Solarium. pentandria monogynja. 577 



less impure according to circumstances. One of the seeds is 

 well rubbed for a minute or two round the inside of the ves- 

 sel containing- the water, generally an unglazed earthen one, 

 which is then left to settle ; in a very short time the impuri- 

 ties fall to the bottom, leaving the water clear and, so far as 

 I have been able to learn, perfectly wholesome. These seeds 

 are constantly carried about by the more provident part of our 

 officers and soldiers, in time of war, to enable them to purify 

 their Mater. They are easier to be obtained than alum, and 

 are probably less hurtful to the constitution. 



3. S. colubrina* Willd. sp. i. 1052. 



Scandent ; tendrils simple. Leaves from oval to oblong, 

 obtusely acuminate, triple-nerved, polished. Berries many- 

 seeded. 



* The following note was written by Dr. Roxburgh several 

 years before this species was discovered in Bengal, and attached to 

 8. X us- vomica. I have therefore thought it best to insert it 



here.— W. C. 



There is a tree, but exceedingly rare on this coast, which the 

 Telingas call Naga musadi (Xaga, or Tansoopaum in the Telin- 

 ga language, means the Cobra de Capella, or Coluber naga of 

 Linnaeus, Tansoo means dancing, and pawn a serpent ; this sort 

 being famous for erecting its head and moving it from side to side 

 at the sound of music.) I have only once met with this tree, and 

 then it was pointed out to me by a Telinga Doctor. The tree had 

 been cut down, and carried away, from the only remaining one that 

 I could find, some shoots had sprung up. The leaves upon these 

 were opposite, short-petioled, obtuse lanceolate, three-nerved; 

 about two, or two and a half inches long, by three-fourths of an inch 

 broad. The petioles were very short, and connected at their in- 

 sertions by a membrane. I took up this root with the greatest 

 care, cut off the upper part from whence the shoots grew, and 

 planted it in my garden, but it soon perished. From the above 

 circumstances I am inclined to think there is a species different 

 from the third, or scandent species called Colubrina, as well as 



VOL.1. KU 



