120 SCIENCE OF AGRICULTURE. 



sieve, only, but the quantity or percentage in the 

 whole soil. How much in a given measure of land. 

 I would substitute for the first rule something like 

 the following. Take a measure of the soil, (and this 

 measure may be from a hall pint to a bushel, accord- 

 ing to the size of the stones, pebbles, &c. mixed 

 with the soil, throw it into a tub or other vessel of 

 cold water, stir it well and let it settle, skim off all 

 vegetable fibres, &c. that may swim, wash and take 

 out the larger stones; all that can be easily separat- 

 ed in this manner, and put them back into the mea- 

 sure. Stir the water and pour it through a sieve into 

 another vessel, wash the coarse sand and pebbles 

 remaining in the sieve until all the pulverized earth 

 adhering to it is added to that which before passed 

 through the sieve, let it settle, decant the clear wa- 

 ter, dry the sediment and proceed as directed in the 

 rule. Put all the contents of the sieve back with 

 the stones into the measure, and fill it up with water. 

 Pour off the water, measure it, and in this manner 

 ascertain what part of the measure is occupied by 

 stone, and make all subsequent calculation accord- 

 ingly. If the half bushel measure containing the 

 stones will hold exactly eight quarts of water, then 

 if your analysis of the fine parts give six per cent, 

 of geine, put it down in the table three per cent., 

 and so on. It would seem to me, also, well to save 

 the undecomposed vegetable fibres, etc. dry, weigh 

 and let their weight occupy another column in the 

 table, as it must at least one year add to the value 

 of the soil. Managing in this manner, we should 

 no longer be misled by tables, that shew the best 

 alluvial soils, less geine than most other varieties. 

 Nearly the whole of these soils pass through the 

 sieve Hence a cubic foot of this soil in situ may 

 contain more geine than a cubic foot of a rocky soil 

 in situ, which in the table shews a much higher per- 

 centage of the same constituent* 



