.^2S PERSISTENCE OF ARSENITE. 



The manured plots were better than the unmanured ])lots. 

 The manured part of the control plot was better than any of the 

 others, including" the isolation plots, which received neither 

 manure nor poison. The manure in the poisoned plots clearly 

 counteracted the effect of the poison to some extent ; but whether 

 this was due to the influence of the manure in the ordniary way 

 or to some interaction between the arsenical and the organic 

 compounds in the manure it is impossible to say. 



In his notes, Mr. van der Merwe several times records that 

 the ])lants near the borders of the plots were better than in the 

 centre, and suggests that this may have been caused by leaching 

 due to the slight rills which marked the line of plots providing 

 better drainage around the borders. Judging from the general 

 appearance of the plots at the time, this explanation at least 

 seems plausible. 



The injurious effect of the arsenic was most noticeable 

 during June and July, while the different plants were young. 

 Of the grain crops, severity of injury was apparent in the order 

 named : — Oats, barley, wheat, and rye. Oats were very sensitive, 

 and would make a valuable " indicator " of arsenical poison in 

 the soil. Barley followed it very closely. Wheat was consider- 

 ably less sensitive than oats or barley. Rye showed a considerable 

 degree of resistance, so nnich that it would be of little use as an 

 indicator ; but this fact establishes its value as a crop to be grown 

 when arsenical soil-poisoning is suspected. Peas were much 

 more sensitive than beans — in fact, peas would also serve as a 

 good indicator of arsenical soil-poisoning. Although the other 

 crops recorded above showed the effect of the poison, they were 

 not specially sensitive, and were so irregular that no close com- 

 [)arisons could be made. 



As the rains diminished, and the season advanced into spring 

 and summer, all of the plants made more rapid progress. 

 Although there was a heavy mortality amongst the original plants 

 in tlie 500, 400, and 300 lb. plots— 50 to 70 per cent, in oats and 

 barley — the remaining plants did so much better on account of 

 having more room, that on first glance they obscured the effect 

 of the arsenic ; but on areas sufficiently large to give reliable 

 harvesting results by weight, the yield would have been low in 

 Ijfojjortion. 



The plots were continued through 1916, 1917, and 1918, 

 with the same crops as indicators. The results are in the same 

 order, but distinctly less marked each succeeding season. So far 

 during the present season (June, 1918) the effect of the arsenic 

 in the stronger plots can still be traced in oats and barley ; but 

 it is doubtful whether it will show to any extent as the crops 

 approach maturity. 



In the experimental plots, amounts of arsenic, far in excess 

 of any possible bait-accumulations in the manure, did not have 

 a noticeable effect on the crops grown. It therefore seems justi- 

 fiable, on the basis of the above analyses, and the plot tests 



