96 



with magnesium carbonate. By the end of May, 1924, all the 

 plants in the 2, 3 and 4 tons per acre of magnesia lots were 

 killed. One ton per acre showed some ill effect at first, but the 

 crop recovered. With magnesium carbonate there was no failure^ 

 but, on the contrary, a slight proportional increase all round. 



The crops were cut on August 18th, and the comparative 

 results were : — 



Magnesia and Magnesium Carbonate upon Wheat — Stackyard Field Soil, 



1924. 



The Table shows that an increase of crop is given with a 

 half-ton and 1 ton of magnesia, but that 2 tons per acre or more 

 will absolutely kill a wheat crop, whereas higher amounts of 

 magnesia as carbonate will improve the crop. Magnesian lime- 

 stone, however, is ineffective, at least in the first year. 



These experiments on lime and magnesia (3 and 4), confirmed, 

 as they have been, on different soils of the farm, and with and 

 without mineral manures, leave no doubt that there is a very 

 marked difference between the effect of caustic lime and that of 

 carbonate of lime, and again, between lime and magnesia. Caustic 

 lime has clearly been proved to be a far more active form than 

 chalk, and, while its addition, within reason — say up to 2 and 3 

 tons per acre — will produce much benefit on land requiring lime, 

 magnesia, in the caustic state, will in that amount prevent the 

 growth of the crop. The further information is now given that 

 ground limestone, be it magnesian or not, exercises no influence, 

 for a time at least. 



These experiments have now been, in the main, so fre- 

 quently repeated, and with like general results, as to leave practi- 

 cally no room for doubt as to their bearing on agricultural prac- 

 tice, and on the respective use of caustic lime, chalk, caustic 

 magnesia or carbonate of magnesia. Incidentally, as I have 

 pointed out elsewhere, they have a marked bearing on the practi- 

 cal treatment of land which contains magnesia in excess of the 

 lime present. 



