116 PROCEEDINGS OF THE AMERICAN ACADEMY 



for this species, but does not figure it. On the other hand, Berthold 

 finds a cluster of apical cells in the tip of Champia parvula, of which 

 he gives a diagram. The only thing bearing on the other problems 

 suggested by our preliminary examination of Champia is Wille's hint 

 that the diaphragms are derived from the filaments. 



My observations on the apical growth in Champia agree in the main 

 with Berthold's, as far as his go. I do not find so great regularity here, 

 however, as Berthold would give one to understand to exist. Accord- 

 ing to my observations, on looking down upon a tip of the plant, or 

 in examining cross sections of it, a number of rows of cells are to be 

 seen converging towards a common point, the apex (Figs. 4, 5, 6). 

 Three or four of these rows meet at the apex, into the angles formed 

 by them are pushed an equal number of other rows, and the remaining 

 space is filled with a third series. I have found the number of these 

 rows to vary from eleven to fifteen, but always to equal the number 

 of longitudinal filaments in the branch, as Berthold has pointed out 

 (Figs. 6 and 8). The reason for this will be seen when the origin of 

 the latter is understood. 



At the head of each of these rows, that is, at the part nearest its 

 apex, there is a cell somewhat larger than those directly beneath it 

 (Figs. 3 and 5). This cell gives rise to others of the row by anticli- 

 nal division ; that is, by forming partitions at right angles at once to the 

 surface of the frond and to the axis of the row. So each of these 

 cells at the heads of the rows is a true apical cell (Scheitelzelle of the 

 German botanists). Each one is in the shape of a triangular pyramid 

 with rounded sides. The apex of the pyramid is directed inward, 

 while the base lies at the surface of the frond. The length of the 

 pyramid is 10 to 15/x, while the width at the base varies from 5 to 10/a. 

 The cells formed by the division of the apical cells which are at the 

 middle of the cluster, meet in the middle line below the apical cell 

 (Fig. 3). And all of these daughter cells, whether they meet in 

 the middle or not, appear somewhat crescent-shaped when close to 

 their apical cell. They do not divide usually until removed several 

 cells from it by division of the apical cell. They then divide, each by 

 a partition parallel to the surface of the frond, at about a fourth or a 

 third of the length of the cell from its inner end (Fig. 3). It is 

 probable that the branches have their origin at this point, as explained 

 later. The outer cell thus formed divides again into two or three cells 

 (Fig. 6), and these may again divide. This division is by means of 

 partitions which are at right angles to the surface of the plant and 

 oblique to the original cell wall. The result of all this is the irregular 



