298 F. M. HOWLETT. 



In each of these cases a faint attraction (strongish in (2)) for diversus was noted, 

 but apparently not for zonatus or ferrugiiieus. No other samples of eucalyptus 

 were attractive, and I therefore neglected result (1). 



I was unable to get analyses of the citronella used, and relied at first upon a 

 partially correct statement that the main constituents of the oil were citronellal, 

 citral, citronellol and geraniol. All these were tried many times, both alone and 

 mixed in various combinations and proportions, but none had any attraction whatever. 



In March of 1913 I got more accurate information regarding the constituents of 

 oil of citronella. In the new (1911) edition of Thorpe's " Dictionary of Applied 

 Chemistry '' these are given as follows : — • 



Geraniol, citronellal, camphene, dipentene, limonene ; with small amounts of 

 linalol, borneol, methyl eugenol, methyl heptenone, and sesquiterpenes. 



Now the previous experiments had shown quite clearly that the attraction of 

 citronella was not diminished by admixture with other substances of a more or less 

 similar character, and I felt justified therefore in ruling out of consideration all those 

 constituents in the above list which also occurred in reasonable quantity in any of 

 the oils which I had previously found unattractive. These were as follows : Geraniol, 

 citronellal, camphene, dipentene, limonene, linalol, borneol and methyl heptenone. 

 That is to say, out of the list given by Thorpe, the only definite constituent of 

 citronella which had not been tested in the series of trials of different oils was methyl 

 eugenol. This substance occurs in fair quantity in oil of bay, which had not been 

 included among the oils tried. I obtained some bay oil and exposed it, and was 

 pleased to find that it was very definitely attractive, five ferrugineus settling in the 

 bowl in the course of twenty minutes, although the weather was distinctly 

 unfavourable. It proved subsequently to be a considerably better lure than 

 citronella for hoih ferrugineus and zonatus. 



As already mentioned, some degree of attraction for diversus had also been shown 

 by {a) a particidar sample of eucalyptus ; I decided to neglect this result for the time 

 being, as the origin of the sample was unknown, and no other sample showed any 

 attraction ; (6) two mixtures containing clove oil. 



Clove oil consists of 70-90 per cent, of eugenol, but it also contains eugenol acetate 

 and furfurol, as well as methyl alcohol and salicylic acid. Bay oil contains eugenol 

 and methyl eugenol, as well as chavicol, estragol and two terpenes. Citronella contains 

 methyl eugenol, but its composition varies very much and it is largely adulterated. 



This seemed to confirm the connection between attractiveness and eugenol or 

 some compound connected with or derived from it. Since eugenol itself constitutes 

 from 70-90 per cent, of clove oil, it was to be presumed that the attractiveness lay 

 in some other compound, as otherwise we might expect clove oil to be more attractive 

 than citronella, whereas the reverse is the case. It seemed probable, then, that the 

 attractive substances were derived from eugenol and were relatively abundant in 

 citronella as compared with clove oil. The substance attractive to ferrugineus 

 might be methyl-eiigenol. 



As I was not at once able to obtain any methyl-eugenol, I experimented on the 

 effect of boiling clove oil with amyl alcoholic potash (KOH dissolved in amyl 



