127 
The Co-Efficient of Humidity. 
But variation from a specific type is axiomatic in biology and 
the determination of the mean or “ mode ” corresponding to the 
type has become a familiar problem in biometrics. Such a point of 
view is necessary in dealing with soils as plant-habitats. The 
number of veins in a beech leaf, as shown by Karl Pearson, 1 varies 
considerably, but by examining a large number of leaves a type, 
represented by the mode, can be established and the greatest 
number of individuals is aggregated round this mode. Beech leaves 
from another locality will possess an appreciably different mode 
and the range of variation will be extended by increasing the 
number of observations. But the extreme deviations from the mode 
will occur very rarely, in accordance with Galton’s Law of 
Frequency. 
Whether or no the water-content is the best method of expressing 
the soil-moisture may be left in abeyance for the moment. Whatever 
is selected as the criterion of the humidity of the habitat its values 
will show variation about a mean. If the deviations follow the 
normal curve of frequency, supposing a sufficient number of 
observations to have been made, the mean affords a definite index 
of the habitat, and a habitat that satisfies such a test is as much an 
entity as is a biological species. The extreme deviations will possess 
considerable significance, for they will indicate the physiological 
limits of the existence of the plant association. Above and below 
these limits the association can no longer exist and even before 
these points are reached it is likely to be invaded by members of 
other associations. Of the two, the minimum value probably has 
the greater significance for ecological purposes. If it is determined 
under such conditions as prevailed during the summer of 1911 it must 
give a close approximation to the true value of Clements’ echard, 
or the non-available water in the soil, perhaps more so than values 
obtained “ under control,” or by cutting out blocks in the field. 
II.—Method of Analysis. 
Before proceeding to explain the method adopted for expressing 
the soil-moisture it is necessary to review briefly the process of 
analysis and the reduction of the results. 
The soil sample taken within the layer of root activity is 
preserved in a tight tin. If in the laboratory it does not appear to 
be uniform it is divided into an upper and a lower layer and suitable 
blocks from the centre of each are rapidly weighed in tared porcelain 
dishes, after larger rootlets have been removed if present. These 
1 The Grammar of Science. 
