ON THE HOURLY OBSERVATIONS OF THE THERMOMETER. 185 
winter months begin with October and end with March, both 
inclusive. Sir David Brewster seems to have joined October to 
the winter months, and April and May to the summer months, 
from having observed a coincidence in the form of the projected 
curves of temperature of April and May with those of summer, 
and the curve of October with those of winter. This classifica¬ 
tion appears natural and sufficiently perfect, and is not opposed 
to the results of the observations at Plymouth, the daily curve 
for October being more flattened and of less range than that of 
April or May, as may be seen by a reference to Plate VII. in 
which the monthly curves are projected. When the register 
terminates it may be perhaps desirable to consider the results of 
other distributions; this may be done with but little additional 
trouble, since the observations are carefully recorded in large 
folio volumes, and all the monthly and annual means computed 
and set down ; including intervals of ten days each. 
The numerical results for the years 1833 and 1834, of the 
hourly, daily, and monthly temperatures, as also for periods of 
three months each, comprising spring, summer, autumn, and 
winter, likewise summer and winter, including six months 
each, are given in the eight following tables. From these we 
may readily deduce: 
1 . The mean temperature of various seasons, and that of the 
whole year. 
2 . The daily progression of temperature. 
3. The two periods of each day at which the mean tempera¬ 
ture occurs. 
4. The relation between the mean temperature of the whole 
twenty-four hours, and that of any single hour, or of two similar 
hours. 
5. The average daily range for each month. 
Lastly, the form of the curves described by the march of the 
temperature between given periods of the day and night. 
