365 



transferred to .1 small Buchner funnel, 5 cm. in diameter, upon 

 which is placed a closely fitting filtei paper. ["he non-combined 

 portion is then rapidly sucked away by the aid ol the filter 

 pump. Ilie thus dried cake is then ti red to a piece ol 



fine calico, the calico folded over and the cake spread with a 

 spatula to cover an area ol aboul 6 cm. by 8 cm., finallj folded 

 into a pad, which is placed between several layers ol absorbenl 

 paper and the whole strongly pressed for three rninuti ["he 



cake is then broken up with a spatula mi a glazed tile or "ii gla 

 transferred to a measuring flask with graduated neck, decomposed 

 with warm water, the cineol lifted into the neck of the flask, 1 ooled, 

 and when the separation is complete the volume measured. If the 

 original oil were diluted a correction is, of course, necessary. 

 We have obtained very concordant results with the same sample 

 of Eucalypus oil and now use the method constantly. 

 In order to test the range of accuracy with this method, using pure 

 cineol, and mixtures of cineol and the phellandrene fraction oi 

 Eucalyptus radiata and E. dives, the following tabulated results 

 were obtained. The cineol content in the mixtures employed had 

 the following range, 80, 66-6, 60, 40, 33-3, and 20 per cent. The 

 more satisfactory results were obtained when the mixtures richest 

 in cineol were diluted. To secure good results the acid should be 

 added slowly, the mixture made very cold, and the petroleum 

 ether rendered quite cold before adding to the cake. 



fhi cineol remaining from these determinations had specifii gravity at 15 = 0-929; rotation a D - 0-2 : 

 refractive index .it j" [-460; 



' n,. cineol from thesi determinations had sp itj << 1 , 0-929; rotation oil; refractive index 



at 20 1-4589 



A more perfect separation b\ the rapid method i> thus shown. 



