SOILS AND MOSSES OF SPITSBBKGEN. 451 
after the soil had been sealed in tins for some months, are of course no 
indication of the actual number of bacteria in the soils at the time when the 
samples were taken, but for purely comparative purposes they no doubt 
correspond to real differences in the bacterial population of the soils. The 
chemical and physical properties of the soils, however, throw no light on 
these differences. The high nnmbers of organisms found in soil 5 may be 
attributed to the sea-birds' droppings with which it was manured ; but, on the 
other hand, soil 4 was similarly manured, and yet yielded fewer organisms 
•than soil 3, which was practically free from humus. The abundance of 
rhizopod tests in soil 1 corresponds closely with what is found in peaty soils 
elsewhere (20). It is possible that this does not really indicate a larger 
active population of these organisms than in other soils^ but is due simply to 
the fact that in a soil, where microbial activity is small, such tests, coming 
either from animals living in the soil, or being brought down from the 
mosses etc. growing in it, disintegrate more slowly than in other soils, and 
so accumulate in considerable numbers. 
The 1922 samples were examined in a fresher condition than those of the 
previous year, and on the whole yielded a considerably larger number of 
living species. The good results obtained from sample 5 (1921), however, 
seem to indicate that these earlier samples had not suffered appreciably as a 
result of storage, and the difference in the results obtained from the two sets 
of samples probably therefore corresponds with an actual difference in the 
population of the soils in situ. Sample 3 (1922) was taken with a view to 
confirming the results obtained from sample 1 (1921), in which both bacteria 
and protozoa were so exceptionally few. Unfortunately it was not possible 
to bring soil from the identical spot, and although the localities from which 
the two samples were obtained appeared verj' similar and supported a similar 
type of. vegetation, the analysis given in the table (p. 456) shows that the two 
soils are not really comparable. 
The abundance of protozoa in samples 1 and 3 (1922) is of particular 
interest, in view of the high acidity of these soils. Such results, however, 
are not unique, for an active protozoal population has been found by 
Mile. Perey in the soil of one of the Rothamsted experimental plots for 
which the pH is 365; and S. M. Nasir (in some unpublished work also 
done in this laboratory) has found that in artificial cultures active flagellates 
can tolerate a pH of 4"5, active amoebae a pH of 3"9, while ciliates can 
grow even when the aciditj^ is as high as that represented by a pH value 
of 3-5. 
3. Mosses. 
Twelve of the samples came from Klaas Billen Bay, and were composed of 
the following mosses : — Hypmnn (5 spp.), Cinclidium stygium, OrtJiotherium 
clu'i/seum, Sivartzia moiitana, Bryxnn spp., Camptotherium nitens, Grimmia 
33* 
