Pressure and Temperature Results for the Great Plateau. 275 
high during the greater part of May, while the minima are at times 
extraordinarily low. So far as Kimberley is concerned it is the only 
depression of the mean curve having both depth and length. 
But by far the most interesting of all the possibly annual pheno- 
mena is the narrow but deep depression recurring year after year near 
the middle of July. Speaking in averages, the, barometer begins to 
fall at Durban on the 11th, is falling for four days, and rising for 
five to a point (on the 18th) rather higher than it originally fell from. 
At Kimberley the fall begins on the 12th and continues until the 15th 
—1.e., it is still falling at Kimberley after the minimum has passed 
over Durban—rising from that time to the 20th. Apart from the 
difference of time, and the somewhat greater range at Durban, the 
depression is pretty much of the same aspect at both places with 
regard to the gradients; resembling in this the ordinary sporadic 
cyclonic disturbances that visit the country ubiquitously. Now it 
may be laid down as a rule, in a general way, that pressure is the 
inverse of temperature, the one being high while the other is low. 
But this rule does not rigidly apply to the July depression, nor for 
that matter to the prevailing conditions during the greater part of 
the month. On the contrary, the temperature begins the month by 
rising and falling with the pressure, next it does not respond one 
way or the other to the changes of pressure, then it begins to fall 
two days after the pressure has begun to fall, and to rise also two 
days after the pressure has risen; afterwards for a time it behaves 
normally. It is curious, too, that although the pressure wave is felt 
first at Durban by fully a day, the temperature wave is felt first at 
Kimberley, just as though the two waves were independent and 
travelling in opposite directions. 
It needs no very keen discernment to see that we have here a very 
different class of depression to that which accompanies the hot winds 
of the coast. This one is common to high and low land without 
bias, and appears to be as much communibus anus as the mean 
pressure or temperature of any month. To show this better Table 
42 has been constructed from all the materials available here 
(namely, eight years from Kimberley proper, three from Kenilworth, 
and fourteen from Durban), giving the daily readings of the barometer 
for nine days in July (10th—-18th) during each year. It will be seen 
that the tendency to a minimum at the mean date not only holds for 
the ten years from which the averages of Tables 7 and 19 have been 
computed, but also for the years before and since. One cannot help 
regretting that the pressure conditions of Port Nolloth may not also 
be examined. 
The atmosphere consists of a quantity of matter which cannot 
