191 2.] 'Summary of Agricultural Experiments. 1043 



6. Extra manuring to affected trees did not do away with the 

 effect of grass, which also was just as marked when the crop was 

 grazed by goats as when mown and left on the surface, as was 

 ordinarily done. 



The effect of aeration of the soil was also insufficient to account 

 for the better thriving of the trees grown on cultivated soil, and in 

 the same way there was no evidence to suggest an alteration in the 

 physical condition of the soil or a change in the character of the 

 soil bacteria sufficient to account for the effect. 



Partly as a result of experiments designed to test the point directly, 

 and also partly as a result of eliminating all other possible explanations, 

 the authors conclude that in the growth of the grass a substance 

 toxic to the roots of trees is formed. This substance, however, easily 

 oxidises, and if the grass is killed or removed, the trees soon recover 

 from its effect, and are actually benefited by grass leachings in cases 

 where these liave been exposed to air. 



Live Stock and Feeding Stuffs. 



Feeding Experiments with Cattle and Sheep (Northumberland C.C. 

 Educ. Com., Bull. 17, Cockle Park Expt. Sta.). — This bulletin gives 

 accounts of several different feeding experiments carried out at Cockle 

 Park in 1910-11. The results may be summarised in the following 

 form : — 



1. Testing of Feeding Standards. — Two lots of cattle were fed on 

 rations exactly the same, except that one received 6 lb. soya cake 

 (or decorticated cotton cake) per head per diem, while the other lot 

 were given 4 lb. of soya cake (or decorticated cotton cake) and 2 lb. 

 Bombay cotton cake. The latter, along with 80 lb. swedes, 8 lb. 

 meadow hay, and 8 lb. oat straw, per 1,000 lb. live weight of the 

 animal, corresponded to the Wolff-Lehmann standard, while the 

 former provided a larger proportion of digestible albuminoids. 



The cattle receiving the richer ration increased during the four 

 months' experimental period at a rate of 13! lb. per head weekly, as 

 compared with 13J lb. in the case of the other lot. Owing, however, 

 to the greater cost of the former ration the cattle were fed on it at a 

 slight loss, while the others on the cheaper ration, with less albu- 

 minoids, gave a slight profit. 



2. Comparison of Decorticated Cotton Cake and Soya Cake. — 

 The analyses of the two cakes were very similar, but the cost of 

 the cotton cake was £8 10s. per ton, against £6 12s. 6d. for the 

 soya cake. 



When tried with two lots of fattening bullocks the average gain 

 per head in the lot receiving decorticated cotton cake was 14 lb. 

 weekly, while in the lot receiving soya cake it was 13 lb. The 

 greater cost of the former food, however, more than neutralised this 

 difference, and while the use of soya cake resulted in a gain of 

 25. 8d. per head, decorticated cotton cake gave a loss of 2s. 6d. per 

 head. This confirms the results of previous trials, which have shown 

 44 that these two cakes have a very similar value for feeding purposes." 



With stirks six months old the financial returns were slightly in favour 

 of the decorticated cotton cake. 



3. Feeding Sheep without Swedes. — In this experiment "three-parts 



