186 SECTIONAL ADDRESSES 
it is parallel with a component thereof; or, conversely, if it does lie 
smoothly, it is always oblique to the threshold. If, as in the former case, 
the edge is parallel to the threshold, there must be a region where, when 
the door is opened, it buckles. In the normal action of the door the point 
of buckling, which is necessary to its opening at all, is structurally pre- 
determined as to its position. How in particular this is accomplished 
will further appear. 
4. Histology of the Door—The tissues of the door are of two kinds 
(always of two courses of cells), hinge and middle piece. The former 
occupies a zone around the sides of the door, the latter a more or less 
extensive area at the middle of the lower edge and extending upward from 
this to some distance. Hinge tissue has a course of deep cells backed by 
a thin course (Figs. 3-8). It is extremely flexible, and can bend sharply 
through an arc of 180 degrees without damage. ‘This results, in large 
part at least, from the character of the deep cells, whose periclinal walls, 
inner and outer, are in the form of bellows. The anticlinal walls of both 
courses are reinforced by numerous cellulose props which prevent collapse 
on bending (Lloyd, 1932). In the middle piece the two courses of cells 
are of equal thickness, and while flexible the tissue has a certain rigid- 
ity and resists flexure in both directions equally. ‘The walls are very richly 
provided with cellulose props. The function of the hinge is to keep the 
door flexed outwardly as far as possible. If this outward stress is met by 
the threshold, the effect of the hinge is to exert a thrust of the middle 
piece against the resisting surface (cornuta type). Aside from the above, 
each type of trap requires special treatment. 
5. Trichomes.—In all species there are glandular trichomes on both the 
outer and inner surfaces of the trap. Those on the outer surface are 
usually sessile, of three cells (basal, mid- and capital cell—the last often 
doubled), the pattern of structure of all the glandular cells wherever 
occurring. These are found scattered over the whole plant surface, and 
are not peculiar to the traps, on whose interior surfaces occur trichomes 
of similar basic structure, but the outer two to four cells forming the 
capital are elongated or, if with a single-celled capital, there is a single 
sausage-shaped terminal or capital cell. Some species, therefore, have 
quadrifid and bifid trichomes (as Darwin called them) devoid of cuticle, 
the latter in the vicinity of the threshold ; or bifid and single trichomes, 
correspondingly placed. They may be few or very numerous, e.g. in 
U. longiciliata but six; in U. Jateriflora, sixteen; in U. vulgaris, 
hundreds. The form is very characteristic, but familiar to anyone who 
has only cursorily examined the interior surface of the trap of any species. 
The distribution of these trichomes is various. There is always 
a segregation of single and bifid, or only bifid trichomes: (a) on the 
inner surface of the threshold bolster, (b) on the surface above the inner 
margin of the threshold, and (c) on the general surface of the interior, 
which may be very thickly studded everywhere, or may be many fewer 
and placed in rather definite positions, e.g. they are frequently absent 
from the flanks and confined to the more peripheral, especially ventral 
region, as viewed laterally. What these differences may mean is obscure ; 
I am inclined to regard such as of no importance whatever. 
