438 



METEOROLOGY. 



May, in which he examined the weather charts for 

 the past twenty years, Dr. R. Hennig, of Berlin, 

 finds that these days are, with rare exceptions, a 

 yearly recurring phenomenon, but do not affect the 

 same parts of Europe ; that the period of their oc- 

 currence varies considerably. It may embrace the 

 whole month, but most frequently takes place dur- 

 ing the second decade, and usually lasts for three 

 or four days ; that the phenomenon generally begins 

 during stormy northwest winds, accompanied with 

 frequent showers of rain, snow, or hail. Night 

 fmsts and hoarfrost sometimes occur during the 

 f.-irly period of this unsettled weather, but generally 

 take" place after the passage of areas of low baro- 

 metric pressure. During this cold period an exten- 

 sive area of high pressure obtains over the ocean 

 adjacent to the western or northwestern shores of 

 Europe. The subject was investigated before by 

 Dove, in 1856, and Von Bezold, in 1882. 



r'roin the examination of the weather charts 

 isMied from the Meteorological Office of Paris, 

 particularly those from January to March, 1895. 

 the Rev. M. Decheirans, of the St. Louis Observa- 

 tory. Jersey, has found that the extremes of heat 

 and cold, observed respectively in areas of low and 

 high barometric pressures, do not occur at the cen- 

 ters of those systems, but in the neighborhood of 

 the mean isobars: also that the descending current 

 of air in an area of high pressure escapes along di- 

 vergent lines, and that it is principally due to this 

 divergence that the cold usual in anticyclones is 

 observed. Similarly, the relatively high tempera- 

 ture in areas of low pressure is due to the con- 

 vergence of ascending air currents. 



The bulletin of the Society of Naturalists of 

 Moscow contains an account of observations made 

 by M. Walther, of Jena, in the transcaspian coun- 

 try upon the temperature of the surface of the 

 ground. The temperature of the air was registered 

 on a day in September, 1897, every hour from five 

 o'clock in the morning till ten o'clock in the even- 

 ing, upon an isolated hill about 35 feet high ; and 

 the temperatures of the sand, the clayey soil, and 

 an olive-brown dolomite were taken by means of a 

 special thermometer Independently of the angle 

 of incidence of the rays of the sun, M. Walther 

 took note of a considerable influence of the wind. 

 The removal of the thermometer situated in the 

 wind to a sheltered spot gave a rise of temperature 

 of 6.5 C. for the sand, 4.5 0. for the clay, and 4.5 

 C. for the rock, the temperature of the air remain- 

 ing the same. The temperature of the air reached 

 its greatest height at two o'clock, when it was 33.5 

 C. : that of the clay soon rose to 46.5 C. ; and even a 

 little before two o'clock the temperature of the sand 

 was 4s., r ) C. The rock did not reach its highest 

 temperature till three o'clock. The setting of the 

 sun had considerable effect; and shortly after it 

 the temperature curve of the soil cut that of the 

 air and continued to be several degrees below it. 

 Kain had a still greater effect. Ahnger observed 

 in the desert on the same day that the temperature 

 11 after a heavy shower from 30 C. to 10" C. M. 

 Walther relies on sudden changes like these to ex- 

 plain the disintegration of recks and their breaking 

 uj> by parallel fissures. 



Precipitation. In a bulletin of the Weather 

 Bureau of the United States Department of Agri- 

 culture, on the rainfall of the United States, pre- 

 pared by Mr. A. .1. Henry, observations obtained 

 from the largest and most trustworthy registers are 

 compared, and annual, seasonal, and other charts 

 arc discussed. Special attention is 'given to the 

 rainfall of the crop-growing season as being a mat- 

 ter of most vital interest. As regards the monthly 

 distribution' of rainfall by districts and types ac- 

 cording to natural boundaries, the conditions favor- 



able for rainfall are found to be nearness to the 

 ocean, proximity to the track of storms, and the posi- 

 tion of mountain ranges. The rainfall of the North 

 , Pacific coast is discussed as an example of the com- 

 bined effect of all three of these conditions. There 

 a wet season prevails from October to March, and 

 the summer is nearly rainless, except in northern 

 California and parts of Oregon and Washington. 

 About half the yearly fall occurs between December 

 and February. In a narrow belt on the northwest 

 coast, extending from Cape Flattery to about the 

 middle of the Oregon coast, and some distance in- 

 land, the annual amount of rain is more than 50 

 inches, and at some points is 100 inches. To the east 

 and north of this, the annual rainfall diminishes, 

 and is least on the lowlands and valleys between 

 the Coast Range on the west and the Sierra Nevada 

 and the Cascade Range on the east, where the 

 amount is in some cases not more than 10 inches, 

 though in years of plentiful rainfall it may rise to 

 20 inches in the best-watered parts. East of the 

 Rocky mountains the annual rainfall ranges from 

 10 to 18 inches, and increases slowly to 00 inches 

 on the Florida and Gulf coasts, and to from 40 to 

 50 inches in the Eastern States. It, however, gradu- 

 ally declines from the Atlantic coast westward and 

 from the Gulf coast northward. The large majority 

 of excessive rains are said to occur west of longi- 

 tude 105 west, and principally in the summer 

 months, in connection with afternoon thunder- 

 storms. They occasionally take place on the track 

 of the West India hurricanes, and are more abun- 

 dant on the Gulf and South Atlantic coasts than at 

 inland places. The greatest rates of rainfall per 

 hour, estimated from periods of five minutes at the 

 Weather Bureau stations that possess self-register- 

 ing gauges, were 9 inches at Bismarck, N. Dak., 8.4 

 inches at St. Paul, Minn., and 8.2 inches at Ne\v 

 Orleans. 



A formation of small cumulus clouds over a. 

 fire has been observed by Mr. R. de C. Ward, at the 

 Harvard College Observatory, Arequipa, Peru, and 

 is .described by him in the " United States Weather 

 Review.'' Behind the western flank of Mount 

 Charchani, and about 15 miles away, a column of 

 smoke was rising from a considerable fire of bush- 

 wood, at a probable height of about 14,000 feet 

 above sea level. While looking at the smoke, Mr. 

 Ward noticed the formation of a small cumulus 

 cloud directly over it. and between 4,000 and 5,000 

 feet above it, when the sky was almost clear and 

 there was little wind. The cloud soon disappeared, 

 and was succeeded by another, which again disap- 

 peared within five minutes. Eight distinct cloud- 

 lets were seen thus to form and disappear within 

 half an hour, after which the smoke vanished. Two 

 previous similar cases are referred to by Mr. Ward 

 one recorded by Prof. Espy, and the other de- 

 scribed in " Science " of Jan. 8, 1897. 



The report of the International Meteorological 

 Committee on cloud observations records nearly 

 3,000 measurements of heights and velocities made 

 at Upsale Observatory. Sweden, during the year be- 

 ginning May, 1896, l'.635 of which were taken by 

 photography. The discussion of the results shows 

 that the annual deviation of the mean height of the 

 clouds is very pronounced, with a maximum during 

 June and July and a minimum during winter. 

 During the summer season the mean height of the 

 cirrus is 8,176 metres, and that of the cumulus U>s"> 

 metres. The heights of the upper and middle 

 level clouds are lower than at the Blue Hill Obser- 

 vatory in Massachusetts, while the lower forms are 

 at nearly the same level ; this is accounted for as 

 probably a natural effect of the difference of posi- 

 tion of the two stations. The velocity of the upper 

 clouds is greater than that of the lower, and the 





