AT NIGHT, FROM THE EARTH, ETC. 
1G3 
At the time of two of these five sets of observations the sky was nearly covered with 
cloud ; at one other time one-third part of the sky was clouded over and much haze 
prevailed; and at the remaining times the sky was not bright, and there was a per- 
ceptible haze. The mean temperature of the air was 30°‘9 ; the mean reading of the 
thermometer on long grass clear of snow was 1 0, 8 less than that on short grass, also 
clear of snow ; and the readings were nearly alike when the two were covered with 
snow. The effects of snow on long grass was to cause the thermometers on the grass 
to read higher by 8 0- 6, but this difference does not fully represent the non-conduct- 
ing power of snow, as the temperature of the air varied very little whilst these experi- 
ments were being made ; had the temperature of the air fallen or risen much whilst the 
snow had been on the ground, the temperature of the earth under the snow would 
have continued nearly at the same reading, and in that case the effects of the snow 
would have been much more decided. The next favourable opportunity I had of 
repeating my experiments on snow was 1845, February 12*. Snow had previously 
fallen to the depth of three inches, and during the night, which was cloudless, the 
reading of a thermometer which was placed on long grass was —6°, whilst that 
covered by snow was 28°'0 ; the effect of the snow was therefore to keep the grass 
warmer by 34°, and therefore vegetation was kept warmer by this amount than it 
would have been had there been no snow. The reading of a thermometer placed on 
the snow was — 12°, and the difference between the readings of two thermometers, the 
one placed on snow and the other under the snow, was 40°. After this time the read- 
ing on grass clear of snow rose to 24°, and that on snow increased to 15°, without 
causing any variation in the reading of that under the snow, which still read 28?. 
Snow would therefore appear to be a very perfect non-conductor of heat. The lowest 
reading of a thermometer on flax was — 12°^ ; that on short grass was 5°, being 11° 
higher than that on long grass ; this heat (see remarks following Table XLV.) repre- 
sents the difference between the quantities of heat conducted from beneath the surface 
to long and to short grass, and it would represent also the greater quantity of heat 
lost from the earth as covered by short grass than that by long grass ; and if to this 
1 1° we add the heat conducted beneath the surface to long grass, which did not differ 
much from 6°, it would appear that heat to the amount of 17 ° was conducted to the 
short grass from the earth beneath it. Snow being so perfect a non-conductor of 
heat, evidently prevents to a high degree the loss of heat by radiation from bodies 
covered by it ; and it also prevents the loss of heat from such bodies by conduction, at 
times when the temperature of the air is lower than they are. Raw wool, flax, straw, 
and other bodies which are bad conductors of heat, act in a similar way and prevent 
the injurious effects of cold to bodies covered by them, to which injurious effects 
vegetation is liable in this climate in every month of the year, as it is liable to a 
temperature at night below the freezing-point of water in every month. 
* For these observations, see the Greenwich Magnetical and Meteorological volume for the year 184.5, 
pages 256 and 257. 
