— 123 — 



As it had been found that the separate fractions reddened magenta 

 solution which had been decolorised by sulphurous acid, we first tried to 

 separate the ketones with sodium bisulphite solution from the aldehydes 

 present. The bisulphite-compound formed a mucilaginous mass, easily 

 soluble in water. The portions which had not reacted with the bisulphite 

 were removed by extracting with ether and the aldehyde was separated 

 out from the bisulphite compound by boiling with soda. This purification- 

 process was repeated several times in order to obtain a product which 

 should be as free as possible from other combinations. The aldehyde 

 which was obtained after repeated fractionation had an aromatic odour 

 and possessed the following properties: — b. p. 86 to 87° (6 mm.), 222 

 to 224° (761 mm.), d 20 o 0,9938, <* D — 38°48', n D20 o 1,48393, mol. refr. found 

 47,20, calculated for CnH 16 46,71, for CnH 16 /f 48,38. 

 0,1532 subst: 0,4458 g. C0 2 , 0,1375 g. H 2 0. 



Found Calc. for CnHi O 



C 80,54 °/o 80,49 °/o 



H 9,97 °/o 9,75 °/o. 



A semicarbazone was readily formed; repeatedly recrystallised from 

 alcohol (in which it dissolves sparingly), it melted at 223 to 224°. The 

 oxime was liquid, b. p. 135 to 137° (7 mm.). 



The aldehydic character of the compound CnHi 6 was proved by its 

 behaviour towards sodium bisulphite solution and magenta-sulphurous acid. 

 By oxidising it with chromic acid in a solution of glacial acetic acid we 

 arrived at the acid corresponding to the aldehyde; but this reaction gave 

 by-products which prevented the experiment from proceeding smoothly. 

 With ammoniacal solution of silver nitrate (1,5 g. aldehyde to 3 g. nitrate 

 of silver), oxidation was effected very readily, and an equally favourable 

 result was obtained by exposing the aldehyde in dilute soda solution to 

 the oxidising action of the oxygen. The acid, repeatedly recrystallised, 

 melts at 91 to 93° and is optically active; [«] D — 33° 17' in 13,8% alcoholic 

 solution. 



0,1496 g. silver salt: 0,0558 g. Ag. 



Found Calc. for CnHi 5 2 Ag 



37,29°/o 36,91 °/o 



Judging by its behaviour the aldehyde is probably identical with nor- 

 triq/doeksantalal, which was obtained by Semmler 1 ) as a decomposition- 

 product of tri-c^/doeksantalal. In order to obtain further confirmation of 

 this theory, we decomposed the aldehyde into teresantalic acid. 12 cc. of 

 the aldehyde was boiled for 1 1 k hours with 24 cc. acetic acid anhydride 

 and 2,5 g. sodium acetate in an acetylation-flask. The reaction product 

 was poured into water, absorbed with ether and fractionated. The resulting 

 product (b. p. 110 to 130° at 10 mm.) was not uniform, but was probably 



*) Berl. Berichte 43 (1910), 1890. 



