CONSTITUTION OF NITRIC PEROXIDE. 3 



reversible relation with porphyrexine^ (C 6 H 9 N 8 ) :NOH, porphyrexide 

 seems to belong to the class of oxi me- peroxides. Piloty and Schwerin 



<lo not indeed recognise this. :md have instead come to the conclusion 

 thai the nitrogen of the group :.\0 in porphyrexide and in mononitric 

 peroxide musl be quadrivalent. In the light of Haga's experimental 



results, this view of the matter lias become untenable, since porphy- 

 rexide and the peroxylaminesulphonates appear to belong to the same 

 class of nitroxy-compoundsj as Piloty and Schwerin themselves have 

 pointed out. 



In the few cases in which it has been possible to determine 

 cryoscopically the molecular weighl of a glyoxime-peroxide, this has 

 been found to include :XO twice. This result may be owing to the 

 glyoxime constitution of these peroxides, but even so there is still no 

 peroxide, except mononitric peroxide, and possibly porphyrexide, the 

 molecular weight of which is such that it contains the group :XO 

 only once. The occurence of many nitroso-compounds in a colourless, 

 solid form, and in a bluish-violet liquid form, does not lend much 

 assistance in deciding the molecular weights of the two forms of a 

 peroxylaminesulphonate, since they contain not :N0 but 'XO. But it 

 must not be left out of sight that Piloty (who thinks otherwise, and has 

 been followed by Schmidt, Bamberger and others) has succeeded in 

 showing that the white form of these compounds contains the group 

 •\<) twice, and that the deeply-coloured modification contains it only 

 once. But here the latter is the form which must be treated as the 

 chemically active one, whilst the double weight found for the white 

 form has to be left uninterpreted chemically, as, for example, in the 

 case of the formula (C 8 H 17 NO) 2 for nitroso-octane. 



In the paper by Hantzsch and Semple (Joe. cit.) there occurs, but 

 in a foot-note only and without comment or explanation, the 

 punctuated formula <>X: (S0 3 K) 2 ; whether this is to lie regarded as a 



