212 SECTIONAL ADDRESSES | 
actuation of the trap is accordingly by pressure on the bristles, not seen 
by von Luetzelburg, who gave a good description of the plant as a whole. 
U. RENIFORMIS. 
This plant, well known in cultivation in our greenhouses, is a large, 
orchid-like species from the American tropics. Its habitat is in the 
wet moss and other compact epiphytic growth on stems of big grasses, 
trees, etc. The stolons are thick and white, and bear, for an Utricu- 
laria, very large reniform leaves, long flowering scapes with large, 
purplish flowers, and a lot of very minute traps, smaller than the traps of 
minute species, e.g. U. gibba. With this as typical are associated an array 
of species, severa of which are similarly large or, at any rate, of impres- 
sive size, such as U. longifolia, U. montana, U. Lundu, U. Endresu, 
U. Gliickii, with some much smaller, even diminutive, species, of which 
I have examined U. Dusenii, also seen in cultivation. They are all neo- 
tropical, none of the type occurring in the Old World. 
Except as to minutiz of structure, there is really very little to say in 
comparison with the vulgaris type. Instead of the elaborately branching 
antennz of the floating forms, in the terrestrial forms we are considering 
the antennz are merely tapering horns, curved backwards or forwards, 
from an overhang curved downwards over the entrance. In this respect 
they ally themselves with gibba rather than with vulgaris. The stalk of 
the trap, which in lateral view is well rounded, is usually closely approx- 
imated to the entrance. The door is like that of U. vulgaris, but has 
a more massive middle piece, more so in some species than in others. 
Whether this difference indicates anything as to the delicacy of action or 
not it is difficult to say. The traps of U. reniformis do not seem to engulf 
air by any means as readily as those of the floating forms, and this may 
advantage them, growing as they do merely in a very wet environment 
and not submersed. In all cases actuation of the trap is achieved by 
contact with four latch-lever bristles. I have already pointed out that in 
form the trap of U. caerulea looks like that of the reniformis type, the differ- 
ence being revealed in the absence of the door bristles and in the possession 
of bifid trichomes instead, as in the American species, of quadrifids. 
The threshold is so placed that the face of it is directed obliquely 
outwardly, giving a very characteristic form to the bolster of tissue which 
bears it. This position suggests that the trap is less easily actuated. 
Three peculiar American and one African species can naturally be 
mentioned in this connection, all sufficiently peculiar to deserve specific 
examination. These are U. Lloydiit, U. nana, U. longiciliata, and U. 
Kirku. 
U. Lloydii Merl in MS. is a small plant of terrestrial habit, bearing 
traps which resemble the globulariefolia type in the possession of a steep 
ramp leading up to the entrance from the level of the stalk. The traps 
are dimorphic, the two forms being, in certain details of structure, 
very strikingly different. This dimorphism was first noticed by Merl, — 
who drew my attention to it before I had examined the plant at all. 
There can be no doubt of the dimorphism, which, in lack of conclusive 
