562 SUMMARY AND CHAP. XII. 



ledons should be compared, and it will be seen that 

 they are essentially alike. Ordinary circumnutation 

 is converted into a nyctitropic movement, firstly by an 

 increase in its amplitude, but not to so great a degree 

 as in the case of climbing plants, and secondly by its 

 being rendered periodic in relation to the alterna- 

 tions of day and night. But there is frequently a 

 distinct trace of periodicity in the circumnutating 

 movements of non-sleeping leaves and cotyledons. 

 The fact that nyctitropic movements occur in species 

 distributed in many families throughout the whole 

 vascular series, is intelligible, if they result from the 

 modification of the universally present movement of 

 circumnutation ; otherwise the fact is inexplicable. 



In the seventh chapter we have given the case of 

 a Porlieria, the leaflets of which remained closed all 

 day, as if asleep, when the plant was kept dry, appa- 

 rently for the sake of checking evaporation. Some- 

 thing of the same kind occurs with certain Graminese. 

 At the close of this same chapter, a few observations 

 were appended on what may be called the embryology 

 of leaves. The leaves produced by young shoots on 

 cut-down plants of Melilotus taurica slept like those of 

 a Trifolium, whilst the leaves on the older branches 

 on the same plants slept in a very different manner, 

 proper to the genus ; and from the reasons assigned 

 we are tempted to look at this case as one of reversion 

 to a former nyctitropic habit. So again with Desmo- 

 dium gyrans, the absence of small lateral leaflets on 

 very young plants, makes us suspect that the imme- 

 diate progenitor of this species did not possess lateral 

 leaflets, and that their appearance in an almost rudi- 

 mentary condition at a somewhat more advanced age 

 is the result of reversion to a trifoliate predecessor. 

 However this may be, the rapid circumnutating or 



