CHEMISTRY. 391 



and therefore an indefinite compound is formed, which we call a solu- 

 tion. On examining the heat of formation of chlorides and of oxides 

 (as obtained by Thomsen) he finds that that oxide (or chloride) which 

 has the greatest heat of formation is the least soluble. Thus the heat 

 of formation of the chlorides of Mg, Ca, Sr, Ba increases in the order 

 of the metals as given ; and the solubility of the chlorides of these 

 metals decreases in the order given; again the heat of formation of the 

 oxides increases in the order Ba, Sr, Ga, Mg, whereas the solubility of 

 these oxides decreases in the same order. 



Mr. Durham contends that if his views be admitted, crystallization 

 can be satisfactorily explained, and regular structure follows : 



In such a compound as BaCl2 . 6H2O, the atoms of the molecule must 

 be arranged somewhat in this way : 



H.O 



His theory affords also a simple explanation of the freezing of water : 

 In water attraction exists between the H2 of one molecule and the O of 

 another, and vice versa; now, if the heat of the liquid be diminished suf- 

 ficiently'^, that attraction will cause cohesion of the molecules, and will 

 produce solid water or ice, the regular structure of which is caused by 

 the symmetrical arrangement of the atoms. Hence the various condi- 

 tions of matter, solid, liquid, and gaseous, may be due to the chemical 

 affinity of the constituent atoms, modified in various ways by the kinetic 

 energy of the system. 



These views are opposed to that which depicts chemical affinity as a 

 sort of arbitrary force acting in units or bonds; on the contrary, affinity 

 acts between all atoms of matter, whether of the same or difi'ereut kinds, 

 in varying degrees of intensity and quantity, producing combinations 

 of more or less stability, graduating from the so-called mechanical mixt- 

 ure of clay and water up to the irresolvable molecules of the perma- 

 nent gas, condensing by ks action the gas into the liquid, and the liquid 

 into the solid. In short, there are no hard and fast lines in nature, but 

 every phenomenon graduates by almost imperceptible degrees into an- 

 other, (ligature, xxxiii, 615.) 



A General Method for the Determination of Molecular Weights, by F. 

 M. Baoult- — The author has previously shown that tbe molecular 

 weights of orgauic bodies soluble in water can be determined by the 

 amount of reduction in the temperature of its freezing point. Further 

 investigations now enable him to ireneralize this method and to main-. 



