CV1 PROCEEDINGS OE THE GEOLOGICAL SOCIETY. [May 1909, 



mica-schist and phyllite, but when the salinity is taken, into 

 account it will be found equally characteristic of those proceeding 

 from the basalt and the Cretaceous rocks. This chlorine is certainly 

 not derived from the rain, it is native to the soil, and yet cannot be 

 traced to ancient desert-deposits. Evidently we have still much to 

 learn regarding the distribution of this element, and in any case it 

 would seem that some at least, possibly a large part, of the sodium 

 associated with chlorine, which is included in our first table, may 

 fairly claim to be taken into account. 



In the next place, since the streams from which this water was 

 taken are much shorter than those afforded to our study by the two 

 Americas, it may be supposed that the silica which they contain as 

 a result of disintegration has not been exhausted to the same extent 

 by organic agencies. It might, therefore, afford us a hint as to the 

 ratio which might be expected to exist between the amounts of 

 silica and sodium present in inland waters, not; far from their 

 source in disintegrating rocks. Such a hint might prove more 

 trustworthy than the conclusions we have drawn from theoretical 

 reactions. Deducting the sodium supposed to be associated with 

 chlorine, the ratios in the several analyses are as follows : — 



Rock. Sodium. Silica. 



Granite 3"24 1733 



Mica-schist ... 6-14 15-00 



Cretaceous ... T56 2'87 



Bock. Sodium. Silica. 



Basalt 3-40 7"67 



Phyllite 4-60 14-93 



Thus the water from the granite gives a ratio of Na : Si0 o = 6 : 1, 

 that from the basalt of nearly 2:1. This suggests a doubt as to 

 whether the silica yielded by the decomposition of ferro-magnesian 

 minerals is really set free (as we have supposed) in a soluble form. 

 In the case of biotite, which we have not considered, since it is 

 not sufficiently abundant to greatly affect our results, the silica is left 

 on treatment with hydrochloric acid as an insoluble residue which 

 retains the form of the original mineral. 



The average ratio, taken from the above table without properly 

 weighting, is 3:1. If we apply this ratio to the 56,830 tons of 

 silica previously found, we shall obtain 18,940 as the number of tons 

 of corresponding sodium, and 114 millions of years as the age of the 

 Ocean. 



The question of submarine supplies of river-water may next 

 engage our attention. Prof. Joly has excluded all desert-regions 

 from his estimates as being non-contributory. At first sight this 

 would seem a necessary precaution, but there is much to suggest 

 that some deserts do in fact drain into the sea, not of course by 



