K.—BOTANY 211 
short lengths of the door edge on either side of the middle piece. In 
either event the result is the same. The inward movement of the door 
under the pressure of a column of water consists first in a longitudinal 
flexure of the middle piece, which, as the reversal of door curvatures 
advances, becomes curved inwardly. The flexures then move into the 
lateral hinge areas together with the outer hinge, reversing the curvature 
of the whole door. At the top of its swing the door edge is simply curved 
in the arc of a circle (approximately), and the opening, as seen by the 
observer stationed so that he may look into the trap along its axis, appears 
circular, or nearly so, the threshold forming the lower arc, the door edge 
the upper. I have satisfied myself that this is a correct record of the 
movement, by taking motion pictures at high speed (160 frames per 
second), which I shall have the pleasure of showing you. This seems to 
be a simple enough matter, but, as it has been described otherwise, it is 
not superfluous to have spent some effort in getting at the facts. 
It is an interesting cell structure which allows such free movement, 
which is at the same time completely reversible, so that the door springs 
at once into its original position on the slacking of the water column. The 
whole movement occupies 4; second. Furthermore, one may play with the 
door with a not too sharp needle point or with a minute glassbead, moving 
the door hither and yon, with no harmful effect. ‘There is an admirable 
elasticity and flexibility of the tissues which fit its needs (Lloyd, 1932). 
U. resupinata-—In describing the above I have erred to a purpose in 
regarding all the species of the type U. vulgaris as submersed, freely 
floating forms. There are some exceptions, all (?) American species. 
U. resupinata will serve to illustrate a small group of species which are 
terrestrial to the extent that they grow anchored in the bottom of ponds 
where the water is shallow, or in similar situations (wet sand, etc.). In 
U. resupinata the traps are dimorphic, larger ones growing on the terete 
green leaves, and supplied with the vulgaris type of appendages (branched 
antennz and lateral bristles), and small ones on the underground parts, 
with appendages much reduced or absent. The middle piece of the 
door is somewhat more massive than in vulgaris, but, aside from this, there 
is no notable difference. 
Another exceedingly curious plant is U. neottioides, belonging to 
Kamienski’s segregate Avesicaria—a poor name for a plant with abundant 
traps. The plant grows in running water, attached to more or less solid 
substrata. The stolons in contact with hard surfaces grow and appear 
after the fashion of Podostemon. From these arise free stolons with 
linear, leaf-like branches, and near the axils occur the traps, which are 
peculiar in having the entrance and stalk at opposite ends. One may 
imagine this to be correlated with the movement of water, the trap being 
stream-lined and the entrance where the back-swirl occurs. The structure 
of the door and threshold is practically identical with that of U. gibba. 
The antennz are very small and bear one or two short branches.” The 
7 It would be gratifying to see good material of U. rigida, another species 
growing in ‘ swift-running water’ and attached to the substrata. It is said to 
be devoid of traps (Stapf, Flova of Tropical Africa). Later: the material has 
been seen at Kew, but no traps were found. 
