352 TVm. A. Norton on the Variations 



opposing causes should be different in different seasons. From 

 the vernal to the autumnal equinox, the dew is less in amount 

 and the nocturnal fall of temperature is greater, and therefore the 

 height of the morning maximum of horizontal force ought to be 

 less, than from the autumnal to the vernal equinox — a conclusion 

 which accords with fact. (See as above.) 



To obtain a theoretical estimate of the relative height of this 

 maximum in these two periods, we have only to seek for the 

 amount of the diminution of the horizontal force during the 

 night, that would result from radiation alone, and then allow for 

 the proportionate effect of the dew, both thermal and directly 

 magnetic. Now it appears, on an examination of the curves of 

 the daily variation of horizontal force, that the diminution of the 

 horizontal force at sundown, which must be chiefly due to radia- 

 tion, is about one-half of one of the division-spaces in the dia- 

 gram : according to this, the tendency of the radiation is to dimin- 

 ish the horizontal force, during a night of twelve hours, (from 5 

 p. m. to 5 a. m.,) the amount of six of these spaces. We have 

 obtained for the proportionate amounts of dew on a single night 

 during the two periods above mentioned in 024, and in 04. 

 These numbers bear to each other the ratio of 6 to 10, or 3 to 5. 

 Now, the curves of the daily variations of the horizontal force for 

 the middle quarters of the year show that during these periods 

 the morning maximum is about one division and a half below 

 the evening maximum. According to this the tendency of the 

 entire effect of the dew must then be to increase the horizontal 

 force 4£ divisions. Increasing this number in the proportion of 

 10 to 6, we have for the effect of dew during the first and last 

 quarters of the civil year, (or from the autumnal to the vernal 

 equinox,) 7£ divisions: which makes the morning l£ divisions 

 higher than the evening maximum. This is a close approxima- 

 tion to the actual state of the case, as shown by the curves. The 

 effect of the dew is partly attributable to the heat evolved, dimin- 

 ishing the loss of heat and of surface temperature, and partly to 

 the direct magnetic action of the dew. If the continual diminu- 

 tion in the loss of temperature at night, from hour to hour, could 

 be partially attributed to any other external cause, the effect of 

 the dew in augmenting the horizontal force would only have to 

 be diminished in the same proportion. 



It remains for us now to consider the secondary variations of 

 the horizontal force which occur during the forenoon. We have 

 already seen that the horizontal force decreases from 4 or 5 a. m. 

 to 10 a. m., although the temperature increases ; and have attribu- 

 ted this fact to the evaporation of the dew and rain that fall 

 during the night. This explanation involves the supposition that, 

 in the average of months, the amount of the evaporation is 

 greatest early in the day. That this is really the case may be 





