_ ON CYPERUS BULBOSUS VAHL, 861 
be sessile there, but much more often terminate very eis er but 
tough flagelliform thread- like. ladiats, which are three or four 
inches long and often numerous from the scale-axils. I am qnite 
at a loss to guess what is intended by Mr. Clarke's figures,* which 
are said to represent the rhizome of C. jemenicus, and are described 
at page 4 of his paper. The size of the bulbils is pretty uniform, 
about three-eighths of an inch long, and their structure somewhat 
like that of the pet, with two or three external hairs or blank 
papery scales; the inner ones, however, are thick, white, crisp and 
eshy, forming a solid sweet edible kernel. The Latter of 
el Coa 
The preparation = the ‘‘ Silandi Arisi’’ for food in Ceylon is 
simple enough. The bulbils are separated from the sand by a 
a fire, the sa 
They are eaten in this state, or more often made into flour by 
— them for three hours and then pounding; for the flour 
mall cakes or puddings are prepared. There is no aromatic 
sat in these little starchy bulbils, as in the tubers of C. 
rotundus 
From this latter abundant and pestilent weed, with its branched 
tuberous rhizome, there can be no difficulty in 
Tacemose, with the | onset spikelets well above those of the lowe 
branchlets, which are rarely elongated, the outline of the bellowed 
ns being saree ovoid, with the divaricate sessile spikelets 
sually in pai 
* Lic. tii, figs. 17, 18. 
