Flora of the Ceylon Littoral. 31 
ones at the base, which thicken with and often become fused with 
the trunk (Fig. 8)—and is therefore unable to grow in localities so 
exposed to the wind and tide. Its roots, however, are provided 
with another adaptation which serves the same purpose as the 
stilt-roots from the point of view of aeration, though destitute of 
mechanical function. The roots of Druguiera grow horizontally 
close below the surface of the mud, and in places bend sharply 
up and then down again so as to project above the surface in the 
form of a knee (Figs. 8 and 9). Here again the whole of the root 
has a thick spongy cortex into which the air can penetrate through 
the part of its surface above the mud. The cortex of this aerial 
portion of the root always splits extensively, placing the lacunar 
system in open communication with the air. According to Karsten 
(loc. cit. p. 55), the exposed portion of the cortex bears numerous 
closely-crowded lenticels. The buried portions of the root are 
richly provided with laterals, a number of which penetrate down¬ 
wards into the soil and help to fix the tree. The viviparous 
embryos of Druguiera, resemble those of Rliizophora, but are cigar¬ 
shaped, and never more than a few inches in length (Fig. 10). 
The commonest species of these two genera, viz.: R. conjugata , 
R. mucronata and B. gymnorhiza, form, as has been said, the great 
bulk of the mangrove vegetation of the Ceylon coast, at least of 
the part examined by us; though they are never found, so far as we 
have seen, on the actual sea-coast line, but are confined to the 
edges of the estuaries and lagoons connected with the sea by 
narrow openings. In the lagoons especially, as at Negombo and 
Balapitiya, the mangroves are extremely well developed, extending 
close up to the mouth (in the latter locality, within about 50-feet of 
the sea). In the estuaries of the rivers, however, as at Bentotta, 
and Kalutara, part of the estuarine shores next the sea, where there 
are villages, have been frequently cleared of the mangrove vege¬ 
tation, and artificial banks made in order that coco-nuts and other 
trees might be planted; and in such cases one frequently has to go 
some distance up the river before the untouched mangrove 
vegetation is met with. 
Schmidt (loc. cit. p. 14) points out that Rhizopliora conjugata 
is very indifferent to the quantity of salt in the water ; we also 
have observed Rliizophora (and Druguiera) in localities in Ceylon 
where the water was practically fresh (e.g. in the further portions 
of the lagoon at Negombo and in the lake at Balapitiya). The 
character of the plankton collected in the localities named bears 
quite a freshwater stamp. 
