SUPPLEMENTARY REPORT ON METEOROLOGY. 149 
B. Sedentary Observations. 
332. Private individuals, fond of science, would find the 
establishment of such systematic observatories as we have 
recommended, to add great importance as well as interest to 
their labours. 
333. Their business would then be like that of detailed sur- 
veying after the great net-work of triangulation has been com- 
pleted. The local meteorology of a country is often of great 
practical as well as scientific importance, and the immense dif- 
ference of climate frequently experienced within a few miles, 
whilst it shows the embarrassing effect of local causes, and the 
danger of drawing general conclusions from insulated or im- 
perfect registers, is itself a curious subject of research. 
334. Ordinary thermometrical and barometrical registers may 
be continued, as has been usual, with additional attention to 
the construction and repair of instruments, the choice of situa- 
tion, and the choice of hours. In respect of the latter, 9 a.m. 
and 9 p.M. are very convenient and fitting hours for thermo- 
metric observations, to which may be added 3 or 4 P.M. for the 
sake of the barometric oscillation. But it is one advantage of 
systematic observation, such as was recommended in the last 
section, that it will render available observations made else- 
where at any hour or hours by a general principle of reduction 
to the mean*. Self-registering instruments (thermometers, 
rain-gauges, anemometers, and the like) are well adapted for 
private observation from the little superintendence they require. 
335. I would, however, particularly urge the propriety of not 
confining individual efforts to the mere multiplication of simple 
registers. The same expenditure of time and money is appli- 
cable to much more interesting purposes. Special series of 
observations and experiments on one or more subjects, such as 
the following, would be of very great interest indeed :— 
1, The temperature of the soil at small depths. If the dif- 
ficulty of procuring long thermometers be an objection, 
water-bottles may be lowered to different depths in sepa- 
rate tubes of wood sunk in a well, which is then filled up 
with earth or sand, and the temperature may be noted by 
a common thermometer on pulling them up, the opening 
of the tubes being well stuffed with hay or woolt. (98.) 
* As was clearly pointed out by Sir D. Brewster, in discussing the Leith hourly 
observations. It must be remembered that such reductions are correct only 
within the limits of the region or climate for which they have been ascertained. 
+ A not unimportant modification of these experiments might be made by 
covering a certain space of the surface of the soil in which the thermometers 
were sunk, with a composition of ascertained radiating and absorptive power. 
