Average 





Sea-water. 



Mud-water 



77-758 



79-019 



10-878 



11-222 



0.217 



0-220 



4-737 



3-252 



2-365 



2-506 





O-206 





0-909 



0-345 



2-686 



3-600 





100-000 



100-000 



REPORT OF TEE CHIEF ASTRONOMER 663 



SESSIONAL PAPER No. 25a 



Sodium chloride 



Magnesium chloride.. 

 Magnesium bromide.. 

 Magnesium sulphate.. 

 Potassium sulphate.. 

 Ammonium sulphate. . 

 Magnesium carbonate. 

 Calcium carbonate.. .. 

 Calcium sulphate.. .. 



In the mud-water calcium sulphate is absent, magnesium sulphate is deficient 

 when compared with average sea water; calcium carbonate is increased, and 

 magnesium carbonate and ammonium sulphate are both present. The high 

 chlorides show that the carbonates are not in excess because of fresh-water inflow. 

 The ratio of magnesium carbonate to calcium carbonate is 1 :3. When the 

 clear water filtered from the mud was boiled for a short time, a crystalline 

 precipitate was thrown down, consisting of 73-3 per cent calcium carbonate and 

 26-7 per cent magnesium carbonate. The formation of both carbonates is 

 ascribed by Murray and Irvine to the reaction of ammonium carbonate chiefly 

 on the sulphates, a conclusion which cannot be doubted, especially in view of the 

 presence of ammonium sulphate in the mud-water. The alkaline carbonate was, 

 of course, derived from decaying animal matter contained in the muds. 



These different experiments teach that hydrous carbonate of magnesium 

 can be precipitated by ammonium carbonate emitted from decaying animal 

 remains; that the precipitation is much slower than in the case of calcium car- 

 bonate and is retarded by the presence of calcium salts in the solution. We 

 thus see how, in the nearly limeless sea water of pre-Cambrian time, the propor- 

 tion of precipitated magnesium carbonate would be high, even, possibly, approach- 

 ing the ratio in true dolomite. Indeed, it is quite possible that precipitates of 

 pure basic carbonate of magnesium later changed to magnesite, were formed 

 in those places in the sea basin where the calcium salts were completely absent 

 from the oceanic composition. 



On the other hand, the abstraction of magnesium from the pre-Cambrian 

 water may have followed a process analogous to that which may be actually 

 observed in the boring at the Funafuti atoll.* Crystals of dolomite may have 

 grown on or near the surface of the pre-Cambrian bottom mud, much as they are 

 now growing in the buried (porous) coral rock of the atoll. 



Clearly our ideas must not be too rigid regarding the detailed history of 

 these early magnesian deposits. We cannot say how far the sea waters in which 

 animal life first began were charged with magnesium salts. We cannot 

 say how far these and the other salts brought in by the early rivers con- 

 tributed to the formation of the extensive dolomites and magnesium limestones 

 known to occur in pre-Cambrian terranes. Our hypothesis holds that the 



*The Atoll of Funafuti, published by the Koyal Society, London, pp. 392, 413, etc. 



