112 PROCEEDINGS OF THE AMERICAN ACADEMY. 



Now the combination of these two effects, namely that when the 

 molecules come very close to each other the attractive forces depend 

 on the orientation as well as on the mean distance apart, and that 

 the molecules may assume a greater uniformity of arrangement, has 

 far reaching consequences that provide the possible explanation ot all 

 the complicated effects which we have found to exist. Ihus, it we 

 consider two possible configurations of the liquid, each having the 

 same volume, but one with a more orderly arrangement of the mole- 

 cules we see that the more orderly arrangement involves a greater 

 effective space open to occupation. One consequence is that the more 

 orderlv arrangement has the greater compressibility. One striking 

 example of this has been found in the case of mercury, ihe com- 

 pressibility of solid mercury has been found to be less than the com- 

 pressibility which the liquid would have at the same temperature 

 if it could be compressed without freezing to the same volume as the 



solid. . i.- -J. 1 



In the detailed discussion of the thermodynamic properties it has 

 been shown that we have here a possible explanation of many compli- 

 cated effects. It explains increasing compressibility with rising pres- 

 sure decreasing compressibility with rising temperature, increasmg 

 thermal expansion with increasing pressure, and decreasing expansion 

 with rising temperature. It is not necessary to go mto the details 

 of the argument again. It is to be emphasized, however, that we 

 have here a mechanism capable of explaining a bewildering array 

 of experimental facts. There must be at least some validity m the 



^°Not''much has been said in the explanation above of the results of 

 possible association, because under high pressures, when the mole- 

 cules find difficulty in adapting themselves to the space at their dis- 

 posal, it seems unlikely for groups of molecules to unite themselves 

 into very close knit units. The molecules, on the other hand, do appar- 

 ently preserve their individuality under these high pressures, and do 

 not break up into simpler compounds. It might be expected, for 

 instance, that pressure would transform ether mto isobutyl alcohol, 

 a substance of the same atomic constitution, but with a smaller volume. 

 Such was not the case, however. But it may be that association does 

 play an important part at low pressures. In this case it would be 

 capable of explaining various irregularities in very much he same 

 way as suggested above. For instance, if association takes place with 

 decrease of volume, the thermal expansion or compressibility may be 



