NOTEMBEB 1, 1907] 



SCIENCE 



599 



PHENOMENAL RAINFALL IN SUVA, FIJI 



The Quarterly Journal of the Royal Meteor- 

 ological Society for July, 1907, contains a 

 discussion of a phenomenal rainfall in Suva, 

 Fiji, August 8, 1906, which came during 

 a thunderstorm. Unfortunately, the exact 

 amount had to be, in part, estimated, owing 

 to the observer's having failed to measure 

 the fall at intervals during the night. The 

 measurements showed a fall of over 37 

 inches, without taking into account the over- 

 flow, which was an unknown quantity. The 

 gauge was twenty-five feet above the ground, 

 and the observer calculates that the total fall 

 must have been fully 41 inches in about 13 

 hours. Considerable uncertainty naturally at- 

 taches to this record, but there can be no doubt 

 that the rainfall was a very heavy one. 



RAINFALL IN THE LAKE REGION 



A STUDY of the average annual precipita- 

 tion in the Lake region, by Professor Alfred 

 J. Henry, of the United States Weather 

 Bureau, appears in the Meteorological Chart 

 of the Great Lakes, No. 1, 1907, and is illus- 

 trated by a chart. Measurements of rain and 

 snow have now been made for a period of 

 thirty-six years (1871-1906) at 21 stations. 

 The period 1871-1906 is taken as the funda- 

 mental period. The total number of stations 

 used was 107, all but 7 of which had more 

 than ten years' observations. The records of 

 ten years and over were generally reduced to 

 the fundamental period. The total annual 

 amount of rain and melted snow is about 31 

 inches. The increase in precipitation due to 

 the presence of the Great Lakes is probably 

 not more than 2 or 3 inches annually. 



VARIATIONS IN LEVEL OF LAKE CHAD 



The Scottish Geographical Magazine, Au- 

 ^st, 1907, summarizes the results of military 

 reconnaissances undertaken in 1906 by the 

 troops in the Lake Chad region, including 

 notes obtained from the natives in regard to 

 the changes of level of Lake Chad. There 

 leems to be a twenty-year periodicity, and at 

 the end of four or five twenty-year periods 

 there seems to come an almost complete desic- 

 cation, and then a great rise of level. An old 



native remembered a drying up which has 

 been placed between 1828 and 1833, while in 

 1851, about twenty years later, the level was 

 high. In 1906 the lake appears to have been 

 very low. 



ROUMANIAN METEOROLOGICAL WOEK 



A RECENT mail has brought renewed evi- 

 dence of the excellent work which the Meteor- 

 ological Institute of Eoumania is carrying on. 

 As lately reported in Science, M. Stefan C. 

 Hepites has retired from the directorshj.p, and 

 has been succeeded by M. I. St. Murat. Vol. 

 XVIII. of the Analele of the institute is a 

 publication of nearly 1,000 pages, 4to, contain- 

 ing, in French and Roumanian, the 17th re- 

 port of the work of the institute (for 1905-06) ; 

 a stiidy of the climatology of Craiova; 

 memoirs on rainfall, earthquakes and sun- 

 shine, and the usual climatological tables. 

 Separate brief reports concerning the hydro- 

 metric and agricultural conditions of January- 

 May, 1907, in Roumania, throw further light 

 on the activities of the Meteorological Insti- 

 tute. 



BRIGHT SUNSHINE IN THE BRITISH ISLES 



" The Distribution of Bright Sunshine over 

 the British Isles " is discussed by Richard H. 

 Curtis in Symons's Meteorological Magazine 

 for September, 1907, and is accompanied by a 

 chart showing the average annual duration of 

 sunshine. The records used in the prepara- 

 tion of this chart are those of the burning 

 recorders. A few records exceed twenty-five 

 years, and series for shorter periods are avail- 

 able for a large number of stations. The 

 short series have been weighted for the length 

 of period they cover. The number of hours 

 of bright sunshine is indicated by " isohels." 

 In 1891 the London Meteorological Office pub- 

 lished Dr. R. H. Scott's " Ten Years' Sunshine 

 in the British Isles." Dr. H. N. Dickson drew 

 the first sunshine map for the British Isles 

 from the data in that paper (Scot. Geogr. 

 Mag., 1893). The Atlas of Meteorology (pi. 

 18) reproduces Dr. Dickson's map. 



symons's meteorological magazine 

 The five-hundredth number of Symons's 

 Meteorological Magazine is that for Septem- 



