102 BACON. 



The following experiments probably throw a little light on the whole 

 question of the sensitiveness of various terpenes to heat, lightj etc. 



Two sealed tubes were heated in a bomb furnace from 270° to 300°, one 

 containing limonene from elemi, the other limonene from orange-peel oil. After 

 five hours the limonene from elemi had decreased in its optical rotation by only 

 1°.3, while that from orange oil had decreased 14°. 7. Wallach " states that by 

 several hours' heating at 2S0° to 270°, hesperidene ((i-limonene from orange-peel 

 oil ) is changed to dipentene. Clover '° found limonene from elemi to be little 

 changed by heat. The above experiments would seem to show that both state- 

 ments are correct, and it seems probable to me that the change in the limonene 

 from orange-peel oil is due to a small amount of a catalyzing impurity in this 

 substance. I am strengthened in this supposition by the fact that whereas 

 limonene is relatively stable in light and air, orange-peel oil containing over 90 

 per cent of limonene is exceedingly sensitive to light and air, and very readily 

 change into a resinous mass. Some experiments which I made on Philippine 

 oranges point in the same direction. The peel of the native orange contains a 

 very large quantity of oil, but this is so unstable that in the course of two 

 minutes a rather thick layer of it will be completely converted into a solid resin. 

 I finally succeeded in obtaining a small quantity of this orange-peel oil by 

 squeezing it onto sponges soaked in sodium carbonate solution, and over 90 per 

 cent of the oil proved to be a d-limonene boiling at 174° to 176°, giving a tetra- 

 bromide melting at 104° and limonene nitrolbenzylamine melting at 93°. I con- 

 sider, therefore, that many cases of observed changes in terpenes on heat- 

 ing may not represent intrinsic properties of the terpene itself, but may be due 

 to small traces of impurity, for in layers of similar thickness, Philippine orange- 

 peel oil, consisting of 90 per cent of limonene, changes to a resin several hundred 

 times as fast as limonene itself. 



DESTRUCTIVE DISTILLATION OP THE "ELEMI" BESIN. 



The residue of elemi from which the volatile oils had been distilled 

 in vacuo was used in this experiment, as there is very little destruction 

 at the temperature at which the terpenes distill at ordinary pressure, and 

 I did not wish to contaminate the resin oil with the terpene. 



One kilo of this residue was distilled in a 2-liter Jena flask. It melts quietly 

 with very little darkening and no foaming, evolution of gas, or other signs 

 of decomposition. As the temperature is increased, considerable crj^eking and 

 evolution of water soon takes place and the whole mass becomes very dark, the 

 oil beginning to distill at a vapor temperature of 200°. The whole distillation is 

 very quiet with no foaming or frothing and very little evolution of gas. TTie oil 

 at first is green, then an intense, fluorescent blue and toward the close of the dis- 

 tillation almost white. The total distilled was 670 cubic centimeters of which 

 40 were the aqueous layer. Three hundred and ten grams of black tar were 

 left in the distilling flask. The oil, shaken out with dilute alkalies, gave 18 

 grams of acids, while after saponification with alcoholic potash a further 7 grams 



"Ami. (I. Chem., (Liebig), (1885), 227, 289. 

 " hoc. cit. 



