416 Transactions of tJte South African Philosoi^liical Society. 



that the cloud-covering on tlie mountain was continuous throughout 

 this period. 



Even on the very improbable supposition that the rain collected 

 by the open gauge fell absolutely vertically, so that the horizontal 

 area of the rain-gauge alone formed the catchment area, while the 

 remainder of the water caught was deposited only on the vertical 

 reeds, &c., by a horizontal current, it will be found that the deposit 

 from the " Table-cloth " is simply almost doubled, amounting to 

 12*97 inches during the period, and representing a daily average of 

 0-23 inch. 



In fact, in whatever way we may care to regard the deposition of 

 moisture to have taken place, it will be found that Dr. Marloth's 

 figures are enormously in excess of the true amount. 



All experiments in this connection ought to admit of but one 

 interpretation if any reliance has to be placed on the results obtained, 

 and a comparison made between the amounts collected by the two 

 gauges. The following considerations Avill show how faulty the 

 experiments discussed are in this respect : — 



The reeds have been assumed to be right circular cylinders of 

 0*09 inch diameter, but it may be regarded as an open question if 

 the moisture soaking in through the cut ends would not cause the 

 reeds to swell in spite of their siliceous skeleton, and thus still 

 further increase the area exposed to the horizontally-moving current. 



Attention ought also to be drawn to the possible effect of the 

 reeds in decreasing the velocity of the wind by breaking it up into 

 small whirls, and so producing practically a calm over the gauge, 

 thus admitting of a vertical deposit of moisture. The action is 

 similar to that of a Nipher's shield for rain-gauges at any consider- 

 able elevation above the ground, which breaks up the eddies tending 

 to carry the rainfall over the gauge, and so renders the amount 

 caught at any elevation practically the same as that falling on the 

 ground. 



The amount deposited from the mist ought to vary with the 

 density of the mist and the velocity of the wind. 



One advantage that the vertical catchment area possesses over 

 that of the horizontal gauge is that the water rapidly runs down- 

 wards into the receiver, and so is protected from possible evapora- 

 tion during any break in the prevalence of the mist. 



In estimating the effect of such deposits from mists on water 

 supply, it ought to be borne in mind that in the apparatus described 

 the reeds have been placed to the best advantage, being on an average 

 1 inch to 1^ inch apart, and so arranged that they do not overlap, as 

 would ])e the case with a close natural growth of these plants ; con- 



