392 



alkalinity decreased from 60.2 to 47, a result which he however 

 suspects being due to an observational error, while in the 

 second it increased from 50.2 mgrs. to 60.8. 



Irvine & Yodng (Froc. Roy. Soc. Edhiburgh vol. 15, 1887 

 p. 316) made some further experiments and found that the 

 seawater from the «German Ocean«, on which they experi- 

 mented, was always capable of dissolving more lime from the 

 remains of different organisms as well as from crystallized and 

 amorphous carbonate of hme. The quantities taken up vary 

 very much, according to the material used, from 32 mgrs. of 

 Coral sand to 649 mgrs. of precipitated carbonate of lime. 



Andehson [ibid. vol. 16. 1889, p. 319) studied the solubility 

 of carbonates in solutions of different salts and found that 

 these, with the exception of CaSO^, dissolved more carbonate 

 of lime than distilled water. In artificial seawater, free from 

 carbonates or carbonic acid, very nearly as much was dissolved 

 as natural seawater will take up in addition to its original 

 alkalinity. The Author arrives at the conclusion, by no means 

 borne out by his experiments, that the solubility of carbonate 

 of lime «has nothing to do with the existence of free carbonic 

 acid or bicarbonates». His experiments prove, however, that 

 certain seawater-salts, especially the MgCl^, play an important 

 part with regard to the solubility of carbonate of lime and that 

 therefore Schloesings formula, quoted above (p. 368), by which 

 it ought to be possible to compute the quantity of CaCO^ 

 corresponding to any given tension, cannot be used in the case 

 of seawater. 



By boiling in vacuo or in a current of COg-free air sea- 

 water rapidly loses its dissociable carbonic acid and becomes 

 turbid ^), but, on the other hand, no trace of turbidity became 

 visible at ordinary temperatures or, indeed, up to 80° in my 

 experiments -with very low tensions (0.7 — 1.1-5), and the direct 

 determinations of the alkalinity showed no declime. 



') By this fact the conclusion of Anderson is disproved. 



