THE SOURCES OF THE NITROGEN OF VEGETATION, ETC. 475 
Seedsmen to the Royal Agricultural Society of England, who bestowed much labour 
and attention upon the selection. From the quantity of each kind received, the largest 
and the smallest were picked out, as were also any that did not look quite healthy. 
Given numbers of the remainder were then weighed, and the average weight, per seed, 
was calculated. A few seeds, each weighing as nearly as possible the mean weight, were 
then selected for planting. 
In order to estimate the quantity of Nitrogen in the seeds sown, in some cases a quan- 
tity of seeds equal in weight and number to those sown was submitted to analysis. But 
the difficulties of grinding, without loss, so small a quantity, and the consideration that 
one small quantity might differ more in composition from another such quantity, than 
either would from the average composition of a large number of well-selected seeds, led 
us generally to estimate the Nitrogen in the seeds sown from the percentage of it found 
in the mixture of a large number ground up together. 
The seeds selected for growing were sown in the pots of soil prepared as already 
described, to the depth of about 1 inch below the surface. With large seeds, such as 
Beans, it was necessary that care should be taken so to deposit them that the radicle 
and plumule should each take its natural direction. If this precaution was neglected, 
the seed was liable to be raised out of the soil after sprouting, which involved the incon- 
venience of opening the apparatus in which the plant was enclosed, in order to re-bury 
the seed. 
In some cases, as soon as the seeds were sown the pots were removed from over the 
sulphuric acid, and placed at once beneath the large glass shades which were to serve as 
the enclosing apparatus. In other instances, the pots were first placed under other 
shades, luted by mercury or sulphuric acid, and standing in the laboratory, and then, 
after a few days, they were removed to their final position. 
G.—The Atmosphere supplied to the Plants. 
As regards the essential conditions of growth, and the circumstances associated with 
it, which must be kept within the control of our means of investigation, the same 
remarks apply to the atmosphere, though with less force, as have already been made in 
reference to the soil (subsection A, p. 470). 
It is true that the constitution of the atmosphere is less complicated, and that we are 
much better acquainted with it than we are with that of ordinary soils; yet the extreme 
mobility of the atmosphere renders the presence in it of exceedingly small quantities 
of substances calculated to influence vegetable growth much more dangerous in quan- 
titative experiments on vegetation than would be their presence in the soil. Thus the 
presence of gaseous impurities, and of solids mechanically suspended, in the atmosphere 
cannot be overlooked. And hence, although it is not necessary to submit the natural 
atmosphere to such radical changes as those to which the natural soil must be subjected, 
some measures must be taken to exclude the sources of error to which allusion has just 
been made. 
