588 PROCEEDINGS OP THE AMERICAN ACADEMY. 



affinity — thus forming the smallest particle or molecule of the com- 

 pound — and on its other sides by a weaker affinity, causing the cohe- 

 sion to the atoms belonging to other molecules — thus forming a solid 

 mass of many molecules. Every one will grant that there are many 

 reasons for believing that this cohesion is often much feebler than chemi- 

 cal affinity. For example, the former is often far more easily overcome 

 by the introduction of heat energy than the latter. This is the case with 

 zincic chloride, for example, a salt which boils unchanged at a temper- 

 ature as low as 710°, giving a vapor which seems to possess the simplest 

 possible formula, ZnCl 2 . 



To those who are indisposed to consider seriously any hypothesis con- 

 cerning the different sides of a single atom, it is only necessary to point 

 out the benefit which has already been derived from the stereochemical 

 hypothesis of van't Hoff and Le Bel. The present discussion carries the 

 logic of three dimensions into another division of chemical argument. 



If one side of an atom is more firmly bound than another, it is only 

 reasonable to suppose that the more firmly bound portion is subjected to 

 the greater pressure, and therefore compressed into the smaller volume 

 or more flattened. It seems probable that this compressed portion 

 must be less compressible than the portion less firmly bound, because 

 universal experience shows that with a given substance at a given tem- 

 perature, the greater the volume the greater is the compressibility. 

 In other words, that part of an atom which is under less pressure 

 must be more compressible than the part which is under greater pres- 

 sure. Hence, a small development of energy by a change of a given 

 small cohesive pressure for another cohesion slightly greater would cor- 

 respond to a large contraction, equal to the change of volume caused by 

 a great change in a pressure of chemical affinity already large. Thus 

 the molecular contraction caused by the compression of cohesion may at 

 times quite mask the atomic contraction caused by the far greater pres- 

 sure of chemical affinity, because the latter is concerned with that portion 

 of the atomic volume which has been previously compressed. There is 

 nothing in this argument to show that the shape of the atom is perma- 

 nently deformed or irregular, as some have supposed; it seems much 

 more probable that in the free state every atom is spherical, and that the 

 inequalities in the atomic surface occur only under the influence of un- 

 equal pressures, such as those which form the subject of this paper. 



These considerations enable us to interpret at once the seeming anom- 

 aly concerning zincic and argentic chlorides. Zincic chloride is easily 

 volatile, boiling, as has been said, at 710°, while argentic chloride was 



