Dots were made with Indian ink along the midrib of both pulvinated cotyledons of a rather old seedling of Oxalis Valdiviana; their distances were repeatedly measured with an eye-piece micrometer during 8 3/4 days, and they did not exhibit the least trace of increase.  It is therefore almost certain that the pulvinus itself was not then growing.  Nevertheless, during this whole time and for ten days afterwards, these cotyledons rose vertically every night.  In the case of some seedlings raised from seeds purchased under the name of Oxalis floribunda, the cotyledons continued for a long time to move vertically down at night, and the movement apparently depended exclusively on the pulvini, for their petioles were of nearly the same length in young, and in old seedlings which had produced true leaves.  With some species of Cassia, on the other hand, it was obvious without any measurement that the pulvinated cotyledons continued to increase greatly in length during some weeks; so that here the expansion of the cells of the pulvini and the growth of the petiole were probably combined in causing their prolonged periodic movements.  It was equally evident that the cotyledons of many plants, not provided with pulvini, increased rapidly in length; and their periodic movements no doubt were exclusively due to growth.

In accordance with the view that the periodic movements of all cotyledons depend primarily on the expansion of the cells, whether or not followed by growth, we can understand the fact that there is but little difference in the kind or form of movement in the two sets of cases.  This may be seen by com-[page 115] paring the diagrams given in the last chapter.  Thus the movements of the cotyledons of Brassica oleracea and of Ipomoea caerulea, which are not provided with pulvini, are as complex as those of Oxalis and Cassia which are thus provided.  The pulvinated cotyledons of some individuals of Mimosa pudica and Lotus Jacobaeus made only a single oscillation, whilst those of other individuals moved twice up and down in the course of 24 hours; so it was occasionally with the cotyledons of Cucurbita ovifera, which are destitute of a pulvinus.  The movements of pulvinated cotyledons are generally larger in extent than those without a pulvinus; nevertheless some of the latter moved through an angle of 90o.  There is, however, one important difference in the two sets of cases; the nocturnal movements of cotyledons without pulvini, for instance, those in the Cruciferae, Cucurbitaceae, Githago, and Beta, never last even for a week, to any conspicuous degree.  Pulvinated cotyledons, on the other hand, continue to rise at night for a much longer period, even for more than a month, as we shall now show.  But the period no doubt depends largely on the temperature to which the seedlings are exposed and their consequent rate of development.