=153. Coefficient of Evaporation.=—At an ordinary room temperature in the shade, samples of soil, if they are subjected to experiment in tolerably thin layers have nearly an equal coefficient of evaporation. That is, the absolute quantity of water evaporated in a given time is almost entirely conditioned upon the magnitude of the surface exposed and the temperature of the surrounding air. Only when exposed in conditions as nearly as possible natural in thin layers to the action of the sunlight and shade do the soils show their peculiarities in respect of the evaporation of moisture. In order to see these peculiarities, samples of soil which have been previously examined must be subjected to examination at the same time with the soil whose properties are to be determined. The zinc box, before described, should be protected with a well fitting cover of thick paper, and the different samples of soil which are to be tested placed therein. This should now be placed in a wooden box, the top of which is exactly even with the top of the zinc vessel. This box containing the vessel should be exposed to the sunlight. After twenty-four hours the zinc boxes can be taken away from position and their loss in moisture determined, and these weighings, according to the condition of the atmosphere, can be continued from fourteen days to three weeks, the temperature of the air of course being carefully determined at each time. At first, all the different soils being saturated with moisture, it will be observed that the loss of moisture is proportionately the same for all. Soon, however, the rapidity of the evaporation in the samples of soil rich in humus and clay will be decreased as compared with the sandy soils, and in general, those which possess a high capillary power capable of bringing the moisture rapidly from the deeper layers to the surface. There soon comes a point when the difference in evaporation is at its greatest; and then there will be a gradual diminution until the samples lose no further moisture. This point, for the different soils, can be determined by frequent weighings of the vessel. =154. Determination of Capillary Attraction.=—Long glass tubes graduated in centimeters may be used for this determination, or plain tubes so arranged as to admit of easy measurements with a rule. The tubes may be from one to two centimeters internal diameter and about one meter long. The fine earth should be evenly filled in little by little, with gentle jolting. The lower end of each tube, before filling, is closed with a piece of linen. The tubes, after filling, are supported in an upright position by a frame AE, Fig. 19, in a vessel B containing water in which the linen covered ends D dip to the depth of two centimeters. The height of the water in the several tubes should be read or measured at stated intervals. The water contained in the supply vessel should be kept at a constant height by a Mariotte bottle. [Illustration: FIGURE 19.