Newton, Leibnitz, and Boscovich to the Atomic Theory. 79 



parts is opposed. The same remark applies also to nitrogen, though 

 from the perfection of the individual atom in this case a spherical and 

 dense molecule of nitrogen is still less to be looked for. As to the 

 number of individuals requisite in either oxygen or nitrogen, to consti- 

 tute such a molecule, it is not as in vapour 36, but only the thinl 

 part of that number ; for in both species, in both and N, the regions 

 of union are regular pentagons ; and to this figure the isometrical 

 polyhedrom belonging is the dodecahedron. And therefore the most 

 perfect molecule of oxygen and nitrogen, will be a dodecatom. Now, 

 though we are not to expect in the atmosphere a dodecatom of oxygen, 

 and much less a dodecatom of nitrogen, yet the demand of oxygen for 

 union, and the differences of the two, and N, are so great, that 

 we may expect, under peculiar circumstances, the development of a 

 dodecahedral molecule, composed of both ; two paxticles of nitrogen, 

 which has positive poles of its own, being caught to form poles, and 

 ten of oxygen, which is itself an equatorial form, being caught to 

 form the equator. But such a molecule will possess great assimilative 

 power, and wherever it meets with vapour or water, as it must do everj'- 

 where in nature, doubtless it will 

 transform a particle in each pole 

 into HO. We thus obtain a 

 magnificent molecule, its formula 

 H0,N0i»N,0H=2(N0«+H0), 

 which is the well known formula 

 of nitric acid. 



And here we may apply our 

 method to deduce a property 

 which has been hitherto obtain- 



able only by experiment. Thus, 

 as this molecule is so delicatel}^ 

 constructed and so large, we can- 

 not expect it in the aeriform, we 

 can only expect it in the dense 

 state ; for such heat as is implied 

 in the aeriform stata would cer- 

 tainly dissipate it in a short 

 time. But as it is composed 

 wholly of aeriforms, the dense 

 molecule which we are to expect must be the lightest which is com- 

 patible with spliericity. Now, the poles of this molecule are triangular^ 

 which leads us to tlu; t(;traliedron, or a tetratom, as the lightest 

 isometrical liquid molecule possible. And this, under tli(i hiw of assimi- 



