Again, about three feet of water annually falls upon the land, while the evaporation from the land is but little more than 20 inches. If this water fell upon a flat surface of soil, with an impervious subsoil of rock or clay, we should have some sixteen inches of water in the course of the year more than evaporates from the land. If a given field be dish-shaped, so as to retain it all, it must become a pond, and so remain, except in Summer, when greater evaporation from a water surface may reduce it to a swamp or marsh.

With 16 or 18 inches more water falling annually on all our cultivated fields than goes off by evaporation, is it not wise to inquire by what process of Nature or art this vast surplus shall escape?

Experiments have been made with a view to determine the proportion of evaporation and filtration, upon well-drained land, in different months. From an able article in the N. Y. Agricultural Society for 1854, by George Geddes, we copy the following statement of valuable observations upon these points.

It will be observed that, in the different observations collected in this chapter, results are somewhat various. They have been brought together for comparison, and will be found sufficiently uniform for all practical purposes in the matter of drainage.

"The experiments upon evaporation and drainage, made on Mr. Dalton's plan, were in vessels three feet deep, filled with soil just in the condition to secure perfect freedom from excess of water, and the drainage was determined by the amount of water that passed out of the tube at the bottom. These experiments have been most perfectly made in England by Mr. John Dickinson. The following table exhibits the mean of eight years:

Year.October to March.April to September.Total each year.
Rain.FiltrationPer cent
filtered.
Rain.FiltrationPer cent
filtered.
Rain.FiltrationPer cent
filtered.
183618.8015.5582.712.202.1017.331.0017.6556.9
183711.30 6.8560.6 9.800.10 1.021.10 6.9532.9
183812.32 8.4568.810.810.12 1.223.13 8.5737.0
183913.8712.3188.217.412.6015.031.2814.9147.6
184011.76 8.1969.6 9.680.00 0.021.44 8.1938.2
184116.8414.1984.215.260.00 0.032.1014.1944.2
184214.2810.4673.212.151.3010.726.4311.7644.4
184312.43 7.1157.214.040.99 7.126.47 8.1036.0
Mean13.9510.3974.512.670.90 7.126.6111.2942.4

"A soil that holds no water for the use of plants below six inches, will suffer from drouth in ten days in June, July, or August. If the soil is in suitable condition to hold water to the depth of three feet, it would supply sufficient moisture for the whole months of June, July, and August.

"M. de la Hire has shown that, at Paris, a vessel, sixteen inches deep, filled with sand and loam, discharged water through the pipe at the bottom until the 'herbs' were somewhat grown, when the discharge ceased, and the rains were insufficient, and it was necessary to water them. The fall of water at Paris is stated, in this account, at twenty inches in the year, which is less than the average, and the experiment must have been made in a very dry season; but the important point proved by it is, that the plants, when grown up, draw largely from the ground, and thereby much increase the evaporation from a given surface of earth. The result of the experiment is entirely in accordance with what would have been expected by a person conversant with the laws of vegetation.

"The mean of each month for the eight years is:

Months.Rain.Filtration.Per cent
filtered.
Inches.Inches.
January 1.841.30 70.7
February 1.791.54 78.4
March 1.611.08 66.6
April 1.450.30 21.0
May 1.850.11 5.8
June 2.210.04 1.7
July 2.280.04 1.8
August 2.420.03 1.4
September2.640.37 13.9
October 2.821.40 49.5
November 3.833.26 84.9
December 1.641.80110.0