2l8 THE MEDICINAL SPRINGS OF SOUTH AFRICA. 



Winclhuk (^19.2°) as being the temperature of the upper layers 

 of the rock, and the maximum temperature of the water (77-5°) 

 as that of the lower layers from which the water derives its heat, 

 the average difference of 58.3° between the surface and lower 

 layers at the " eye "of the springs, gives a depth df origin of at 

 least 2,000 metres. The temperature gradient is taken as 33 

 met.cs per i°C.* No allowance is made for the cooling effect of 

 the rocks before the water reaches the surface, nor for the 

 possible infiltration of cooler surface-waters. Such corrections, 

 if data were available, would give a higher figure than the one 

 calculated above. 



Considerable discussion has centred round the question 

 whether the water of the springs is of meteoric origin, (vadose 

 Thermen), or whether the water is liberated from the magma as 

 water vapour or even hydrogen under high pressure (juvenile 

 Thermen). Although the question has not been solved definitely, 

 the balance of evidence is in favour of the asstmiption that the 

 springs are, at least in great part, of the latter origin, namely, 

 the high and uniform temperature, the regularity of flow, the 

 independence of the water supply elf seasonal variations in rain- 

 fall, and the considerable amount of free carbon dioxide and 

 carbonates dissolved in the water. 



Chemical Nature of the Water. — The composition of the 

 residue from 100,000 c.c. of the water of the Pahlquelle, heated 

 sufficiently high to decompose bicarbonates and hydrates, and to 

 drive oft' water of crystallisation, is as follows* : — 



NaCl 10.7 grams. 



Na^SO^ 12.78 grams. 



Na^CO. 32.OT grams. 



K,SO^ 8.80 grams. 



MgSO^ 3.08 grams. 



CaCO;^ 8.93 grams. 



MgCO, 1.45 grams. 



ALO.^ 0-70 grams. 



SiO. 8.90 grams. 



The water contains appreciable amounts of free carbon 

 dioxide, so that the carbonates of both the alkalis and alkaline 

 earths should be calculated as bicarbonate ions. Lithium and 

 other balneologically important constituents are present in small 

 quantities, but the actual amounts have not been determined. 



In estimating the proximate composition of the mixture of 

 saline constituents of a water the silicic acid is usually regarded 

 as uncombined. Whilst this assumption, even ilf incorrect, in- 

 troduces no appreciable error in the majority of cases, it is 

 wholly unwarranted in the case of hot alkaline springs like those 



* In tlip discussion of the .paper it was suggested that this figure is too 

 h>w. 



t Dr. K. Renz, " I l)er die Eignung der Windhuker heiasen Quellen zu 

 Heilzwecken," from Der Weltkrieg, Windhuk. .Tune, 1917. 



