CHEMISTRY: LEWIS, EASTMAN AND RODEBUSH 27 
That this same phenomenon is exhibited by potassium, sodium and calcium 
is shown in figure 2, where the regular curve which would fit the experimental 
points at lower temperatures is drawn as a continuous curve and the individual 
points represent our experimental determinations of the heat capacity. In 
each of these cases the heat capacity rises well above the theoretical value 3R. 
The deviation from the regular curve at any one temperature is greater for 
calcium than for magnesium and greater for potassium than for sodium. We 
may therefore predict that this phenomenon is more marked the more electro- 
positive the metal, but that it will be found for every substance at high tem- 
peratures. According to modern theories of atomic structure the atom can 
no longer be regarded as the ultimate unit of matter. Besides the atomic 
nucleus there are electrons,' and all of these particles, if capable in any degree 
of independent motion, should according to the earlier theory of the equiparti- 
tion of energy, acquire their full share of thermal energy. If, however, we 
admit that at a given temperature deviations from the equipartition principle 
occur to an extent which is greater the lighter the particle and the more rigid 
the constraint which holds it to a fixed position, then it follows that at a tem- 
perature at which other particles are gaining thermal energy approximately 
according to the equipartition law, the electron which is by far the lightest of 
all particles, will acquire very little energy unless bound by very weak con- 
