128 Composition of Waters of Land-Drainar/e and of Rain. 



In the 4th and 5th columns we have the percentage of filtration 

 and evaporation in the different months of the year, and we find 

 that in the three first and three last months in the year, the water 

 which is disposed of by filtration, greatly exceeds that which 

 escapes by evaporation ; whilst in the summer months, if we may 

 so call the six intermediate ones, the quantity of drainage-water 

 is reduced to a very low point, and in those of May, June, July, 

 and August, it is practically insignificant. Nor must this be 

 thought foreign to our subject, for, inasmuch as the decomposi- 

 tions which occur in the soil, are, in a great degree, modified by 

 the temperature and the degree of moisture ; and as these are 

 concerned in the liberation of elements of vegetation, which 

 might be supposed to be removed by drainage, it is impor- 

 tant to bear in mind that the flow of water through a soil is 

 not uniform in the different months of the year, but is in fact 

 very much greater at those seasons when the activity of vege- 

 tation and of decomposition in the soil is in great measure 

 suspended. 



For our present purposes we shall, therefore, assume that 42'4: 

 per cent, of all the water falling from the heavens filtrates 

 through the soil. It is obvious, that if we kn^w the rain- 

 fall of any locality in inches, it is easy to calculate approxi- 

 mately the number of gallons of water which, in the course 

 of twelve months, will drain from an acre of land. I say ap- 

 proximately, because it has already been seen that it is by 

 the distribution of the rain-fall, rather than the quantity 

 which falls, that the amount of drainage-water is regulated. 

 It is well indeed that we should, once for all, observe, that 

 however elaborately an examination of this question of the 

 composition of drainage-water might be carried out, the results 

 at the veiy best can only be general. It is practically impossible 

 that any collection of specimens should furnish the data for a 

 rigid determination of the truth. We cannot collect the whole 

 drainage of the year of any considerable portion of land, and as 

 we know from reasoning that its composition must be continually 

 varying, samples taken from time to time, however frequently, 

 cannot by possibility be supposed to I'epresent the whole year.* 



* It would be indeed possible, at a considerable expense, to accomplish this 

 object. A given area of surface-soil, lying on an ascertained clay-bottom, might 

 be isolated, by means of a puddled dyke, from the remaindet of the field, with 

 ■which, in other respects, it would altogether accord. A tank, sufficient to hold 

 the drainage of the interval of time elapsing between the collection of the samples, 

 might be constructed in such a May that tlie quantities might be accurately mea- 

 sured. It is obvious that in this way, by taking from time to time samples for 

 analysis, we might ascertain, with tolerable accuracy, the total quantity of various 

 substances removed from a given area of land within the twelvemonth. The 

 same object might be accomplished, though perhaps less satisfuctorilj-, by means 

 of a large Dalton gauge. 



