10 ^president's address. 



silicification in the Namib to chemical weathering under desert 

 conditions.-" 



Closely connected in occurrence and origin with the silcretes. 

 are the ironstones or ferricretes, rocks in which a large part of 

 the cementing material is hydrated ferric oxide. Though such 

 rock is forming to-day wherever water containing iron in solution 

 appears at the surface and deposits iron oxide, either by the 

 direct influence of atmospheric oxygen or through the vital 

 activities of plants,^" there are many bodies of the rock which 

 are undergoing erosion and are of considerable antiquity. The 

 condition which appears to be favourable for its accumulation 

 is an intermittent supply of ferruginous water leaking out at the 

 surface; in countries where rain is well distributed and rather 

 high in amount, the delicate films of oxide of iron are either 

 masked by other material or washed away. 



A characteristic superficial deposit in dry countries is the 

 limestone, variously called desert limestone, surface limestone, 

 calcareous tufa or calcrete. Most rocks contain sufficient lime 

 in the form of silicates or carbonate to yield the bicarbonate to 

 ground water, and where the solution does not flow away, 

 whether on account of the flatness of the ground or the small 

 quantity of water, carbonate of lime is deposited at or near the 

 surface of the ground. The deposit is, of course, general and 

 thickest on flat ground and on calcareous rocks, but under- 

 exceptional circumstances where there is a leakage of water from 

 a limestone formation on an escarpment, such as the escarpment 

 of the Kaap Plateau, thick deposits may be found on steep 

 slopes. The appearance of these tufas with decreasing humidity, 

 up to a certain limit, is well illustrated in this country, both in 

 the Transvaal and further south, as one travels westwards.^' 

 The tufas, especially if the conditions be unfavourable for the 

 formation of thick deposits, will disappear when a humid 

 climate supervenes, and in this respect they differ from the 

 silcretes and ferricretes. The want of any sort of time scale for 

 the formation of these tufas prevents their being of use in judging 

 the period required for the accumulation of particular deposits, 

 and obviously the rate must vary enormously with circum- 

 stances.^' In very dry countries there is little limestone formed, 

 presumably because the amount of ground water is insufficient. 

 Probably the decrease of limestones amongst the sand dunes of 

 the west coast northwards from Saldanha Bay is to be attributed 

 to this. 



The thick deposits of tufa in parts of the Transvaal Bush- 

 veld, the Western Transvaal, and the Kaap Plateau point to the 

 long duration of dry conditions. 



The presence of shells of fresh-water moUusca in the 

 calcareous tufa round some of the pans''^ which are now rarely 

 filled with water, is evidence that conditions were favourable for 

 their existence at times, but shelly deposits in pans are unusual, 

 and the large pans of the Transvaal, such as Lake Chrissie, 

 which generally contain water, seem to be very poor in 



