1920] METEOROLOGY. 417 



extent upon official statistics of food prices, the author endeavors to estahlish 

 a weather period of 15J years during the past three centuries. He predicts 

 unseasonable weather, bad harvests, and hifjli prices, with possible famines, 

 in one or more of the years 1024, 1025, and 1926. 



Lunar and rainfall periods, E. RFesnard (Compt. Rend. Acad. l?ci. [Paris], 

 no (1020), No. Jf, pp. 2 42-2^ .5 ).^'rhi!i article briefly reviews evidence tending 

 to establish a certain correlation between meteorological phenomena, par- 

 ticularly rainfall, and lunar phases and motions. 



[Influence of altitude and topography on temperature distribution in 

 southern California], F. D. Youxg (Cal. Citrogr., 5 {1920), Nos. 5, pp. 136, 

 160, 161, figs. 3; 7, pp. 215, 231, figs. //).— Studies near Pomona, Cal., on tem- 

 peratui'e distribution on a clear, calm night, over relatively level ground, on 

 steep hillsides, at the bases of steep hills, and at different heights above the 

 ground, are reported. 



" It was noted that the temperature at some distance above the ground 

 does not change much during such a night, while the air near the ground 

 cools steadily, causing a considerable difference in temperature to develop be- 

 tween the air strata at the two levels before morning. The surface air, cool- 

 ing steadily all night through conduction of its heat into the colder ground 

 (which in turn is losing heat to the sky through radiation) continues to grow 

 relatively denser and denser than the air at greater elevations. The greater 

 density of this colder surface air operates to keep it in contact with the 

 ground. Over sloping ground the force of gravity tends to cause this thin 

 surface layer of cold air to move down the slope and te^ gather in depressions 

 in somewhat the same manner as water. The flow of air and of water down 

 the same slope differ, ho\^ever, in that air is heated by compression as it moves 

 to lower levels while water is not, 



" From observations in the Pomona Valley it appears that there is little, 

 if any, advantage to be gained by locating on the upper portion of a long, 

 uniform slope with a fall of 150 ft. or less to the mile. In an orchard located 

 even a short distance abwe the base of such a slope, the frost hazard is not 

 likely to be greater than in a similar orchard near the summit of the slope. 

 However, in even slight depressions of whatever shape or direction, on this 

 slope, the frost hazard is likely to be considerably greater. The same is true 

 of groves located directly at the base of the slope, where the ground begins 

 to rise in the opposite direction." 



Elevation above sea level was not an important factor under the conditions 

 of these studies. " The rate of increase in temperature with altitude is likely 

 to be greater after a warm day than after a cold day and to be greater on a 

 night with low humidity than on a night when there is a great deal of mois- 

 ture in the air. 



" An interesting fact brought out in the observations on the steep hillside 

 is that the highest average minimum temperature was found at the 225-ft. 

 station, 50 ft. below the summit of the ridge; the minimum temperature at 

 the .summit was as high as that at the 225-ft. station on only one night 

 during the sea.son. The summit of the ridge is well rounded and the ground 

 sloped away from the summit station in all directions. The steep slope began 

 about 20 ft. from the thermometer shelter. As the same phenomenon has been 

 observed by the writer on hills of varying height in other sections, it is prob- 

 ably safe to say that the frost hazard will be greater on a well-rounded or flat 

 hilltop or ridge than for some distance below on the hillside, if the slope is 

 fairly steep." 



The effect of snow upon the growth of winter w^heat, C. L. Meisingkk 

 {Science, n. ser., 51 {1920), No. 1330, pp. 639, 640).— This is a brief review of 



