23 $ 
SECOND REPORT—1832. 
column is subject, when it is allowed for, the height at the level 
of the sea may be considered known at any instant,—a most 
material advantage to the observer between the tropics, and 
which has conferred much of their accuracy upon Humboldt’s 
beautiful barometric levellings and sections*. 
The great obstacle to the accuracy of barometrical measure¬ 
ment, and the most influential change produced by the hour of 
the day, is to be found in the variable temperature of the strata 
of air intervening between the stations, as it is clear that the 
mean of the upper and lower temperatures may often deviate 
greatly from the true mean of the intercepted column. The 
currents produced from the plains to the mountains during the 
diurnal revolutions of temperature are extremely considerable ; 
and hence the errors arising from the hour of the day greatly 
exceed those which the horary oscillation would produce. This 
has been pointed out and experimentally investigated by M. 
Horner, a Swiss Meteorologist of great activity, who made the 
mountain of the Rigi the scene of his operations f. Still more 
lately M. Gautier, Professor of Astronomy at Geneva, has pub¬ 
lished some interesting observations on the same spot, by which 
he found the error from the hour of the day alone, to amount 
to 14 toises upon a height of 700, the corresponding observa¬ 
tions being made at Zurich J. 
In my papers on the application of the sympiesometer to the 
measurement of heights, already alluded to, I have given the 
results of many comparisons of this method with the geometrical 
one in several parts of Scotland, and a number of heights from 
original trigonometrical operations §. 
I cannot dwell either upon the construction of portable baro¬ 
meters, or upon the precautions required in observation; but I 
strenuously recommend the subject to the scientific meteorolo¬ 
gist, as one which will repay his labour, and which is yet open 
to most important ameliorations. When we consider the accu¬ 
racy and extensive knowledge we have arrived at in the posi¬ 
tion of points of interest on the surface of the globe, with regard 
to the coordinates of latitude and longitude, and how little has 
been done for the third coordinate of elevation, we shall have 
a field before us open to cultivation at every corner. The re¬ 
sults to physical geography of what has already been done by 
the use of the barometer, excite our warmest hopes of its exten¬ 
sion. To mention only one instance ;—the singular discovery 
/ 
* Voyage aux Regions Equinoctiales; Atlas, and Observations Astrono- 
miques. Baron Humboldt has recently circulated a beautiful “ Carte hypso- 
metrique” of the Cordillera. 
f Bibliotheque Universelle, 1831, N.S. iv. 337. X Ibid- v. 337. 
§ Edinburgh Journal of Science , N.S. iv. 91. 329. 
