212 
Mr. Knight on the Causes zvhich influence 
which was disposed, in almost every possible way, within 
their reach; and I always found abundant fibrous roots in the 
rich soil, and comparatively few in the poor. 
The following experiment afforded the most remarkable 
result, and one the least favourable to the hypothesis which 
I have advanced in a former Paper, * and to the conclusion 
which I shall now endeavour to support ; and therefore I 
think it necessary to describe it very minutely. Some seeds 
of the common bean (Viciafaha), the plant with which many 
former experiments were made, were placed upon the surface 
of the mould in garden pots, in rows which were about four 
inches distant from each other. A grate, formed of slender 
bars of wood, was then adapted to the surface of each pot, so 
as to prevent both the mould and the seeds falling out, in 
whatever position the pots might be placed ; and the bars 
were so disposed, as not at all to interfere with the radicles 
of the seeds, when protruding. The pots were then directly 
inverted ; and the seeds were consequently placed beneath 
the mould ; but each seed was so far depressed into the mould, 
as to be about half covered : by which means each radicle, 
when first emitted, was in contact with the mould above, and 
the air below. Water was then introduced through the 
bottom of the inverted pot, in sufficient quantity to keep the 
mould moderately moist ; and, the pots being suspended from 
the roof of a forcing house, the seeds soon vegetated. 
In former experiments, f wherever the seeds were placed 
to vegetate at rest, the radicles descended perpendicularly 
downwards, in whatever direction they were first protruded; 
but under the preceding circumstances they extended hori- 
* Phil. Trans. i8c6, page s. f Ibid. 
