208 
SECOND REPORT — 1832 . 
observations. He has determined that the upper strata of the 
atmosphere contain more carbonic acid than the lower ones ; 
that the quantity undergoes a sensible diurnal variation, being 
greater during the day than during the night; and that the 
quantity generally is greater in dry weather than in damp, when 
it is absorbed by the moisture of the soil. The proximity of 
the ground probably accounts in that manner for the fact of its 
being less in quantity in the lowest strata of the atmosphere, 
which otherwise would be hostile to Mr. Dalton’s views. 
Temperature. 
The thermometer is certainly the most perfect of our meteo¬ 
rological instruments. The range of natural temperatures being 
confined on the surface of our globe within comparatively nar¬ 
row limits, namely, 96® centigrade or 172° Fahrenheit in the 
shade * * * § ; the indications of the mercurial thermometer may be 
considered as absolutely accurate. Notwithstanding, too, the 
difficulties of procuring tubes of perfect calibre or making due 
allowance for its variation, the deviations of thermometers made 
by different makers may, when particular care is taken in their 
construction, be confined within very narrow limits. This I have 
recently had the means of particularly observing in some compa¬ 
risons of standard thermometers belonging to the Royal Society 
of Edinburgh, by Professor Christison and myself f. The rela¬ 
tion of the mercurial to the air thermometer has been investi¬ 
gated since Gay-Lussac’s and Dalton’s experiments, by MM. 
Dulong and Petit, who extended the examination to a great 
range of temperature. More recently M. Auguste has taken 
up the subject, and given a formula of comparison J. M. Par¬ 
rot has re-investigated the subject of the fixed points of ther¬ 
mometers §. He finds that the purity of the water employed 
has a sensible influence on its point of congelation ; and has 
observed one tenth of a degree of Reaumur between that of the 
water of the river Neva and distilled water. He has likewise 
determined that the maximum heat of water in a state of ebul¬ 
lition occurs at and below seventeen lines under the surface. 
Nothing has lately been done in the way of materially impro¬ 
ving self-registering thermometers : Rutherford’s are still the 
best. Magnus has proposed one acting by the expulsion of 
* Arago, Annuaire du Bureau des Longitudes, 1825, p. 186. 
f Captain Sabine ( Account of Experiments with the Pendulum, fyc. 4to,) 
found the difference of above a degree in two standard thermometers by the 
same maker. Such an error however cannot be considered unavoidable. 
X Poggendorff’s Annalen, 1828. 
§ This he has published in a Latin pamphlet, 4to, Petropoli, 1828. 
