262 Richards: Ceramothamnion codii 



trichogyne or trichophoric apparatus was revealed. The poly- 

 spores occur as considerable aggregations, enclosed in a definite 

 wall, and borne in the axils of certain specially developed branches 

 (Fig. 20). It is only in this connection that the upright filaments 

 ever branch ; an irregular whorl of branches is formed with the 

 main axis of the filament continuing above them, and at their 

 bases are formed one or more large favella-like masses of irregu- 

 larly arranged spores. The branches, which vary from two to as 

 many as five in number, usually assume ultimately an equal length 

 with the original axis, sometimes even surpassing it in importance 

 and in so doing often push it to one side, making it quite indistin- 

 guishable. In consequence of this the masses of mature poly- 

 spore often appear to be borne terminally on an irregularly forked 

 filament. By following their development it can be seen, how- 

 ever, that these branches arise as lateral outgrowths from the cells 

 of the younger nodes (Figs. 15, 16, 17), originating in all respects 

 after the manner in which the upright filaments spring from the 

 prostrate one, except that two or more arise from each node. 

 After the beginning of these outgrowths it appears that the axis 

 from which they spring ceases to develop to any extent, all of its 

 vitality being spent in the production of the polyspores and the 

 accompanying whorl of branches. In a few instances a second 

 node below also produces a few side branches, but these never 

 develop to any great length, nor were polyspores seen to be borne 

 in their axils. In the formation of a branch one of the nodal cells 

 takes on the function of an apical cell while the surrounding ones 

 multiply somewhat and as the new internodal cell formed from the 

 apical cell enlarges from around its base a junction with the parent 

 axils, four or five nodes, and occasionally many more are formed 

 before any activity in the direction of the origin of the polyspores 

 is to be observed. For this reason it is often difficult to trace the 

 exact point of origin of the latter. The polyspores, however, were 

 found to originate always from the node of the main axis immedi- 

 ately above the one which gave rise to the whorl of branches 

 (Figs. 16, 17, 18). The cells begin to multiply more rapidly, and 

 form groups of four or five cells ; these ultimately by still fur- 

 ther division and by great increase in size produce the favella-like 

 masses figured in Fig. 20. First one group develops, then an- 



