PROCEEDINGS OF THE POLYTECHNIC ASSOCIATION. 677 



hccat which counteracts the vast force of gravity that presses toward 

 his centre. In another paper, I have attempted to show that all 

 celestial bodies generate heat internally, in the direct ratio of their 

 masses, by the assimilation *nd condensation of ethereal matter, 

 and the conversion of its expansive force into radiant torce. If 

 this is true, the larger planets will, all else equal, be less dense 

 than the sn)aller ones. The mutual attractions and perturbations 

 of the embryo planets, during the countless ages of their proo-res- 

 sive foftufition, must have had some iniluence upon their relative 

 magnitudes and densities, and, perhaps, also upon their intervals. 

 The attraction of an immense exterior mass, like Jupiter, would 

 certainl}^ prevent much of the light matter from reaching the orbit 

 of Mars; this would not only account for the anomalous smalluess 

 but also for the great density of Mars. The second satelh'te of 

 Jupiter occupies an analogous relative position; and, like Mars, it 

 is remarkably small and dense. At the first thought, it would 

 seem that, according to this theory, the magnitudes of the planets 

 should be successively greater from Mercury to Neptune; wdiereas, 

 they only increase outwardly to Jupiter, and then decrease. A 

 little examination will show us that the facts are in strict accord- 

 ance with the requirements of theory. It should be recollected 

 that the tendency of rotation was to make the nebulous wheel 

 thickest in the centre and gradually thinner to the outer edo-e- 

 and that the influence of the resisting medium did not chano-e this 

 form, except between the orbit of J*upiter and the sun. Bevond 

 Jupiter the form remained the same as if there had been no resist- 

 ing medium; that is to say, the wheel became gradually thinner 

 from Jupiter to the outer edge. This is in accordance with the 

 fact that Saturn is smaller than Jupiter, and Uranus and Neptune 

 smaller than either. 



This theory is doubly confirmed by the satellitic systems of 

 Jupiter and Saturn. In each of these, the distribution of matter 

 is 'essentially the same as in the solar system. There is, lir:?t, a 

 central primary, containing many times more matter than all the 

 the secondaries together; second, there are several small seconda- 

 ries near the primary; third, there is one giant secondary, contain- 

 ing more matter than all the other secondaries of that system to- 

 gether; fourth, beyond the giant are one or more smaller bodies 

 with Avide intervals. 



