200 ANNUAL EEPOET SMITHSONIAN INSTITUTION, 1912. 



It is quite different when we discover a coincidence wliicii could 

 have been predicted, and is thus not the result of chance, and espe- 

 ciall}^ when that coincidence is a numerical value. Now, there are 

 coincidences of this last nature which have recently brought confir- 

 mation to our atomic views. 



The kinetic theory of gases has thus received unexpected corrob- 

 oration. New theories have been very closel}^ patterned after the 

 kinetic theory, for instance, the theory of solutions as well as the 

 electronic theory of metals. The molecules of a dissolved substance, 

 as well as the free electrons to which metals owe their electrical con- 

 ductivity, behave just as do the molecules of a gas within its inclo- 

 sure. The parallelism is perfect and can be followed even to numer- 

 ical A'^alues. Thus what seemed doubtful becomes probable. Each 

 one of these three theories, if it had to stand by itself, would seem 

 only an ingenious h3^pothesis for which we might substitute other 

 explanations equally probable. But when, as in each of the tlu"ee 

 cases, a different explanation w^ould be necessary, the coincidences 

 found would be inadmissible as the result of chance, whereas the 

 kinetic theories make the coincidences necessary. Further, the 

 theory of solutions quite naturally leads us to that of the Brownian 

 movements, where it is impossible to consider the thermal agitation 

 as a theoretical fiction, suice it is actually seen under the microscope. 



The remarkable counting of the number of atoms by Perrin com- 

 pleted the triumph of the atomic theory. Wliat carries our convic- 

 tion are the multiple concordances among the results obtained by 

 completely different procedures. But a short time ago we would 

 have thought ourselves fortunate if the numbers found had the same 

 number of digits; we would have asked only that the first significant 

 figure should be the same. That first f.gure we know to-day. What 

 is more remarkable, v/e are now discussing even the most diverse 

 properties of the atoms. In the processes used with the Brownian 

 phenomenon, or in those used for the law of radiation, we do not deal 

 directly with the number of atom§, but with their degrees of freedom* 

 of movement. In that process where we consider the blue of the 

 sky, the mechanical properties of the atoms come into pla}^; the 

 atoms are looked upon as producing an optical discontinuity. Finally, 

 when we take in hand radium, what we observe is the emission of 

 projectiles. Here, were there discordances, no embarrassment would 

 have been felt, but happily there were none. 



The atom of the chemist is now a reality. But that does not mean 

 that we have reached the ultimate limit of the divisibility of matter. 

 When Democritus invented the atom he considered it as the abso- 

 lutely indivisible element witliin which there would be uotliing 

 further to distinguish. That is what the word meant in Greek. It 

 was for that reason that it was coined. Beyond the atom he wished 



