1920 Pease; on Desmarestia 335 



cell .iivision in the axial row probably takes place in D. ligulata in the same 

 way that Soderstrom has alreadj^ demonstrated for D. aculeata. 



Since the original wall Avhich cut ofif the branch cell was not perpen- 

 dicular to the transverse wall of the axial cell, but slightly oblique to it, 

 the branch, at its beginning, was not exactly at a right angle but inclined 

 slightly forward toward the growing point. The lateral walls of both 

 axial cells and lateral cells in the young stages are not cylindrical but 

 barrel-shaped. As a result, the supper side of the basal cell of the branch 

 is pressed against the base of the axial cell next above, and the two walls 

 become adherent. Subsequent changes in the shape and position of the 

 cells due to the development of surrounding tissues so modifj'' their appear- 

 ance at the point of juncture that unless the early stages were traced it is 

 impossible to tell from which axial cell the branch originates (Plate 62, 

 Figs. 2 and 3, and plate 59, Figs. 2, 3 and 6). 



Cortication begins immediately below the growing point, and accord- 

 ing to the observation of the writer does not proceed as indicated by Reinke 

 (1880), w^ho claims that filaments develop which become applied to 

 the axial cell, encircle it, meet and fuse, and by repeated division form 

 a layer of small cells lying close to each other about the original cell 

 of the thallus. The cortical layer originates, as described by Janczewski 

 (1875) and Reinke (1880), as a small cell cut off on the lower side of the 

 basal cell of a lateral branch of the main axis. This new cell does not 

 divide again immediately, but another small cell is cut off from the side 

 of the same basal cell. The first cell to be cut off then divides, either 

 below or on one side. Since the same process is proceeding on the opposite 

 side of the hair branch, a cluster of small cells is thus formed about its 

 base. jNIeanwhile, the opposite branch hair of the pair is also developing 

 a cluster of small cells about its base. At this time there is then a cluster 

 of small cells on opposite sides of the axial cell, and this is the beginning 

 of the flattened thallus. The axial cells at this point have a length only 

 about half their diameter, and each cell bears a pair of branches, hence 

 the branches are very close together, and the cluster of small cortical cells 

 about the base of one lies in contact with the branch below it. These cort- 

 ical cells now divide laterally without any definite sequence, and gradual- 

 ly spread toward each other until the axial cell and the basal cells of its 

 paired branches are completely covered (Plate 63). 



Since elongation of the axial row is taking place at the same time, 

 several stages of this process of cortication can be observed on a single 

 filament. From the point at whicli the first small cell is cut off to the 

 jjoint at wliich complete cortication is attained, from two to six axial cells 

 may intervene, in all intermediate stages of cortication. 



By the time complete cortication is reached, the basal cells of the 



