BY THE USE OF KITES OFF THE WEST COAST OF SCOTLAND, 1902. 
127 
the instrument, including the hairs of the liygrometer, 1:)nt the thermometer was 
protected from rain and, to a certain extent, from drifting particles of water by its 
case. If the thermometer became wet, it would for a time give a wet l)ull) instead of 
a dry bulb temperature; but, under the conditions of deposition of the moisture, 
the difference would not be large. It would, in effect, show the temperature in 
descent less than that in ascent. The readings obtained do not give any clear 
indication of any such effect. 
A further uncertainty as to temperature readings arises from the variation in 
actual temperature of the atmosphere from point to point in the same horizontal 
layer. The primary object of the temperature observations may be said to be the 
determination of the vertical distribution of temperature ; but an inquiry of not less 
interest, though of greater difficulty, would be the examination of the details of 
horizontal distribution of temperature. Two striking instances of horizontal variation 
of temperature may be mentioned. First, the temperature recorded on Ben Nevis 
is generally several degrees below that at the same level over the sea at Crinan, 
60 miles away, and, secondly, the cloud level indicated l)y the disappearance of the 
kite was always much higher than the cloud layer on the hills of the near land, 
sometimes, as on July 16, l)y as much as a thousand metres. This matter will be 
referred to later on. But besides these instances, it may be mentioned that a 
characteristic tyj)e of weather in the neighbourhood was indicated by detached 
cumulus clouds forming first over the land, but sometimes drifting over the sea. 
These well-marked detached cumulus clouds were a very common feature, and they 
certainly indicate a want of horizontal unifoi’mity of condition as regards moisture, 
probably as regards temperature also. Probably contiguous portions of a horizontal 
layer are moving, one upward and the other downward, with considerable difference 
of temperature. The descending column may perhaps be traced in the ruffling of 
the surface, which is characteristic of “ catspaw ” weather. On. several occasions, 
notably on August 8 and August 20, convection currents, whicl) would correspond 
with this state of the atmosphere, produced a noticeable effect on tbe kites, the angles 
of which varied from time to time in a very irregular manner. The apparatus 
employed is not sufficiently refined for the records to identify these differences 
without very special observations ; they have accordingly been disregarded in the 
discussion of the observatioiis. 
Tahidation o f t/ic Traces. 
The meteorograph was raised 40 times between July 4 and August 25 ; single 
ascents were obtained on 23 days, two ascents were made 021 eight days, on one of 
which, July 26, a third ascent was made. The clock stopped at the commencement 
of the ascent on August 21 and before the conclusion of the ascent on August 18. 
In the other ascents moi’e or less satisfactory records of temperature were obtained, 
less satisfactory records of humidity. 
