188 Mr. P. E. Chase on the Influence of 



than in summer. Kreil, indeed, inferred from the ten-year series 

 of the Prague observations, that in winter the lunar-diurnal va- 

 riation either disappears, or is entirely concealed by irregular 

 fluctuations, requiring a long series for their diminution. The 

 method of reduction which he employed was, however, less per- 

 fect than that now used. The second characteristic of the in- 

 equality consists in the earlier occurrence of the maxima and 

 minima in winter than in summer. The winter curve precedes 

 the summer curve by about one and three-quarter hours.'" (Ibid, 

 pp. 12 & 13.) 



XL The curves of semiannual variation at all points where 

 continuous observations have been made, present a striking simi- 

 larity, the amount of deflection being nearly equal in all parts 

 of the globe. (See diagram, Part II. p. 12. Compare St. Helena 

 Obs. 2. cxix. ; Toronto Obs. 2. xvi. xvii.) 



Now it is evident that the mechanically-polarizing currents 

 (Proc. Amer. Phil. Soc. vol. ix. p. 367 et seq.) must be variously 

 deflected at different periods of the day, in consequence of gravi- 

 tation-disturbances arising both from the varying distance of the 

 sun and from solar heat. Although it may sometimes be desi- 

 rable to consider the so-called thermal currents apart from those 

 which would be produced independently of any change of tem- 

 perature, such a distinction is unnecessary in a preliminary quali- 

 tative investigation like the present, because the periods both of 

 maxima and of minima are precisely coincident in the two cur- 

 rents (except as they may be slightly modified by the earth's 

 radiant heat), and because even the thermal currents are occasioned 

 simply and solely by the varying gravitation of fluids of varying 

 density. 



Regarding, therefore, the air and sether over any given mag- 

 netic meridian, during the day-hours the intertropical and during 

 the night the extratropical portions will be most drawn toward 

 the sun, and the following deflections will be thus produced in 

 the portions nearest the equator : — 



6 to 12 A.M. 



12 to 6 p.m. 



6 to 12 p.m. 



12 to 6.A.M, 



Northern zones. S.E 



s.w. 



S.E. 



S.W. 



Southern zones. N.E 



N.W. 



N.E. 



N.W. 



The night disturbances, whether from variations of temperature 

 or from simple fall toward the sun (the distance fallen varying 

 as the square of the time from midnight), will be very slight. 

 The earth's rotation, centrifugal force, and the atmospheric 

 inertia tend to throw each of the phases forward and to increase 

 the magnitude of the westerly, while they diminish the easterly 

 deflections. If these modifications were sufficient to override the 

 slight easterly tendency at 6 to 12 p.m., and to advance the 



