758 



SCIENCE. 



[Vol. II., No. 45. 



number of mistakes made, in each addition. 

 AH observations and suggestions relating to' 

 tlie subject will be most gratefuUj- received. 



Henry Farqdhae. 



Office of U. S. coaBt-anrvoy, 

 Washington, D.C. 



WHIRLWINDS, CYCLONES, AND TOR- 

 NADOES.'^— YI. 



Having seen how storms arise, and exam- 

 ined the general motions of their spiral winds, 

 we must next consider their progression from 

 place to place. It is now a familiar fact, that 

 storms do not remain atationarj', but advance 



forth. The apparently lawless winds of a 

 storm could be reduced to system if they were 

 supposed to blow around a centre which itself 

 has a progressive motion. In nearing the 

 centre, the barometer falls, and the winds in- 

 crease their strength. The manner and cause 

 of the progressive motion must now be ex- 

 amined. 



The four regions where tropical storms move 

 into temperate latitudes — the seas south and 

 east of India and China, and south-east of the 

 United States, in the northern hemisphere ; 

 and those east of Madagascar and (probably) 

 of Australia, in the southern hemisphere — are 



THE REGIONS OE TKOPTCAL CYCLONES. (TAEEN FROM BTIELEB'S ATLAS.) 



at a velocit}' of from five to fifty miles an hour 

 along a line known as their track. Although 

 perceived bj^ Franklin about 1750, this, as 

 well as their whirling motion, first found full 

 and satisfactorj' proof at the hands of Dove 

 of Berlin (1828), and Redfleld of New York 

 (1831). The latter gave the more numerous 

 examples, and was the first to explain the 

 motions of storm-winds at sea. The method 

 of his discoverj' was simple enough. Infor- 

 mation concerning the storm ' was gathered 

 from all attainable records, and the condition 

 of the winds and weather was plotted for cer- 

 tain hours. At once the result stood clearly 



^ ConUnued from No. 44. 



all crossed by storm-tracks, running first west- 

 ward near the equator, then turning toward the 

 pole, and passing around the apex of a para- 

 l3olic curve near latitude 00°, into an obliquely 

 eastward course. The more numerous storms 

 of temperate latitudes have less regular tracks, 

 but are nearly alwaj-s characterized b)' a strong 

 eastward element in their motion ; their chief 

 variations to the right or left being dependent 

 on thermal changes with the seasons, and on 

 the configuration of land and water which 

 they traverse. There have been four causes 

 suggested to determine the progression of the 

 storm-centre : namel}', the general winds of the 

 region, and especially the stronger and less 



