io8 ' BOTANICAL GAZETTE [augusi 



growth and the orientation of the cell division is such as to pro- 

 duce the water form, while those physical and chemical conditions 

 resulting from a partial withdrawal of water by evaporation [i.e.y 

 an increased density of the protoplasm) result in that sort of 

 cell behavior which produces the air form of the plant. Just 

 why this difference in the density of the protoplasm should 

 result in such a different behavior of the cells is a question which 

 this paper does not attempt to discuss. The answer can only 

 come with an increased knowledge of the mechanics of cell 

 division, about which at present we know almost nothing. 



One other phenomenon must be mentioned. The results 

 given were all obtained from plants collected outside, which 

 were usually producing the air forms, or else from seedlings 

 grown inside. When some of these had been kept submerged 

 it was noticed that after several months they started to produce 

 the air type of leaf. This only occurred in some cases, and even 

 those in the same aquarium behaved differently. Also in the 

 case of those outside, which were submerged all winter, during 

 May and June many of them succeeded in producing air leaves, 

 though none were able to develop flowers. It would seem here 

 as if some of the plants after a time became accustomed to the 

 stimulus and refused to respond. Or it may be that as only the 

 air form is capable of fruiting, in the effort to produce flowers 

 the plant has the ability of self-adjustment to its conditions and 

 develops the air form in spite of its environment. Examination 

 of these primordia showed that the cells are smaller and the 

 protoplasm denser and less distended with water. It is possible 

 that the protoplasm is able to adjust itself, perhaps by the expul- 

 sion of water, into that condition in which it exists when in the air. 

 This, of course, is speculation, but investigations are under way, 

 which it is hoped w^ill give us some information on the subject. 

 In conclusion, I have pleasure in acknowledging my indebt- 

 edness to Professor John M. Coulter, Professor Charles R. 

 Barnes, and to Dr. B. E. Livingston, for much kind assistance 

 and advice. 



The University of Chicago. 



