one of the richest forts of marls. The cente fimal proportions of this mixture should be, And in the troy pound, fuppofing the water, &c. to amount to 100 grains, the quantities The fterility of this mixture feems to proceed from a defect of calcareous earth. If we fuppofe the marl poorer in that earth, the defect will be ftill greater. The retentive powers. of the different earths with refpect to water, being expreffed by the quantities which each can retain without fuffering any to drop, as above faid, and the quantities retained by the mixed mass of these earths being proportional to the respective quantities of each, it should feem that in fertile foils, where the fall of rain is from 20 to 30 inches, this power should not exceed 70, nor fall fhort of 50 per cent. It were of great confequence to fettle this point with precision; but to do this would require more numerous experiments. To explain my meaning I fhall give one example. Of the retentive Power of the Fertile Soils, mentioned by Mr. Bergman. This foil contains, as we have already feen, The conftitution of the Irish fertile foils has not been ascertained, nor has the average annual quantity of rain been determined here. Indeed the folution of the queftion propofed by the Academy, does not strictly require it fhould, not having been limited to any particular country: but I fhould fuppofe its beft foil to approach to the nature of that of Upfal, the fall of rain being probably between 24 and 28 inches. In 1792, which was reckoned remarkably wet, it was 30 inches in Dublin. Before I quit the experiments of Mr. Tillet, it will be proper to mention a few made by him, which feem to invalidate the neceffity of the prefence of the three fimple earths in fertile foils. Imo. In his 26th experiment he tells us he employed only pure fand, fuch as is used for making glass, yet corn grew well in it the first year, indifferently the second, and nearly failed in the third. Mr. Haffenfraz having repeated the experiment in pots unperforated, did not find it to fucceed even the first year, therefore the fuccefs of Mr. Tillet was owing to the perforation at the bottom of his pot, through which water, water, impregnated with the different earths, and coal must have paffed. In fact, Mr. Tillet's conclufion is contradicted by univerfal experience. 2. In his 28th experiment, in which powdered limestone only was employed, the corn fown profpered exceedingly during the three years. To the cause mentioned, in treating of the 26th, I must add, that the limestone he used was that of St. Leu, which contains clay, and confequently filex and argill; it is fo porous as to admit from 3-19ths to 1-5th of its weight of water, as Mr. Briffon has fhewn ; and thus is eafily decompofed. The coarse powder to which it was reduced anfwered the fame purpose as coarfe filex; and the finer might nourish the plants. 3. In his 30th experiment he employed mere potters clay; the grain grew tolerably well the first year, but perifhed the fecond; on the third it flourished moft. It is hard to draw any specific conclufion from this experiment, for it is plain that if the texture were not much loofer than that of clay, the corn could not grow at all, as was the cafe in his 6th and 8th experiments, F 2 riments, already mentioned, and as Mr. Haffenfraz, who repeated this experiment, obferved. Rain-water might however supply a fmall quantity of calx fufficient for a small produce of corn. I pass over his experiments on old mortar, as the three earths were evidently contained in it, though in unknown proportions. Soils on the declivity of hills, ought to be more retentive of water than those on plains, as is evident. CHAP III. TO DETERMINE THE COMPOSITION OF A SOIL. Imo. IN dry weather, when the foil is not over moist nor dry, let a furface of 16 square inches be cut through to the depth of 8 inches; this may be effected by a right angled spade, formed for this particular purpose. Of the parallelopiped |