1884.] The Conditions of Chemical Change in Gases. 57 



and by nitrous oxide. He concluded that the division of the oxygen 

 between the combustible gases changes per saltum, and that the ratio 

 between the carbonic acid and steam produced might always be 

 expressed by whole numbers, but not always by small integers. 



v. Meyer found that the presence of an inert gas, nitrogen, 

 favoured the formation of carbonic acid. He also found that more 

 steam was generally formed when the explosions were made in 

 narrow tubes than when they were made in wide tubes. 



In 1876 Horstmann* published a paper in a local journal at 

 Heidelberg in which he showed that Bunsen's results were vitiated 

 by his having exploded the gaseous mixtures in a eudiometer saturated 

 with aqueous vapour. At the high temperature of the explosion, 

 steam oxidises carbonic oxide, so that the proportion of carbonic acid 

 found after the explosion partly depended upon the initial tempera- 

 ture of the eudiometer, and the quantity of steam consequently 

 present. 



Horstmann showed that there was no discontinuous alteration in 

 the ratio of the carbonic acid to steam formed in the explosion 

 either in presence of aqueous vapour or without. 



In the same year, 1876, in ignorance of Horstmann's paper, I 

 repeated Bunsen's experiments, and came to the same conclusion as 

 Horstmann. I found that the aqueous vapour in the eudiometer 

 reacted with the excess of carbonic oxide at the high temperature 

 reached. I repeated the experiments, drying the eudiometer and the 

 gases carefully before exploding. My results gave no evidence of 

 any change per saltum in the division of the oxygen. 



In testing the dried gases it was discovered that an electric spark 

 does not ignite a dry mixture of carbonic oxide and oxygen. f A 

 trace of aqueous vapour was found to render the mixture explosive, 

 all other conditions being the same. A mixture of the gases 

 imperfectly dried with freshly fused potash was found to be unaffected 

 by a spark at pressures below 500 millims. At a pressure of 500 

 millims. the mixture was ignited by the spark and burnt slowly. When 

 left in contact with anhydrous phosphoric acid, either over mercury 

 or seaJed up in tubes, a mixture of two volumes of carbonic oxide with 

 one of oxygen may be subjected, under atmospheric pressure, to 

 powerful sparks from a coil or Leyden jar without exploding or 

 igniting. The addition of a trace of steam, of dry hydrogen, dry 

 ether vapour, dry pentane vapour, dry sulphuretted hydrogen, or dry 

 hydrochloric acid vapour, was found to render the mixture inflam- 

 mable, all the other conditions being the same. Dr. BotschJ has 

 published a paper in which he denies the non-inflammability of dry 



* " Verh. des Heidelb. Naturf . Med. Vereins," N.S., i, 3. 

 t " British Assoc. Report," 1880. 

 J " Liebig. Annalen," 1882. 



