Francis E. Lloyd — 242 — Carnivorous Plants 



brane, the cuticle, which excludes solutes. The four-armed hairs of 

 the internal surface absorb water more rapidly than it can find its 

 way in. Nold (1934) had advanced the theory that the potential 

 difference existing between the outer and inner surfaces of the traps 

 accounts for the movement of water outwardly. He localized this 

 difference in the only parts of the walls free of cuticle, namely the 

 outer spherical and the inner four-armed glandular cells, the only 

 ones through the walls of which the water can pass. That this shall 

 pass outwardly, he believed, is assured by the difference in pH at 

 these places. At the inner surface this is 6.2 and for the outer 6.6, 

 determined by the Folin colorimeter; or 7.5 and 8.2 with the quin- 

 hydrone electrode, differences which seem non-significant for water 

 movement. Nold seems to have shown, however, that the loss of 

 water from the trap increases inversely with the pH. of the outer 

 medium, the normal behavior taking place at ^H 5-7. The traps 

 are damaged at lower and higher pH values. Yet it has been shown 

 that Utricularia can prosper in water of ^H 4 (Emil Wehrle, 

 1927) and U. minor in "weakly alkaline water" (Nold). In any 

 event, since a difference of potential between inner and outer sur- 

 faces is known to cause a water loss, but since also organs are known 

 which show such differences without water movements, it is scarcely 

 possible to regard Nold's hypothesis as proven. This judgment is not 

 weakened by inspection of the evidence advanced. 



There is a further point in the mechanism of the trap about which 

 opinions had been expressed previous to 1929, without the provision 

 of proof. I refer to the method by which the watertightness of the 

 door is procured. That watertightness is a necessary condition for 

 the successful action of the trap was first recognized by Brocher, 

 and by his successors in investigation, all of whom placed faith in 

 the contention that it is due to the tight appHcation of the door 

 selvage to the threshold, aided by the mucilage present. Withycombe 

 realized the inadequacy of this explanation, and supplemented it by 

 arguing that the door edge rests against the outer edge of the "middle 

 layer," the pavement, seeing in this a valvular seat. An examination 

 of the action of the trap and certain details of the emplacement of 

 the door led me to suspect that the explanation was a lame one. 

 This led to the discovery that the entrance of the trap is guarded, 

 not merely by one valve, the door, but by two, the door and a second 

 valve, the velum, attached to the threshold and finding its seat against 

 the door selvage, thus blocking the chink. This second valve has 

 been seen in some 75 species, in slightly various form to be sure, 

 but always present {24, 25). This discovery led to a minute examina- 

 tion of the structure of the trap in all material available from various 

 parts of the world. The results of this survey, made on many living 

 species and on still more preserved ones, he in the field of anatomy, 

 which in the presentation thus far has received only minor mention. 

 This is now to be taken up in the following, in which it will emerge 

 that Withycombe was quite right in principle if wrong in his under- 

 standing of the mechanism. I myself erred similarly in 1929. 



The physiological anatomy and histology of the trap of Utricu- 

 laria vulgaris and closely related forms will now be considered. Within 



