ANALYSIS, REFINING, AND COMPOSITION. 



25 



otherwise the physical properties of the residue would have been 

 higher in the third distillation than in the fourth. Since the original 

 heavy oils had been completely removed, the higher physical proper- 

 ties of the residue in the third and fourth distillations must be due 

 to the heavy oils formed during the course of the distillations. The 

 difference in the values of the physical constants between the last 

 fraction and the residue in figure 11 is therefore an approximate 

 measure of the change in composition due to the temperature of the 

 distillation. The lack of any marked difference in properties be- 

 tween the first and second fractions indicates that there was no 

 appreciable decomposition into sub- 

 stances with lower boiling points. 



STEAM DISTILLATION OF CRUDE TURPENTINE. 



In order to obtain, for comparison, 

 the change in composition due to dis- 

 tillation with steam, all the material 

 represented by figure 2, except the 

 residue, was twice distilled by steam 

 through a Geer dephlegmator. The 

 curves for these two distillations were 

 practically identical and are shown in 

 figure 12. Reasoning in the same way 

 as above the difference in physical 

 properties between the last fraction 

 and the residue indicates the change in 

 composition due to distillation. This 

 change is appreciable in the case of the 

 steam distillation, but is somewhat less 

 than in the direct distillation. 



.92 33 



U18 U80 



DIRECT DISTILLATION OF GUM TURPENTINE. 



Fro. 12. Alteration in physical prop- 

 erties of a crude wood turpentine 

 due to steaat distillation. 



The same distillation was made on 

 a gum turpentine (the one whose 



analysis is shown in figure 2). In this, as in case of the wood tur- 

 pentine, the third and fourth distillations gave nearly identical curves, 

 which are shown in figure 13. The amount of change in composition 

 is also very nearly the same as in the case of the wood turpentine 

 shown in figure 11. 



It is evident, then, that, under the conditions prevailing in the dis- 

 tillation shown in figures 11 to 13. the amount of change in composi- 

 tion during analysis is very slight, either in steam or direct distilla- 

 tion. This change apparently consists in the polymerization of a 

 portion of the terpenes to substances with higher physical constants, 

 58637 Bull. 10513 4 



