89 



the blacUlcr are observed to have resumed the position shown in 

 the middle i)hotograph of hgure 4. If, now, we prick the wall 

 the utricle expands instantly to tlu- form j)iclure(l in figure 4C and 

 remains expanded. 



The extreme rapidity of the trapping is made evident by a 

 motion picture, two successive frames of which are shown in 

 figure 7. It will be seen that in the upper photograph a worm 

 is outside the trap : in the lower the greater part of it is within. 

 It happened in this case that the worm was about to divide and 

 when the trap was sprung division was completed, one portion of 

 the worm entering, the other remaining outside ! 



It is not necessary to wait for an animal to spring the trap : 

 we may use the tip of a fine needle for the purpose. If the bladder 

 is detached from the plant and a slender glass rod is presented 

 to it, it will swallow it, actually jumping at anything that trips the 

 mechanism (figure 8) as observed by Czaja (1922a). We are 

 driven to the conclusion that an inrush of water carries the prey 

 into the trap. If tripped in air the bladder will swallow a bubble of 

 it (figure 6). When plants are lifted from the water it is usual 

 to find that many of the bladders contain air — the traps are sprung 

 as the water flows from the plant. Ekambaram and Brocher 

 actually heard the clicking sound made by the springing traps. 



We are now in a position to consider the facts related above. 

 There are two possible 'explanations' of the inrush of water. It 

 might be supposed that sudden turgor changes in the cells of the 

 lateral walls cause them to become less concave. Such a change 

 would not be unique for their are many cases recorded of just such 

 an alteration of turgor. The evidence, however, is all in favour of 

 the view that water is pumped out of the bladder during the transi- 

 tion from the tripped to the set condition. It has been shown 

 that a tiny hole in the wall of the trap or the insertion of a fine 

 bristle under the edge of the door is sufficient to prevent the setting 

 of the trap (Czaja). 



If there is indeed a pumping out of water leading to a 'negative 

 pressure' within the utricle it is necessary to assume that the door 

 is water-tight or that any leakage that occurs is so slight as to be 

 unimportant. When traps in the set condition are placed in water 

 containing india ink no leakage can be detected. We have watched 

 such traps for days and in no case have we seen leakage or noticed 

 an automatic springing of the mechanism. 



