Chase. 1 370 [April. 



limited area, there is a well-known cyclonic tendency, the portion of 

 the eddy which is nearest the equator, (jcni'mlly flowing eastward. 

 Mr. Galton* has ingeniously shown that in descending cyclones, the 

 direction may be reversed, and I should expect a similar reversal to 

 be of frequent occurrence in the neighborhood of some of the power- 

 ful ocean-currents, at points where they tend to produce backward 

 eddies. Such points are found midway between the Sandwich Is- 

 lands and California, about 3.5° west of Chili, near the west coast of 

 New Holland, in the Indian Ocean, northeast of Madagascar, and in 

 other places. 



The efi"ect of ocean-currents in producing cyclones, and directing 

 their course, is well illustrated by the repeated observations that have 

 been made in the Gulf Stream. Prof. Lesley's interesting account of 

 the series of storms encountered by the Canada on her one hundredth 

 voyage,"!" exhibits the natural consequences of the friction of two belts 

 of air at different temperatures, moving in opposite directions. The 

 warm air over the Gulf Stream, and the cold air over the Arctic cur- 

 rents that flow nearer to the American continent, are both borne 

 very nearly in their normal directions, but with the approach of winter 

 their parallelism becomes almost vertical, the cold belt becomes wider 

 from its encroachment upon the land, and the vortices that arise from 

 their concurrence are frequently brought down to the surface of the 

 ocean, instead of taking place in the higher regions of the air, as 

 they usually do during summer. 



While sudden, violent tempests that are occasioned by local dis- 

 turbances over a limited area, are almost necessarily cyclonic, I am in- 

 clined to adopt Espy's theory with regard to long storms, that usually 

 " the wind will blow in towards a line rather than towards a point," 

 and in fixvor of this hypothesis as well as of the periodicity of weather- 

 changes, I would suggest the following explanation. 



The normal currents of the atmosphere (Figs. 2, 3) are subject, 

 as we have seen, to a daily disturbanc(; by the sun's action (Fig. 1). 

 This disturbance, like the moon's tidal action, is cumulative, and 

 has a constantly increasing tendency to overcome the aerial polarity. 

 The gathering wave follows the sun until it is saturated with vapor, 

 and as soon as it becomes powerful enough to influence the normal 

 current, it must produce a shifting of the wind, and a deposition of 

 moisture. The equilibrium of temperature is then restored, to be 

 subjected anew to the same constant disturbance and the same 

 stormy culmination. 



* Phil. Mag., Sept., 1863. 



t Proc. Amer. Philos. Soc, April 1, 1864. 



