204 On Permanent and Temporary 3Ieadows and Pastures. 
subsequently each year, to give in the form of additional 
manure the actual quantity of potash required by the crop. 
This will vary from 89 to 223 lbs. per acre per annum, as the 
productive power of the soil, resulting from its other physical 
and chemical conditions, may be equal to a yield of from 1'59 
to 4*38 tons of dry hay per acre. As it may not be known 
what the production will be, I should advise at first a dose of 
134 lbs. of potash per acre. If this is insufficient, the following 
year we must be guided by the amount of crop obtained. 
On soils where the richness exceeds "2 per cent., the quantity 
of potash may be lessened, but unless the percentage is very 
high, say from "3 per cent, to '4 per cent., it is always well to 
give at least 89 lbs. to the acre, before sowing down the grass- 
land. 
Magnesia. — Soils deficient in magnesia are rare : still some 
do exist, and when analysis shows that the quantity of magnesia 
is less than from '2 per cent, to "3 per cent., it is necessary that 
some should be added to the manure. It has been seen that 
10 tons of dry hay contain 61'82 lbs. of magnesia on an average. 
So that it will be well on such soils to add 2 6" 76 lbs. per acre 
to the manure. 
Nitrogen. — I have previously demonstrated that the growth 
of grass enriches the soil in nitrogenous matters, and that this 
accumulation of nitrogen in the soil is often considerable. 
Should we then conclude that nitrogenous manures are useless on 
pasture-lands, or that it is possible to economise in the applica- 
tion of this element, which is the most costly of all those which 
enter into the composition of chemical manures? 
We have seen also that hay contains more nitrogen than the 
majority of other crops. It would therefore seem reasonable to 
infer that some of this nitrogen comes from the atmosphere, 
even in face of the theory, held by the majority of scientific men 
and agriculturists, that the soil is the only source from which 
vegetation gets its food. The latter opinion is hardly tenable 
any longer, in face of the facts which 1 have established ; 
but yet, on the other hand, experience shows that nitrogenous 
manures, — especially nitrates and salts of ammonia, — exercise 
a very favourable action on the growth of graminaceae on grass- 
lands. In the very elaborate work of Messrs. Lawes and Gilbert, 
recently published in the ' Journal ' of the Royal Agricultural 
Society, I find some valuable information on this point. 1 have 
taken from it the following figures, which are an average of 
agricultural experiments carried on for over 20 years, and carried 
out with the care, the competence, and the skill, of which these 
agricultural savants have given so many proofs. 
