Relation to Chalk and Mineral Salts. 81 
The sample of contortum used had a remarkably low acidity, only 
about half that usually shown by this species. 
The plants were grown in 600 c.c. conical Jena glass flasks, in 300 c.c. 
of solution. Ten plants 5 cm. long were grown in each flask, and every 
culture was duplicated, so that the results are averages for twenty plants. 
The cultures lasted five weeks, from the middle of March to the end of 
April, and were kept in an unheated room. All cultures subsequently 
described were carried out in precisely the same way. The results are 
calculated in percentages of the growth in distilled water; what that is for 
the various species may be seen from the following figures : 
S. contortum ) 4*6 cm. 
"S. recurvum , 2*25 cm. 
.S. rubellum , 0-75 cm. 
The small growth of rubellum makes it impossible to regard small differences 
in its case as significant. 
The results of the cultures in chalk solutions are given in Table VII. 
Table VII. 
CaC 0 3 in mg. 
per litre. 
Growth of Sphagnum 
contortum. recurvum. rubellum . 
O/ O/ O/ 
50 
/o 
84 
Vo 
47 
Vo 
0 
75 
57 
49 
0 
100 
44 
5 i 
0 
125 
30 
24 
0 
150 
11 
16 
0 
I 75 
4 
14 
0 
200 
0 
11 
0 
225 
0 
9 
0 
250 
0 
9 
0 
From this it appears that contortum is the least sensitive, though 
recurvum shows a very slight growth in high concentrations. The geotropic 
reaction persists in contortum up to 150 mg., while in recurvum it disappears 
at 100; the latter is also much more severely bleached: rubellum is 
the most severely affected. 
The investigations of Paul on this point have been described in 
detail. 
The experiment quoted in Table VII shows : (1) that the species 
studied are sensitive to chalk in different degrees; (2) that the more acid 
species are the more sensitive. Paul’s thesis is thus confirmed in principle. 
But the very exact parallel which he finds must be criticized on several 
grounds. In practice it is not possible to determine either acidity or 
sensitiveness with the degree of accuracy which Paul suggests. 
As far as can be gathered from his paper, it would seem that the 
samples used for acidity determinations were the same as those used for 
G 
