63

merits, mixed three-eighths of potters clay withthree-eighths of parings of limestone and two-eighths of fine sand; the only difference betweenthis mixture and that of the first experimentwas, that in the first experiment coarse sand wasused, and in this J?ne, yet the former was fruit-ful in the highest degree; but in this the grainprospered indeed the first year, but sickened inthe second, and failed in the third: the propor-tions have been already stated. Here we havea clear proof of the necessity of an open texturein soils, without which the best proportions areuseless.

SECOND.

In his thirteenth experiment he employed amixture of two-eighths potters clay, four-eighthscoarse sand, and two-eighths marl. The comgrew well the first year, poorly the second, anddecayed the third. The composition of themarl is not mentioned; but supposing it to con-tain 70 per cvvt. of calx, and 30 of clay, ofwhich the one-half is argill, it would form one