460 MATHEMATICS AND NATURAL PHILOSOPHY. 



magnetic variations to thermic causation. "We may ascribe to the general and' 

 almost exclusive prevalence of the thermic hypothesis, aud to its influence on mag- 

 netic reasonings, that the well-known erroneous opinion was so confidently pro- 

 mulgated by a deservedly high magnetic authority, that a line must exist sur- 

 rounding the globe, in which the needle would be found to have wo diurnal varia- 

 tion. "We have now, on the contrary, reason to be assured, by the facts of the an" 

 nual inequality thus discovered, that there is no such line-; but that everywhere 

 in the regions of its supposed existence a diurnal variation subsists, having op- 

 posite characteristics in opposite parts of the year as influenced by the sun's posi- 

 tion on either side of the .equator, and disappearing only at. the epochs when the 

 sun passes from south to north or from north to south Declination. 



Lunar Variation. — But if thermic relations have failed to supply a connecting 

 link between the sun and those magnetic variations which are, without doubt, re- 

 ferable to the stm as their primary cause, the failure of that hypothesis is made 

 still more obvious by the existence of variations governed by the moon's position 

 relatively to the place of observation. We are indebted to M. Kreil, now holding 

 the same position in Austria that I have filled in England, for the first suggestion 

 of the existence of a lunar-diurnal variation of one of the elements, viz., of. the 

 Declination, founded on observations at Milan and Prague ; and in the Phil. Trans, 

 for 1856, Art. XXII., will be found an exposition of the facts of the moon's diurnal 

 influence on each of the three magnetic elements at Toronto, viz., on the Declina- 

 tion, Inclination and Total Force. In the case of this investigation, notwithstand- 

 ing the smallness of the values concerned, the instrumental means supplied to the 

 Colonial Observatories have been found competent to determine, with an approxi- 

 mation sufficient for present theoretical purposes,, the character and amount/or 

 each element of the regular daily effect of the moon on the terrestrial magnetic 

 phenomena, the existence of which does not appear to have been even suspected 

 at the time when the Report of the Committee of Physics was drawn up. The 

 discovery of the moon's influence on any of the magnetic elements is due, as already 

 stated, to M. Kreil ; but Toronto- is the first, and as yet the only, station, at which 

 the numerical values at every lunar hour of the lunar-diurnal variations of the 

 three elements have been published. Corresponding statements to that which has 

 been given for Toronto, will be found for St. Helena and Hobarton, in the volumes 

 of those observatories, which are now in preparation. All the results at the three 

 stations present the same general characters. The lunar influence does not ap- 

 pear to participate in the decennial inequality which is found in all the solar va- 

 riations (Phil. Trans. 1857, Art. I.). The lunar-diurnal variation of each of the 

 elements is a double progression in the twenty-four hours, having epochs of maxi- 

 mum and minimum symmetrically disposed. In character, therefore, it differs 

 from what might be expected to take place if the moon were possessed of inherent 

 magnetism, i. e. if she were a magnet, as it is usually termed, per se ; and accords 

 with the phenomena which might be expected to follow if she were magnetic 

 only by induction from the earth. On the other hand, it is believed that the 

 amount of the variation, as observed at each of these stations, very far exceeds 

 what can be imagined to proceed from the earth's inductive action reflected from 

 the moon. In this theoretical difficulty we are naturally thrown back to seek a 

 more extensive knowledge of the phenomena than we have yet obtained, and to 

 the generalization which will follow, when sufficient materials for it have been 

 procured. Insubordinate particulars, a difference, which is apparently systematic 



