METEOROLOGY. 475 



August, wo have the iiioro iutt'use aud relatively shorter droughts of the second 

 series, with heat, gales, and hurricanes; on the other hand, when they accumulate 

 aliout March and April, that is, the month of the equinox, and the one following, 

 aud about September, the mouth of equinox, and October following it, we have 

 droughts of the first series that are less severe, but much longer than the droughts 

 of the second series."' 



Incidentally the recorded dates of "red rain," whicli are evidences of 

 drought, were collected. xVll of the 09 instances found fitted into the 

 19-year cycle. 



Kite experiments at the Weather Bureau, C. F. Marvin ( U. 8. 

 Dipt. Ayr., Weather llnreau, Moiitlily Wcedher Ixeview, 31 {1S96), Xos.4, 

 pp. 113-123, Jigs. 26; '>, pp. 156-166, figs. 30; 6, pp. 199-206, figs. S ; 7, 

 pp. 238-255, figs. 16). — The history of the use of kites for the purpose 

 of studying" atmospheric conditions is briefly reviewed. It is shown 

 that probably the earliest attempts to use tandem kites for this j)uriiose 

 were made by Alexander Wilson, of Glasgow, in 1749. The first in the 

 United States to use the cellular kites constructed after the model of 

 Hargrave, of Sydney, Australia, were 0. TI. Lamson, J. B. Millet, and 

 S. A. Potter. 



A detailed account is given of the scientilic methods pursued by the 

 Weather IVureau in the construction and testing of kites of various 

 kinds. Data for tests of the strength of various kinds of wire and string 

 unknotted and knotted in various ways are tabulated and discussed. 

 A careful analysis is also given of the forces acting on kites — pull, drift, 

 resolution and combination of forces, wind pressure on plane surface, 

 center of pressure, edge pressures, resultant pressure, pressure on thin 

 curved surfaces, effect of waviness or fluttering, whirls or eddy etfects, 

 lateral stability, longitudinal stability, and conditions that modify the 

 angular elevation of the kite, and an elaborate description and discus- 

 sion is given of the forms and construction of various kites experi- 

 mented with by the Weather Bureau, giving mathematical formulas and 

 methods and results of observations on the behavior and efficiency of 

 the kites tested. 



Monthly Weather Revie^w ( U. 8. Ihpt. Agr., Weather Bureau, 

 Montldy Weather Review, 24 {1896), Nos. 4, pp. 105-144, charts 9 ; 5^ pp. 

 145-190, charts 12, figs. 3; 6, pp. 191-228, charts 6; 7, pp. 235-255, charts 

 8). — Besides the usual monthly summaries of observations and lists of 

 recent publications on meteorology, these numbers contain an illus- 

 trated article on kite experiments at the Weather Bureau, by C F. 

 Marvin (see above). In addition, Xo. 5 contains articles on the destruc- 

 tive forces of hurricanes and the conditions of safety and danger, by F. P. 

 Alexander; report on the tornadoes of May 1*5 in the State of Michigan, 

 by N. B. Conger, and notes by the editor on long-range forecasts, frosts 

 in California, total snowfall for the season 1895-90, Pontgeu rays and 

 cloudy condensation, and the tornano of May 25, 1890, in Cook County, 

 Illinois. No. contains notes by the editor on Mexican climatological 

 data, and on kites, balloons, and clouds; and Ko. 7, articles on kites in 



