253 



/ 



1 



thick and strongly cuticularized. Instead of stretching across 

 the cell at right angles with the radial walls, and parallel with the 

 lower tangential walls, they are strongly inclined upward, many 

 making a sharp angle. Here, where the wall is so plainly cuti- 

 cularized, the process of division is rapidly going on ; new walls 

 are seen in both directions, tangential and radial. But, as before 

 stated, the tangential divisions are limited for the most part to 

 four or five layers of cells, forming the coUenchymatic cylinder 

 outside the palisade cells. The radial divisions, of course, keep 

 pace with the increasing circumference, and the outer tangential 

 wall must increase at a rapid rate; it must increase not only in 

 surface, but it does keep pace in thickness, for no section can be 

 made which does not show all the cells of the epidermis fitted 

 with outer walls of uniform thickness. 



Now if we turn to the sections studied, we shall see a peculiar 

 development which must in some way serve to aid in this series 

 of rapid changes. We have before mentioned a curvature in the 

 wall in nearly all cells in the process of division (which process is 

 detected by the thinness of the new wall). Besides the thin new 

 wall just forming, there is another element which, so far as I 

 know, is peculiar to this variety. This is a sphere or spherical 

 body consisting of partly cuticularized cellulose, extending from 

 the outer wall down into the cell, to which in all cases is attached 

 the new wall. That is to say, this spherical body projecting down 

 into the cell was never found unless below it was the new radial 

 wall which was just forming to increase the number of cells in 

 the circumference. New cells were sometimes seen without the 

 sphere, but the sphere is never found without the wall. Now by 

 a study of the surface of these cells the following facts are recog 



nized : 



These spheres are found principally or more frequently on 



the tangential walls of tlie grown epidermal cells. At the stage 

 where they are most numerous, these cells have a long diameter 

 parallel with the axis of the stem. The tangential diameter is 

 the one which must enlarge so rapidly to keep pace with the in- 

 crease In circumference, and these bodies occupy such a position 

 as to be readily made use of in the new outer wall. 



If we suppose this to be the function of these bodies, the 



