198 
SECOND REPORT-1832. 
barometer, which greatly exceeds in accuracy those which had 
previously been proposed. Yet may the speculations of these 
philosophers, and the discussions to which they give rise, be 
more important to the science than the labours of a professed 
meteorologist, who has made, with minute scrupulosity, all the 
ordinary entries in his Journal, daily for a life-time. 
We must not be supposed to give to theory a pre-eminence 
over observation. Had the meteorologist just supposed, instead 
of observing all the ordinary instruments, perhaps not upon the 
best construction, and at hours dictated by convenience or by 
accident, directed his attention even to the merely mechanical 
examination of any one phenomenon, not to the mixed result of 
a chaos of heterogeneous principles ;—had he, with an eminent 
chemist of our own time, determined the specific gravity of air 
every day, and watched the unsuspected variations which that 
amount undergoes;—had he directed his observations to the de¬ 
tection of the lunar atmospheric tides had he examined by 
reiterated experiments, under every varied condition, the solu¬ 
tion of the beautiful problem of the barometrical measurement 
of heights ;—had he taken advantage of lofty and mountainous 
situations to study the formation and dissolution of clouds and the 
influence of humidity and temperature in their phases, or of a low 
and flat country for determining the amount of solar and terre¬ 
strial radiation in sheltered spots and under different aspects of 
the heavens ;—had he in any of these, or in one of a hundred 
other equally fertile paths of inquiry, added to our knowledge 
of the connexion of cause and effect in this intricate subject, 
he would have conferred, at perhaps even less expense of time 
and labour, an infinitely greater boon upon the science which 
he wished to advance. 
The mere local meteorology of a country may frequently be 
a very interesting object in relation to its physical geography 
and agriculture, and as such may be prosecuted by the syste¬ 
matic establishment of Registers on a small scale ; but for the 
great facts of the science the adequate support of a few great 
Registers in any country would suffice, provided such be sus¬ 
tained on the most liberal scale and on the most accurate prin¬ 
ciples, by great Societies, or, still better, by Governments. In¬ 
struments must be provided on the best possible construction, 
placed in the best situations, and observed at the best times, 
and with undeviating regularity, by fit observers. The critical 
hours vary in different climates, and should be determined with 
the greatest care by preliminary experiments; and no greater 
error has been committed in the establishment of such Registers 
than the indiscriminate observation of all instruments which 
