. The mean of 0.133 the greatest and of 0,077 the least quantities in the table, equal to 0.105, is probably very near the true mean. The difference between this and the result of the formula is a less fraction of the whole oscillation than occurs in a similar comparison in lat. 56° between the result of the formula and the mean of Professor Forbes's own observations; the errors in the two cases are however of opposite kinds.

The manner in which the observations in table III. were made, has been already noticed. The calculated altitudes of the sun at the period of the first observation on 7th January, and at that of the last on 30th April, were respectively 51° and 88° 17'. Supposing the atmosphere equally pervious, the ratio of the intensity at the former angle to that at the latter, is theoretically a little greater than 9-tenths, a ratio approaching much nearer to equality than it would have done if the altitudes had been smaller with the same common difference. It is remarkable that on the 14th January the photometer indicated 37, the greatest intensity recorded, and on the following day 32, the only other observation as great as the latter of these having been on the 30th April. It suggested itself to me at one time as probable that the apparent effect of the sun's rays might be influenced by the temperature of the instrument employed to measure it, but there can be no doubt that this and every other disturbing cause is excluded in the results obtained by means of the actinometer. The varying quantity of undissolved vapour lying in the path of the rays, offers perhaps the best explanation of the anomalous variations in their intensity, as was suggested by Dr. Richardson in explanation of the apparently great power of the sun's rays at Fort Franklin. Successive strata of thin vapour at a great height, may be conceived to intercept many of the heating rays, without causing a diminution of transparency sensible to the eye.

I regret extremely that the tables contain no observations on terrestrial radiation at night. The method of using the actinometer in the measurement of solar radiation, as already described, is applicable also to terrestrial radiation. Of course in this case the glass plate between the bulb of the instrument and the sky would need to be removed: it would also be necessary to lengthen the time of observation, and,