Oils and Fats, 



14 



[July, 1911. 



answers every test, including ol course 

 Schimmel's " increased test," and which 

 possesses a geraniol-content ranging from 

 57 to 61 per cent according to the drums. 

 The price of this newly-introduced 

 variety of oil, the nppearance of* which 

 will probably be received with general 

 satisfaction, is about 10 per cent in 

 excess of the ordinary quality. 



In our October Report of 1909* we 

 referred to experiments which had been 

 made by J. F. Jowitt of Bandarawela, 

 Ceylon, in the cultivation and distil- 

 lation of Andropogon (Cymbopogou) 

 grasses. The oils obtained by Jowitt 

 have since been examined at the Impe- 

 rial Institute in London by Samuel S. 

 Pickles, the botanical origin of the 

 corresponding grasses being at the same 

 time again carefully determined by 

 Stapf. The result of this examination, t 

 which was published at the end of last 

 year, is especially interesting because, in 

 addition to the familiar citronella and 

 lemon-grass-oils, the oils from the wild 

 mana grassj as well as from delft-grass, 

 are here described for the first time. Up 

 to the present nothing was known 

 concerning the properties of these oils. 



Of mana- grass, which probably consti- 

 tutes the mother-plant of citronella- 

 grass, Stapf distinguishes two varieties : 

 Cymbopogon Nardus var. Linncei (typi- 

 cus) and C. Nardus var. confertiflorus. 

 In Ceylon the natives differentiate still 

 further, having a series of names for 

 each slightly different plant (compare 

 the table on p. 10), but Stapf traces all 

 these back to the two varieties men- 

 tioned. The oils of these two grasses 

 have no attributes which are character- 

 istic for each particular variety ; on the 

 contrary, in both case3 the same range 

 of differences in yield and properties of 

 the oils distilled at different times is 

 observable, and this is probably corre- 

 lated to the mode of cultivation and 

 manuring, as well as to the season when 

 the oil was distilled. 



The yields varied from0'06 to 45 per 

 cent.; they were at their lowest in grass 

 distilled in May, the succeeding months, 

 generally speaking, producing a larger 

 yield. All the oils ranged in colour 

 from bright to deep yellow, and with 

 few exceptions they had an agreeable 

 citronella-like odour, although mostly 



* Report, October, 1909, 42. 



t Cyrabopogon Grass Oils in Ceylon. Circu- 

 lars and Agricultural Journal of the Royal 

 Botanic Gardens, Ceylon, Vol. "V., No. 12, 

 November, 1910, 145. Also Comp. Bull. Imp. 

 Inst. 8 [1910], 144. 



J Comp. Report April, 1907, 32, 33. Also 

 see Report, October, 1909, 42. 



a little acrid. They gave a clear solu- 

 tion with 1 to T4 vols, of 80 per cent, 

 alcohol. When 10 volumes of the 

 solvent was added the solution showed 

 a more or less marked opalescence and 

 in two cases a slight turbidity ; only in 

 one instance (No. 14) did the solution 

 remain clear even when diluted. In 

 addition to the total- geraniol content 

 (geraniol citronella) the geraniol content 

 was estimated separately by the phthalic 

 anhydride method*, the citronella-con- 

 tent being calculated from the differ- 

 ence. This test showed that the relation 

 of geraniol to citronellal varied as irregu- 

 larly in these two grass oils as did the 

 other properties, the variation being 

 similar in each case. In some, such as 

 the Lenahatu oils, geraniol predominated; 

 in others, as for instance, the Maha- 

 Pengiri oils, citronellal. These and all 

 other details are best seen in the table 

 on the next page. 



Setting aside samples 11 and 12, the 

 constants of the two oil varieties lie 

 within the following limits : — 



Oil of C. Nardus 

 var. Linnaei. 

 dl5° ... 0-894 to 0-926 

 aD ... +4°54 to— 6°32' 

 Total 



gerniol... 43 5 to 64'7% 



Oil of C. Nardus 

 var. Confertiflorus. 

 0-900 to 0-929 

 H 12°12' to— 2°11' 



39-1 to 64-2%. 



As already stated above, there is 

 therefore no particular difference be- 

 tween the oils of the two mana-grasses. 

 Samples 11 and 12 show greater devi- 

 ations ; they attract attention because 

 of their high sp. gr., their very pro- 

 nounced dextrorotation and their low 

 total-geraniol content. These samples, 

 however, are abnormal, and they have 

 therefore been left out of account in 

 drawing up the limits of value, because, 

 according to Stapf, the corresponding 

 grasses are to be regarded as degenerate 

 forms. 



None of these mana grasses is thought 

 to possess any commercial interest. 

 Apart from the fact that the oil-yield 

 is generally small, the oils themselves 

 are of such poor quality that they can- 

 not compete with the citronella oils of 

 commerce. We cannot form an opinion 

 on the question how far this judgment 

 is correct, not having had an oppor- 

 tunity yet of handling oils of this kind. 



From Lenabatu-grass, now called 

 Cymbopogon Nardus, Rendle, lenabatu, 

 Stapf, Jowitt prepared four oils which 

 behaved like the ordinary commercial 

 oils. The yield varied from 0'42 to 

 0"56% and the constants lay with- 

 in the following limits : d 15 e 0'913 



* Report, October, 1899, 24. 



