ON" THE NUMERICAL RELATIONS OF GRAVITY AND MAGNETISM. 



121 



currents produced by these disturbances apparently give rise to magnetic perturbations, 

 which, although far less regular in their character than those of the barometer, furnish 

 some indications of obedience to similar laws. (Ibid., pp. 434-8). 



IV. 



If any direct magnetic disturbance is occasioned by differences of gravitation towards 

 the sun, it is reasonable to suppose that the instantaneous transmission of the gravitating 

 force excites a magnetic current at the same instant in the sether. If M is the sun's mass 

 and It its distance, its attraction is M4-B, 2 =354936-^230()()' .00067. Since this attrac- 

 tive force is added to that of the earth at midnight and subtracted at noon, the solar 

 attraction-disturbance produces a daily difference in the tendency of a particle of au- 

 to wards the earth's centre equivalent to .00134 of the total force. The solar-diurnal mag- 

 netic disturbance between the same hours, as shown by Table II, amounts to .00138 of 

 the total force. 



V. 



Each particle of air may be considered as a planet revolving about the sun in an orbit 

 that is modified by elasticity, terrestrial attraction, &c. In consequence of these dis- 

 turbances, there is alternately a fall of twelve hours towards the sun between midnight 

 and noon, and a rise from noon to midnight. From the nature of accelerating forces, the 

 mean attraction-intensity disturbance should be found at 12 u -¥*/ '2=8 h - 29' from midnight, 

 and, if my hypothesis of a connection between gravitation and magnetism is correct, the 

 mean daily disturbance of magnetic intensity should take place at the same time. Such 

 is very nearly the case, the; morning mean occurring about 8 b 44' after midnight, and the 

 afternoon mean 8 lk 18' before midnight. The average daily mean is therefore 8 b ' 31' from 

 midnight, differing but 2' from the theoretical mean. The apparent retardation of 13', in 

 consequence of inertia and rotation, will be referred to in discussing Props. VIII and IX. 



During the intervals between the extremes and means, the sethereal currents and eddies 

 are so variously affected as to render the task of precisely calculating either the particular 

 values of the several disturbances, or their aggregate, extremely difficult, if not altogether 

 impossible. We may, however, by a reference to the following table, easily discover some 

 additional evidences of the dependent relation of magnetism to gravity. The theoretical 

 variations are computed on the hypothesis that the magnetic differences vary as the square 

 of the time, counting from midnight, the commencement of the fall. The barometric dif- 

 ferences from the mean are given in ten-thousandths of an inch, the hourly thermometric 

 differences in degrees of Fahrenheit, 

 vor,. x\\\ — 16 



