of Edinburgh , Session 1885-86. 
951 
the kindness of Mr Omond and his assistant Mr Rankin, I was able 
to make the observations continuous, and also to add a considerable 
number of observations at low temperature, taken during the past 
winter. 
The readings of wet and dry bulb thermometers given were 
taken with the instruments in daily use at the Observatory. These 
thermometers are by Messrs Adie and Wedderburn, and are 
sufficiently delicate for all the purposes of the investigation. The 
diameter of the spherical bulbs was 0 - 40 inches. The wet bulb was 
covered with thin muslin, as served out by the Scottish Meteorological 
Society, and every care was taken to ensure its being properly 
moistened. The thermometer screen was an ordinary double-louvred 
Stevenson, open below, and of the usual size. The observed read- 
ings have been corrected in accordance with the Kew certificates 
attached to the instruments, and checked by comparison with the 
readings of one of Aitken’s screens. 
The direct hygrometer used was Professor Chrystal’s, described in 
the note referred to ( Proc . R. S. E., vol. xiii. p. 199). After a little 
practice this instrument can be made to give very satisfactory 
results, deposits having been obtained on the silvered surface at 
temperatures as low as 1 0, 5 F., by using mixtures of salt and snow. 
The dew-point was in each case taken to be the temperature at 
which a uniform coating of moisture was deposited on the silvered 
surface, when its temperature was being reduced. The box con- 
taining the thermometer was first cooled quickly to within a few 
degrees of the point where deposition was expected to take place, 
then its temperature was allowed to fall slowly till moisture began 
to appear and disappear in irregular patches, when the cooling was 
still further checked. Almost immediately after, and at a quite 
definite point, a uniform deposition made its appearance, varying 
slightly at its outer edges ; the reading was then taken. Readings 
taken from the clearing off of the deposit were of course unsatis- 
factory, as the moisture deposited may take some time to evaporate, 
during which the thermometer may rise considerably. 
The deposit appearing and disappearing in patches may have been 
due to the unequal temperature of the mixture in the box of the 
hygrometer, but as this took place even when the temperature was 
allowed to fall with great slowness, I am disposed to think it was 
