196 ' report — 1859. 



propose to illustrate the subject by examples of lunar action in the spring 

 and autumn months. Thus in the early part of May, it will be interesting 

 to remark the amount of Lunar Influence exerted at the period of low tem- 

 perature which embraces Dr. Marller's three cold days, viz. the 11th, 12th, 

 and 13th, and which on an average of 86 years' observations at Berlin was 

 found to be more than 2 degrees colder than the calculated mean of the season. 

 The following Table of mean temperatures of the first twenty days of May 

 for 43 years at Greenwich, will show the amount of depression which oc- 

 curred at that station. The means are in each case for the civil day. 



1st, 50-3 2nd, 51-5 3rd, 50-9 4th, 51-5 5th, 51 8 



6th, 51-9 7th, 523 8th, 52-1 9th, 51-0 10th, 50-9 



11th, 51-6 12th, 51-3 13th, 51-0 14th, 506 15th, 51-9 



16th, 531 17th, 540 18th, 535 19th, 530 20th, 539* 



If we now examine a lunar curve (fig. 3) of the mean temperatures of the 

 11th, 12th, and 13th days at Greenwich, — they are purposely taken, though 

 not the coldest, — it will not fail to be noticed that the general line of the 

 wave, notwithstanding its pronounced character, follows the model curve, 

 with the exception of a remarkable rise on the second day before first 

 quarter, and on the second day before last quarter. It bears also a very close 

 resemblance to the curve of temperature for the year 1859. 



A lunar curve of the mean temperature for the month of May during 43 

 years has also been formed, and found to agree with the model curve ; the 

 mean of the means of the day of first quarter and five days after is 54'0, of 

 the day of full moon and five days after 51 0- 9. The amount of cloud on a 

 seven years' average for the second day after full moon is 4"9 ; for the second 

 day after first quarter 8*1; and the mean amount of cloud at the syzygies and 

 quarters for the day of the moon's change and the day preceding and follow- 

 ing is as follows : — 



At New Moon 6-8 



At First Quarter 6*9 



At Full Moon 5*7 (the minimum). 



At Last Quarter 6"3 



10 represents an entirely clouded sky. 



On viewing these results one cannot but recall to mind the belief of the 

 French gardeners in the ravages of "La Lune rousse" towards the end of 

 April or beginning of May ; and the explanation of the phenomenon given 

 by M. Arago, — that it was, without doubt, due to the absence of cloud. 



The observations of mean temperature at this period, however, and the 

 relative number of days of the lunation on which they occur, deserve a more 

 minute consideration. To facilitate it, Tables have been formed of the mean 

 temperatures of five consecutive days at full moon, and five corresponding 

 days at first quarter: and for the purpose of reference, a Table of the mean 

 temperatures of the month of May for 43 successive years is appended. 



On referring to these Tables, it will be at once perceived that the mean 

 temperature of each of the five days at full moon (see Table II.) is far below 

 that of the five days at first quarter (Table I); and also that the number 

 of observations which occurred in the 43 years at the two periods is very 

 different. 



* From observations kindly furnished by Principal Forbes, of St. Andrews, it appears that 

 the 9th, 10th, and 14th days of May were the co'dest at Edinburgh on an average of 40 years. 

 The 12th, 13th and 14th days, allowing for the estimated march of temperature, were the coldest 

 days at Greenwich. It is possible that the epoch of greatest depression would in a longer 

 series of years coincide with that at Berlin. 



