Causes and Effects of Variation in the Bange of Temperature. 343 
monthly value of mean pressm^e arises physically or fortuitously is not 
yet clear. 
The general tendency of some of the results given in Table 2 is 
perhaps what would have been expected. Thus, e.g., we should expect 
that the greater cloudiness of the sky, the heavier rainfall, and the 
greater number of rainy days in any given month would go with a 
diminished percentage of sunshine. It is not so obvious, however, that 
these should also go with a uniform decrease of both maximum and mean 
temperature and an increase of both minimum temperature and range of 
temperature, especially when it is remembered that some of the hottest 
Kimberley summer days are to a greater or less extent characterised by 
clouded skies. Nor does it necessarily follow, as a mere piece of intellec- 
tual reasoning, that the range of dew-point temperature, or the range of 
relative humidity, should increase as the range of temperature decreases. 
In any case, whether the results are, or are not, much as might be 
expected, the actual measures are of interest. In particular, the mutual 
relationship of sunshine and cloud is worth attention. It will be seen that 
the sum of the percentages of sunshine and of cloud is not a constant 
quantity, but that it becomes greater as the range of temperature 
becomes less, rising from a total of 106 per cent, when the range of 
temperature is 32°-6 to a total of 114 per cent, when the range of tem- 
perature is 24°-3. 
The minimum temperature does not rise at the same rate as the 
maximum temperature falls in the interval between the average month's 
greatest range of temperature to its least. Thus it comes about that the 
average month of least range of temperature is also the average month of 
least mean temperature. This is important. Professor Haughton's 
lectures have been quoted as teaching that the mildness of the climate 
of the West of Ireland is due to the abundant rainfall. It has been 
suggested that this mildness of temperature is a direct consequence of the 
heat disengaged in the process of condensation from aqueous vapour to 
rain. The suggestion does not hold water. The Kimberley results show 
that although the nights are a little warmer in the wet months the days 
are colder by a greater number of degrees. That is to say, whatever the 
" blanketing " action of the clouds may be, or whatever the quantity of 
heat disengaged in the process of condensation, the radiation that is 
checked by, and the latent heat that is received from the clouds, are 
together less than the direct solar heat that is shut off by the clouds 
from the earth's surface. 
During the winter months the diurnal curve of dew-point temperature 
is quite unlike what it is in the summer. In the winter half-year, April 
to September, the vapour tension is higher at XIV. than it is at VIII., 
whereas for the rest of the year the case is the other way about. A sepa- 
