98 



Sir J. B. Lawes and Prof. J. H. Gilbert. [Jan. 9, 



more marked development, both, of swellings on the main roots, and 

 of nodules on the root-fibres. The very meagre development of 

 nodules both with lupin-soil-extract seeding, and in the lupin-soil 

 itself, in 1888, when, as will be seen farther on, the res alt was so 

 different in 1889, suggests the question whether the lupin-sand of 

 1888 had been too much dried, and so sterilised. 



The Analytical Results. 



We will now turn to the evidence afforded by analysis, as to the 

 difference in the amount of growth, and especially as to the difference 

 in the amount of nitrogen assimilated, in the peas grown under the 

 different conditions. 



The following table shows the percentages of ash, and of nitrogen 

 determined by copper oxide (each calculated on the dry substance), 

 in the stems and leaves together, and in the roots, of the plants in the 

 different pots. 





Per cent, in dry substance. 



Ash. 



Nitrogen. 



In stems 

 and leaves. 



In roots. 



In stems 

 and leaves. 



In roots. 



Pot 1. Sand, without soil -ex- 

 Pot 2. Sand, with soil-extraet. . 

 Pot 3. Sand, with soil-extract. . 



Per cent. 

 19-70 

 16 07 

 13-87 

 9 17 



Per cent. 

 28 67 

 36 75 

 23 26 

 20 44 



Per cent. 

 2-904 

 4-900 

 4-006 

 4-534 



Per cent. 



2- 574 



3- 195 

 3-357 

 2 -791 



It is remarkable how much lower is the percentage of ash in the dry 

 substance of the more normally- developed plants grown in the 

 garden soil, than in that of those grown in the sand with plant- 

 ash added. There can be no doubt that the amount of soluble 

 mineral matter provided in the quantity of ash used was excessive ; 

 and less has been supplied in the experiments of 1889. The per- 

 centage of ash in the dry substance of the roots is, however, in all 

 cases high, but d.oubtless some adherent sand would be included. 



The differences in the percentage of nitrogen in the dry substance of 

 the differently grown plants are consistent with the known characters 

 of growth. Thus, the lighter colour, and the comparatively restricted 

 growth, of the planto in pot 1, indicated nitrogen exhaustion j and the 



