298 
Transactions of the Boyal Society of South Africa. 
So far as one year's observations can be trusted, then, the results of 
the wind observations at Lourengo Marques shov^ — as the longer period at 
East London showed — that the current statements, and the physical 
atlases, are wrong in the prevailing directions they assign to the winds of 
the south and south-east coast of South Africa. The "prevailing south- 
east rain-bearing wind" is, in fact, a myth; and as Captain Campbell 
Hepworth has pointed out (though little notice seems to have been taken 
of his testimony), south-east winds may blow fifty or a hundred miles to 
seaward, yet they are deflected as they approach the coast, so as to take 
the direction of the coast-Une.''' 
Of more immediate consequence to the main subject-matter of this 
paper, however, is the diurnal oscillation of the vane at Lourengo Marques. 
In the Eeport from which the greater part of the above information has 
been extracted the directions and velocities have been given for every 
alternate hour of each day, i.e., at I., III., V., VII., &c. I have tabulated 
the directions for each given hour and worked out their Cartesian com- 
ponents. The component velocities have not been dealt with here, the 
business being simply with the directions at particular hours. Eesults 
are given in Table 5. 
* M. W. Campbell Hepworth, " Weather Forecasts and Storm Warnings on the "Coast 
of South Africa." Eead before the Meteorological Society in February, 1883. Reprinted 
in " Notes on Maritime Meteorology," 1907. 
XV 
Fig. 2. 
