In other words frost in the ordinary meaning of the word is a problem IN LOCAL AIR DRAINAGE. It is true that there are times when with thorough ventilation and mixing of the air strata the temperature will fall rapidly and damage from frost result; but such conditions are perhaps more fittingly described as cold waves or freezes, as distinguished from frosts. Thus, in California during the first week of January, 1913, when there was much air movement, the citrus fruit crop was damaged to the extent of $20,000,000. The condition is generally referred to as a frost, but it was quite different from the usual frost conditions in that section. It is, however, interesting to note that improved frost-fighting devices were used with much success and the total savings aggregated about $25,000,000. The orange growers also had the benefit of accurate forecasts and expert advice and were thus able to provide fuel and labor in advance. Passing over at present the larger disturbances, we shall consider only the frosts of still nights. And it should not be forgotten that the accumulated losses of these frosts may equal the losses of the individual freezes, for the latter occur at long intervals, while the quiet frosts of the early fall and the late spring are recurrent, destroying flowers, fruits and tender vegetation in many sections, year after year. Air may flow in any direction, but attention has been centered more upon the flow in a horizontal than in a vertical direction. Thus none of the wind instruments used at Weather Bureau stations gives any record of the up and down movement of the air. In frosts of the usual type this vertical displacement is all-important. True, there may be brought into the district, by horizontal displacement, large masses of cold air and the temperature thus materially lowered; but the marked INVERSION of temperature occurs only when these horizontal currents or winds are lulled. On windy nights, as is well known, there is less likelihood of frost than on quiet nights, because of the thorough mixing of the air vertically. There is then no tendency for stratification and the formation of levels of different temperature, followed by low surface temperature. In general, the temperature falls as one rises in the air; but, at times of frost, it is found that the higher levels are warmer than the lower ones. The coldest stratum is found about ten centimeters (four inches) above the ground; while at a distance of ten meters temperatures are as much as five degrees higher than at the ground.