142 



chloride were absolutely useless for constitution -determinations, and 

 were not suitable for identifying purposes. 



In addition to the alcohol C 10 H 18 O, Semmler obtained by the 

 action of formic acid on sabinene a terpene C 10 H 16 (b. p. 169 to 173 ; 

 d 2 o° 0,829 to 0,831; a D -f- 13 to -f- 14 ; n D 1,471), which, on the 

 strength of its physical properties, he considers a pentamethylene derivative. 

 There would then have taken place in this reaction a disruption of 

 the three-membered ring in sabinene, and the semicyclic linking would 

 have remained intact. Attempts made to attach formic acid to the 

 hydrocarbons camphene and /?-(pseudo-) phellandrene which contain 

 such a linking, appear to confirm this view, as these bodies did not 

 attach formic acid. 



Semmler gives a representation of the reactions described above 

 in the following formulae: — 



H C CH, 



CH, 



CH 



CH — (HO)C 



A 



Sc^CH — 



H 2 C CH 2 

 Alcohol C 10 H 18 O (I) 



CH 



HC 



' )CH-C^ 

 CH/ 



CH 2 



)C = CH 2 



H 2 C CH 2 

 Pseudo-terpinene 



C = CH, 



H 2 C CH 



Sabinene 



CH 



CH 



CH 



CH 



CH 3 

 CH 



CH 



C = CH — 



H 3 C CH 

 Alcohol C 10 H 18 O (II) 



HC CH, 



)CH-c/ 



>C = CH 



H 3 C CH 



Cyclopentadiene 



The production of terpinene nitrosite has already been described 

 by Wallach 1 ) on a previous occasion. The behaviour of this body 

 towards alcoholic potash had led him to entirely different (not yet 



1 ) Liebig's Annalen 



(1888), 35. 



