AT NIGHT, FROM THE EARTH, ETC. 
199 
readings are affected from this cause, and to an unknown amount ; the heat of the 
air is known, and therefore it can be accounted for ; the amount of heat evolved 
during the change of vapour into water appears to be about 4° from all the experi- 
ments I have made ; the heat radiated from dense clouds, near the earth, must be 
very nearly the same in amount as that radiated from the earth, as the clouds when 
at a low elevation must possess very nearly the same heat as that of the lower atmo- 
sphere ; but when such clouds are high their temperature must be less than when at. 
a low elevation, and they will radiate less heat to the earth than they will receive 
from it (see the following experiments upon the different results deduced from high 
and low clouds, of the same modification and covering the same extent of sky) ; and 
the amount of heat from the other sources is unknown. The whole effect of all 
these checks upon the production of a great cold at night, by the radiation of heat 
from bodies on the surface of the earth, cannot be estimated, yet, notwithstanding their 
operation, the reading of a thermometer placed on the ground has been frequently 
very low. The reading of a thermometer thus placed represents the amount of heat 
received from all the above sources, diminished by the amount radiated from itself. 
The place in which the observations were taken is not favourably situated for the 
production of a great cold, from radiation of heat at night, it being surrounded, at no 
great distance, by large trees, and consequently the humidity of the atmosphere is 
great. 
From the circumstance of low readings always having taken place when the sky 
has been cloudless and bright, we may readily infer that the temperature of space 
must be very low indeed. 
The reading of a thermometer placed on grass is much affected by the heat con- 
ducted to it from the earth beneath ; yet, notwithstanding, its readings were always 
less than those of the thermometer in the reflector, in the ratio of 1000 : 858 (see 
Table XLV.), therefore it is necessary to multiply the results derived from the latter 
by 1T7 to reduce them to results that would have been derived from the former. 
By examining the numbers in the columns under the lowest thermometrical readings 
in Table CVII., it appears that long grass, and therefore vegetation is liable to be 
affected at night from the influence of radiation by a temperature below the freezing- 
point of water every month in the year, for even in July 1844, the only exception in 
that year, the thermometer read 35°'5, whilst that in the reflector read 39°'3 ; in the 
year 1843, in July, the reading in the reflector was as low as 35 0, 2, and it seems very 
probable that long grass temperature at this time was at or below 32°; and as all 
the readings would have been lower if the experiments had been made in the open 
country, it seems certain that vegetation is always liable to the temperature of 32° in 
this country. 
The next table is formed by taking the means of all the numbers contained in each 
division of the Tables LXXVIII. to XCVIII. 
