33 CALCAREOUS MANUUKS-l'KAC I'lCE. 



The products oi' the spaces A and B, in 1828, were not only estimated as 

 usual from the measurement of tlie corn in ears, (whicli estimated quanti- 

 ties are those in the table,) but they were also shelled on the day when 

 gathered, and the grain then measured, and again some months after, when 

 it had become thoroughly dry. Care was taken that there should be no 

 waste of the corn, or other cause of inaccuracy. The result showed nearly 

 double the loss from shrinking in the corn not marled, and of course a pro- 

 portional greater comparative increase of product in that marled, besides the 

 increase which appears from the early measurement exhibited in the table. 

 The grain of A, not marled, when first shelled, measured a very little more 

 than the quantity fixed by estimate — made as usual by measurement of 

 the ears, and lost by shrinking 30 per cent. The marled grain, from B, 

 measured at first above 4 per cent, more than the estimate, and after 

 shrinking, fell below it so much as to show the loss to be 16 per cent. 

 The loss from shrinking in this case was greater than usual in both, 

 from the poverty and consequent backwardness of the part not marled, 

 and the uncommonly large proportion of replanted and of course late 

 corn on the whole. 



The two last experiments, as well as the 4th, were especially designed to 

 test the amount of increased product to be obtained from marling, and to 

 show the regular addition to tlie first increase, which the theory promised 

 at each renewal of tillage. As to the main objects, all the three experi- 

 ments have proved failures — and from the same error, that of marling too 

 heavily. Although for this reason, the results have shown so much of the in- 

 jurious effects, still, taken altogether, the experiments prove, clearly, not 

 only the great immediate benefit of applying marl, but also its continued 

 and increasing good effects when applied in proper quantities. 



Experiment 1 2. 



On 9 acres of sandy loam, marled in 1819 at 400 bushels, (25 per cent.,) 

 nearly an acre was manured during the same summer, by penning cattle. 

 With the expectation of preserving the manure, double the quantity of 

 marl, or 800 bushels in all, was laid on that part. The field in corn in 

 1820-in wheat, 1821 -and at rest 1822 and 1823. 



Results, 1824. In corn, the second rotation after marling. The effects 

 of the dung have not much diminished, and that part shows no damage from 

 the quantity of marl, though the surrounding corn, marled only half as 

 thickly, gave signs of general, though very slight injury from that cause. 



Experiment 13. 



Nearly two acres of loamy sand were covered with barn-yard manure, 

 and marl, (45 per cent.,) at the same time, in the spring of 1822, and the field 

 put in corn the same year, followed by wheat. The quantity of marl not re- 

 membered— but it must have been heavy (say not less than six hundred 

 bushels to the acre) as it was put on to fix and retain the manure, and I 

 had then no fear of damage from heavy dressings. 



Result, 1825. Again in corn— and except on a small spot of sand almost 

 pure, (nearly a "blowing sand," or liable to be drifted by high winds in dry 

 weather,) no signs of disease from over-marling were seen, then or after- 

 awards. 



