STRUCTURE OF BAMBOO LEAVES, 87 
perennial, but, as a rule, with a different duration of life. In Bambusa arundinacea, for 
instance, the first culms which attain full size are, say, 20 years old when they flower and 
die, whereas the culms subsequently produced year after year are younger. 
In common with many other Grasses, Bamboos are eminently gregarious, they readily 
get the upper hand in a forest and make it very difficult for other trees to come up. But 
there are many other gregarious trees and shrubs, Beech and Spruce in Europe, and 
Dipterocarpus tuberculatus, Roxb., Shorea robusta, Gaertn. f., several Oaks, Pines, and 
many species of Strobilanthes in India, which are eminently gregarious plants, which, so 
far as known, have no peculiarities in the structure of their leaves analogous to those 
which give to Bamboo leaves their peculiar character. 
The uniformity of the structure of Bamboo leaves is very remarkable, considering the 
great morphological differences in that order. Of Indian Bamboos four genera, Dinochloa, 
Melocalamus, Melocanna, Ochlandra, have no endosperm in the ripe seed, a discovery 
which will, I hope, soon be published by Dr. Otto Stapf, as the result of his researches, 
in continuation of his paper on Melocanna bambusoides. Two genera, Gigantochloa and 
Oxyienanthera, together with Ochlandra travancorica, Benth., have monadel phous 
stamens. Arundinaria and Phyllostachys have 3 and Ochlandra 6-120 stamens. In 
spite of this great variety in regard to important morphological characters, the 
structure of the leaves is the same in all genera and species, so far as they have been 
examined. 
I believe I am justified in recommending the thorough study of Bamboos to younger 
botanists. The main part of the work, however, must be done in countries where 
Bamboos are indigenous, and the first operation must be to establish facts. 
The Tribe PHARE&. 
Somewhere (I cannot now trace the paper) it has been stated that the leaves of Olyra 
have a structure similar to those of Bamboos. This genus Dr. Stapf, in ‘ Flora Capensis, 
vii. p. 319, has placed under a new tribe PAa«rec, and he has kindly informed me that 
three other genera, two of which appear under Panicee in Bentham and Hooker's 
‘Genera Plantarum, in his opinion belong to the same tribe. This has induced me to 
examine specimens of these four genera. ‘The material at my disposal was scanty, but it 
may be useful briefly to state the result of my observations. 
l. Olyra latifolia, Linn. A perennial grass 3-8 feet high, of tropical and subtropical America and 
Africa. Bulliform cells and apparent cavities as in Bamboos, most of the former as well as a large 
proportion of the linear epidermis-cells filled with silica. Short epidermis-cells filled with silica 
numerous and conspicuous. Upper epidermis smooth, lower uneven. Parenchyma-sheath aad 
: mestome-sheath of vascular bundles as in Bamboos. 
2. Olyra floribunda, Raddi. Brazil. Bulliform cells, parenchyma-sheath, and mestome-sheath of vascular 
bundles as in Bamboos. Apparent cavities narrow, the thin-walled cells in them containing 
chlorophyll. Chlorophyll-cells with distinct folds. Upper epidermis uneven. Midrib: three vascular 
bundles on ines underside. 4 
