186 METEOIIOLOGICAL OBSERVATIONS. 
Secondly^ That the mean of the maxima and minima 
differs ^^-q from the average of the twenty-four hours. 
(The mean of the means of the maxima and minima which 
occurred during the series was 49°.28 varying only 
from the standard of comparison.) 
Thirdly, That of the averages of the temperatures of 
two hours, those of 5 a. m. and 5 p. m. noon and midnight, 
11 A. M. and 11 p. M., 10 A. M. and 10 p. m. approach 
nearest to the mean of the day. The average of 5 and 5, 
differing from it only^^J^ of a degree; of 12 and 12, 
; of 11 and 11, ; of 10 and 10, |. 
Fom^tldy, That of the averages of the temperatures of 
three hours, those of 7 a. m,, 1 p. m. and 10 p. m., ap- 
proximate most closely to the average of the whole twenty- 
four, the mean of the former differing from the latter. 
Fifthly, Of all the combinations of hours I tried, none 
gave the true mean temperature of the day so well as the 
mean of 5 a. m. 10 a. m. % p. m. and 10 p. m., which cor- 
responded with the average of the day within j/o^ ^ 
degree. 
Such are the simple facts regarding the most important 
results to be obtained from this series of observations : — I 
abstain from farther remarks at present, and shall now con- 
clude by mentioning the other results, less important, de- 
rived from the same source. 
1. At an average, the maximum of temperature occurred 
at 4 p. M. and the minimum at 5 a. m. 
% The average daily range was 9.93, its maximum be- 
ing 23°, which happened in August. Its minimum 5° in 
February. 
3. The temperatures of 7 a. m. and 10 p. m. were ex- 
actly the same.. 
Leith, ) 
J?/%1823. I 
