1 80 Prof. Schonbein on the Allotropic Modifications of Oxygen, 



the peroxide of that metal = Ag0^, and the peroxide of hydrogen 

 = HO +©, catalyze each other into metallic sijver, water and 

 inactive oxygen. Other ozonides, such as PbO + 9 or MnO + O, 

 on being brought in contact with HO + 0, are transformed 

 into PbO or MnO, HO and 0. Now the peroxide of barium 

 = BaO + 0, acts like HO + 0. If you pour water upon an 

 intimate mixture of AgO, or AgO^ and BaO^^ a lively disengage- 

 ment of inactive oxygen will ensue, AgO, AgO^ and BaO^ being 

 reduced to metallic silver and baryta. In concluding the first 

 part of my letter, I must not omit to state the general fact, that 

 the oxygen disengaged in all cases of reciprocal catalysis of oxy- 

 compounds, behaves in every respect like inactive oxygen. 



There is another set of chemical phsenomena, in my opinion, 

 closely connected \dx\\ the polar states of the active oxygen con- 

 tained in the two opposite classes of peroxides. You know that 

 a certain number of oxy-compounds, for instance the peroxides 

 of manganese, lead, nickel, cobalt, bismuth, silver, and also 

 permanganic, chromic, and vanadic acids, furnish with muriatic 

 acid chlorine, whilst another set, such as the peroxides of ba- 

 rium, strontium, potassium, &c., are not capable of eliminating 

 chlorine either out of the said acid or any other chloride. This 

 second class of oxy-compounds produces, however, with muriatic 

 acid, the peroxide of hydrogen ; and it is quite impossible in any 

 way to obtain from the first class of the peroxides HO^, or from 

 the second chlorine. 



You are aware that, from reasons of analogy, I do not believe 

 in the doctrine of chlorine, bromine, &c. being simple bodies, 

 but consider those substances as oxy-compounds, analogous to 

 the peroxides of manganese, lead, &c., in other terms, as " ozo- 

 nides.^' Chlorine is therefore to me the peroxide of murium 

 = MuO -f 0, hydrochloric acid =Mu04-H0, and, as already 

 mentionegi, the peroxide of barium =BaO -F0, that of hydrogen 

 = HO -1-0, and the peroxide of manganese =MnO-f 0. Pro- 

 ceeding from these suppositions, it is very easy to account for 

 the difi^erent way in which the two sets of peroxides are acted 

 upon by muriatic acid. 



From reasons as yet entirely^ unknown to us, HO can be 

 chemically associated only with 0, and with no other modification 

 of oxygen, to constitute what is called the peroxide of hydrogen; 

 and in a similar way MuO (the hypothetically anhydrous muri- 

 atic acid of older times) is capable of being united only to to 

 form the so-called chlorine, which I denominate peroxide of 

 murium. If we cause MuO + HO to react upon BaO -H ©, 

 MuO unites with BaO, and HO^withO; but if you bring 

 together MuO -t- HO M'ith Mn + 0, part of MuO is associated 

 to MnO, another part to 0, water being eliminated, according 



