OXIMJDOSULPHONATES OR 8ULPHAZOTATES. 37 



NHO(SOJv), and XH(S03K),-N0(0K)(S0,K)^-M-itl, tl.e nitroo-en 

 qninquevalent. He believed that he had evidence of the latter salt 

 beino- always formed from the nitriJe. ^(SOgK).,, which for him way 

 trisnlphammonate, NH2(S()J\)3. Claiis's observations and conclusions 

 were examined at lenirth by Raschig and satisfactorilv refuted bv him: 

 they need not, therefore, be discussed by us. 



Having rejected Claiis's views, liaschio- gave the hvdroxvlamine 

 constitution to the neutral salts, but proposed another constiriition 

 for the two alkaline salts, in which hke Claus he made the nitroo-en 

 quinquevalent : 



(S03K),NK<^ >NH(SOJv), and (SO,K),N'K<^ )>A^Fv(SO,K),. 



The character of this constitution need iiof be noticed here : we shall 

 probably retui-n to it in our paper on the nalphazilatea. 



One argument given by Raschig for regarding as distinct the 

 acids of the neutral and alkaline salts was, that althougli acids at once 

 convert the latter salts into the former, alkalis effect the opposite 

 change only with difficulty. Weak alkali, he found, requires weeks 

 and to be in excess in order to effect the conversion ; while a con- 

 centrated solution of alkali must be, in order to act quicklv, boilino- 

 hot and also in excess. Our experience, recorded on page 31, lends 

 no support to this finding. The delay he observed in the appearance 

 of crystals of the salt, a fact to which he attached weight, must have 

 been due to the salt remaining in supersaturated solution, although he 

 himself dismissed this explanation of the matter as inapplicable, aver- 

 ring that this salt when obtained does not show the phenomenon of 

 supersaturation, without however testing the matter bv adding- ;i bit 

 of the salt to excite crystallisation. Tlie salt does not indeed super- 

 saturate its solutions to the degree shown bv the neutral potassium 



