in the Magnetic Forces of the Earth. 361 



ture at B, in any short interval of time, becomes the same as at 

 A. This will happen at the hour of the most rapid rise of tem- 

 perature, or about 9 a. m. After this the hourly increment of 

 temperature will be less for B than for A, and the same portion 

 of the isothermal line will move southward. This somherly 

 movement will continue beyond the time of maximum tempera- 

 ture (3 p. m. at Philadelphia), and until the fall of temperature at 

 B becomes less rapid than at A. This will happen about 7 p. m. 

 It is manifest, from the concave form of the curve of daily varia- 

 tion of temperature, during the night, that after this the decrease 

 of temperature at B will be continually less than at A, and there- 

 fore that the isothermal line will move northward, to the east of 

 station A, and southward to the west of it. This motion will 

 continue until 5 a. m. ; and beyond this, as we have already seen, 

 until towards 9 a. m. In obtaining these results, we have taken 

 it for granted that the law and rate of the mean daily variation of 

 temperature is the same at B as at A. This doubtless is not 

 strictly true, and therefore the epochs of maximum and mini- 

 mum of declination should be somewhat different from the times 

 above specified. If we neglect this difference, it appears, that on 

 the supposition which has been made, the needle would move 

 toward the east from 9 a. m. to 7 p. m., and toward the west from 

 7 p. m. to 9 a. m. The actual state of things di tiers from this in 

 two or three points; during the last half of this period of west- 

 erly movement, or nearly so, there is actually an easterly move- 

 ment, and during the first half of this period of easterly move- 

 ment there is actually a westerly movement ; and the evening 

 minimum occurs generally some two or three hours later, (about 

 10 p. m.) These discrepancies, (with the exception of the last, 

 which is comparatively trifling,) disappear if we compare the 

 curve of declination (fig. 25) with that of horizontal force (fig. 3), 

 instead of that of temperature, as we should do. If this be done, 

 it is found, as we have already seen, that the points of maximum 

 variation of the horizontal force, or of inflexion in the curve, fall 

 at the epochs of the maximum and minimum of declination. 



To understand the movements, in detail, of the line of equal 

 surface magnetic action, upon which the daily horizontal* move- 

 ments of the needle depend, we have only to compare the change 

 of the horizontal force during the hour following the time con- 

 sidered, with the change that occurs an hour later during the same 

 interval of one hour ; for the latter is the change that occurs at a 

 Place an hour to the east of the station of the needle, cotempo- 

 raneously with the change at the station itself. When these two 

 changes are equal the line in question is stationary. When they 

 a re both decrements, if the first is greater than the second, as 

 from 8 a, m. to 10 a. m., the line rises, to the east of the station, 

 a nd the needle moves westvvardly ; but if it is less, as from 6 a. ml 



