260 USE OF THE LEGUMINOS^ FROM 



and constitutes in the fertilizer industry the best and 

 highest-priced manure. M. Desbassyns has demonstrated in 

 Reunion the effects of Lima beans on particularly poor soils, 

 the yields having been increased in remarkable fashion after 

 four years' rotation. As we know that the soils of this 

 colony are usually rich in nitrogen, we cannot but think 

 that the cause of their improvement lies in the modification 

 of the soil layer by another fertilizing agent. It is certain 

 that the transformation of the mass of foliage into humus, 

 through several years' successive decomposition, must have 

 acted in a favourable manner on the arable layer. 



Modern research has thrown new light on this subject. 



Professor Dietrich Meyer, of Magdeburg, cultivated 

 wheat, barley, oats, lupin, vetch, &c., in vessels filled with 

 soil previously dosed with the nutritive elements which 

 are soluble in weak acids. He observed that the soil in 

 which the LeguminoScC had grown contained more soluble 

 phosphoric acid, potash, and lime than that in the vessels 

 containing the cereals. It is to be presumed that in opera- 

 tions on a large scale a crop planted in soil which has 

 carried a leguminous plant benefits from the dissolving of 

 the mineral elements brought about by the latter, which, 

 according to the researches of Bernard- Dyer, possesses more 

 radicular acidity than others, such as cereals, crucifers, &c. 



The humus resulting from the decomposition of the 

 Leguminos^ is naturally superior to that of the Graminse, 

 Panicum maximum, Cynodon dactylon, the reason being 

 that lime, magnesia, potash, phosphoric acid, &c., account 

 for three-quarters of the mineral matters contained in the 

 Leguminos^, silica only accounting for one quarter; whereas 

 the ash of the Gramin^e usually gives 75 per cent, silica and 

 only 25 per cent, lime, potash, &c. We are indebted to 

 M. Boname for the establishment of this fact, the result 

 of numerous analyses made at the Station among both 

 Leguminosas and Gramin^e. Our thanks are also due to 

 M. Joulie for the comparison of the ash of Gramin^e and 



