PHILOSOPHICAL SOCIETY OF WASHINGTON. 143 



mic motions generating radiant light and heat must be referred 

 We may thus picture to ourselves the monochromatic lines of the 

 spectrum as exhibiting a second order of occult or insensible kinetics, 

 in quality and range as different from and as much below the 

 kinetics of the molecule, as this differs from and is below the kinetics 

 of tangible masses. 



The Origin of Atomic Motions. — With regard to the nature and 

 origin of the atomic motions, it appears tolerably clear that they 

 are primarily derived from the shocks of the molecules or systems 

 of which they are the components ; and that there is at every 

 molecular collision a transfer or exchange of energy teuding to 

 equalize the internal momentum of pulsation with the external 

 momentum of translation. The primum mobile is therefore the 

 falling together of molecules uuder the influence either of gravi- 

 tation, or of chemical affinity. While it is difficult to realize the 

 precise manner in which molecular aud atomic motions are re-dis- 

 tributed during the brief instants of impact, it appears in the high- 

 est degree probable that the atoms describe elliptical orbits, which 

 may become circular, but never rectilinear. Were the atomic 

 motions mere oscillations, it would appear unavoidable that under 

 the stress of special impacts, some of them must occasionally be 

 detached, — as in the case of molecular evaporation. But the ulti- 

 mate molecule is unchangeable and " indivisible : " — held together 

 in bonds incomparably stronger than those of hardest steel. And 

 the loss of an atom may be regarded as an impossible catastrophe. 

 Moreover, from the utter irregularity of direction in molecular 

 encounters, obliquity of impact on the rapidly changing atoms, 

 would appear almost a necessity : and hence would result as neces- 

 sarily — elliptical paths of excursion. 



In this constant play of atoms derived from repeated collisions, 

 we must believe that these atoms are whirled in ever varying rota- 

 tions — simultaneously with their orbital revolutions; but as these 

 double motions form but parts of their common fund of kinetic 

 energy, it is not probable that any special phenomena will ever dis- 

 tinctly reveal such axial motions ; — unless indeed it be hereafter 

 shown that polarity is the resultant of concerted directions of rota- 

 tional or orbital axes, or of both. 



The Amplitude of Atomic Orbits. — Of the actual or relative 

 diameters of these orbits we are as ignorant as we are of the sizes 



