314 
plate accompanying Blane’s paper it is called ‘Nardus Indica,’ 
Blane being of opinion that it was the Nardus Indica of the 
ancients. .In 1795 Jones* disposed of that theory, and also 
established the specific name by which it is now generally known, 
except in so far as it is always spelt erroneously “ Jwarancusa.” 
The first letter of the name as used by Jones is distinctly J, not 
I. The substitution of I for J has altogether obscured the 
derivation of the name, which is from Jwara (fever) and ankusa 
(the hook used by the elephant driver to restrain his elephant), 
hence “fever-restrainer’’ as Blane and, more recently, Maddent 
have correctly rendered it. The grass was subsequently found by 
Dr. Boyd near Hurd war, and as his specimens were distributed with 
Wallich’s plants, it has become fairly well known. Its further 
history is of little interest, and may be gathered from the synonymy 
given on p. 804. In the Panjab it is known under the same name 
as C. Schoenanthus, viz., Khavi,t and is probably also used for 
the same purposes. Its affinity with C. Schoenanthus is, indeed, 
very great, and the two are, as Hackel has already pointed out, not 
always distinguishable with certainty. Theareaof C. Jwarancusa 
extends from the outer hillzone of the United Provinces into 
Kumaon and Garhwal, and westwards as far as Kashmir and the 
north-eastern Panjab. At high altitudes, as in Kumaon and §piti, 
or in the dryer parts of the Panjab, it becomes dwarfed and narrow- 
leafed and forms a “ transition state” to C. Schoenanthus. The 
latter is a characteristic desert plant, able to exist with a minimum 
supply of water. On the other hand, C. Jwarancusa is dependent 
on an, at least temporarily, abundant supply of water, and prefers 
the neighbourhood of rivers, or actually grows in the beds of 
torrents. It is not impossible that the distinguishing characters 
of C. Jwarancusa as compared with C. Schoenanthus, that is the 
robust state, the long, flat and relatively broad leaves, and the 
more composite panicles, are mainly due to edaphic influences. 
3. Cymbopogon Nardus, Rendle. 
(Andropogon Nardus, L.) 
Citronella Grass. 
FOUNDATION OF THE SPECIES AND EHARLY HISTORY.— 
If the history of Linnaeus’s Andropogon Schoenanthus is bewil- 
dering, that of his A. Nardus, the other aromatic Andropogon 
known to him, is perfectly clear. In this case Linnaeus has been 
quite consistent, and his references, with the exception of those to 
Mattioli and Baaee are unobjectionable. Moreover, there is still 
at the British Museum, in excellent preservation, Hermann’s 
specimen of ‘ Pengriman’ on which the species finally rests. 
Paul Hermann, chief medical officer in the Dutch Hast India 
Company’s service, resided at Colombo between 1672 and 16/7, 
and all his collections were made in the neighbourhood of that 
town. This fixes sufficiently the origin of the specimen which, in 
* Jones in Asiat. Research. vol. iv. (1795), p. 109. 
+ Madden in Trans, Edinb. Bot. Soc., vol. v. (1857), p. 138. Madden has given . 
here also another version, viz., Jwarandsaka (fever en, 
; Baden Powell, Punjab Prod, (1868), p. 383, 
