596 ECOLOGY 



produced m a saturated atmosphere, appearing to show that transpira- 

 tion results in the formation of air leaves, and freedom from transpiration 

 in the formation of water leaves. This theory is strengthened by the 

 experimental production of air leaves in concentrated solutions of 

 potassium chlorid or of calcium nitrate, the high osmotic pressure of the 

 medium having the effect of transpiration in increasing the concentra- 

 tion of the cell sap, precisely as in Stigeocloniiim. Why the dissected 

 form should result if the cell sap is dilute, and the entire form if it is 

 more concentrated, cannot now be told, the exact mechanics of form 

 changes being but little understood. 



While the cells in Stigeodonium are almost always plastic, the leaf of 

 Proserpinaca loses its plasticity after attaining a length of three or four 

 millimeters, the form then in development continuing to maturity re- 

 gardless of habitat changes. Furthermore, some leaves appear to be 

 fixed from the outset, those following the cotyledons and those develop- 

 ing on horizontal autumnal shoots being dissected, while the late stem 

 leaves are likely to be entire regardless of environment. More puz- 

 zling still is Sium in which the cotyledons are followed by palmate 

 leaves of mesophytic aspect, regardless of conditions, these being fol- 

 lowed by dissected pinnate water leaves, and later by simply pinnate 

 air leaves. The submergence of an old plant results not in the de- 

 velopment of dissected pinnate water leaves, but of palmate leaves 

 like those first appearing in the seedling. In the water lilies both 

 submersed and floating leaves are developed in the water, so that 

 variations in the water relation hardly can be assumed to be causative 

 factors, as in Proserpinaca. 



Observations like those noted in the preceding paragraph have led to 

 the theory that the plasticity of amphibious plants is more apparent 

 than real, the phenomena of leaf variation representing mere stages in 

 development. The palmate leaves of Sium and the narrow submersed 

 leaves of Sagittaria and of Castalia from this viewpoint are regarded as 

 juvenile leaves, while the pinnate leaves of Sium, the floating leaves of 

 Castalia, and the sagittate air leaves of Sagittaria, are supposed to be 

 adult. The theory has been carried even farther, representing that the 

 stages in the life of the individual (the ontogeny) repeat similar stages in 

 the life of the race (the phylogeny), illustrating what is known as rr/- 

 pitulation. For example, a remote ancestor of Sium might be imagined 

 to have been a palmate-leaved mesophyte, and a more recent ancestor, 

 a pinnately dissected hydrophyte, while the present species might be 



