Chat XII COXCLUDING EEMARKS. 661 



their proper -vertical position. Some curious facts 

 have l>een given under this head, showing that hori- 

 zontally extended leaves suifered more at night, when 

 the air, which is not cooled by radiation, was prevented 

 from freely circulating beneath their lower surfaces ; 

 and so it was, when the leaves were allowed to go to 

 sleep on branches which had been rendered motionless. 

 In some species the petioles rise up greatly at night, 

 and the pinnae close together. The whole plant is 

 thus rendered more compact, and a much smaller 

 surface is exposed to radiation. 



That the various nyctitropic movements of leaves 

 result from modified circumnutation has, we think, 

 been clearly shown. In the simplest cases a leaf 

 describes a single large ellipse during the 24 h. ; and 

 the movement is so arranged that the blade stands 

 vertically during the night, and reassumes its former 

 position on the following morning. The course pursued 

 differs from ordinary circumnutation only in its greater 

 amplitude, and in its greater rapidity late in the 

 evening and early on the following morning. , Unless 

 this movement is admitted to be one of circumnu- 

 tation, such leaves do not circumnutate at all, and this 

 would be a monstrous anomaly. In other cases, leaves 

 and cotyledons describe several vertical ellipses during 

 the 24 h. ; and in the evening one of them is increased 

 greatly in amplitude until the blade stands vertically 

 either upwards or downwards. In this position it con- 

 tinues to circumnutate until the following morning, 

 when it reassumes its former position. These move- 

 ments, when a pulvinus is present, are often compli- 

 cated by the rotation of the leaf or leaflet ; and such 

 rotation on a small scale occurs during ordinary cir- 

 cumnutation. The many diagrams showing the move- 

 ments of sleeping and non-sleeping leaves and coty- 



