top of the board had gained fourteen grains in weight from the moisture, or dew, that had formed upon it, while the same amount of cotton on the under side of the board had only increased four grains. He tried further experiments by making little paper houses, or boxes, to cover a certain portion of grass or vegetation. He found that while there would be a heavy dew on the grass outside there was little or none within the inclosure. These experiments were conducted in various ways and closely watched to see that none of the phenomena were in any way connected with falling rain. It has been determined that substances like grass and green leaves of all kinds, hay and straw, while they are poor conductors of heat, are excellent radiators. In another chapter we have referred to this quality of straw, that is taken advantage of by the inhabitants of hot countries in the manufacture of ice and in our own land for storing it. Perhaps everyone who has lived in the country has noticed that on a summer's morning when the grass is laden with dewdrops a gravel walk or a dusty road will be perfectly dry. This is due to the fact that the gravel will retain heat and not radiate it, for a much longer time than grass or green leaves. Dew begins to form upon the grass very soon after the sun is set because the moment the sun's rays are withdrawn the heat is rapidly radiated by the blades of grass, which cools the earth under it and the air above and surrounding it, so that if the air is anywhere near the moisture saturation point on cooling at the surface of the ground it will readily give up a part of its moisture, which condenses in drops upon the blades of grass. If the night is still and clear and there is much moisture in the air, the dew will be heavy, but if the night is cloudy there will be little or no dew formed. The clouds form a screen between the earth and the upper regions of the atmosphere, which prevents the heat from radiating to a sufficient extent to form dew. For the same reason no dew will form under a light covering spread over the ground even at some distance above it. The covering acts as a screen, which prevents the heat from radiating to the dew point. From what has gone before it will be seen that if the atmosphere is not charged with moisture up to the point of saturation it will require a greater amount of depression of temperature to cause condensation, and this is why we usually have heavier dews in June when the air is more highly charged with moisture than we do in August when it is dry. This also accounts for the ice clouds, called cirrus, being formed so high up in the atmosphere during dry weather. There is so little moisture in the air that it requires a very great difference of temperature to cause condensation to take place, and the necessary depression is not reached in these cases except at an altitude of several miles.