OEGOLOGY OF SOME PLANT COMMUNITIES.
39
Attention was next directed to the water content of the soil. In
all our earlier estimates samples were taken to a depth of 8 inches
only. On May 13th samples were also taken from each area at a
depth of 18 inches, but the results were identical in each case with
those taken at 8 inches. Erom the study of root systems to be des-
cribed below, it afterwards appeared that it might have been better
to study the water content at much greater depths, and we hope to
complete a study of this question at a later date. Nevertheless the
study of water content to a depth of 8 inches only, yielded results of
considerable interest, and we think that the figures obtained are likely
to bear a close relation to the water content at lower levels, especially
in view of the identical results obtained at a depth of 18 ' inches. The
samples were usually taken in air-tight, screw-top aluminium jars 1
and the water content calculated, after drying at 100 C., as percent-
ages of the wet soil.
On some occasions the soil temperatures were recorded at the same time, but no appreciable differences were found in any of the areas and these records were therefore discontinued. The records of water content are as follows : —
 | A table should appear at this position in the text. See Help:Table for formatting instructions. |
As rain fell on 13-5-1920, after No. 6, the records were dis- continued.
An examination of these figures indicates (l) that the ratio of water content in the four areas remains similar at all seasons, the order being always II, I, III, IV, and (2) that in the hot weather the actual water content in each case appears to converge to a minimum, with the possible exception of area II. The next step was to en- deavour to connect the water content with the physical structure of the soil, and this proved a more difficult problem than was anticipated. In the first place the soils were separated into three grades of fineness by passing air-dry weighed samples of each through sieves of 30 and 60 meshes to the inch. As in each case the proportion of small stones, etc., too large to pass through the 30 mesh sieve was very
