EVOLUTION OF COLLOIDS 127 



mutual molecular affinity ; others unite 

 violently with great evolution of heat. 



As to molecular valency some pairs, or sets, 

 of molecules will unite one to one, others 

 two to one, and so on. Also, the relative 

 numbers so uniting are fixed by the chemical 

 nature of the reacting molecules, and, curiously 

 enough, sometimes there are two ratios of 

 union, just as it has been seen in the previous 

 chapter that the same element may have 

 two valencies, the lower one arising as a result 

 of partial self-saturation. 



The numerical values of molecular valencies 

 often reach far higher figures than those 

 of atomic valencies. It was pointed out that 

 atomic valencies probably did not exceed a 

 doubtful hexi-valency in a few elements. 

 But an undoubted molecular valency in one 

 saturated substance of twenty-four fold that 

 of the other substance united with it, is seen 

 in at least one group of molecular unions. 



Doubt has been cast upon the chemical 

 nature of these molecular unions, because it 

 is difficult in many cases to prove that there 

 is always a definite ratio in the numbers of 

 the molecules aggregating together. 



In all such cases it must be remembered 

 that it is difficult to obtain the point where 

 interaction between the substances is com- 



