116 Major-General Sabine on the Cosmical Features 



approached, and progressively increases as that line is receded 

 from on either side. 



IV. We have next to notice a magnetic variation which has 

 for its period a lunar day, thereby establishing the fact of the 

 existence of a sensible magnetic influence exercised by the moon 

 at the surface of the earth. It is derived in the following man- 

 ner : — Hourly observations having been made at the several hours 

 of mean solar time, and having been subjected to processes by 

 which the effects of the magnetic disturbances and of the solar 

 diurnal variation have been wholly or for the most part elimi- 

 nated, the residual values are rewritten in lunar monthly Tables, 

 entering each value in the column of the lunar hour nearest to 

 the actual time of observation. The monthly means of the 

 several lunar hours then show the mean lunar-diurnal variation 

 in that lunation. This has been done for the three magnetic 

 elements at stations situated in both hemispheres and in the 

 tropics, as well as in the higher latitudes. The result has been 

 to give everywhere and in each of the three magnetic elements 

 the same highly characteristic general form for the curve of the 

 lunar-diurnal variation, that form being essentially and systemat- 

 ically different from that of the solar-diurnal variation. For 

 the sake of brevity, I will confine myself, as before, to the phe- 

 nomena of the declination. In so doing, we find four culmina- 

 tions, two of maximum easterly, and two of maximum westerly 

 deflection, all nearly equal in magnitude, and separated by nearly 

 equal intervals of lunar time. Thus at Kew, for example, we 

 obtain from the mean of four years of observation (1858 to 1861 

 inclusive) the curve represented by the unbroken line in PL III., 

 where it will be seen that we have at the lunar hours of 1 and 

 13 extreme westerly deflections amounting to about 10" from 

 the zero line, and at the lunar hours of 7 and 19 extreme east- 

 erly deflections of nearly the same magnitude, the zero line being 

 passed through four times in the twenty-four hours at nearly 

 equal intervals. I have figured in the same Plate and figure, by 

 a broken line, the curve of the corresponding phenomena at Ho- 

 barton; because, without entering into particulars, which would 

 require more time than can be given, the juxtaposition of the two 

 curves illustrates both the correspondence and the differences in 

 the lunar-diurnal variation in different parts of the globe. The 

 correspondence shows itself to the eye by the general agree- 

 ment in the form of the curves. The principal difference con- 

 sists in the circumstance that, Kew being in the northern mag- 

 netic hemisphere, the deflections represented are those of the 

 north end of the magnet ; and Hobarton being in the southern 

 magnetic hemisphere, the deflections represented are those of 

 the south end of the magnet. It will be seen that the deflec- 



