Pressure aud Temperature Results for the Great Plateau. 261 
and in a climate not differing greatly, excepting at midsummer, from 
that of the dry regions of South Africa. Between Kimberley and 
Durban, making every allowance for the difference of latitude, the 
fall is not quite a degree in 500 feet. The monthly comparison 
numbers for Mount Abu and Karachi are given below in Table 37.” 
Considered side by side with the comparison of Table 25 they are 
most instructive, Abu being a fairly typical isolated mountain, and 
its observatory standing at about the same elevation above Karachi 
as Kimberley, on its typical table-land, is above Durban. 
We come now to the remarkable numbers tabulated in columns 
8-13 of Tables 1-25. Their meaning has been explained near the 
beginning of this paper. They show that the curve of daily maximum, 
and the curve of daily minimum temperature for the year have almost 
the same characters at both Durban and Kimberley, the respective 
positive and negative excursions occupying the same intervals of 
time at both places. Fifty-four per cent. of the registered maxima 
are above the mean maximum, and 52 per cent. of the registered 
minima are above the mean minimum. We infer from the given 
ratios that when the maximum falls below its mean amount it 
does so deeply and sharply, but its rises are more leisurely and more 
prolonged: that is to say, the general aspect of the graph is that 
of a prolate cycloid on the positive side of its base. The minima 
follow much the same course, although the cycloid shows a dis- 
position in the winter months to shift to the negative side of the 
base. 
Compare these temperature numbers with those for the pressures 
given in the concluding columns of Tables 1-24, and in columns 
17, 18, 19, of the monthly summaries. Here the position is that the 
barometer at both Durban and Kimberley is as many times above 
the mean as it is below ; 7.e., the average shape of the negative parts 
of the curve of departure from the mean daily pressure is an exact 
reflection of the average shape of the positive parts. In other words 
the curve displays the same peculiarities of detail either way up. 
Such being the case at both Durban and Kimberley, we must con- 
clude that the mean pulsation is essentially undulatory. The idea is 
to some extent justified by an analysis of the pressures year by year 
deduced from hourly observations at Kenilworth. It is found that 
the annual mean pressure is practically constant, although the mean 
barometer during any month may differ greatly from that of the 
same month of another year. 
As stated at the outset, the Durban pressures are for ten years, 
while only eight years are available at Kimberley ; moreover, the 
* From Blanford’s Climates and Weather of India. 
