cubic centimeter of the atmosphere is the amount corresponding to saturation. Then, if the temperature still continues to fall, the vapor must begin to condense. Whether it condenses as dew or cloud or fog or rain will depend upon how and where the cooling takes place.

208. The formation of dew and frost. If the cooling is due to the natural radiation of heat from the earth at night after the sun's warmth is withdrawn, the atmosphere itself does not fall in temperature nearly so rapidly as do solid objects on the earth, such as blades of grass, trees, stones, etc. The layers of air which come into immediate contact with these cooled bodies are themselves cooled, and as they thus reach a temperature at which the amount of moisture which they already contain is in a saturated condition, they begin to deposit this moisture, in the form of dew or frost, upon the cold objects. The drops of moisture which collect on an ice pitcher in summer illustrate perfectly the formation of dew. If condensation takes place upon a surface colder than the freezing temperature, frost is formed, as is observed, for instance, on grass and on window panes.

209. The formation of fog. If the cooling at night is so great as not only to bring the grass and trees below the temperature at which the vapor in the air in contact with them is in a state of saturation, but also to lower the whole body of air near the earth below this temperature, then the condensation takes place not only on the solid objects but also on dust particles suspended in the atmosphere. This constitutes a fog.

210. The formation of clouds, rain, sleet, hail, and snow. When a warm moist current of air near the surface of the earth rises high into a region of less pressure, it undergoes a marked lowering in temperature on account of its expansion (see § 184, last lines). If the resultant temperature due to the expansion is below that at which the amount of moisture already in the air is sufficient to produce saturation, this excessive moisture immediately condenses about floating dust particles and forms a cloud. If the cooling is sufficient

to free a considerable amount of moisture, the drops become large and fall as rain. If this falling rain freezes before it reaches the ground, it is called sleet. If the temperature at which condensation begins is below freezing, the condensing moisture forms into snowflakes. When the violent air currents which accompany thunderstorms carry the condensed moisture up and down several times through alternate regions of snow and rain, hailstones are formed.

211. The dew point. The temperature to which the atmosphere must be cooled in order that condensation of the water vapor within it may begin is called the dew point. This temperature may be found by partly filling with water a brightly polished vessel of 200 or 300 cubic centimeters capacity and dropping into it little pieces of ice, stirring thoroughly at the same time with a thermometer. The dew point is the temperature indicated by the thermometer at the instant a film of moisture appears upon the polished surface. In winter the dew point is usually below freezing, and it will therefore be necessary to add salt to the ice and water in order to make the film appear. The experiment may be performed equally well by bubbling a current of air through ether contained in a polished tube (Fig. 168).

FIG. 168. Apparatus for determining dew point

212. Humidity of the atmosphere. From the dew point and table given in § 204, p. 183, we can easily find what is commonly known as the relative humidity or the degree of saturation of the atmosphere. Relative humidity is defined as the ratio between the amount of moisture per cubic centimeter actually present in the air and the amount which would be present if the air were completely saturated. This is precisely the same as the ratio between the pressure which the water vapor present in the air exerts and the pressure which it would exert if it were present in sufficient quantity to be in the saturated

condition. An example will make clear the method of finding the relative humidity.

Suppose that the dew point were found to be 15° C. on a day on which the temperature of the room was 25° C. The amount of moisture actually present in the air then saturates it at 15° C. We see from the P column in the table that the pressure of saturated vapor at 15° C. is 12.7 millimeters. This is, then, the pressure exerted by the vapor in the air at the time of our experiment. Running down the table, we see that the amount of moisture required to produce saturation at the temperature of the room, that is, at 25°, would exert a pressure of 23.5 millimeters. Hence at the time of the experiment the air contains 12.7/23.5, or .54, as much water vapor as it might hold. We say, therefore, that the air is 54 per cent saturated, or that the relative humidity is 54 per cent.

213. Practical value of humidity determinations. From humidity determinations it is possible to get much information regarding the likelihood of rain or frost. Such observations are continually made for this purpose at all meteorological stations. They are also made in greenhouses, to see that the air does not become too dry for the health of the plants, and in hospitals and public buildings and even in private dwellings, in order to insure the maintenance of hygienic living conditions. For the most healthful conditions the relative humidity should be kept at from 50 per cent to 60 per cent.

Low relative humidity in the home causes discomfort and colds, and leads to waste of fuel estimated at from 10 per cent to 25 per cent. The average home heated to 72° F. by steam or hot water is estimated by health authorities to have a relative humidity of 30 per cent, and even as little as 25 per cent with hot-air heating. This is less than the average humidity of extensive desert regions. Higher humidity in the home would diminish the cooling effect due to rapid evaporation of the perspiration from the body, and would make us feel comfortable if a lower temperature were maintained (see § 214).

Born in Albany, New York; taught physics and mathematics in Albany Academy and Princeton College; invented the electromagnet (1828), discovered the oscillatory nature of the electric spark (1842) by magnetizing needles in the manner described on page 263, and made the first experiments in self-induction (1832); was the first secretary of the Smithsonian Institute and the organizer of the Climatological Service, a branch of the Weather Bureau which collects the weather observations of nearly five thousand unpaid coöperative observers located in all sections of the United States