551 



it can take into accounl not only the course of the isotherm for 

 small concentrations, but tiie whole course. And besides it has the 

 advantage that it does not bind itself beforehand b^' a preconceived 

 opinion on the question which can at present mostly not be decided, 

 of what nature the sorption phenomonon is (solid solution real sur- 

 face adsorption, dissociable chemical combination, or two or three 

 of these possibilities at the same time). The simultaneous determi- 

 nation of the two curves does, however, supply a collection of facts 

 important for the decision of this question, which every theory has 

 to take into account. 



2. Experiments. 



The purest animal carbon of MbRCK was used for the investigation 

 It was placed in air-dry condition in a wide-moufhed glass jar; its 

 water content was determined at 230° C. after 3 hours' drying. It 

 is not impossible that in this way the water percentage is found 

 slightly too high, the weight of the carbon having possibly been 

 slightly diminished by oxidation. As in most hygroscopic substances 

 of this kind it remains somewhat arbitrary rvhat is considered to be 

 "dry" substance. 



For the determination of the sorption heats quantities of from 

 5 to 12 grammes of carbon were weighed in air-dry condition, 

 which can easily be done accurately, as the substance is not parti- 

 cularly hygroscopic in this condition ; the carbon cannot be weighed 

 accurately when quite dry. In crystallisation dishes these samples 

 of carbon were brought in exsiccators over sulphuric acid-water 

 mixtures of different strengths; we then waited till equilibrium had 

 been approximately established. In this way samples of carbon were 

 obtained in which the water was very uniformly distributed. Wliere 

 the water-content of the air-dry carbon was known, the increase 

 or decrease of weight of the sample of carbon yields its water con- 

 tent at the known vapour tension. 



This caibon was jdaced in a glass tube, which was closed with 

 a tight-fitting rubber stopper ami placed in a calorimeter vessel 

 filled with water. The experiments were made in a room in which 

 the temperature was particularly constant. After temperature equili- 

 brium had been established, the course of the thermometer was 

 followed; then the contents of the tube were emptied into the water 

 of the calorimeter vessel, after which the temperature was again 

 observed. After from 2 — 4 minutes the generation of heat did not 

 increase appieciahly any longer. 



