The next step is to thoroughly oxidise the precipitate. This cannot be satisfactorily accomplished by utilising the oxygen of the atmosphere, as is done in other cases, because that method entails the production of a certain amount of oxide of iron, which prejudicially affects the purity of colour of the finished article. Of the chemical oxidising agents which are available, the most satisfactory in point of cost and efficiency is chloride of lime (bleaching powder). For each cwt. of green copperas, mix about 20 lb. of bleaching powder into a thin cream with water, and add it, in small quantities at a time, to the precipitate, constantly stirring so as to ensure the absorption of the whole of the chlorine by the blue. Without the addition be made gradually and under agitation, the chlorine will be generated more quickly than it can be absorbed, entailing a waste of gas and a noxious vapour to be breathed by the workmen. Sometimes the bleaching powder is added at an earlier stage, viz. to the green copperas solution, and in that case the blue assumes a violet tone. After the addition of the bleaching powder solution to the bluish-white precipitate, it is acidified with hydrochloric acid, which develops the blue. When the whole has settled, the supernatant liquor is drawn off, and the blue powder is well washed and strained on a filter, then placed in pans and dried very gradually indeed in the dark, at a temperature never exceeding 130° F. The slower the drying the better is the gloss of the pigment. It is most essential that iron be excluded during the final grinding operation, or it may cause ignition of the mass, and its conversion into oxide of iron would speedily follow. It has been proposed to treat the white precipitate (obtained in the usual manner from green copperas and yellow prussiate) by the chlorine contained in aqua regia (nitro-hydrochloric acid). The copperas, however, must be as free as possible from basic sulphate (oxide), which is ensured by keeping a little metallic iron in the acid solution of copperas. It is also desirable to effect the precipitation with crude prussiate, so as to avoid absorption of oxygen and premature development of the blue colour. Habich considers that the mistake is generally made of using too little copperas, and he has found that when 90 lb. of copperas have been added to 100 lb. of yellow prussiate, a drop of iron solution in the filtered liquor produces no precipitate, while the white precipitate has carried with it a certain proportion of prussiate, which can be washed out. He therefore proposes to avoid this waste by pouring the copperas solution into the prussiate solution, with constant agitation, till no further precipitate goes down, then adding one volume of the copperas solution equal to one-ninth of that already used. After fifteen minutes stirring, it is certain that all the prussiate carried down is decomposed.