20 — 



up to io°/ , are marked with a green seal with the words "go per cent, 

 purity". The export of all oils adulterated to a greater extent is 

 prohibited. 



The export firms have the right to appeal against the results 

 declared by the inspector; in such case a further sample is taken 

 and sent to B amber; the result of his examination is final. For this 

 test certain fees are payable, which are refunded if an error is proved 

 against the inspector. 



Such stations for examining the oil would be established at Galle 

 and Colombo, the two export ports which come exclusively under 

 consideration for citronella oil. The inspectors must of course have 

 sufficient experience for carrying out the tests, and be thoroughly 

 reliable. They should be placed under the supervision of a higher 

 official, who would have the power of removing the seals at any time 

 and taking a second sample from the drums, which would then be 

 tested by B amber. A check of this kind would have to be made 

 frequently and unexpectedly. 



The expenses in connection with this new arrangement might be 

 met by a corresponding export tax. 



It is to be hoped that the Government measures indicated here 

 will now also become an established fact. This would no doubt meet 

 with general approval, in the interest of the honest trade in Ceylon 

 citronella oil. But in view of our present experience, we doubt whether 

 B amber's method will be found suitable for the examination in 

 question. 



In our previous Report (p. 29 et seq.) we have already stated 

 that B amber's method is useful for the qualitative test for adulter- 

 ations, although in our experience it is not free from the risk that 

 unadulterated oils will also frequently be objected to, because they 

 may not pass B amber's test; further down we refer again to such 

 an oil. But it appears to us risky that it is also intended to indi- 

 cate the adulteration quantitatively, as this method, as shown by us 

 at the time, does not give reliable results in this respect. It may be 

 anticipated that this will very often give rise to differences of opinion. 



As the real value of Bamber's method would lie in the quanti- 

 tative estimation of the adulterant, but as it does not give reliable 

 results, we scarcely believe that this method is to be preferred to the 

 "raised Schimmel's test" 1 ) recommended by us, which latter 

 consists of this, that the citronella oil in question, after the addition 

 of 5 °/ Russian petroleum, is tested for its solubility in 80 per cent, 

 alcohol. A pure citronella oil should under these conditions dissolve 

 in 1 to 2 vol. 80 per cent, alcohol, and even when as much as 10 vol. 



A ) Compare Report April 1904, 32. 



