220 EXPERIMENT STATION RECORD. 



of evaporation during the latter period, and inversely proportional to the amount 

 of precipitation and the relative humidity of the air after the heading stage had 

 been reached. 



Briefly stated, a good season for summer cereals is distinguished by a rela- 

 tively large amount of precipitation and a rather low temperature from the time 

 of germination to heading and a relatively large amount of heat and light with 

 smaller precipitation after the period of heading. The weight of the grain ap- 

 peared to be directly dependent upon the intensity of solar radiation and 

 invej-sely proportional to precipitation, humidity of the air, and cloudiness in 

 the period from flowering to ripening. In the case of barley, the yields were 

 directly dejiendent upon precipitation during the entire period of growth, espe- 

 cially prior to heading and during the j)receding fall and spring. A high tem- 

 perature during the period of growth lowered both the quantity and quality of 

 the yield. In the case of winter cereals the critical period appeared to be 

 during September and October, and sometimes during August. The quantity 

 and quality of winter cereals depended mainly on the precipitation during the 

 period from seeding to the beginning of winter. 



The rainfall of Nevis and Antigua, A. H. Kirby {West Indian BuL, 10 

 (1910), ^^o. 3, pp. 273-28-'f). — This is a study of rainfall statistics of these two 

 islands undertaken in view of a suggestion that the rainfall, particularly of 

 Nevis, is diminishing. The study furnished no evidence that the rainfall of 

 the islands is greatly diminishing, but it is shown that the precipitation has 

 been below normal during the last few years. 



The general conclusions reached are " that there are periods of years of 

 diminished rainfall, over large areas, that are intermittent in their occurrence. 

 There is no evidence to show whether these happen regularly or not. 



" For both islands, during a period of years in which the rainfall is deficient, 

 the times at which the precipitation may be relied upon to be nearest to the 

 normal amount occur at or near the equinoxes and at the end of the year. 



" On the contrary, during such a period, the mouths whose rainfall suffers 

 the greatest diminution are those which are most remote from the equinoxes, 

 with the exception of those that immediately precede the close of the year. 

 This, in effect, means that a period of diminished rainfall owes its existence 

 mainly to abnormally small precipitation during the beginning and middle of 

 the year. This does not preclude the possibility of assistance being given, in 

 attaining the general result, by the fact that the rainfall of the other months has 

 remained near the average for several years. 



"All the above considerations help to indicate that the conditions which regu- 

 late changes in the rainfall for periods of several years have a wide area of 

 operation." 



Investigations on the nitrogen content of atmospheric precipitation in 

 Flahult, Sweden, H. von Feilitzen and I. Lugner {Filhling's Landiv. Ztg., 59 

 (1910), No. 7, pp. 2^8-252). — Observations are reported which show that the 

 total amount of nitrogen brought down to the soil in precipitation (rain and 

 snow) at Flahult during the year 1909 was 5.18 kg. per hectare (4.61 lbs. per 

 acre). This agrees closely with data obtained at Rothamsted and other places 

 and shows tliat the amount of nitrogen available for plant growth from this 

 source is very small. 



Nitric acid and ammonia in the rainfall at Tonkin, Aufray (Bui. Econ. 

 Indo-Chine, n. ser., 12 (1909), No. 81, pp. 595-616).— Data from observations 

 covering the years 1902 to 1909 are reported, showing that duiiug this period 

 the nitric acid in the rain water varied from 15.48 to 70.37 kg. per hectare (13.78 

 to 62.63 lbs. per acre) annually. The ammonia varied from 5.4 to 18.5 kg. per 

 hectare (4.8 to 16.5 lbs. per acre) annually. 



