*24 Tranxdctioiix. 



20 atmospheres, and the conditions would be similar to those of the 

 last experiment. The values of the conductivity are given below, further 

 explanation being unnecessary, the behaviour being quite similar to that 

 under experiment E. 



* In this case the decomposition of the hydrate took place at 3°, being accompanied 

 by the same jjhenomena of fluctation, &c., as in the last experiment, but on a smaller 

 scale. There \\'as also a second break at about 6°, the conductivity thereafter varying 

 much less with temperature. 



Conclusions. 



1. At both 0° and 18^^ solutions of COg in water are feebly ionized (about 

 ^i|^ in the case of a A solution at 18°) ; the conductivity increases with 

 the pressure or concentration of CO2 from 1 to 30 atmospheres, the propor- 

 tional increase, however, becoming less the greater the concentration. Thus 

 at 0° the conductivity at 4^ atmospheres is about double, at 14 atmo- 

 spheres about treble, and at 30 atmospheres less than quadruple its value 

 at 1 atmosphere pressure. 



This behaviour disproves the statement ascribed to Pfeiffer, that " change 

 of pressure produces no alteration in the conductivity." 



2. Temperature has a large effect on the conductivity, which at 8° to 9° 

 is increased per degree by about 2-9 per cent, of itself. 



3. After a CO2 solution under high pressure has been kept for a few 

 •days there is a tendency for the solution to acquire a kind of chemical fixity 

 ■or supersaturation, as shown by an increased value of the conductivity. 

 Probably the COg enters slowly into combination with water, the compound 

 in its turn being only slowly decomposed. 



4. A compound of CO, with water is formed under very varying con- 

 ditions at low temperatures and high pressures. 



{a.) The conductivity of the hydrate-bearing solution is considerably 

 less than that of the free solution under similar conditions. 



(6.) It crystallizes out with difficulty, there being a great tendency to 

 .a state in the solution resembling supersaturation with respect to the 

 hydrate. 



On one occasion the pressure was raised to 30 atmospheres and reduced 

 again to between 25 and 20 atmospheres before the hydrate would form ; 

 on another it was formed in an apparently arbitrary way after the pressure 

 had been for some time at 18 atmospheres ; it was also formed at 15 atmo- 

 .spheres, and on another occasion could not be induced to form at all. 



This tendency to supersaturation, and the element of chance consequent 

 upon it, probably explains W'roblewski's varying observations, giving 

 12-3 atmospheres in one place and 16 atmospheres in another as the pressure 

 luider which the hydrate formed at 0°. 



