Jri-T 17, 1903.] 



SCIENCE. 



91 



jury to his health" (Nat. Geogr. Mag., May, 

 p. 203). " Great province. This is only 

 150 miles from Manila, with air as bracing 

 as Adirondacks or Murray Bay. Only pines 

 and grass lands. Temperature this hottest 

 month in the Philippines in my cottage porch 

 at three in the afternoon, 68° F. Fires are 

 necessary night and morning" (Cablegram 

 from Governor Taft to the Secretary of War, 

 dated April 1.5). These are two recently 

 published statements regarding the climate 

 of the highlands of Benguet, in the northern 

 part of the island of Luzon. Such broad 

 general statements are misleading. They tend 

 to spread false notions regarding the pos- 

 sibility of the acclimatization of the white 

 race in the Philippines, and of outdoor work 

 by white men in a tropical climate. Altitude, 

 as in the highlands of Benguet, or of India, 

 gives some relief in the way of lower tem- 

 perature than at sea level. It means the ab- 

 sence of some tropical diseases which prevail 

 on the lowlands, or a more rapid recovery 

 from these diseases than at sea level. But 

 all experience shows that altitude does not 

 solve the acclimatization problem. A tropical 

 sun is always a tropical sun. A tropical 

 climate is always a tropical climate. It 

 should be the aim of all Americans who send 

 lis accounts of Philippine climates, avoiding 

 generalities based on first impressions, care- 

 fully to study the effects of the climate upon 

 white men. The experience of English, 

 French, Germans and others in the tropics 

 furnishes evidence enough of the inaccuracy 

 of much that has been written of the climate 

 of the Philippines. 



THE RECENT FLOODS. 



The disastrous floods of March, April and 

 June on the Mississippi and Missouri Rivers 

 naturally attracted much attention in the 

 daily papers. Along the lower Mississippi 

 River the March-April flood of the present 

 year was the greatest on record if stages of 

 water alone are considered, although the act- 

 ual volume of water was probably less than in 

 the flood of 1897, the greater heights in 1903 

 being the result of the extension of the levee 



system. The rise of the Ohio and lower Mis- 

 sissippi Rivers was steady during February, 

 owing to several heavy general rains, and dur- 

 ing the last two days of the month another 

 storm moved northeastward through the Ohio 

 valley, making it certain that floods would 

 occur. Other heavy rains occurred on March 

 7 and 8. The flood warnings of the Weather 

 Bureau were timely and accurate. The stages 

 which were forecasted, and those which were 

 actually recorded, between Cairo and New 

 Orleans are shown in the following table, 

 taken from an article by H. C. Frankenfield 

 on ' The Weather Bureau and the Recent 

 Floods ' (Nat. Geogr. Mag., July, 1903). 



c. „•;„«<, Forecast Actiml 



Stations. stage (ft). Stage (ft.). 



Cairo 50.5 to 51 50.6 



Memphis 40.0 40.1 



Helena 51.0 51.0 



Arkansas City. 53.0 53.0 



Greenville 49.0 49.1 



Vicksburg 52.0 51.8 



Xew Orleans... 21.0 20.4 to 20.7 



At Cairo the forecast was four days in ad- 

 vance, and at New Orleans twenty-eight days 

 in advance of the crest. 



At the end of May and early in June the 

 floods on the lower Missouri and the upper 

 Mississippi were greater than any on record 

 except that of 1844. They resulted from 

 hea^-y daily rainfalls over Kansas, coming in 

 connection with persistent low pressures over 

 the eastern slope of the Rocky Mountains. 

 Similar conditions prevailed eastward into 

 northwestern Missouri and Iowa. At St. 

 Louis warnings were issued on June 5 of a 

 stage of thirty-eight feet in about four days. 

 That stage was exactly reached on the fifth 

 day. 



RAINFALL AND SUNSPOTS. 



Dr. W. J. S. LocKYER continues his investi- 

 gation of rainfall and sunspot cycles (Nature, 

 Vol. 68, pp. 8-10). Smoothed rainfall curves 

 for the British Isles, Brussels, Madras, Bom- 

 bay, Cape Town and the Upper Ohio valley 

 show a long-period variation at all the sta- 

 tions, and further, the occurrence of the great- 

 est rainfall generally in the years 1815, 1845 



