IRRITABILITY. 381 



at a comparatively early stage in their development, are those 

 which are most affected in their growth by continuous dark- 

 ness ; whereas those, such as sheathing leaves, which are 

 naturally protected more or less from exposure to light by 

 others investing them, attain a relatively more perfect ex- 

 pansion. 



The radial or bilateral leaves, we have said, resemble 

 internodes in that they become excessively elongated in 

 darkness, and the question naturally arises why do dorsi- 

 ventral leaves behave otherwise ? It seems that growth in 

 breadth is in all cases hindered or prevented by darkness. 

 Not only do dorsiventral leaves afford examples of this, but 

 stems also : for example, the broad leaf-like internodes of 

 Cactacese such as Opuntia are developed in darkness as 

 slender cylindrical or prismatic structures. 



Several explanations of the remarkable effect of the 

 absence of light in diminishing or preventing the expansion 

 of leaf-blades have been offered, and it will not be uninstruc- 

 tive to consider them. 



G. Kraus endeavoured to explain it by his well-known 

 " self-nutrition " theory ; he ascribed it to the fact that, in 

 darkness, leaves are incapable of performing one of their most 

 important functions, namely, the construction of organic sub- 

 stance. This explanation has, however, been shewn to be 

 quite inadequate by the observations of Batalin, of Rauwen- 

 hoff, of Godlewski, and of myself) As a matter of factfleaves 

 continue to grow when they are placed for a time in darkness, 

 in blue light, or in an atmosphere which , contains no carbon 

 dioxide, under conditions, that is, which render impossible the 

 normal formation of organic substance in them. 



Batalin suggested as a possible cause of the smallness of 

 leaves grown in darkness, that the process of cell-division is 

 arrested under these circumstances. But this suggestion is 

 shewn to be valueless by the fact that radial or bilateral leaves 

 grow excessively in darkness, and that even some dorsiventral 

 leaves grow considerably. Further, Prantl ascertained that 

 the average number of cells in a leaf of the embryo of 

 Phaseolus, whilst still in the seed, was 343, whereas that in 



