Mr. J. P. Harrison on the Moon's Influence over Cloud. 143 



them without any disengagement of gas ; make them unclean, 

 and then the adhesive force of the solid becomes more energetic 

 for the gas than for the liquid, and there is a consequent separa- 

 tion of gas from the solution. 

 King's College, London. 

 July 1867. 



XVIII. On the Moon's Influence over Cloud. 

 By J. Park Harrison, M.A. 



To the Editors of the Philosophical Magazine and Journal. 

 Gentlemen, 



IN Mr. Ellis's communication on the tendency to disper- 

 sion of cloud under a full moon, in the last Number of 

 the Philosophical Magazine"*, sufficient account has not, I think, 

 been taken of the probable effect of the interruption of conti- 

 nuous observation on more than 400 days in the seven years 

 (1841-47) during which two-hourly observations were made at 

 Greenwich. These, for all scientific purposes, blank days were 

 not equally distributed over the lunation; ten would perhaps 

 be found to occur at conjunction, and sixteen at opposition. 

 And the imperfect means derived from three or four observations 

 on these days cannot be " corrected " so as to make them avail- 

 able or admissible in- an inquiry of this nature. The amount of 

 cloud during sixteen or eighteen hours, for example on a day 

 of full moon, cannot be supplied from the general mean of cloud 

 for those hours on other days of the lunation. 



In addition to this, Mr. Ellis has grouped several days together, 

 and so merged and eliminated any distinctive character they may 

 individually have possessed. It is quite true that the means of 

 seven days at new and full moon, and at first and last quarter, 

 at Greenwich, are nearly identically the same, and consequently 

 approach very closely the mean of the whole lunation. But the 

 curve of cloud which I constructed from the means of the more 

 complete observations eight years agof, for comparison with 

 curves of mean temperature according to the age of the moon, 

 shows that there certainly are clear and cloudy periods — a fact 

 which receives confirmation from the results obtained by Schiibler 

 and AragoJ. 



Thus, for example, at the period of last quarter, on the day 

 itself and on the three days before and after it, the mean amounts 



* Phil. Mag. July 1867. t Ibid. March 1859. 



% From long series of observations at Paris, Stuttgard, Munich, and 

 Augsburg, the greatest number of clear days and the fewest rainy days 

 were found to occur after last quarter. (Annuaire du Bureau des Longitudes, 

 1833, p. 161, &c. See also Smythe's translation of Arago's works.) 



