158 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1925 



adapted to the larger scale atomic and molecular phenomena. There 

 are, of course, occasions when the behavior of matter under high 

 pressures intimately involves the inner structure of the atom, but in 

 general the pressures with which I am concerned are too small to 

 produce large changes in the atoms themselves. We are concerned 

 here with pressures from 10,()()0 to 20,000 atmospheres. These are 

 capable of producing comparatively large and significant changes 

 in the properties of many substances, but they can not, of course, 

 be compared with the pressures of perhaps billions of atmospheres 

 which occur in the interior of the stars, and which are supposed to 

 be a factor in atomic evolution. 



The significance of high pressure as a tool of atomic research 

 lies in the extreme simplicity of the change it produces in the ex- 

 ternal conditions, merely pushing the atoms closer together so that 

 they are compelled to give a somewhat more intimate account of 

 their ow^n inmost selves. The changes produced by pressure are 

 obviously much simpler than those produced by a change of tem- 

 perature. Xo one regards our understanding of a phenomenon as 

 satisfactory unless we are able to give some account of the changes 

 produced by temperature; still less should we be satisfied if we can 

 not account for the changes produced by pressure. Further, by com- 

 bining the effects of temperature and pressure, we extend enormously 

 the field in which we may look for phenomena suggestive for our 

 theories, for the field is now two instead of one dimensional. 



It is not necessary to apologize for activity in atomic phenomena, 

 even if they are not on the extreme end of the exploring line, be- 

 cause they are the phenomena which are perhaps of most immediate 

 concern to us in daily life, and we have as j'et hardly begun to 

 understand them. The extent of our ignorance may be emphasized 

 by the fact that none of our theories is able to predict accurately 

 any of the high-pressure phenomena which I shall summarize, in 

 spite of their extreme simplicity. 



Perhaps the most fundamental of the changes produced by pres- 

 sure are those of volume itself. I have measurements of the com- 

 pressibility of 5 gases, 15 liquids — most of them simple organic 

 liquids, but including water and mercury — and some 45 different 

 solids, 35 of them metallic elements, several minerals, and several 

 varieties of glass. I may mention here also the compressibility of 

 11 of the alkali halides determined by Dr. J. C. Slater, in the Jelfer- 

 son Laboratory. All of these compressibilities have been determined 

 at several temperatures, so that we know the effects of pressure on 

 thermal expansion. The melting curves of 87 substances, including 

 metallic elements and organic compounds have been determined. 

 These melting measurements include a determination of the change 

 of melting temperature when pressure is raised, the change of volume 



