Although it was not until 1822, when Dutrochet pointed out the true significance of the pulvini, Hill recognized that these structures were concerned with the movements of the leaflets, not only in the case of _Abrus_, but also in _Mimosa_. He remarked that "It is on the operation of light upon these interwoven clusters of fibres [which are placed at the bases of the main rib, and of the several foot-stalks of the lobes], that the motion of the leaves in gaining their different positions depends; and consequently, the motion itself is various according to the construction of these fibres. "In the Abrus they are large, and of a lax composition; consequently the lobes are capable of a drooping, an horizontal, and an oblique upward position: in the Tamarind, and the broad-leaved Robinia, they are more compact, and hence all the motion of which those leaves are capable, is an expanding open and a closing sideways; which the direction and course of the fibres also favours: in the Parkinsonia they are smaller, and yet more compact; and the consequence of this is, that its lobes have no farther possible motion, than the expanding upwards." Again, "The clusters of fibres are as a kind of joints on which their lobes are capable, under the influence of light, of a certain limited motion." Further, with regard to _Mimosa_, he remarks that "To propagate the motion when the leaves are in a state to shew it, there requires a perfect and confirmed state of those clusters of fibres lodged at their base." Hill then describes the experiments upon which he based his conclusions; these shew that he was fully awake to the importance of keeping the conditions of an experiment, other than those of light, as near constant as possible, and that the position assumed by the leaves depends upon the intensity of the light. His final experiment was to place the _Abrus_ in a bookcase in such a position that the sun shone full upon it; when the leaves were fully expanded he closed the doors and found that in an hour "The lobes were all drop't, and it was in the same state that it would have shewn at midnight. On reopening the doors the elevated position of the leaves was assumed in twenty minutes." Hill offers the same explanation of the movements of _Mimosa_ as of those exhibited by _Abrus_, the reason for their greater conspicuousness in the former plant being due to the fact that in _Mimosa_ "As there are no less than three sets of these clusters [of fibres which are placed at the bases of the foot-stalks], the effects of the same principle are naturally much greater than in the Abrus where there is only one."