468 Analj/ses of Books, CDec, 



'* From these it appears, that the minimum intensity happened 

 nearly at the time the sun passed the magnetic meridian, and 

 rather later than in May, which was also the case with the time 

 of the sun's passage over the meridian :* the intensity increased 

 until about six o'clock in the afternoon, after which time it 

 appears to have decreased during the evening, and to have been 

 decreasing from an early hour in the morning. 



" The general agreement of these intensities with those 

 deduced from the observations made in doors, is as near as 

 could be expected, considering that an interval of twenty days 

 had elapsed between the two sets of observations. From this, 

 and the agreement in the manner in which the westerly and 

 easterly points of equilibrium approach and recede from the 

 north in the two cases, which I have before pointed out, we may 

 conclude, that there is nothing anomalous in the action which 

 takes place on the needle under the different circumstances of 

 its being placed in doors or in the open air ; and that the apparent 



*' * The diurnal variation, both in the direction of tlie needle and in the magnetic 

 intensity, appears to have a reference to the position of the sun with regard to the mag- 

 netic meridian ; it is therefore probable, that the sun is the principal cause of both these 

 phaenomena. The circumstance of the situation of the magnetic pole in what appears 

 to be, independent of elevation, the coldest region of the globe, supported as it is by the 

 fact of a diminution of temperature causing an increase of magnetic intensity, would 

 lead us to infer, that the effect produced by the sun is principally to be attributed to the 

 heat developed by it ; but should any periodical effects, corresponding to the time of 

 the sun's rotation about its axis, be observable in the diurnal variation, we must sup- 

 pose that the sun, like the earth, is endued with magnetism, and look for a cause of this 

 magnetism, common to all the planets. Being engaged more than two years ago in 

 making some experiments on the effects produced on the needle by unpolarized iron, I 

 discovered that a peculiar polarity was imparted to the iron by simply making it revolve 

 ^bout an axis ; and this naturally suggested the question to me, whether the magnetism 

 of the earth, and consequently, diat of the other planets and the sun, might not be 

 owing to their rotation ? From the effects which I have observed to be produced on iron 

 by its rotation, it appears probable, if the magnetism of these bodies be not caused by 

 their rotation, that at least the effects will be modified by, and, to a certain extent, 

 depentlent on such rotation. Since first observing the fact, that simple rotation will 

 cause a peculiar polanty, if I may be allowed the expression, in iron, I have made a 

 great variety of experiments on the subject, which have enabled me to trace the laws 

 according to which this polarity in the iron affects a magnetic needle, independently of 

 the effect produced by the mass. It would lead me to too great a length here to state 

 the several eftects that are produced by the rotation of iron, or the laws which govern 

 them ; but I will briefly mention one. Let us imagine a plane to pass through the 

 centre of a horizontal needle, at right angles to the meridian, and making an angle 

 with the horizon equal to the dip ; then, if the plane of a circular plate of iron coincide 

 with this plane, and the plate be fixed on an axis passing through its centre at right 

 angles to its plane, so that it can be made to revolve in its own plane, the direction of the 

 needle will be different, according as the several points of the plate are brought into any 

 particular position by making it revolve in one direction or the opposite, excepting in • 

 four positions of the centre of die plate. If the centre of the plate be successively placed 

 to the east or west of the centre of the needle in the same horizontal line, and over the 

 needle in the plane of its meridian, then the deviation of the needle due to the rotation 

 of the plate will be in contrary directions in the two cases, the plate revolving in the same 

 direction in both. These and other peculiar effects arise entirely from the rotation of 

 the iron, and are not produced by any friction on the axis. As the effects are not very 

 considerable, to render them conspicuous it is necessary to make use of a plate eighteen 

 inches in diameter, and to have its ce-itre within sixteen inches of that of the needle. If 

 the needle is under the influence of magnets, as in the foregoing obscryationsj the effects 

 produced by the iQtj^tio^ of t^e plate im CQn§44ei^.blc.'' 



