Dixon — On the Vegetative Organs of Vanda teres. 451 



epidermal cells of V. teres. Each, cell is roughly cubical. Its outer 

 wall is domed and greatly thickened, and on the highest point of 

 the dome there is a knob (PI. xii., fig. 9) ; the outer four-fifths of the 

 wall is strongly cuticularized, while the inner fifth gives the reactions 

 of cellulose. The thick cuticle is raised up and forms a circular wall 

 round the stomata. These domed and greatly thickened epidermal 

 cells are found on the lamina and on the outside of the tubular petiole. 

 On the inner surface of the petiole the epidermis is composed of lenti- 

 cular cells, wHch is uninterrupted by stomata. 



Through the kindness of Mr. F. W. Moore, keeper of the Royal 

 Botanic Gardens, Glasnevin, I obtained material of Bendrohium tereti- 

 folium, which enabled me to compare the structure of its remarkable 

 cylindrical leaves with those of Vanda teres, and also to make out some 

 points in their development. The manner of growth of Dendrobium 

 teretifolium is quite different from that of Vanda teres. The stem is 

 thin and wiry, and when young is covered with membraneous sheathing 

 scale-leaves, which at first completely cover the internodes, but owing 

 to the subsequent elongation of the stem they persist only as mem- 

 braneous collars round the base of the internodes. At about the ninth 

 node a foliage leaf is formed and not a scale-leaf. This is cylindrical 

 and may attain a length of as much as 60 cm., and is placed in such a 

 way as to apparently form a continuation of the stem. When this 

 leaf dies, or sometimes even before, a lateral bud develops, on the stem, 

 which terminates in the foliage leaf. This lateral bud develops some 

 ei"ht or nine internodes, all of which are sheathed with scale-leaves, 

 in the same manner as the parent stem when young, and finally puts 

 forth a cylindrical leaf, which seems in every case to terminate its 

 growth. This bud is usually developed in the axil of one of the last 

 four scale-leaves, but, apparently by subsequent elongation of the stem, 

 is displaced, so that it often seems to spring from near the middle of the 

 internode above. 



A tranverse section made of one of the cylindrical foliage leaves 

 is almost perfectly circular in outline (PI. xiv., fig. 23). If the 

 section be made somewhere near the middle of the leaf there are 

 to be found five larger fibro-vascular bundles arranged round the 

 centre of the section ; outside these smaller bundles occur more or 

 less irregularly. Occupying the middle of the section is a curious 

 slit-like structure, about 0-15 mm. in length (PL xn., fig. 8). By 

 examining a succession of sections it is easy to determine that this 

 structure, which appears like a slit in a transverse section, is in reality 

 a long passage, the opposite walls of which are in close contact. The 



2 H 2 



