STRUCTURE OF BAMBOO LEAVES. 79 
their leaves. Karelstschicoff (p. 181) mentions them and so does Duval-Jouve (* Histo- 
taxie, pp. 306, 307). Max Güntz (p. 49) describes their strueture, and Marshall Ward 
(* Grasses,’ p. 68) states that, contrary to the general assumption, there are frequent 
though minute transverse bundles joining the parallel veins. In many systematic works 
the species of Bamboos are classified into those with transverse veins (Phyllostachys, most 
species of Arundinaria) and those without them. The fact is, that in the leaves of the 
second class the transverse veins are obscured by the chlorophyll parenchyma. They 
can, however, in all cases be made conspicuous by bleaching reagents or by scraping 
away part of the tissue. The following figures show the transverse veins in longitudinal 
sections: Pl. 11. fig. 3, Arundinaria Japonica; Pl. 13. fig. 25, Dinochloa andamanica ; 
and Pl. 12. fig. 18, Dendrocalamus giganteus. In transverse sections of other grasses 
the veins are shown in Duval-Jouve (‘Histotaxie,’ tab. 17. fi g. 11, Stipa altaica, and tab. 19. 
fig. 6, Aira latifolia). Regarding the anatomical structure of these transverse veins I 
am not quite clear, and would recommend the subject to the attention of younger 
botanists. Güntz (p. 49) quotes a paper by Alfred Koch (Strasburg, 1884), according to 
which they mostly consist of tracheids. In the midribs transverse veinlets frequently 
anastomose between vascular bundles. 
The bands or girders of sclerenchymatous fibres which, with the vascular bundles, 
constitute the longitudinal nerves and those which occupy a large portion of the midrib 
have already been noticed; it remains to mention those which we tind in the margin of 
Bamboo leaves. Here, along the margin, there is always a colourless band, the edge of 
which bears a line of thick-walled pointed hairs or (on the inner margin of the leaves of 
Melocanna bambusoides) a large number of closely-set slender hairs. This colourless 
band under the epidermis consists entirely of thick-walled sclerenchymatous fibres 
(Pl. 12. fig. 14, Dendrocalamus giganteus). It has already been stated that Bamboos, 
as a rule, have no girders of sclerenchymatous fibres above the lower epidermis under 
the bulliform cells. 
The leaves of Bamboos, like those of most other Grasses, owe their toughness mainly 
to the numerous girders of sclerenchymatous fibres which run under the epidermis 
through the entire length of the leaf. When a leaf is macerated in nitric acid and 
potassium chlorate, these fibres constitute a large portion of the mass of isolated 
elements. They are very gradually narrowed into sharp pointed ends, and the walls of 
the larger ones are distinctly pitted. Their length varies from 0'6 to 1 mm. 
GROUPS OF LARGE THIN-WALLED COLOURLESS CELLS. 
Under the heading **Parenehyme simple incolor,” Duval-Jouve includes what has 
here been called the parenchyma-sheath of vascular bundles, as well as the tissue of 
large cells frequently containing starch, which forms part of the midrib. Besides these, 
which have been discussed in the last section, we tind in many species, generally on both 
Sides of the midrib, groups of large thin-walled cells, without intercellular spaces snd 
without any apparent contents : Dendrocalamus strictus (Pl. 12. fig. 19), A a 
