were removed and the pressure drawn down. The 

 digester was then inverted by rotating in its seat 

 and the hot liquor drained through the relief line. 



The " direct extract" was dark brown in color 

 and the rosin was in perfect solution. For analy- 

 tical purposes the liquor was aliquoted. This 

 presented some difficulties; the liquor had to be 

 kept hot to prevent precipitation of the rosin soap 

 and the syrupy solution was not easy to handle. 

 Some of the aliquot portions were strengthened 

 while hot with additional caustic soda to make the 

 precipitation of resinate more nearly complete. On 

 cooling, the direct extract gave bulky precipitates 

 of fairly white rosin soap. The dark, supernatant 

 liquors were decanted and drained through cotton 

 or filter paper, and analyzed for unprecipitated 

 rosin. The soap, more or less contaminated with 

 humus and liquor, was dissolved in water and the 

 rosin determined in the usual way. 



The chips retained from one to one and a half 

 times their weight of liquor after draining. This 

 represented a serious loss of rosin. In order to re- 

 cover most of this soap, the chips were washed by 

 covering with hot water and boiling for some time. 

 The " first wash liquor" was drained as before, 

 cooled and aliquoted for determination of rosin 

 content. The color of the solution was dark brown. 

 No soap precipitated on cooling, although about 

 90 per cent could be precipitated by strengthening 

 moderately with caustic soda. 



To further clean the chips enough to allow 

 sampling and careful examination, the wood was 

 washed a second time by boiling with fresh water. 

 This " second wash liquor" contained but little 

 rosin. In actual practice the chips would be ready 

 for the introduction of pulping liquor after draining 

 the first wash liquor from the digester. 



The extracted chips after washing as above were 

 discharged from the digester and allowed to dry 

 somewhat. The chips had been appreciably soften- 

 ed by the alkali and darkened by partial attack of 

 the lignin. After thorough mixing, the weight was 

 taken and the mixture sampled. The sample was 

 reduced to small chips and boiled with successive 



30 



