SUPPLEMENTARY REPORT ON METEOROLOGY. 147 
mended, the next business connected with a public observa- 
tory is to furnish the secular constants of meteorology, such as 
the mean annual temperature, the mean annual pressure at the 
level of the sea, the limits of variation of these, the mean ela- 
sticity of vapour, the mean and total quantity of sunshine, the 
mean direction and integral quantity of wind, and similar 
facts, for a given place at a given time. These observa- 
tions are of a less elaborate kind than the preceding ; but the 
fate of all volunteer experiments shows, that to determine such 
quantities with minute precision, is incompatible with any but 
an official system of registration, which shall be conducted for 
very many years on exactly the same system, with instruments 
of the same kind, with unremitting attention not only to the 
fidelity of the observations, but to the perfect repair and com- 
parability of the instruments. Even should such a system not 
be started on the very best possible plan, it is better that it 
should continue uniform, than undergo perpetual change. If the 
hours or locality be not the best possible, the diurnal laws being 
known, the result of any hour may be converted into a mean 
result; and as to locality, it is much more important to preserve 
the consistency of results, than to avoid trifling disadvantages. 
327. We have seen that it requires very many years’ observa- 
tions to ascertain the mean climateric state of any point of the 
earth’s surface. In this country, for no one such point probably 
is it accurately known. Let the instruments be the simplest, but 
the best; the observations, however few, perfectly regular; and 
we have a certainty that we are laying up valuable facts at least 
for another generation, if not for the present one. 
328. Instead of labouring to collect a multiplicity of imper- 
fect registers, let us begin by laying the foundation of a rational 
and accurate science. If we can but determine accurately the 
secular elements of pressure and temperature at a hundred well- 
chosen points of the earth’s surface, we shall do more than has 
ever yet been done to draw a proper system of isothermal and 
isobarometric lines, which shall hand down to posterity the 
actual mean condition of our globe. 
329. Observations must not only be made, but they must be 
wholly reduced and printed, like the astronomical observations 
conducted by Mr. Airy*. They must be ready to be applied to 
the purposes of science, and synoptic tables of results widely 
distributed. In the deduction of Laws from Results, much use- 
less labour may be spared (and the remark is especially appli- 
cable to private observers), by employing graphical projections 
* For this purpose excellent printed forms have been supplied to the Ant- 
arctic expedition. 
L2 
