" 7.4 to 20.2
Savannah 13th to 30th " 5.2 to 17.3
Austin, Texas 10th to 30th " 4 to 24
Clarkesville, Tenn. 4th to 30th " 10.3 to 20.5

AUGUST.

Bloomfield, N. J. 9th to 14th " 5 to 15
Austin, Texas 6th to 12th " 0 to 19
Philadelphia 10th to 15th " 8 to 14
Jacksonville, Fla. 10th to 15th " 6 to 8

Observations by Lieut. Gillis, at Washington, give mean differences
between wet and dry thermometers, from March, 1841, to June, 1842, as
follows:

Observations at 3 P. M.:

_Jan._ _Feb._ _Mar._ _Apr._ _May._ _June._
3 deg..08 4 deg..40 6 deg..47 5 deg..37 7 deg..05 8 deg..03

_July._ _Aug._ _Sept._ _Oct._ _Nov._ _Dec._
8 deg..89 5 deg..29 5 deg..63 4 deg..61 4 deg..77 2 deg..03

A mean of observations for twenty-five years at the Radcliffe
Observatory, Oxford, England, gives a difference between the wet and dry
thermometer equal to about two-thirds the difference, as observed by
Lieutenant Gillis, at Washington.

On the 12th day of August, 1853, in Austin, Texas, the air was perfectly
saturated at a temperature of 76 deg., which was the dew-point, or point of
the thermometer at which dew began to form. The dew-point varies
according to the temperature and the humidity of the atmosphere; it is
usually a few degrees lower than the temperature of evaporation--never
higher.

From observations made at Girard College, by Prof. A. D. Bache, in the
years 1840 to 1845, we find, that for April, 1844, the dew-point ranged
from 4 deg. to 16 deg. lower than the temperature of the air; in May, from 4 deg. to
14 deg. lower; in June, from 6 deg. to 20 deg. lower; in July, from 4 deg. to 17 deg.; in
August, from 6 deg. to 15 deg. lower; and in September, from 6 deg. to 21 deg. lower.
The dew-point is, then, during the important months of vegetation,
within about 20 deg. of the temperature of the air. The temperature of the
dew-point, as observed by Prof. Bache, was highest in August, 1843,
being 66 deg., and lowest in January, 1844, being 18 deg.; in July, 1844, it was
64 deg., and in February, 1845, it was 25 deg.. Its hourly changes during each
day are quite marked, and follow, with some degree of regularity, the
changes in the temperature o