294 BULLETIN OF THE BUSSEY INSTITUTION. 
soil-nitrogen, as I have explained in another paper.* In the 
assay of any loam by the proposed plan the conditions would 
usually be tolerably favorable for nitrification, and perhaps very 
nearly equally so in all cases, while in the field the conditions are ~ 
in some instances favorable for nitrification and in others adverse, 
and the loams are, practically speaking, fertile or sterile accordingly. 
It is of course conceivable that some useful modification of 
the process may hereafter be devised, perhaps for example by the 
use of plants of such extreme sensibility, that they will grow in 
some classes of soils and not in others; but for the present I am 
forced to the conviction that considered merely as an ‘‘ assay ” 
process teaches nothing but what could be learned, with vastly 
less trouble, by means of chemical analysis. I have found, in 
short, that while it is easy to prove by the assay that soils re- 
puted poor really contain large quantities of plant-foods which 
only need opportunity in order to become available for the support 
of crops, it is extremely difficult to judge by means of it between 
several poor or several rich or several mediocre soils in such wise 
that the verdict shall be of the least use for the farmer. 
The plan of the proposed assay was as follows : — 
Instead of growing plants directly in the soil whose character is 
to be determined, the idea was to take a quantity of pure silice- 
ous sand or other absolutely inert material as the standing room in 
which the plant should grow; and to mix with this sand a small 
portion (or several small portions) of the soil to be assayed, 
precisely as if this soil were a chemical substance of unknown 
composition whose action towards the plant we wished to test. 
A series of glass jars (ordinary wide-mouthed green-glass pre- 
serve jars in most instances) were charged with the mixture of 
sand and soil, seeds were sown in each jar, and the contents of the 
jars were kept moist with pure water until the young plants were 
well started. Henceforth the jars were watered methodically with 
highly dilute solutions of chemical substances, each jar being 
watered with its own special solution, and the several solutions 
being compounded in such wise that to some one at least of the 
jars there should be supplied whatever kind of plant food the soil 
might lack. ft 
In the earlier trials a number of different chemicals or mixtures 
of chemicals were employed in this sense, but one or another of 
* Bussey Bulletin, 2. 280. 
+ Compare Bussey Bulletin, L. 54. 
