rEBKUAET 2, 1906.] 



SCIENCE. 



175 



dom among the atoms of different sorts 

 which will reproduce the calculated degrees 

 of freedom. 



Dextrose 



Levulose 



Mannite 



Dulcite 



Eesorcin 



Pyrocatechin 



Pyrocatechin in 



alcohol 



Phenol 



" in alcohol 



37 



42 



51.5 



46 



29 



35 



26 

 33 

 23 



The supposition made before in the case 

 of the sugar molecules, assigning different 

 degrees of freedom to the two groups of 

 carbon atoms, must be made again in this 

 case. 



When we turn to the liquids which con- 

 tain other atoms than those of carbon, 

 hydrogen and oxygen, we find that for a 

 number of them, containing atoms of the 

 chlorine group, the calculated degrees of 

 freedom can be reproduced by assigning 

 one degree of freedom to each atom of car- 

 bon and of hydrogen, and three to the re- 

 maining atom or atoms. 



In the case of the ordinary solid oxides, 

 the general rule is that the calculated de- 

 grees of freedom can be reproduced by 



assigning 3 degrees of freedom to each 

 metallic atom and 1 or 2 to the oxygen 

 atom. 



The chlorides and sulphides of the metals 

 conform to the choice of 3 degrees of free- 

 dom for the metallic atom, and 3 also for 

 the atoms of chlorine or sulphur. 



A question of a peculiar kind and of 

 special interest comes up when we examine 

 the heat capacities of dilute aqueous solu- 

 tions of electrolytes. Julius Thomsen has 

 shown that these solutions exhibit the very 

 remarkable peculiarity that the apparent 

 molecular heat of the solute diminishes as 

 the dilution increases, so that it even be- 

 comes negative after a certain dilution is 

 reached. Now it is evident that this can 

 not be accounted for except by supposing 

 that the water is so associated with some 

 part, at least, of the solute as to have its 

 own heat capacity diminished. The heat 

 capacity is evidently an additive property 

 of the solution. When we think of the 

 solute as associated with some of the Avater, 

 we may conceive of the solution as made 

 up of the following parts: (1) the water 

 lying outside the groups of water molecules 

 affected by the solute; (2) the undisso- 

 eiated molecules of the solute with the 

 water molecules joined with them; (3) the 

 dissociated ions of the solute with the water 

 molecules jpined with them. Each of these 

 parts will have its own heat capacity. 



This conception of the composition of a 

 solution leads to a simple formula for its 

 heat capacity, of the form 



H — W + A{1 — p) +Cp, 



in which W is the heat capacity of the 

 water used to make up the solution, p is 

 the dissociation factor (it being understood 

 that one gram-molecule of the solute is 

 used in making up the solution) and A and 

 C are constants. The A equals the heat 

 capacity of the undissociated molecule 

 added to the difference between the heat 



