SOME CONTF.MI'ON.lKy .ll>rAXCi:S IX /7/V.S7CV nil 4(« 



condensed phases the atoms are so close together that they distort 

 one another — a perniissiMe idea if used with discretion, yet an atomic 

 thcor>' coiiltl easily liecome a meaningless form of words if this device 

 were employed without limit. Of all the properties of matter, mass 

 alone appears to be entirely exempt from change. For this reason 

 all atom-mtxlels involve mass as an essential property of the atom; and 

 this is the oidy respect in which they all agree. 



Few and simple, therefore, must be the properties of the atom; yet 

 we must not rush to the other extreme, and contri\e atoms simplitied 

 into usclessness. The chemists know of eighty-eight different ele- 

 ments, sufHciently unlike to be distinguished; and we all know how 

 great is the contrast between carbon and gold, hydrogen and lead, 

 fluorine and helium. It is scarcely likely that such differences as 

 these can be explained by atoms which are simply hard pellets differing 

 only in size and shape and weight, like those of Lucretius and Newton, 

 or by atoms which are abstract centres of force, like those of Boscovich. 

 We are forced to invent atoms more complicated that these; and from 

 this it is not far to say that we must imagine a structure for the atom; 

 and from this scarcely farther to say that we must imagine an atom 

 built of parts. 



At this point we meet with a clamor from a number of excellent 

 people, many of them otherwise quite innocent of Hellenic culture, 

 who have it firmly fixed in their minds that atom is the Greek word 

 for indivisible; whence they conclude that when the ph>'sicist speaks 

 of subdividing his atoms, he is contradicting his own terms, he is 

 violating the rules of his own game, and has forfeited his right to be 

 heard.- The premise may be right, but the conclusion is absurd. If 

 some of the properties of gold are explained by assuming it made of 

 atoms with fewer properties, and later the explanation is impro\ed 

 and extended by assuming these atoms made of still smaller particles 

 with still fewer properties, the second step is not less legitimate than 

 the first. It may l)e contended, with some reason, that the name 

 atom should be transferred at once to the smaller particles. At best 

 this would be one of the changes which are desirable in principle but 

 cause more trouble than they are worth. The contention is, however, 

 weakened by the fact that some at least of the smaller particles of 

 which we imagine gold atoms to be made are not imagined to be 

 peculiar t(j gold, but are conceived as particles of a fundamental sub- 

 stance common to all elements. That the "atoms" of the many 



' I should have put this passage even more strongly, but that Schuster tells that 

 Kelvin himself inveighed on one occasion against the idea of subdividing atoms. 

 He was answered by a young man who said, "There you see the disadvantages of 

 knowing Greek." This seems as good an answer as any. 



