258 MR. PERCY GROOM ON BUD-PROTECTION IN DICOTYLEDONS. 
Special Work *. 
The following observations refer chiefly to colleters. It may therefore be briefly stated 
that it is to Hanstein alone that we are indebted for our knowledge of colleters of 
Dicotyledons. He considers that colleters were always trichomes, whilst he supposed 
that the mucilage part of their secretion first appeared as a colligenous layer in the cell- 
wall. 
My own observations refer to plants belonging to Rubiacee, Apocynacee, Asclepia- 
dea, Guttifere, Dilleniacee. 
RUBIACEA. 
Cosmibuena obtusifolia is a native of tropical 8. America. The opposite stalked 
leaves have between them two large leaf-like stipules which possess no midrib. On 
closer examination it is seen that the bases of the stipules are continuous with a short 
outgrowth from the inner face of the leaf-base. So the stipules may be described 
as sheathing, but the axillary portions of the stipules are very small. This sheath bears 
a number of villi, which thus form a complete zone round the stem. 
Тһе Bud.—The apex of the shoot may be seen to be surmounted by two (rarely three) 
stipules glued tightly together by the “ blastocolla.” On separating these two stipules 
the next two leaves are seen directed vertically upwards, just within, with their inner 
faces closely adhering (Pl. LIX. fig. 1); these leaves by their growth force the apices of 
the stipules apart and so emerge. Thus the stipules do not protect the leaves to which 
they belong; they protect the younger leaves. The stipules subsequently drop off, and 
the scars are covered with cork. The actual growing-point is at the base of a relatively 
deep pit, the wall of which is formed by the concentric sheaths belonging to several 
successive pairs of leaves and stipules. The growing-point is of course covered with 
secretion. | 
Development of the Stipules.—In the earliest stage I succeeded in observing that there 
was an annular outgrowth of tissue common to leaf and stipule: this was thicker where 
the young leaves had already made themselves visible. Тһе stipule was merely repre- 
sented by a tiny triangular protuberance. 
Structure and Development of the Villi.—Each villus is, roughly, pear-shaped, and 
consists of a single layer of palisade-like secreting-cells, which cover a mass of thin- 
walled “ conducting parenchyma ” cells. Each villus arises as an elevation of a few cells. 
The young epidermal cells at this spot are, in general, more elongated at right angles to 
the surface, and possess larger and more deeply-staining nuclei than the adjoining cells 
(fig. 2). Тһе cells below them are smaller and stain more deeply than the rest of the 
paronohy ma. The epidermis of this slight protuberance divides radially only: these 
divisions and those of the subjacent cells lead to the formation of a considerable 
outgrowth. At this stage the young external secreting-cells possess large nuclei 
and a deeply-staining granular protoplasm. The size of the nuclei and the depth 
* : 5 . 
Unfortunately » owing to my leaving England, I was unable to complete the work as I should have wished. It 
was not till too late that км how easy it was to һе deceived by the use of Hanstein's reagent in testing for 
resin ; so that I cannot positively state that resin is present in all the secretions, though gummy mucilage is, 
