1895-96.] Dr J. Halm on the Temperature of the Air. 
271 
I think this is the proper place to make a few remarks about 
the influence of horizontal convection currents on the temperature 
of the air. We know that the principal difference in the daily 
curves of temperature between continental and maritime climates is 
due to the periodic currents carrying colder air from the sea and 
mixing it with the more heated elements over the ground. Appa- 
rently the loss of heat produced by such a mixture will be 
greatest when the difference between the temperatures of the 
heated air and the current is greatest, — that is to say, at the 
time of maximum of the air temperature. Denoting this moment 
by v, and putting the quantity of temperature lost by convection 
= —dk cos (£ - v), we have in this case the following differential 
equations : — 
df 
dz 
, , . , . , . d cos Z 
- hit - u) - hit -t) + K -t— 
dt 
dz 
- h(t - t r ) - dk cos (£ - v) 
r- 
J 
and consequently the integral 
t, = t 0 + a cos cos 8 cos (£ - v) - da cos (fc — v — v), 
which can easily he proved to show a decrease of the daily range, 
and a time of maximum earlier than under the conditions of 
continental circumstances, features well known to be peculiar to 
places near the coasts.* 
* I have to correct a mistake in this part of the paper already published, 
caused by an erroneous theoretical consideration, which, however, does not 
alter the result derived from the integral. 
At stations where the thermometer is too much protected by houses, etc. , 
and where the sun’s rays have not sufficient access to the neighbouring 
ground, we generally observe a diminution of the daily range connected with 
a retardation of the time of maximum temperature. It is not difficult to 
find the cause of these phenomena — first, in the diminution of solar radiation 
acting upon the neighbouring soil ; and, secondly, in the continuous currents 
of hotter air, which are caused by the upward direction of the currents on 
the roofs of the houses, and which mix with the colder air near the pro- 
tected position of the thermometer. Hence we have to alter the above equa- 
tions by taking a smaller value of K, and changing the sign of the expression 
dk cos((-v'). On a much larger scale the same phenomenon appears in 
mountainous countries, where currents of heated air flowing up a valley 
during day-time give the same peculiar character to the daily curve of its 
temperature. 
