Another method employed for determining whether the conditions found in nature were favorable for dew rising from the ground on dewy nights was by observations of the temperatures indicated by two thermometers, one placed on the surface of the grass and the other under the surface, among the stems, but on the top of the soil.  The difference in the readings of these two thermometers on dewy nights was found to be very considerable.  From 10 deg. to 18 deg.  F. was frequently observed.  A minimum thermometer placed on, and another under, the grass showed that during the whole night a considerable difference was always maintained.  As a result of this difference of temperature, it is evident that vapor will rise from the hotter soil underneath into the colder air above, and some of it will be trapped by coming into contact with the cold grass.

While the experiments were being conducted on grass land, parallel observations were made on bare soil.  Over soil the inverted traps collected more dew inside them than those over grass.  A small area of soil was spread over a shallow pan, and after being weighed was exposed at the place where the soil had been taken out, to see if bare soil as well as grass lost weight during dewy nights.  The result was that on all nights on which the tests were made the soil lost weight, and lost very nearly the same amount as the grass-land.

Another method employed of testing whether vapor is rising from bare soil, or is being condensed upon it, consisted in placing on the soil, and in good contact with it, small pieces of black mirror, or any substance having a surface that shows dewing easily.  In this way a small area of the surface of the earth is converted into a hygroscope, and these test surfaces tell us whether the ground is cooled to the dew-point or not.  So long as they remain clear and undewed, the surface of the soil is hotter than the dew-point, and vapor is being given off, while if they get dewed, the soil will also be condensing vapor.  On all nights observed, these test-surfaces kept clear, and showed the soil to be always giving off vapor.

All these different methods of testing point to the conclusion that during dewy nights, in this climate, vapor is constantly being given off from grass land, and almost always from bare soil; that the tide of vapor almost always sets outward from the earth and but rarely ebbs, save after being condensed to cloud and rain, or on those rarer occasions on which, after the earth has got greatly cooled, a warm moist air blows over it.  The results of the experiments are given, showing, from weighings, the amount of vapor lost by the soil at night, and also the heat lost by the surface soil.