vita a» 
. hd 
EVIDENCE OF LUNAR INFLUENCE ON TEMPERATURE. 251. 
_ form, with the dates attached, for 22 years, in the results of the Dublin Ob- 
_ servations. And although it was scarcely to be expected that monthly 
maxima and minima would coincide to any extent with the annual mean tem- 
peratures (or means of the means) of the eight days at the different lunar 
periods, in the first half of the lunation they appeared in a great measure 
to do so. 
At the periods of the syzygies and quarters, upon the days of the change, 
and for the three days before and after, the proportion in which maxima and 
minima mean temperatures occurred in each month, is shown approximately 
in the following Table; the signs + and — before the figures indicating the 
predominance of maxima or minima :— 
Taste II. 
Months. C-) ») e) c 
November ........- +5:4|/4+2:1 = —2:1 
December ..... see) tos2 = —o3:52|—5:4 
WANUATY vvesacaede*s +2:1)/4+2:1)/—3:2)—2:1 
MEGIMATY s-..ccsewe ss +3:21+3:2)—3:2 = 
Marcela vivertt. (ace. +2:1)/—2:1})/—2:1/—3:2 
aj ial eek ag pe eee +5:4/—3:214+5:4/+3:2 
Maly .isiscvcoedseaseas —3:2)/4+3:2)/—3:2)/43:2 
JUNE: {Phew sank used. —5:4|—5:4]/—38:1}+2:1 
HAW cade cats pepsde isp —3:2/4+2:1/+5:4|/—3:2 
August.......sesseee —3:2)/—5:4/4+3:2 = 
September ......... —3:2 = —3:2/4+2:1 
October ...... ecovee) — 221} 43:21 4+ 3:2) —2:1 
. Thus, in winter, for six consecutive months, maxima predominate at @; 
_ for the other six months minima, and that in much the same proportion. 
In four of the winter months, viz. December, January, February, and March, 
in which maxima preponderate at the period of @, minima are in excess at O. 
In July, August, and October, the converse holds good. At the quarters, a 
‘similar reciprocal action takes place in the months of October, November, 
January, April, June, and July. 
_ 4, Not the least striking fact which has been elicited during the progress 
_ of this investigation, is the systematically unequal distribution of the maxima 
and minima mean temperatures over the several days of the lunation. At 
Dublin, for the period under consideration, the greatest number of high and 
w temperatures at each quarter, excepting at O, occur upon the day fol- 
a the change. In the annexed Table this oe be clearly seen, as well 
») t fe) t 
sgl 62 Alt Rete eA 
8 21 24|15)2 23) 7 {11 1415/18/21 13) 8 | 8 |15{15,27/26)24|12 
C 
18/9/15 18)19 25/26/20)16 
_ The other points to which I wish to direct attention in this Table are the 
excess of maxima and minima upon the third day before and third day after 
