— 46 — i 



It must however be borne in mind that with every method, when used 

 by itself, it is only possible to estimate one of the two constituents and 

 not, as has been the practice hitherto, to estimate both constituents by 

 means of subtraction from the so-called total-geraniol. The reason of 

 this is that in the case of citronella oils, in addition to geraniol and 

 citronellal, esters and alcoholic constituents also occur, and that there is 

 consequently a considerable difference between the so-called total geraniol 

 and the sum of geraniol plus citronellal. For this reason we intend in 

 future, when dealing with Ceylon as well as with Java-citronella oil, to 

 refer no longer simply to "total geraniol", but to "so-called total geraniol", 

 as has been done in the present article. 



A fairly accurate idea of the correct constitution of the oil may be 

 obtained by determining, in addition to the so-called total geraniol-content, 

 the geraniol-content by the phthalic anhydride method and the citronellal- 

 content by oximation. By applying this method we found the following 

 values: For the Ceylon oil I: 61,7 p. c. so-called total geraniol, 6,5 p. c. 

 citronellal, 34,4 p. c. free geraniol; for the Java oil: 83,8 p. c. so-called total 

 geraniol, 26,6 p.c. free geraniol, 46,6 p.c. citronellal; after 4 weeks 43,4 p.c. 

 citronellal. 



We have also saponified the oils and the mixtures, when the Ceylon 

 oils were found to show a comparatively high ester ester content, which, 

 however, is partly only apparent, because citronellal also uses up potash. 

 But in any case the results possess a comparative value, provided the 

 manipulation be always carried out under similar conditions. 



Ceylon oil mixture Acid v. = 0,6 Ester v. = 1,5 



Java oil mixture . „ „ = 0,9 „ „ = 2,8 



Ceylon oil I . . . „ „ =0,9 „ „ = 44,8 = 15,7 °/o geranyl acetate 



Ceylon oil II ... „ „ = 1,5 „ „ =33,6 



Java oil „ „ =1,5 „ „ =13,0. 



The above probably explains in part the difference between so-called 

 total geraniol and the sum of geraniol plus citronellal. 



In the table (page 47) given below are set forth the results obtained 

 by the various methods of estimation. This review may contribute to make 

 plainer the particulars given in the foregoing pages, while it also affords the 

 best means of forming an opinion of the value of each separate method. 



Durrans 1 ) has attempted to decide by experiment whether in the 

 acetylation of citronella oil the added sodium acetate plays the part of 

 a catalysing or of a dehydrating agent 2 ). He admits having failed to 

 arrive at a conclusive result, but he believes that from his experiments 



!) Perfum. and Essent. Oil Records (1912), 123. 

 2 ) Comp. Report October 1911, 35; April 1912, 51. 



