LECTURES AND ESS1YS. 409 



These are the largest known yields ever attained for so long a period by 

 any system of farming, and moreover appear to be the limit of yield for 

 that soil and locality, as no other combination produced equal results. 

 !r Note, first, the comparative uselessness (for wheat) of mineral manures 

 alone; second, that successive additions of 200 pounds each of the effective 

 manure (ammonia salts) gave for 12 years a descending scale of increase of 

 ten, eight, one and five-eighths and one-half bushels; and for thirty-two 

 years, of eight and seven-eighths, eight and five-eighths and three and one- 

 half bushels; third, that these differences were less marked for the entire 

 period, as shown by the lower line, but that the same general principle held, 

 as through the first 12 years, of a decreasing effect upon the yield of each 

 successive addition of manure. 



(The fourth 200 pounds, from its slight influence, was discontinued for 

 the last twenty years.) 



Clearly, then, successive additions of fertilizers are not answered by corre- 

 sponding additions to yield, although through a liberal range the difference 

 is inconsiderable. 



Much speculation has been indulged in as to comparative effects of varying 

 amounts of fertilizer, which it seems these results should help to set at rest. 

 The well known fact that residues of ammonia salts accumulate but slightly 

 in the soil, and of nitrates not at all, taken in connection with this law of 

 action of fertilizers, would teach the use of commercial fertilizers in only 

 moderate amounts. 



Mr. J. Valentine of England, the great champion of commercial fertilizer, 

 places the profitable limit at 18s ($4.37) per acre. 



A correlative truth to which I desire to call attention, and on which some 

 erroneous views are held, is this : The limit of profitable increase, and the 

 line of greatest profit fluctuate with the market price, an increased yield alone 

 is not the remedy for low prices. 



This record of twenty years proves clearly that the last bushels of increase 

 are secured at greater cost that the first, and that excessive yields are pro- 

 duced only at inordinate expense, and waste of fertilizer. This being true 

 the highest yields mentioned could have been produced at a profit only when 

 prices ruled exceedingly high. 



The last one-half bushel of increase was secured at the same expense of 

 fertilizer (200 lbs.), as was the first ten bushels, and at present prices of fer- 

 tilizer this last half bushel cost $7.50 — enough to absorb the profit from 

 many, if not all below it. Whatever the prices of wheat or of manure, the 

 facts seem to be that the last one-half bushel of possible increase costs as 

 much as did the first ten bushels; that for high yields the last bushels are 

 the most costly, and that yields that are profitable with wheat at $2.00 per 

 bushel are only secured at a loss when it sinks to $1.00 per bushel. 

 i The following diagram illustrates this relation between cost and income 

 applied to the yields represented by the upper line of diagram No. 1; that 

 is, the average yields for the past twelve years of the first twenty years of 

 Lawes and Gilbert's experiment. 



Assuming the cost of an acre of unmanured wheat to be $10 — nearly the 

 reported cost in Michigan — and adding to this the market price of the ferti- 

 lizer applied gives the line of cost, shown in the diagram by the dotted line, 

 K M. 4 The other lines show the income per acre of the several plats at dif- 

 ferent prices'per bushel. 



