EARTH'S MAGNETISM TO THE SOLAR AND LUNAR PERIODS. 109 



it was made, of above twenty scale division (twelve times the resulting range of 

 horizontal force in the lunar hour angle curve), was rejected, and an interpolated 

 quantity substituted ; this elimination, however, was not found to affect the pe- 

 riods of maxima and minima; it reduced the range, and rendered the curve 

 somewhat more regidar. In the obsei-vations for 1845, as a farther check on these 

 eliminations, a different test number was employed, namely, forty scale divisions, 

 or nearly twenty-five times the resulting range of the lunar hour angle curve. 



24. In both years, the minimum occurs at SO"", or 5^ before the meridian 

 passage ; the maximum at 14'', or about l^"" after the inferior passage : in both 

 years, a minimum occurs at 8"; in 1844, a maximum occurs before 4h; and in 

 1845, before 3''. The maximum at 3'', for 1845, differs little fi-om the maximum 

 at 14''; but the maximum at 4*, in 1844, is considerably less than the maximum 

 at 14''. The coincidences of these results may be considered extraordinary, when 

 it is known that the range in 1844 is only O'OOOBll, and in 1845 only 0-000213, 

 or less than the effect of one degree of Fahrenheit on the magnetism of the bifilar 

 bar. 



25. Several questions spring from this result of the connexion of the intensity 

 with the moon' s hour angle. Does the range of the lunar hour angle curve vary with 

 the moon's declination ? If so, then we do not eliminate the lunar effect from 

 the solar day curve by a monthly or any other summation. Do the periods 

 of maxima and minima vary throughout the lunar month in the lunar hour angle 

 curve, as they do through the year in the solar day curve ? These questions 

 I shall endeavour to examine at another opportxmity. 



Makeestoun, December 26. 1845. 



