690 PROCEEDINGS OF THE AMERICAN ACADEMY. 



effect of cohesion was largely eliminated and the remaining effect could 

 be safely ascribed to action of the pressure of chemical affinity on the 

 different compressibilities of the elements compared in each series. 



The thorough understanding of this matter is so important that it is 

 worth while to cite other cases illustrating it. A particularly illuminat- 

 insf series of examples is that of the halides of zinc and cadmium. The 

 bromides and chlorides of both these metals, as well as the metals them- 

 selves, may be easily volatilized, and their boiling points are well known ; 

 hence more is known in these cases than in almost any other concerning 

 the relative intensities of the internal j^hysical pressures. The work of 

 chemical compression is again taken as equal to the heat of foimation, 

 because only a small part of the heat of formation is usually due to the 

 heat of molecular cohesion, and the change of heat-capacity is small. 



Below are given the necessary data concerning zinc and cadmium. 

 The data are all reasonably accurate, the specific gravities being espe- 

 cially so, thanks to the care of Dr. Baxter and his assistants, who kindly 

 determined three of them for use in the present paper. 



From these results it is seen that in each case the contraction which 

 takes place during the formation of the zinc halide is less than that which 

 takes place during the formation of the cadmium salt by about five milli- 

 liters per mole, although their heats of formation are nearly equal. A 

 part of this difference is undoubtedly to be ascribed to the cause which 

 gives the cadmium salt a higher boiling point in each case, namely, a 

 greater cohesive pressure in the case of the cadmic salts. A part may 

 also be expected to be due to the fact that metallic zinc, having a higher 

 boiling point (920°)* than cadmium (778°) — and therefore a greater 

 molecular latent heat of evaporation or greater internal physical pressure 

 — must be supposed to be in the more compressed state. Other indica- 

 tions also point to the same conclusion ; for example, zinc has the smaller 

 coefficient of expansion, the smaller molecular volume, the greater hard- 

 ness, and the larger " energy quotient," (a quantity having the dimen- 

 sions of pressure which is obtained by dividing the atomic heat by the 

 atomic dilation).! Since each of these latter qualities is not a function 



chlorides boil considerably over 1200°, and not far apart. Lithic chloride seemed 

 to be slightly the more volatile of the three. The melting points are all near 

 together, — of the bromides and iodides as well as of the chlorides, ^and in similar 

 salts of this kind it is not very dangerous to infer similar boiling points from tlieir 

 known melting points. 



* Zeitschr. pliys. Chem., 42, 118 (1902); Compt. Rend., 131, 380 (1900). 



t These Proceedings, 37, 1 (1901). 



