154 Mr. J. P. Harrison on Lunar Influence over 



only at Dublin and Greenwich, but at the Cape, Madras, and 

 Toronto. 



And lastly, maximum or minimum mean temperatures were 

 found to prevail on certain days of the lunation more than on 

 others, — for example, more maxima than minima on the third 

 day before new moon, and on the second day after first quarter, 

 and more minima than maxima on the second day after third 

 quarter, both at Dublin and Greenwich, for a series of years 

 embracing more than 400 lunations*. 



These facts were communicated to the British Association in 

 the years 1857 and 1858, with the object of bringing them as 

 early as possible before the notice of meteorologists ; and they 

 have since been confirmed by Mr. J. C. Bloxam, from an ex- 

 amination of observations extending over sixteen years in the 

 Isle of Wight. Thus in the years 1841-56 the temperature at 

 Newport was found to be 0°'8 lower at the third or last quarter 

 than at the first : and during five days at first quarter (from 

 the third day before to the second day after that phase), there 

 was an increment of temperature amounting to 0°"226per diem; 

 the corresponding days at last quarter showing a decrement of 

 temperature amounting to 0°*110 per diem. The values for N. 

 and S. declination in the summer and winter months were also 

 found to point to the existence of lunar influence in a very 

 marked mannerf. I niay add that it has now been ascertained 

 that at Greenwich the mean temperature of the second day after 

 last quarter is, on the average of forty-three years, 0°'9 below 

 that of the second day after first quarter. 



2. Before proceeding with the subject to which I wish to draw 

 attention in the present communication, it will be necessary to 

 explain that the curve of mean temperature for ten years at Green- 

 wich alluded to in paragraph 1, has been remodelled. It is now 

 formed from the observations of mean temperature for the ten 

 years commencing November 1845 and ending October 1855 

 (Plate I. fig. 2); and the mean temperature of the ten preceding 

 years, commencing November 1835 and ending October 1845, has 

 been formed into a second curve at the same station (fig. 3). The 

 temperatures of the two decades presented a remarkable contrast. 

 Prom the means of each year from 1845 to 1855, as given by Mr. 

 Glaisher in the Philosophical Transactions, it was found that the 

 mean temperature of the years 1845-55 was 49°'4; of the years 



* The ratio in which maxima occurred at Greenwich from 18 14 to 1835, 

 and at Dubhn from 1830 to 1852, was as follows : — On the third day be- 

 fore new moon, as 3 : 2; on the second day after first quarter, 2:1. The 

 ratio which minima bore to maxima on the second day after third (or last 

 quarter) was 3:1. 



t Meteorology of Newport iu the Isle of Wight, p. 143. 



