of Terrestrial Magnetism. 113 



which we have recently spoken merges into an annual mean solar 

 diurnal variation, evidencing its cosmical character and its con- 

 nexion with the sun, first by its period being a solar day, and 

 secondly by its amount in different years being subject to a 

 variation which corresponds in period and epochs with the 

 decennial variation of the solar spots. In both these respects 

 its characteristics are the same as those of the disturbance 

 diurnal variation ; and thus a community of origin is indicated. 

 But it has other features which constitute a well-marked distinc- 

 tion between the two classes of phenomena ; and thus we have 

 also reason to infer that, together with community of origin, 

 there is a difference in the mode of action by which each is pro- 

 duced. When, by the elimination of the larger portion of the 

 disturbances, we have obtained in the residual values a satisfac- 

 tory approximation to the true solar-diurnal variation, we find 

 that the general form of the curve so obtained is substantially 

 the same at all stations in the middle latitudes of the same 

 hemisphere. In the northern magnetic hemisphere it has an 

 eastern extreme about 8 a.m., and a western extreme between 1 

 and 2 p.m., returning to its mean position about 7 p.m. During 

 the hours of the night the needle is comparatively tranquil. In 

 Plate III. figs. 1, 2, 3, 4, the curves of the mean solar-diurnal 

 variation are represented for Toronto, Kew, Philadelphia, Nert- 

 chinsk, all stations in the northern hemisphere, but varying 

 greatly in respect to the circumstances of land and sea, near or 

 remote. In having but one easterly and one westerly extreme, 

 the solar-diurnal variation resembles one of the forms of the 

 disturbance-variation, viz. that figured in the first Plate as repre- 

 senting the easterly disturbances at Kew and Hobarton; but 

 there is this essential distinction between them, that in the 

 disturbance-variation the local solar hour belonging to any point 

 in the curve — the apex, for example, as the most marked feature 

 — varies apparently without limit in different meridians, whilst 

 in the solar-diurnal variation, the time as well as the form is 

 constant in the extra-tropical latitudes of all the meridians of 

 the same hemisphere*. 



* The easterly and westerly deflections spoken of in the text are magnetic 

 east and west, which are not necessarily nor always the same as geogra- 

 phical east and west. In the lands visited by the arctic expeditions, situ- 

 ated to the north of the point where the dip of the needle is 90°, the de- 

 clination is 180°, or, in other words, the marked or north end of the magnet 

 is directed towards the geographical south. In such localities the deflec- 

 tions towards the magnetic east are in fact deflections towards the geogra- 

 phical west, and vice versa. At 8 a.m., for example, the north end of the 

 magnet is deflected towards the magnetical east, or, otherwise expressed, 

 towards the geographical west; and at 2 p.m. towards the magnetic west, 

 or geographical east. But in the same meridian, and in localities situated 

 on the opposite or south side of the point of 90° of dip, the marked or 



Phil Mag. S. 4. Vol. 24. No. 159. Aug. 1862. 1 



