RICHARDS. — SIGNIFICANCE OF CHANGING ATOMIC VOLUME. 583 



show that the more closely the data are scrutinized, the more emphatic 

 is their support of the somewhat revolutionary hypothesis of atomic com- 

 pressibility. The fact that this should be the case is in itself a strong 

 support of the hypothesis ; for fallacious arguments lose weight on closer 

 scrutiny instead of gaining weight. 



As has been already pointed out, the chief omission in all other work 

 on the subject of changing atomic volume * has been the lack of atten- 

 tion to the compressibilities of the substances concerned. If the pressure 

 of afRuity causes a compression of the atomic volume, it is clear that a 

 given affinity will produce a greater change of volume when the atomic 

 volume is easily compressible than when it is only slightly compressible. 

 This matter will be considered at greater length in this paper than has 

 ever been possible before, since a number of appropriate compressibilities 

 have just been determined in this laboratory for the first time, as well as 

 several especially important specific gravities. 



Another question, which must also receive consideration, is the varying 

 intensity and nature of the affinities causing combination and cohesion. 

 Do both of the two affinities, the chemical and cohesive, which together 

 determine the stability of solids and liquids, exert internal pressure; and 

 what effect does each of these have upon the volumes of substance ? 



These questions are so closely related that they may best be discussed 

 in a single thesis. Indeed, owing to the complexity of the subject and 

 the scarcity of exact data, the verdict of each answer is needed to sustain 

 that of the other. 



In the first place, the question of compressibility may be taken up. 



* It is a pleasure to call attention here to the entirely independent work of J. 

 Traube in this direction. By an interesting coincidence he publisiied in Drude's 

 Annalen, 5, 550 (June 20, 1901), the following statement : "Das Atomvolumen eines 

 Elementes andert sich vielmehr von Stoff zu Stoff; es ist urn so kleiner, je grosser 

 die Anziehung zu den benaehbarten Atomen ist," while on June 15 of the 

 same 3'ear there appeared in tlie Proceedings of the American Academy, 37, 17, my 

 version of the same relationship : " The atotiiic volume is not constant, but a function 

 of pressure and temperature and probabli/ of electric stress." By pressure was meant 

 the internal pressure caused by affinity. In his pamphlet, " Uber den Raum der 

 Atome " [Ahrens's Sammlung. chem. und chem.-techn. Vortage IV. 256 (Stutt- 

 gart, 1899)] Traube anticipated several points contained in his paper of 1901, 

 and in my work in the same year, but this pamphlet was wholly unknown to me 

 at the time. The fabric of Traube's reasoning is complicated by his hypothetical 

 assumptions of "Covolumen" and " Kernvolumen," "gebundener" and "freier 

 Aether;" but nevertheless he deserves the credit of having appreciated the 

 importance of many of the facts. The question of priority is of little consequence, 

 especially since Miiller-Erzbach (1881) had priority over both. 



