Temperature of the Am. 
265 
and considerably later in the other months, being about 2 h 40 m in June. If we group the months into the 
astronomical quarters, we arrive at the following result : — 
Spring, maximum temperature, occurs at l h 40 m p.m. Makerstoun mean time. 
Summer, l h 30 m 
Autumn, l h 50 m 
Winter, l h 0 m 
Or, that the maximum temperature occurs later in the day at the equinoxes than at the solstices. It will re- 
quire other years observations to prove the generality of this fact; but it seems to bear some relation, and 
that apparently not of difficult explanation, to the amount of the diurnal range of temperature. 
The mean temperature for the year occurs at 8 h 22 m a.m. 
6 h 59 m p.m. 
The interval between these periods is 10 h 37 m 
The critical interval varies with each month, and is greatest in June, being ll h 46 m , and least in Feb- 
ruary, being 8 h ll m . 
TABLE IV. — Errors of the Approximate Mean Temperatures for each Month and Quarter in 1843, 
deduced from one or two Daily Observations. 
True 
Approximate Means (+) greater, or ( — ) less than true Means. 
Periods. 
Mean. 
Max. 
18* 
18 h 
19 h 
20* 
20 h 
21 h 
22 h 
22k 
23t 
18i> 
and 
and 
and 
and 
and 
and 
and 
and 
and 
and 
and 
8''. 
Min. 
5>>. 
6-. 
6\ 
7 h. 
8*. 
8 h . 
9 h . 
10". 
10>>. 
0 h . 
o 
o 
0 
o 
o 
0 
0 
O 
O 
o 
O 
O 
C 
January 
38-72 
+ 0-3 
+ 0-15 
-0-02 
-0-10 
-0-36 
-0-55 
-0-48 
-0-54 
-0-65 
-0-22 
+ 0-23 
-0-18 
February 
32-98 
+ 0-4 
+ 0-08 
-0-38 
-0-34 
-0-65 
-0-90 
-0-73 
-0-59 
-0-63 
-0-06 
+ 0-41 
-0-80 
March 
39-07 
+ 1-2 
+ 0-03 
-0-61 
-0-36 
-0-92 
-1-73 
-0-88 
-0-28 
-0-51 
+ 0-45 
-0-12 
-0-23 
April 
44-55 
+ 0-7 
-0-52 
-1-20 
-0-34 
-0-31 
-113 
+ 003 
+ 0-74 
+ 0-33 
+ 0-84 
-0-17 
-0-40 
May 
46-57 
+ 0-4 
-0-33 
-0-62 
+ 0-15 
+ 0-25 
-0-42 
+ 0-40 
+ 0-55 
-0-11 
+ 0-43 
-0-21 
+ 0-56 
June 
51-97 
+ 0-1 
+ 0-13 
-0-06 
+ 0-60 
+ 0-64 
+ 002 
+ 0-62 
+ 0-61 
-0-65 
-0-24 
-0-27 
+ 1-02 
July 
56-83 
+ 0-3 
-0-53 
-0-94 
+ 0-11 
+ 0-42 
-0-33 
+ 0-52 
+ 0-72 
+ 0-08 
+ 0-47 
-0-13 
-0-16 
August 
57-29 
+ 0-6 
-0-49 
-1-12 
-0-08 
-0-12 
— 1-21 
+ 0-10 
+ 0-73 
+ 0-06 
+ 0-98 
-0-68 
+ 0-63 
September 
55-22 
+ 0-5 
-010 
-0-92 
-0-13 
-0-67 
-1-99 
-0-59 
+ 0-09 
-0-63 
+ 0-67 
-0-44 
+ 0-28 
October 
42-27 
+ M 
-012 
-1-05 
+ 0-86 
-1-27 
-1-86 
+ 0-93 
-019 
-0-39 
-0-46 
+ 0-79 
- 1-32 
November 
39-59 
+ 0-2 
-0-19 
-0-91 
-0-79 
-0-84 
- 101 
-0-70 
-0-45 
-0-51 
+ 0-08 
+ 0-67 
- 1-10 
December 
45-26 
-0-3 
-0-23 
-0-53 
-0-52 
-0-58 
-0-66 
-0-27 
+ 0-0.7 
+ 0-02 
+ 0-24 
+ 0-36 
-0-49 
Spring 
43-40 
+ 0-8 
-0-28 
-0-82 
-0-19 
-0-33 
-1-10 
-0-16 
-0-34 
-0-11 
+ 0-57 
-0-18 
-0-03 
Summer 
55-36 
+ 0-3 
-0-29 
-0-70 
+ 0-21 
+ 0-31 
-0-50 
+ 0-41 
+ 0-58 
-0-17 
+ 0-40 
-0-36 
+ 0-50 
Autumn 
45-69 
+ 0-6 
-0-13 
-0-96 
-0-59 
-0-92 
-1-66 
-0-74 
-0-18 
-0-50 
+ 0-41 
+ 0-34 
-0-71 
Winter 
38-99 
+ 0-1 
0-00 
-0-31 
-0-32 
-0-52 
-0-70 
-0-50 
-0-36 
-0-42 
-0-02 
+ 0-3.3 
-0-49 
The Year 
45-86 
+ 0-5 
-0-17 
-0-70 
-0-25 
-0-36 
-0-98 
-0-24 
+ 0-09 
-0-30 
+ 0-34 
+ 0-04 
-0-18 
The 12 months. 
Mean of Errors 
0-5 
0-24 
0-70 
0-36 
0-59 
0-98 
0-52 
0-46 
0-38 
0-43 
0-37 
0-60 
Range of E 
rrors 
1-5 
0-68 
1-18 
1-65 
1-91 
2-01 
1-81 
1-33 
0-98 
1-44 
1-47 
2-34 
The quantities given as the true means are from Table III. ; they are, therefore, only approximate, but 
they must be very near the truth. The errors of the mean from the maximum and minimum thermometer are 
obtained from Table II., after deleting the means for Sundays. The means for the odd hours were obtained 
by taking the mean of the even hours between which the odd hours lie. 
The couple of hours best fitted for observations., in order to obtain the best approximation to the monthly 
means, must evidently be determined by 
1st, The smallness of the mean of the monthly errors, 
2d', The smallness of the range of the monthly errors. 
MAG. AND MET. OBS. 1843. 
3 x 
