5- 



Journal of Agriculture, Victoria. [lo Sept., 1912. 



sampling to that depth. Attention was then confined to the upper 18 

 inches, where the d fferences of soil were naturally less pronounced. 



In conducting the analyses, water was determined by drying to constant 

 weight in a steam oven. Nitrates were determined according to the method 

 described in a previous article by Mr. W. C. Robertson, Chief Deputy 

 Chemist. 



Fallowing and Soil Moisture. 



Table I. gives the percentage of water in the top 18 inches of soil at 

 each date of sampling. The results are calculated on the dry soil. 



Table I. 



October 16 

 November 28 

 January 17 

 February 28 



Worked 

 FaUow. 



per cent. 

 25-22 

 19-62 

 20-66 

 21-67 



Neglected 

 Fallow. 



Per cent. 

 24-27 

 15-95 

 11-61 

 12-38 



Oat 

 Crop. 



Per cent. 

 24-95 

 15-22 

 10-01 

 10-06 



In viewing the results the conditions regulating the loss of water on 

 good fallow, bad fallow, and on crop may be briefly recapitulated. Water 

 rises in a moist .soil somewhat rapidly bv capillarity. In a good. fallow, 

 the upper layer of soil is kept loose and dry by cultivation, so that its 

 capillary power is lost. This loose, dry layer protects the water within 

 the capillary zone underneath from sun and wind. In a bad fallow, the 

 surface is compaci;, and also damper, so that capillary action is c»ntinued 

 right through to the top, which is unprotected from drying influences. A 

 crop requires much water for its growth, which must come from the soil. 

 Experiments bearing upon the water losses through plants were previously 

 reported.* 



Influence of Weather. 



Table II. gives the weekly rainfall at Sparrovale from 19th August 

 until the close of the experiments. 



Table II. 



In the weeks omitted no rain fell. September was a wet month, and 

 the plots showed a fair degree of uniformity at the first sampling on i6th 

 October. Witn more moderate but fairly distributed falls in the next 

 l)eriod, the water contents of the soils varied considerably at 28th Novem- 

 ber. The first fallow was hoed over thrice during the period, and the oat 

 <Top was growing rapidly. These circumstances explain the tendency of 

 the results at the second sampling. On 27th December, the crop was cut, 

 and yielded at the rate of 2 tons 4I cwt. hay per acre. From this date 

 until the last samples were taken (two months), a total of 80 points of 



• June, 1912. 



