and the formation of Gypsum and Magnesian Rocks. 377 
to those magnesian limestones which include beds, sneneaiees 2 
organic remains of pure carbonate of lime. In any case 
must suppose a long continued filtration of solutions of sii 
— chlorid through the heated limestone under certain condi- 
tions which seem at least improbable. 
63. The theory of the formation of magnesian sediments — 
be readily understood from the experiments which have bee 
described in the earlier parts of this paper, but before Scene 
to its consideration I wish to call attention to the results of the 
n the waters of the Elbe and Thames chlorids greatly 
a (in the latter with gypsum), with small amounts of 
magnesian salts, and the evaporation of these waters eee: give 
Tise to lakes containing a large proportion of common salt. ‘In 
the Seine on the contrary, aula of lime predominates, while 
the waters of the Rhine, the Danube, the Arr and the Arve 
saan but small amounts of chlorids and large proportions of 
sulphates of lime and magn 
64. In other rivers oe ie alkaline salts; the Loire at Orleans F 
according to Deville, rene: in 100,000 13°46 of soli 
matters, of which 85:0 p. ¢. is carbonate of lime, 30:0 p. c. silica, 
wpile two-thirds of the more oe salts consist of carbonate 
of soda. In the waters of the Garonne, with as large a propor- 
tion of silica, and more te of lime, the carbonate of soda 
equals ohe-founte of the soluble 5 while 100,000 parts of the 
SECOND SERIES, Vor. XXVIII, No. 84.—NOV., 1859. 
48 
