300 MINNESOTA BOTANICAL STUDIES. 



Massed around the central axis are small round cells. These 

 seem to have developed, at least in some cases, from the upper 

 branch {Plate XL VI., figs. 2, a, and j, a) of the original branch 

 coming from the central axis. Schmitz and Hauptfleisch 

 speak of these rounded cells as rhizoids, while Harvey does 

 not discuss their origin, but speaks of them as coating the axis. 

 In the material studied the branching was not always dichoto- 

 mous nor did the branches run radially outwards. The fila- 

 ments in the center of the frond are loosely scattered, being 

 massed together closely to form the cortex. Examining a cross 

 section of the frond we find a large round central cylinder 

 {Plate XL VI., fig. 1). Massed around this cylinder are small 

 round cells. The cells in the center of the section are not con- 

 nected with filaments showing that the filaments of which these 

 cells are cross sections run parallel to the central axis. The 

 tissues are loosely arranged, but towards the periphery dichoto- 

 mous branching can be observed and these branches held to- 

 gether by a gelatinous secretion form the periphery. 



Holdfast. — The holdfast is strong, although quite incon- 

 spicuous. It does not appear to be disc-like, but rather to be 

 composed of branches {Plate XL VI., fig: 7). There is a brown 

 cellular substance which is developed beneath the holdfast. 

 Apparently the holdfast can be developed at any point where 

 the horizontal branch may come in contact with the substratum 

 or at the ends of the small branches which radiate downward 

 from the horizontal branch. 



Asexual Reproduction. — Some branches appear slightly 

 fleshier than others. When sectioned there are found distributed 

 around them, the tetragonidia. These are developed from the 

 peripheral cells. In one section {Plate XL VI, fig. 8) we may 

 find the younger gonidangia the contents of which are not yet 

 divided, those which have divided obliquely forming two masses, 

 and the mature tetragonidangium containing four gonidia. As 

 has been previously stated they do not divide perpendicularly but 

 obliquely. Plate XL VI., fig. 8, illustrates a portion of a longi- 

 tudinal section drawn from the side of the axis cylinder to the 

 periphery. The same structures would be seen on the other 

 side of the axis cylinder. Long narrow paraphyses extending as 

 far again as the tetragonidangia are found which function as a 

 protection for the gonidia. They are developed from certain 

 peripheral cells and are made up from fourteen to sixteen cells. 



