BY R. GREIG-SMITH. 79 



obtained from E. polybractea or E.sideroxylun; one sample came 

 from Mr. H. G. Smith, and the other from Mr. G. I. Hudson. 



The 01. Eucalypti, P.B., was purchased from a chemist, who 

 put it up in ounce bottles at Is., with a label which, after 

 enumerating the various ailments for which the oil is recom- 

 mended, continued with "The extract is a thorough deodorant 

 and disinfectant. A few drops on a cloth in a sick-room renders 

 the air refreshing. In fever-rooms the floor should be sprinkled 

 with it." 



The Essential Eucalyptus Oil was purchased from the same 

 druggist. Tt was sold in four ounce bottles at 2s. The oil was 

 of a pale straw-colour and was apparently a crude oil. The label, 

 inter alia, aftirmed "The oil is a thorough deodorant, disinfect- 

 ant, and an antiseptic of great value." 



The most striking point brought out by the investigation is 

 the irregularity in the action of the oils from the same tree. 

 Two specimens of E. auslraliana, rect., for example, gave widely 

 different bactericidal activities: one was almost inactive, while 

 the other was among the most active of the oils. E. polybractea, 

 crude, was almost a similar case. Since there is so much differ- 

 ence between specimens of the same oils, it is not surprising to 

 find that there is no regularity among the groups. One cannot 

 group the oils by their bactericidal activities. If aromadendral* 

 is the most toxic of the components of the oils, those members 

 of the groups which have it as a typical component should be 

 the most toxic, but such is not the case. The aromadendral oils 

 of "-roups nib and iv are no more bactericidal than the oils of 

 the other groups; indeed the oil of E. aliens, the only member 

 of group iv. tested, is a very poor disinfectant. 



Cineol was inactive when tested in oil, and if an ointment 

 were made with 01. Eucalypti P.B., which would be used by any 

 druggist dispensing it, the preparation would have no disinfecting 

 action towards the ordinary pus organism. 



■ I did not have enough aromadendral to test by tin- method adopted in 

 this research, but as 1 shall show in a following paper, it is a strong dis- 

 infectant in aqueous suspension. 



