88 REPORT— 1840. 
observed to be condensed to a known fraction of its bulk, when 
the length of the column of mercury is also a known fraction of 
the total barometric column. I brought one of these instruments 
to this country two years ago, but I have not yet made trial of it. 
142. Mr. Cooper has proposed a barometer acting by the 
elasticity of the air in a floating vessel regulated by weights, 
which constantly immerse it to the same depth*. The author 
considers it capable of showing a difference of elevation of three 
or four feet. It is intended that the apparatus should in every 
case be employed at a constant temperature of 75°, to which it 
is artificially brought: I conceive that this process is attended 
with inevitable disadvantages. 
143. SirJohn Robison has proposed to use tubes or long phials 
containing air, immersed in water at the top of a hill, instead 
of an air barometer or sympiezometer. The portion which be- 
comes filled with water, when re-examined, would indicate the 
previous rarefaction of the airf. In this and every similar case 
the temperature of the included air is a matter of great uncer- 
tainty, and prevents the possibility (as contemplated) of trust- 
ing such instruments to inexperienced assistants. 
B. Mean Height of the Barometer. 
144. Several considerations would lead us to the inference, 
that the mean pressure of the atmosphere at the level of the sea 
should vary with the latitude; but it is to experiment alone 
that we can look for any indication of a law. Humboldt ap- 
pears first to have remarked, that the height of the barometer 
is lower at the equator than in temperate latitudest; and, 
excepting this fact, little more has been known until the late 
excellent researches of Schouw§, though the partial observa- 
tions of occasional navigators indicate this fact, as well as avery 
considerable depression of the barometer towards the pole|]. 
145. M. Schouw’s statement is the following :— 
Mean pressure at Level 
Sea, in French Lines. 
Lat. 0° to 15° high temp. with a rainy season 337—7 
>» 15 — 30 very dry; rainsrarely . . . 338—9 
9» o0.— 45 temperate. +2. are ee ee ee 
» 45 — 65 coldandrainy. . .. . . 33753 
* Philosophical Transactions, 1839, p. 425. 
+ British Association, Eighth Report, Sections, p. 37. Brunner has de- 
scribed an air-barometer, Poggendorff, xxxiv. 30. 
t Tableau Physique, p. 89, quoted by Kamtz. 
§ Annales de Chimie, tome liii. (1833). See also Poggendorff, xxvi. 395. 
|| See the authorities cited in Humboldt’s Note to Arago, Comptes Rendus, 
ii.570. Some valuable comparisons of barometers at different northern observa- 
tories are to be found in a late number of the Comptes Rendus (1840. 2me 
Semestre). 
