1920] SOILS FERTILIZERS. 319 



the yp.ir tho iiiiiiiimini (ciiipciMturcs are from 8 to 10° liijilicr on the mesas than 

 in tiio valley, and the )iiaximum temperatures are ordinarily 1 to 2° higher. 

 . . . The total precipilation for the year at the Yuma Weather Bureau sta- 

 tion was only 2.9 in. [as compared with 3.r>7 in. at the experiment farm], 

 which indicates the very local nature of the small rains in this rej^ion." 



k Method of constiiuting a diagram to show probaI)le drouiihts, \'. W 

 TcHiKOFF (Engiii. Ncws-Rec, 83 {1019), No. 12, p. 55J,, fig. 1).—A table pre- 

 jpared from meabnirements of rainfall during 10-day periods from April to Octo- 

 ber, inclusive, for 32 years (1883 to 1914) at Kherson, Russia, is piven, which 

 Bhows the computed percentaj,'e probability of drought, that is, 10-day pericjds 

 with not more than 5 mm. rainfall (termed dry), or not more than 10 nrm. 

 (termed unsatisfactory). A diagram is also given which is constructed from 

 this table I)y letting " the abscissas represent the 10-day periods and the ordi- 

 nates the number of cases with precipitation of less tlian 5 and 10 mm. The 

 line parallel to the abscissa, through the points of 30 and 70 per cent, limits 



Pthe dry and nondry periods. The line through the point of 50 per cent indicates 

 the equal probability of dryness and nondryness." Dry periods are represented 

 by shaded portions of the diagram. 



The utilization of cirrus ciouds in the forecasting of weather, (J. IIeboul 

 and L. Duxoykb {Comi>t. Rend. Acad. Sci. [Parl'i], 110 {1920), No. 12, pp. 7U- 

 747; ah.s. in Rev. 8ci. [Paris], 58 {1920), No. 7, p. 219).— From a study of ob- 

 servations on cirrus clouds the authors conclude that in 70 per cent of the 

 cases the appearance of such clouds betokens the approach of bad weather. 



Tropical control of Australian rainfall, E. T. Quayle {Commonircalth 

 Bur. Met. Aust. liul. l.'>, pp. 2.',, pis. 12; rev. in Nature [London], 105 {1920), No. 

 2631, pp. 152, 153). — This bulletin deals with " tlie influence of tropical condi 

 tions upon the development and rainfall production of storms in the temperate 

 belt, and the increased range and accuracy of Australian rain forecasts made 

 possible by taking this into account." The main feature of the author's theory 

 is that high minimum temperatures in the tropical regions of Australia, which 

 usually pei'sist for a few weeks at a time, are due to cloudiness, extra liumidity, 

 or northeast winds, and that they cause " such a flow of air to the southern 

 parts of the continent that the approaching cyclonic 'lows' are compelled to 

 part with rain." 



Tlie value of Aveather records in fumigation, K. S. Woglum {Cat. Citrogr., 

 5 {1920), No. 4, pp. 110, 111, figs. 2).— This article indicat(>s briefly the various 

 ways in which the efliciency of fumigation is influenced by temperature, 

 humidity, pressure, wind, and the state of the sky, points out tlie need of 

 weather observations with this purpose in view, and describes the necessary 

 instrumental equipment for the use of practical oi'chardists. 



SOILS— FERTILIZERS. 



Soil studios (North Dakota Sta. Bill. 1.16 {1920), p. ^N).— Studies conducted 

 since 190(» on the relation of total and water-soluble magnesium for soils sam- 

 pled at varying depths, in connection with soil fertility studies, and also to 

 compare the acidity of soils continuously cropped in wheat witli tliat of virgin 

 soils or .soils under rotation, showed that \uiiformity for all soils was found 

 for nitn)gen at a depth of from 18 to 24 in., for phosphorus at from 20 to 30 in., 

 magnesium at from 6 to 12 in., and for calcium and carbon dioxid at a depth of 

 from 30 to 30 in. 



"Excepting in the virgin soil series the total magnesium is in excess of the 

 total calcium to a depth of between 18 and 30 in., and water-soluble magnesiinn 

 is greater than the water-soluble calcium in tlie wheat soils and practically the 



