ANALYSIS, KEFINING, AND COMPOSITION. 51 



The following conclusions have been read KM 1 for the refining of 

 crude wood turpentines like those of group 1 : 



(1) A fractional distillation without chemical treatment is suffi- 

 cient for the production of good grades of turpentine and pine oil. 



(2) The separation of light oils with boiling points lower than 

 those of the turpentine is unnecessary. 



(3) Better separation of the turpentine from the pine oil, or, in 

 other words, a greater purity of the final products is obtained with 

 a column still than with an ordinary pot still. 



In the refining of crude wood turpentines like those of group 2 

 the following conclusions apply: 



(1) A chemical treatment is required before a distillation can pro- 

 duce a good grade of refined turpentine. 



(2) The color and odor of the refined turpentine obtained from 

 the chemically treated crudes by distillation are much improved by 

 the separation of the "light oils" with boiling points lower than 

 those of the turpentine. 



(3) Although no data are given to show what results might be 

 obtained in an attempt to separate the " light oils " by distillation 

 in an ordinary pot still, yet it is probable that these light oils could 

 not be successfully separated. 



COMPOSITION OF WOOD TURPENTINES. 



THE RELATION BETWEEN PRODUCTION AND COMPOSITION. 



If the crude turpentines are arranged according to the maximum 

 temperatures of their production, the order would be 4, 2, 5, 1, and 3. 

 The proportion of dipentene in the refined turpentines prepared 

 from these crudes increases in the same order as the maximum tem- 

 peratures of production. (See figs. 37, 38, 43, 14, and 33.) There 

 would have been a more proper comparison between maximum tem- 

 peratures and proportion of dipentene in the crude, rather than in 

 the refined, turpentines, but the analyses of the crude turpentines do 

 not in all cases indicate the proportion of dipentene present, as, for 

 instance, in crude turpentines 1 and 3. (Figs. 21 and 30.) 



As was to be expected, there is also a relation between maximum 

 temperatures and the amount of " light oils '' (with boiling points 

 below 157 C.) ; for example, in crude turpentines 1 and 3, produced 

 at (unknown) higher temperatures, light oils are present. 



A similar relation holds between the maximum temperatures of 

 production and the quantity of the products removable by treatment 

 with caustic soda. 



These relations between composition and temperature of production 

 are given in Table 2. 



