2 8o SCIENCE PROGRESS. 



During the past few months the interest of chemists and 

 physicists has been specially drawn to the first two gases 

 in this table, since it is on the observed value of the ratio 

 of the specific heats that it is assumed that the gas argon is 

 monatomic, and therefore elementary. The argument in 

 favour of the monatomicity of the gas appears to be two- 

 fold. On looking down the table it will be seen that, with 

 the exception of gases with more than one halogen atom in 

 the molecule, gases of the same atomicity show a decided 

 tendency to have the same value for y. Thus the diatomic 

 gases are all near 1*4, the triatomic near 1*3, the pentatomic 

 near 1*28, and so on. Admitting this as an empirical law, 

 we conclude that argon has probably the same atomicity 

 as mercury vapour, and mercury vapour is known from its 

 chemical reactions and its vapour density to be monatomic. 

 This argument receives support from another drawn from 

 the actual value of y. The values found in two experi- 

 ments were 1*61 and 1*65, but these are probably both too 

 low. In consequence of the difficulty of preparing the gas 

 in large quantities, the experiments were made in tubes 

 much narrower than the narrowest Kundt found could 

 safely be used, and the effect of using too small a tube is to 

 lower the velocity of sound, and hence to lower the observed 

 value of y. It is scarcely conceivable that the y of any gas 

 could be greater than i'6y, so we may conclude that argon 

 has very nearly this value. Now, using the formula on 

 p. 274 to calculate /3, we find it comes out zero, so that when 

 the gas is heated the energy given to it all takes the form 

 of energy of translation of the molecule — there is none 

 whatever taken up to give relative motion to the parts of 

 the molecule. It does not, of course, follow that the 

 molecule possesses no internal energy ; all we can say is 

 that if it has such energy, the circumstances under which 

 the experiment was made were such that during the heating 

 or cooling of the gas there was no change in the internal 

 energy. If the molecule is anything more than a mere 

 mathematical point it must be capable of rotation, but it is 

 not unlikely that the energy of rotation would change more 

 slowly than the energy of translation, as Watson {Kinetic 



