345 



the submersed spadix, and rise to the surface of the water, 

 where they float, wafted by every breeze, till they have 

 fertilized the female flowers. Each of these male flowers is 

 6-valved, the two external valves being larger and coloured, 

 the four internal smaller, white or cream-coloured, all 

 reflexed. Stamens 2. Filaments diverging. Anthers obtuse, 

 bursting transversely across the summit ; Pollen composed of 

 diaphanous, globular granules. — Female spatha sheathing, 

 about as long as the germen, bifid. Germen inferior, ovate 

 at the base, tapering gradually upwards into the long, fili- 

 form tube of the perianth, which, as well as its 3-partite 

 li?nb, is reddish; segments obtuse. Stigmas protruded just 

 above the limb of the perianth, 3-partite, the segments 

 cuneate, irregular, often bifid, papillose, white. Pericarp 

 enclosed within the spatha, membranous, rounded or 

 slightly compressed, tapering to a point. Seeds numerous, 

 attached to the inside of the pericarp by small filiform stalks. 



[I quite agree with Dr. Wight in considering this to be a 

 true Vallisneria, although the stem is elongated, and the 

 flowers sessile. The same gentleman observes that all the 

 four Vallisnerice of Dr. Roxburgh [V. alternifolia, octandra 

 PI. Corom. V. 2. t. 165, V. verticillata* and spiraloides,) 

 differ very much in their fructification from each other. 

 Figures of them all will soon be provided for the present 

 Work. 



V. alternifolia seems to be a frequent plant in many parts 

 of India, particularly in Madras and about Calcutta, growing 

 in pools of fresh water, and flowering during the rainy 

 season. In Hindostanee it is called Jangi {Naididpanee in 

 the Tamul, Dr. Wight) ; and according to the late Dr. 

 Hamilton, f it is used for the purpose of refining sugar. 

 " It is well known," says this author, " that one of the 

 most common processes for refining sugar, is by filtering 

 water slowly through small quantities of it, contained in 



* Probably the Serpicula verticillata, Roxb. Corom. v. 2. t. 164. 

 f In Brewster's Journal of Science, v, 1. p. 34?. 



