422 



of those lines. It is thus seen that only about one-fifth of the oils so far 

 investigated appear to contain the constituent which on alteration forms this 

 deposit, and these were all distilled from species belonging to closely related 

 groups, the principal oil constituents of which are cineol and pinene. 



That the deposit consists of a polymerised body is shown from its general 

 character and exceedingly high molecular weight ; it also shows a fairly constant 

 composition with most species in which it occurs, as is evident from the results 

 of the analyses, and the closely agreeing molecular weight determinations. 



Its composition and chemical behaviour, as well as its alteration when 

 heated, apparently show it to be associated with the neutral resins. After three 

 or four years the polymerisation appeared to be complete, as no further 

 formation of the deposit took place in the filtered oils after that period. 



EXPERIMENTAL. 



The substance was purified by filtering off the deposit which was thus 

 obtained as a jelly-like mass ; it was spread upon porous plates to absorb the 

 adhering oil, finely ground, dissolved in chloroform and precipitated by the 

 addition of an excess of alcohol. The process was repeated several times, when 

 in most cases the finely-ground powder was quite white and had little cohesion. 



When purified as above all the samples contained minute traces of 

 manganese or alumina, and occasionally iron, but neither lime nor magnesia 

 was detected. In those cases where a reaction for iron was obtained the powdei 

 was slightly tinted, and it did not appear possible to remove this by repeated 

 solution and reprecipitation. 



The powder was quite neutral, was insoluble in alkalis and in all the usual 

 organic solvents except dry chloroform. It was soluble in sulphuric acid with 

 the formation of an orange-brown solution, but on addition of water was again 

 precipitated apparently unchanged. 



The melting point was very high, in most cases about 280-290 C., 

 although that of E. punctata melted fairly well at 265-270 C. The melted 

 substance was a brown brittle resin, mostly soluble in alcohol and in hot aqueous 

 alkalis ; the portion soluble in alcohol melted at 125-130 C. When heated in 

 the tube to slight decomposition it evolved an odour reminding one of an 

 aromatic resin, and gave off a small quantity of a volatile liquid which was 

 acid to litmus. 



The empirical formula obtained with the deposits of E. oleosa, E. populi- 

 folia, and the undetermined species from Queensland, was C 7 H IO O in each case, 

 but with those of E. punctata and E. occidentalis it was C 8 H I2 O. Molecular 

 weight determinations were carried out by the Landsberger boiling point method, 

 using chloroform as solvent. The results were fairly uniform for material 

 of this character, although the molecular weight was so high. 



With the deposit from E. globulus the following was obtained: 1-0416 

 gram, gave the figure 1-368 for ev as the mean of several readings, so that the 

 molecular weight was 1979. 



With that from E. occidentalis 1-1256 gram, gave the figure 1-492, molecular 

 weight = 1960. With that from the undetermined species from Queensland : 

 1-685 gram, gave the figure 2-2, molecular weight = 1992. Calculating from 

 the empirical result with the deposit from the Queensland species, the molecular 

 formula is C J!!6 H l8o O l8 , and with that from E. occidentalis C I28 H I9J O I6 . 



The specific gravity of the powder was 1-08. It also gave a saponification 

 number 87, and an iodine value 105, although no absorption of bromine was 

 shown in the chloroform solution. 



