173 



The tubes were kept at lo5°-]40° for four aud a half hours: uo gas was 

 over beyoud that due to simple expausion. aud uot the slis'htest trace of 

 either chloriue or ozone Avas found in any generator tulie. No oxygen 

 could lie discovered. The mixtures upon testing were found to contain a 

 considerable amount of chlorides. The temperature was raised to and 

 kept at iri(>° for three hours and no chlorine or oxygen was produced. 

 The quantity of chlorides seemed to be increased. At 173° all of the tubes 

 began to evolve oxygen and so long as this temperature was maintained 

 a steady but slow stream of oxygen was produced. No trace of chlorine, 

 chlorine oxide or ozone was produced as high as 18(t°. 



At this point the Avork was stopped for lack of time. Thus far a few 

 conclusions may be provisionally advanced: 



The conditions under which oxygen is ordinarily produced are not 

 ideal, and the moisture always present materially intluences the reactions. 

 This moisture makes possible the production of oxygen at a lower tempera- 

 ture than in the case of dry materials, also the formation of chlorine or 

 chlorine oxide, or both, as Ioav as 125° and before oxygen is evolved. This 

 may be due to hydrolysis of the potassium chlorate or chloride, thus allow- 

 ing oxidation by the manganese dioxide. It is possilile and even pix)bable 

 that no chlorine would be evolved at any temperature within the ordinary- 

 range of heating, if the materials Avere entirely free from moisture. In 

 such a case, McLeod's explanation must fail, since if it be true, the forma- 

 tion of free chlorine is a necessary step in the CA-olution of oxygen. 



This point. Avith others mentioned, Avill lie more fully investigated by 

 future work, and it is hoped that some facts of interest may l>e brought 

 out during the investigation. 



Action of Heat ox MrxTiiREs of Mang.4nese Dioxide With 

 PoT.ASSiu.M Nitrate and With Potassium BniHROMATE. 



J. H. Raxsoai. 



The fact that different mett^llic oxides mixed with potassium chlorate 

 cause the latter to evolve oxygen at considerably loAA'er temperatures than 

 Avhen heated alone has long been knoAvn, though the nature of the chemi- 

 cal action involved is not Avith certainty established. No work has been 

 done, so far as I am aAvare, to see what the effect of these oxides might 

 be on other .sub.stances decomposable by heat. 



