The Roihamstcd Agricultural Experiments. 173 
This striking effect, it must be remembered, has been brought about 
in what was originally perfectly uniform meadow-land, entirely by 
differences of manuring, which have so changed the conditions of 
the struggle for existence in each individual plot, as to bring about 
the success of a different combination of species in each case, and 
thus to produce a number of distinct floras living side by side, 
bounded by mathematically straight lines and apparently never 
trespassing across them. 
Of course the component species of these floras are all 
inhabitants of meadowland of one kind or another, and several are 
■common to the majority of the plots, but it is a striking fact that 
some of the plants which occur on plots subjected to the most 
extreme conditions, particularly on the starved plots, are not only 
not found on the ordinary grass land adjacent, but are nowhere 
common in the surrounding country side. Thus for instance 
Poterium Sanguisorba, the Salad Burnet, characteristic of dry chalk 
downs is found only and is very common on the two most starved 
plots, i.e. the plot which has been unmanured since 1856, and the 
plot which has been exhausted by treatment with superphosphate of 
lime alone since 1859 ; while Briza media, the Quaking Grass, 
universally taken as a sign of poor land, occurs in any quantity only 
on these two and on the plot which has no potash. Neither of these 
plants are conspicuous, at any rate, in the countryside. It would 
be an enquiry of some interest to determine what are actually the 
nearest habitats of these species, from which presumably the 
colonisation of the two plots must have occurred. 
The two most starved plots bear a varied herbage (47 species) 
consisting mainly of weede ( i.e. plants which are of little use as 
constituents of hay, and which stock generally refuse) and are very 
poor in grasses, particularly in the valuable ones. 
Manuring with combined nitrogen, on the other hand, either in 
the form of ammonium salts or nitrate of soda, though in the 
absence of the other mineral manures it gives but a poor and weedy 
crop, encourages the grasses at the expense of the leguminous 
plants. 
A complete manure, consisting of the necessary mineral consti¬ 
tuents together with combined nitrogen, enables the grasses largely 
to suppress the competition of the “ weeds ” and completely to 
exclude the leguminous plants. With increase in the amount of 
combined nitrogen the “weeds ” become still fewer, and the weight 
of hay rapidly increases, though the number of species of grass 
becomes very restricted. Thus in one very heavily “over-manured ” 
