ACTION OF GYPSUM ON CLOVER NOT KNOWN. 325 



of sulphuric acid than on the sulphate of magnesia plot, 

 and absorbed a correspondingly larger proportion. But 

 this additional quantity of sulphuric acid absorbed did 

 not increase the amount of produce ; on the contrary, 

 on the plot manured with sulphate of magnesia, which 

 had received less sulphuric acid than the gypsum plot, 

 the amount of vegetable matter was 8 per cent, higher 

 than on the latter. 



These facts show that we are still in the dark about 

 the action of gypsum ; and it will yet require a great 

 many and most accurate observations before we are 

 likely to arrive at a satisfactory explanation. 



So long as the notion was generally entertained that 

 plants derived their food from a solution, the effects of 

 a soluble salt upon vegetation could, of course, be at- 

 tributed only to the constituents of that salt. But now 

 we are aware that the earth itself performs a special part 

 in all the processes of nutrition ; and there might, there- 

 fore, be grounds for supposing that the action of gypsum 

 upon arable earth, or of the latter upon the former, 

 might furnish a key, to some degree at least, to explain 

 the" effect of gypsum upon the growth of clover. A 

 series of experiments made by me upon the alterations 

 which a saturated solution of gypsum in water under- 

 goes in contact with different arable soils, give very 

 remarkable results, whch I will now state, without 

 venturing to draw any definite conclusions from them. 



I found that a solution of gypsum in contact with 

 all the arable soils which I used, underwent decompo- 

 sition, part of the lime separating from the sulphuric 

 acid, and magnesia and potash taking its place, quite 

 contrary to the ordinary affinities. 



The experiments were made as follows : 300 

 grammes of each earth were mixed with a litre of pure 

 water, and 300 other grammes of the same earth with 

 a litre of a saturated solution of gypsum. After twenty- 

 four hours the fluid was filtered, and the filtrate tested 

 for magnesia. Pure distilled water took up from all 

 the experimental earths, sulphuric acid and chlorine, 

 besides traces of lime, magnesia, and soda, and occa- 



