THE CHEMICAL COMPOSITION OF LAKE WATERS 151 
sulphate or carbonate, or magnesium sulphate, may crystallise out, 
according to the composition of the solutes and the temperature. 
The proviso as to temperature is necessary, because the solubility of 
sodium sulphate increases in an exceptional degree with temperature ; 
in fact, there are waters which are unsaturated for this salt in summer 
and deposit it in winter. Beyond 30 per cent., sodium chloride 
begins to separate out. Magnesium chloride remains in solution to a 
point not far short of actual desiccation.^ 
The following analyses of terminal lake waters illustrate the 
diversity which the composition of such waters can assume, and the 
ffreat difference between them and river waters : — 
CS (N_ 
M 1 § 
^ oA 
Lake Van, 
Armenia 
(Chancour- 
tois).3 
Lake TJrmi, 
Persia 
(Giinther ^ 
and Manley). 
!^ 5 - -r^ 
h- 1 
Great Salt 
Lake 
( Waller ).6 
Caspian Sea 
(Lebedint- 
zeff).^ 
00 
0 -u 
Q 
Dissolved matter : 
Parts per thousand. 
58-0 
21-0 
148-5 
2-2 
199-2 
12-6 
35-0 
^ (Ca. 
o M Na. 
c ° § Cl,Br . 
II S64 . 
a l(AlFe)o03 
0-39 
2-1 
31-7 
0-27 
28-2 
37-1 
0-04 
0-02 
0- 57 
37-6 
1- 2 
20-7 
27'1 
11-9 
0-85 
trace 
0-32 
2-5 
34-0 
0-78 
57-3 
5-0 
0-66 
3-3 
35-8 
40 "9 
15-7 
2-9 
0-27 
0-45 
1- 0 
2- 1 
32-9 
1-7 
55-7 
6-5 
|o-oi 
2-6 
5-7 
24-5 
0-60 
0-93 
41-8 
23-8 
1-2 
3-7 
30-6 
1-1 
0-21 
55-5 
7-7 
In these lakes deposition of sodium salts has not yet begun, or is 
at an early stage. Lake Sarat represents a concentrate of normal 
river waters, Lakes Van and Palic of alkaline waters. In Lake LTrmi 
and Great Salt Lake we see the effect of an elimination of calcium 
sulphate and find sodium chloride accumulating ; both are lakes of 
very high salinity. It will be observed that the dissolved solids of 
these two lakes closely simulate those of the ocean in composition ; 
but this is more or less accidental, and is by no means to be regarded 
1 These statements are to be regarded as simplifications. As a matter of fact, 
the crystallisation of concentrated natural waters is an exceedingly complicated 
affair, owing to the numerous double salts which can and do separate from solution. 
The physical chemistry of the subject for the special case of sea-water has lieen 
worked out with remarkable thoroughness, and may be referred to in A^an't Hoff's 
Ozeanische Sahahlagerungen (2 vols., Brunswick, 1905-09). 
2 Bujor, Ann. Sci. Unw. Jassy, i. p. 158, 1901. 
2 Gomidtes rendus, xxi. p. 1111, 1845. 
4 Proc. Roy. Soc, Ixxv. p. 312, 1899. 
^ Jahrh. geol. Reichsa7istaU, vii. p. 361, 1901. 
6 J. E. Talmage, "The Great Salt Lake," Scott. Geogr. Mag., xvii. p. 635, 1901, 
^ Karaboghaz Expedition Eeports (in Kussian), St, Petersburg, 1902. 
8 Ghallejiger Eeports {Phys. Chew,. Ghall. Exp., Part I. p. 203), 1884. 
