on the Ammonia-cobalt Bases. 259 
perature, however, the chlorid passes as fast as it is formed into 
ehlorid of Purpureocobalt, which may thus be the only final 
which chemically pure chlorid of cobalt and ammonia were em- 
ployed, the process of oxydation went on wef slowly, without 
the precipitation of any trace of sesquioxyd of cobalt. The 
iste or reactions to indicate the presence of chlorid or oxyd 
of Purpureocobalt or Roseocobalt. On boiling with chlorhydrie 
acid, however, the chlorid of Purpureocobalt was thrown down 
mm abundance, and no other substance could be detected in the 
chiorid of Purpureocobalt. We believe that in this case a com- 
bination of the oxyd and chlorid existed in the solution, so 
that, as we have already suggested in speaking of the chlorid of 
d 
equation 
d li reagents added 
ps not be able to overcome the affinity between the oxyd 
and chlorid. It is easy to see that by boiling with chlorh ae 
acid, the combination of oxyd and chlorid will give the chlori 
We have already mentioned that the chlorid of Roseocobalt 
4 Sach 
'S circumstance explains why only the chlorid of Purpureo- 
Cobalt 18 obtained by foiling a Soripletely oxydized ammoniacal 
Solution of chlorid of cobalt, even when this solution contains 
dently described the salts of that base, as if they correspon 
'o, and contained the same radical as chlorid of urpureocobalt. 
