TRANSACTIONS OF SECTION M. 647 
out showing that on land of moderate fertility a ratio of 600 to 700 would be 
required for the wheat areas. This was roughly double the English ratio. 
Again, as regards soluble salts, the drier areas commonly held a slightly higher 
percentage than British soils, and while in Western Australia ‘alkali’ rarely of 
itself caused infertility, his experience of alkali lands, which he had investigated 
for the Victorian Government, indicated that such salts increased the liability 
of crops wilting. On consideration they would expect this. Again, the physical 
character of the soil had an important effect, and the sandy character of much 
of the western lands gave it an advantage over the heavier soils in a dry season. 
This was contradictory to his experience in the English Midlands with a 32-inch 
rainfall. Fifteen inches of rain absorbed by the surface five feet of soil would 
add something less than 20 per cent. of water calculated on the dry soil if it 
were absorbed without loss. But the annual rainfall was spread over several 
months, and the fact seemed to be that with a 15-inch rainfall the sandy soil 
could hold all the rain which fell, and the greater absorbent power of the 
clay soil was then of no advantage. It was indeed a disadvantage, as the finer- 
grained soil could not yield up so much of its absorbed water before wilting set 
in, and in the drier seasons and districts the ‘sand plain’ gave superior results 
to the forest land. In regard to cultivation methods, the author quoted figures 
from his experiments showing the large saving of soil-moisture by early cultiva- 
tion and maintaining a soil mulch. The water saved would usually equal from 
5 to 7 inches of rain in the surface five feet of soil. In Western Australia good 
results from fallowing were more easily obtained than in Victoria, where the 
more frequent summer rains tended to cake the surface, rendering fresh working 
of the land necessary. The water saved showed itself in the crop yields, and 
the results of a Kellerberrin farmer last season, showing 17 bushels on sand plain 
fallowed, and 5 bushels on similar land ploughed from stubble, could be regarded 
as typical under a 12-inch rainfall. The British farmer did not sufficiently 
realise the use of the soil mulch in protecting his winter-ploughed lands from 
the drying winds of spring. Lastly, as to the selection of drought-resistant 
plants, much had been done through acclimatisation, selection, and cross-breed- 
ing, but a careful analysis of the various factors which in wheat constituted 
drought-resistance remained to be carried out before they could claim that plant- 
breeding for this object was placed on a scientific basis. Under the dry con- 
ditions of Australian wheat-growing a safe yield rather than a heavy yield was 
the primary consideration. This necessitated the selection of early or middle- 
early varieties, thin seeding, and in the great majority of cases the non-use of 
nitrogenous manures. 
(iv.) The Capillary Power of Soils. By Heser Green, D.Sc. 
The conventional mechanical analysis supplies data about the sizes of the particles 
of the soil; the information actually required concerns the behaviour of the soil 
with respect to the movements of air and water therein. These latter are dependent 
on the sizes and distribution of the free spaces between the particles and only 
indirectly on the sizes of the particles. 
This suggests a direct measurement, if possible, of the factors determining these 
physical characters and conditions of the soil, and the magnitudes to be considered 
are :— 
S, the pore space, expressed as a fraction of the total volume of the soil; and 
6, the water-content, similarly expressed. 6/S is then the fractional saturation 
of the soil. 
Pa and Pw, the permeabilities to air and water. Incidentally the ratio of these 
two gives us an indication of the amount of colloidal matter present in the soil and 
of its tendency to swell when wet. 
K, the capillary power. This taken with the previous factors gives a measure 
of the rate at which water will percolate from a wet to a dry region in the soil. 
These factors (S, 6, P, and K) have been previously defined and methods for 
their measurement described.! * 
Of these §, 8, and P are simple properties with obvious physical meanings, but 
K (the capillary power) is of a more complex character and may be defined as the 
1 Heber Green and G. A. Ampt, Jour. Agr. Sci., 1911, 4, p. 1. 
