Dmiiial Meteorological Conditions at Kimberley. 155 



air, albeit its pressure might be greater ; exactly in the same way as 

 the weight of a sealed bottle of air could not be increased whatever 

 internal pressure we might impart by heating it. A moment's con- 

 sideration will show that this increase of atmospheric pressure must 

 exist, even without any actual access of air, and must be greatest 

 somewhere near the place where the diurnal wave of temperature is 

 rising most rapidly. The sides of the bottle in the hypothetical case 

 are now represented by surrounding masses of air whose thermal 

 condition is changing less rapidly. No movement can take place on 

 account of heat received until the differences of pressure set up are 

 sufficient to overcome the inertia and viscosity of the surrounding air. 



There is an independent maximum of moisture in the first 

 harmonic component an hour or so after noon, arising probably out 

 of the enhanced evaporation from water, vegetation, and the ground, 

 due to insolation, sunlight, and wind, at the time. There are 

 dependent maxima, in the second harmonic component, coinciding 

 with the germane maxima of barometric pressure, and answering, 

 no doubt, to the same causes. The semi-diurnal wave of pressure, 

 then, appears to be largely mechanical, with an actual alteration in 

 the quantity of air overhead, and of the moisture it contains in the 

 same degree. There is a tri-daily wave agreeing with the cloud- 

 period, and also forming part of the tri-daily wave of pressure. It is 

 large in proportion to the air wave, relatively to the corresponding 

 semi-diurnal waves. 



If we compare these phenomena with those existing elsewhere, 

 we find that the independent wave of moisture, having a diurnal 

 period, and a maximum just after XIII., found only by analysis for 

 Kimberley, becomes materialised over the open ocean, so that the 

 observed diurnal curve of the "elastic force of vapour " (delightful 

 description !) is practically the same as that of the temperature, 

 though rather earlier in phase, whereas nearer land the double 

 diurnal oscillation makes its appearance- — ^the summit of the curve 

 being, as it were, pressed downwards. 



According to Table 30 the harmonic phase-times over both the 

 open sea, and near land, are also very nearly those of the tempera- 

 ture for the first two terms. The amplitude of the first term for the 

 open sea is nearly double that for near land ; but the amplitudes of 

 the second terms are practically unaltered. 



The third terms are very unlike both in amplitude and epoch : 

 near land the epoch is nearly that of the temperature on land, while 

 the amplitude is nearly equal to that of the second term ; out at sea 

 the epoch falls nearly three hours earlier, and the amplitude almost 

 vanishes. 



