Mai!. 1898.] BOTANICAL SOCIETY OF EDINRUIJGH 



115 



of plants, namely, phosplioiic acid, potash, and nitrogen, 

 and also in two ingredients which are of great importance 

 in promoting fertility, namely, organic matter and lime. 



The question before us, however, is not one of relative 

 fertility, which has to do with the number of bulbs that 

 can be grown per acre, but only one of quality as regards 

 the matter of colour. It is evident that the poverty of 

 soil Ko. 1 is due in great measure to the large proportion 

 of stones in it. It would be necessary to remove the 

 excess of these in order to compare the actual fine soil 

 contained in this sample with the others. In the sub- 

 joined tables are given the analyses of the fine soils from 

 which the stones have been altogether removed. 



Seeing tliat the lightest coloured daffodils were grown 

 on soil No. 1, the deepest coloured on soil No. 3, and that 

 those on soil No. 2 were of an intermediate shade, it is 

 evident that if these differences were due to differences 

 in the chemical composition of the soils, we must look for 

 the explanation in the amounts of these constituents that 

 either increased or decreased regularly in proceeding from 

 soil No. 1 to soil No. 3. We see that, in fact, the organic 

 matter, the phosphoric acid, the lime, and the peroxide of 

 iron all varied in that order, while none of the other con- 

 stituents varied regularly in the opposite direction, so that 

 we are led to infer that the increasing depth of colour may 

 be due to the greater amount of one or more of these four 

 constituents. It m.ay be that only one of the four was 

 concerned in producing the colour change; but if so, we 



