176 Dr. H. F. Parsons on some further 
during which on some days the mean temperature of the air has 
not exceeded 17°, and a minimum of 9° F. has been recorded, 
bears out the remark made in my former paper, that it can very 
rarely happen that the ground is frozen at a depth of one foot. 
The curves of the 4 ft. thermometer during the three years 1892- 
94 are shown in a diagram,* in which the curve for 1892 is repre- 
sented in blue, for 1893 in red, and for 1894 in black. The 1894 
curve is formed of the weekly means of the daily readings instead of 
single readings weekly, but the difference is almost inappreciable, 
as it rarely happens that the mean of seven daily readings differs 
from the reading of the middle day of the seven by more than a 
tenth of a degree Fahrenheit. In each year the general course 
of the curve has been similar. The minimum is reached a few 
weeks after the commencement of the year, in the latter part of 
January or February; but in two out of the three years there 
has been a double minimum, the first being attained in January, 
and the second about the end of February or in March. From 
the end of March to the beginning of July there is a rapid and 
almost continuous ascent. Later in July there has been in each 
year a slight decline, from which the temperature has risen 
again to its yearly maximum in August; so that there seems to 
be a tendency to both a double maximum and a double mini- 
mum, though it is possible that this might not be borne out by 
a more extended series of observations. (The Greenwich curve 
for the mean temperature of the air shows a trace of a double 
minimum in January.) From August the earth temperature 
falls steadily to January, the fall being steeper and more regular 
than the ascent. 
In 1892 the lowest temperature attained at 4 ft. depth was 38-8° 
on Jan. 24th, and again 39°6° on March 18th; and the 4 ft. tem- 
perature did not begin to rise rapidly until the middle of May, the 
spring being cold and backward. The highest point attained was 
60-0°, on Aug. 21st. There was a cold October, with a rapid fall 
of the 4 ft. thermometer, after which the temperature remained 
stationary for several weeks. 
1893 began with a week of intense cold, the thermometer in 
the shade attaining a minimum of 12°, on Jan. 4th. The tem- 
perature at 4 ft. depth sank to a minimum of 39°, on Jan. 22nd. 
After this, however, milder weather set in, and the spring and 
summer were very warm and dry. The 4 ft. temperature rose 
steadily and rapidly, and in several weeks of the spring and 
summer was 5° or more higher than it had been in corresponding 
weeks in the year before. The highest temperature attained 
by the 4 ft. thermometer was 63°5°, on Aug. 20th, 384° higher 
than in either the preceding or the following year. Although 
late autumn was cold, a good deal of the extra heat received in 
* Not reproduced. 
