EXPERIMENTS ON THE GROWTH OF PLANTS. 205 
On the alluvial sandy clay banks by the seaside, or poor siliceous clays 
inland, the F. elatior rears its tall coarse form. In Gloucestershire the 
banks of alluvial mud thrown up to prevent the encroachments of the water 
in the Severn estuary are always occupied by this grass, which I look upon 
only as the extension of the pratensis from the rich flats within this boundary. 
The F. pratensis is a grass which is usually recommended for admixture 
in forming new pastures, on which account there can be but little doubt that 
it was used in the glades laid down within the last few years at the entrance 
of Oakley Park, the seat of the Earl Bathurst. When first sown it came up 
true enough, though with a disposition to reediness ; the last four years it has 
become wholly F. elatior in all its features, and is now in such large coarse 
‘“‘hassocks” as to be dissightly as a lawn, and much impairs the hay or pas- 
ture. The secret of all this appears to be that here it was sown on the forest 
marble sandy clays, the texture of which as a soil is similar to that in the 
favourite habitat of this form of grass, and this too, though in a less degree, 
no doubt, favoured the changes as observed in my botanical garden. 
Here then we see, in these forms of Fescue, plants which assume what 
have been taken as specific characters, not only from change of circum- 
stances giving rise to varieties which have been obtained from different 
generations by seeding, but these have assumed the form of varieties from 
the same seed and plants, and absolutely becoming F. pratensis, and after- 
wards F’. elatior from the typical /. doliacea: and so certain is this in my 
experimental garden, that the result of twice sowing these three forms from 
seed from different seedsmen has been the permanent establishment of F’. 
elatior in all three plots. 
Bromus mollis varieties. 
My experiments and observations upon the annual forms of Brome, though 
still in progress, yet seem to warrant a diminution in specific names; for 
example, B. mollis and B. racemosus of authors are sure to be intermixed to 
a greater or less extent from the same seed; thus the seed of the B. mollis 
will have a sprinkling of the racemosus, whilst seed of the latter will present 
exceptional examples of the former; and, besides, all distinction is lost in 
every shade of intermediate form by which the hairy and smooth varieties 
are connected. 
Again, as regards B. commutatus, this is by far too common a grass in pas- 
tures subject to floods and in irrigated meadows, in which situation the 
B. mollis is quite exceptional. Now, as I have watched the laying out 
of poor pastures as irrigated meadows, I have always observed that two or 
three years is often sufficient to change the B. mollis which was alone before 
into B commutatus. Of course it may be considered that this was in virtue 
of that law of substitution of one species for another which so universally 
occurs on a change of soil and other conditions ; but I incline to the belief 
that much of this is after all duc to a change of form and specific character, 
and as regards the grass under consideration our chain of evidence is nearly 
complete when it is stated that the B. commutatus from the irrigated mea- 
dows, most certainly in experiments in my garden, has resulted in fine exam- 
ples of B. secalinus, a form not before known there, and therefore not liable 
_ to have led me into error, as would be the case where the different varieties 
are wild natives near the spot. 
I have not been able to experiment upon the whole of the forms of what 
- I would term the B. mollis group, but I suspect that the B. arvensis which I 
this year found so abundantly on the chalk about Avebury, in Wiltshire, is 
but a form of the same; and though in all probability a foreign one intro- 
duced with “seeds,” yet its individuality may have been implanted by 
