47? ATMOS 
the greateft antagonift of heat, and mod effectually coun¬ 
teracts the operation of the tolar rays in producing it. 
This power it f'eems to exert in all diltances, at the furface 
as well as in the higher regions. From fome experiments 
made by M. Pifhet, it appears, that even in places expo- 
fed to the rays of the fun, the heat, at five feet diftance 
from the ground, is greater only by one or two degrees 
than at fifty feet above the furface, though the ground was 
at that time 13 0 or 20 0 warmer than the air immediately 
in contact with it. Inconiiderable as this difference is, 
however, it does not hold as we go higher up ; for, if it 
did, the cold on the top of the mountain of Saleve, which 
is 3000 feet above the level of the lake of Geneva, would 
be 6o° greater than at the foot of it; whereas in reality it 
is only io°. In the night-time the cafe is reverfed; for 
the ftratum of air, at five feet from the ground, was found 
by M. Pitlet to be colder than that at fifty. Betides this, 
different tfrata of the atmofphere are found to polfefs very 
different and variable degrees of cold, without any regard 
to their fituation high up or low down. In the year 1780, 
Dr. Wilfon of Glafgow found a very remarkable cold ex¬ 
iting clofe to the furface of the ground ; fo that the ther¬ 
mometer, when laid on the furface of the (now, funk ma¬ 
ny degrees lower than one fufpended twenty-four feet 
above it. It has been likevvife obferved, that in clear 
weather, though the furface of the earth be then moll; li¬ 
able to be heated by the fun, yet after that is let, and du¬ 
ring the night, the air is coldelt near the ground, and par¬ 
ticularly in the valleys. Experiments on this f’ubjedf were 
made for a whole year by Mr. James Sex, who has given 
an account of them in the Phil. Tranf. vol. lxxviii. He 
fufpended thermometers (conftrudted in Inch a manner as 
to (hew the true maximum and minimum of heat that 
might take place in the oblerver’s abfence) in a Ihady nor¬ 
therly afpeff, and at different heights, in the open air. 
One of thefe was placed at the height of nine feet, and 
the other at that of 220 feet, from the ground, and the 
obfervations were continued, with only a few days onfil- 
sion, from July 1784.to July 1785. Thegreateft variations 
of heat were in October and June; in the former the ther¬ 
mometers generally differed molt in the night, and in the 
latter moftly in the day. From the 23th to the 2.8th of 
OCfobcr, the heat below, in the night-time, exceeded in 
a fntall degree the heat above ; at which time there was 
frequent rain mingled with hail. From the 1 ith to the 
14th, and alfo on the 31ft, there was no variation at all; 
during which time likevvife the weather was rainy; all the 
reft of the month proving clear, the air below was found 
colder than that above, fometimes by nine or ten degrees. 
In the month of June, the greateft variations took place 
from the 1 ith to the 1 5th, and from the 23th to the 30th ; 
and at both thefe times there appeared to be two currents 
of wind, the upper from the fouth-vveft and the lower 
from the north-eaft. Sometimes thefe were rendered vi- 
lible by clouds, in different ftrata, moving in different di¬ 
rections; and fometimes by clouds moving in a contrary 
direction to a very lenfible current of air below. On 
cloudy nights the lowed thermometer fometimes (hewed 
the heat to be a degree or two greater than the upper one; 
but in the day-time the heat below conftantly exceeded 
that above more than in the month of OCtober. 
To determine whether the nocturnal refrigeration was 
augmented by a nearer approach to the earth, two ther¬ 
mometers were placed in the tnidft of an open meadow, 
on the bank of the river near Canterbury. One was pla¬ 
ced 011 the ground, and the other only fix feet above it. 
The thermometer, at fix feet diftance from the ground, 
agreed nearly with the former at nine feet; but the noc¬ 
turnal variations were found to correfpond entirely with 
the clearnefs or the cloudinefs of the iky : and, though they 
did not always happen in proportion to their refpettive al¬ 
titudes, yet, when tiie thermometers differed in any refpecr, 
that on the ground always indicated the greateft degree of 
cold. The difference betwixt thefe two thermometers, at 
the itnall diftance of fix feet from each other, being found 
P H E R E. 
no lefs than three degrees and an half, the number of ther¬ 
mometers in the meadow was augmented to four. One 
was funk in the ground, another placed juft upon it, and 
the third fufpended at three feet above it. Three others 
were placed on a riling ground where the land was level 
with the cathedral tower, and about a mile diftant from 
it. One of thefe was likevvife funk in the ground, ano¬ 
ther placed juft upon it, and a third fufpended fix feet 
above it. Witli thefe feven thermometers, and the two 
firlt mentioned, which were placed in the city, he conti¬ 
nued his obfervations for twenty days ; but, as the wea¬ 
ther happened to be cloudy during the whole of that fpace, 
excepting for feven or eight days, no confiderable varia¬ 
tion happened excepting on thole days. The refult of the 
experiments was, that the cold was generally greater in the 
valley than on the hill; but the variations between the 
thermometers on the ground and tliofe fix feet above them, 
were often as great on the hill as in the valley. 
Thus it was perceived, that a difference of temperature 
took place at the diftance of only three feet from the 
ground; but live length of the thermometers hitherto 
made ufe of rendered it impoffible to make any experiment 
at a fmaller diftance. Two new ones, therefore, were 
formed by bending down the large tube, the body or bulb 
of the thermometer, to an horizontal pofition, while the 
ftem remained in a vertical one; by which method the 
temperature might be obferved to the diftance of a fingle 
inch. Sometimes, in clear weather, thefe two horizontal 
thermometers were placed in the open air, one within an 
inch of the ground, and the other nine incites above tt. 
When the variation among the other thermometers was 
confiderable, a difference was likevvife perceived between 
thefe ; the lower one fometimes indicating more than two 
degrees lefs heat than the upper one, though placed fo 
near each other. From thefe experiments Mr. Sex con¬ 
cludes, that a greater diminution of heat frequently takes 
place near the earth in the night-time than at any altitude 
in the atmofphere, within the limits of his inquiry, that 
is, 220 feet from the ground ; and at fuch times the great¬ 
eft degrees of cold are always met with near the furface 
of the earth. This is a conftant and regular operation of 
nature, under certain circumltances and difpofttions of the 
atmofphere, and takes place at all feafons of the year; 
and this difference never happens in any confiderable de¬ 
gree but when the air is ftill, and the Iky perfectly un¬ 
clouded. The moifteft vapour, as dews and fogs, did not 
at all impede, but rather promote, the refrigeration. In 
very fevere frol'ts, when the air frequently depolits a quan¬ 
tity of frozen vapour, it is commonly found greateft; but 
the excefs of heat which in the day-time was found at the 
loweft ftation in fummer, diminifhed in winter almoft to 
nothing. 
It has been obferved, that a thermometer, included in 
a receiver, always links when the air begins to be rarefied. 
This has been thought to arife, not from any degree of 
cold thus produced, but from the hidden expanfion of the 
bulb of the thermometer in confequence of the removal 
of the atmofpherical preflure. But from fome late expe¬ 
riments related, Phil. Tranf. vol. lxxviii. by Mr. Darwin, 
it appears, that the atmofphere always becomes warm by 
comprellion, and cold by dilatation from a compreffed (late. 
Thefe experiments were: i.Theblaft from the air-gun 
was repeatedly thrown upon the bulb of a thermometer, 
and it uniformly funk it about two degrees. In making 
this experiment, the thermometer was firmly fixed againft 
a wall, and the air-gun, after being charged, was left for 
an hour in its vicinity, that it might previouily lofe the 
heat it had acquired in the atft of charging; the air was 
then difeharged in a continued ftream on the bulb of the 
thermometer, with, the effect already mentioned. 2. A 
thermometer was fixed in a wooden tube, and fo applied 
to the receiver of an air-gun, that, on difehargihg the air 
by means of a ferew prefling on the valve of the receiver, 
a continued ftream of air, at the very time of its expan¬ 
lion, palTed over the bulb of the thermometer. This ex¬ 
periment 
