382 Scientific Intelligence. 



the higher points, metallic spirals, were distributed at intervals 

 within the tube. If at a given time one of these salts is observed 

 to be melted and another not, the temperature of the tube ob- 

 viously lies between the two melting points. In the first experi- 

 ment, in which the temperature never reached 530° (since silver 

 iodide, fusing point 530°, was never melted and even lead chlo- 

 ride, fusing point 501°, was softened only in the middle of the 

 tube) the layer of charcoal being 66 cm., the evolved gas during 

 the entire hour was completely absorbed by potassium hydrate ; 

 showing that no reduction had taken place. In the second, in 

 which the temperature remained below 585° (the fusing point of 

 silver pyro-phosphate) a minute quantity of non-absorbable gas, 

 burning with a blue flame, was obtained. This temperature 530° 

 -586° marks the limit of the reduction. In the third, silver 

 iodide was completely fused, silver pyrophosphate sintered to- 

 gether, but lithium chloride (fusing point 602°) remained unal- 

 tered. The evolved gas contained 12"6 per cent carbon monox- 

 ide. In the 4th experiment the layer of charcoal was reduced to 

 10 cm. and it was found that no action took place even at 602°; 

 so that in the fifth experiment, carbon monoxide gas was 

 observed only when the temperature lay between the melting 

 point of potassium iodide (634°) and that of potassium bromide 

 (703°). The glass tube was then replaced by one of porcelain 

 8 mm. in diameter, which was heated in a Fletcher gas furnace,, 

 using an air blast. In the sixth the carbon layer being 30 

 cm. long, the temperature did not reach 814° (the fusing point 

 of Na 2 C0 8 ) 90 c. c. of gas were collected in 1 6 minutes and the 

 CO present in the evolved gas was 12 # 7 per cent. In the 7th,. 

 with a layer of charcoal 24 cm. long, and the temperature about 

 861° (the fusing point of Na„S0 4 ) the gas evolved at the rate of 

 1-7 bubbles per second consisted of 58*7 CO and 41-3 C0 2 . In 

 the 8th, the carbon layer was only 10 cm. long, silver was melted, 

 (954°) but copper was not (1054°) and the gas (1-2 bubbles per 

 second) consisted of 94-2 CO and 5*8 C0 2 . In the 9th, the carbon 

 layer was 57 cm. long, the temperature above 861° but below 

 954°, the gas passed at the rate of 3 bubbles per second and con- 

 tained 90-7 per cent of CO. In the 10th experiment a platinum 

 tube 5 mm. in diameter was used, the carbon layer being only 2 

 mm. long. In 4-25 minutes 70 c. c. of gas were collected (about 3 

 bubbles per second) of which 18*1 per cent was CO. In the 10th 

 experiment, the tube was filled with alternate layers of charcoal 

 and of copper turnings, the former amounting in all to 10 cm., 

 the latter to 34 cm. The temperature in the greater part of the 

 tube was between 861° and 954°. The gas passed at a rate of 2 

 bubbles per second and contained 36 '5 per cent of CO. From 

 these experiments the authors conclude : 1st, that the reduction 

 of carbon dioxide to carbon monoxide by carbon has its lower 

 temperature-limit at about 550°. 2d, that if according to theory,, 

 the reduction at a given temperature is proportional to the fre- 

 quence of the contacts between the carbon dioxide and the car- 



