138 GIBBS. 



one hour, probably' because minute i^articles had been melted. Several 

 tubes after exposure were tested for the presence of carbon dio.xide by 

 dipping the point under a clear solution of barium hydroxide and allow- 

 ing a small quantity of the solution to be sucked in. Small quantities 

 of carbon dioxide were indicated in every case. About 2 grams of pure 

 phenol sealed, in a dry, purified atmosphere containing a little more 

 oxygen than atmospheric air, in a 300 cubic centimeter flask with two 

 outlet tul>es, was exposed to the direct sunlight for a month and then 

 tested for carbon dioxide by aspirating the gases through a barium 

 hydroxide solution. All necessary precautions to avoid the introduction 

 of the carbon dioxide of the atmosphere were taken. The tips of the 

 outlet tubes of the flasks were not broken until the rubber tube con- 

 nections were in place, a clear solution of barium hydroxide contained 

 in a U tube attached, and soda lime tubes and a wash bottle containing 

 a concentrated solution of potassium hydroxide, to prevent the entrance 

 of carbon dioxide, connected. A considerable amount of carbon dioxide 

 was indicated by the heavy precipitate of barimn carbonate. 



After the tube K had been sealed off, the atmosphere in the apparatus 

 ■was replaced by pure oxygen, the tube J being wrapped in black paper 

 to exclude the light. ISTo coloration of the phenol could be observed. 

 The tube J containing the phenol crystals was then packed in broken 

 ice and the oxygen current ozonized by connecting the terminals of the 

 ozonizer F with an induction coil. The reaction between the ozone and 

 the cold phenol crystals was at first very slow, no appreciable coloration 

 being visible for several hours. However, when it had begun, the action 

 proceeded with increasing velocit}^, no ozone escaj)ing from the reaction 

 tube containing the phenol after the reaction was well under way. How- 

 ever, before that time a small quantity of supercooled, liquid jjhenol 

 farther along in the apparatus, at the point /, colored instantly upon 

 the contact with the ozone, but none of the latter escaped at N, as was 

 proved by repeated tests. In both cases the first color was bright yellow ; 

 this color gradually extended throughout the mass and changed from 

 yellow to pink and finally to a dark, reddish-brown with gradual lique- 

 faction of the crystals. 



Another experiment was carried out in the same way, except that the 

 phenol in the tube J was not cooled with ice and was in the form of a 

 supercooled liquid. In this instance the phenol colored with great rapid- 

 ity and no ozone escaped to the tube /. The liquid phenol in /, in this 

 experiment, was not protected from the diffused daylight of the room, 

 and nevertheless remained colorless after exposure to the rapid current 

 of oxygen throughout the entire experiment which lasted for fourteen 

 •days.^* On bubbling the oxygen escaping at T through a potassium 



" The phenol was subjected to tlie action of tlie oxygen current for this length 

 of time for the reason that the ozonizer \\as rather feeble and produced only a 

 small degree of ozonization. 



