On the Pithophorace^. 23 



in length, and an upper one with the power of increasing apically in 

 the longitudinal direction of the mother cell, and of bipartition. By in- 

 crease and bipartition in accordance to the law now indicated, a single 

 series of cjdindrical cells is formed — the cells being longer or shorter 

 according to the nature of the species and of the outer circumstances, — 

 and this series of cells forms that part of the cauloïd which I call, in 

 its description, its principal filament. 



Only in very rare, exceptional cases the principal filament of the 

 cauloïd remains unbranched — perfectly branchless, full-grown specimens 

 I have found now and then hi P. œqualis nob. (pi. 1, fig. 5) and P. 

 Cleveana nob., and almost branchless in P. kewensis nob. (pi. 2, fig. 5). 

 In common cases ramification takes place if not in all at least in most 

 cells of the principal filament, and this very soon ; generally long before 

 the principal filament has attained its full development as to length.. 

 The oldest cells, — consequently those situated lowest, nearest to the 

 mother spore of the specimen, — are the first which develop branches; 

 and afterwards the formation of branches proceeds from the lower and 

 older cells to the upper and j^ounger ones. — i. e. acropetally, — but 

 not quite to the top cell, this being as a rule unbranched. 



The formation of the first cell in every branch takes place in the 

 following manner. That cell of the principal filament from which the 

 branch is to be formed, sends forth from one of its sides, a small space 

 below the top, a small process, which is at first shaped like a truncated 

 cone with a strongly rounded top, and which does not point straight 

 outwards, but somewhat upwards (pi. 3, fig. 1 b). This process is formed 

 by an increase as to the surface of the membrane, beginning round 

 a central point, in consequence of which the membrane in this place by 

 and by gets convex. Sometimes this increase of the surface takes place 

 only in the inner layer of the membrane, which then, by its continued 

 increase, breaks the outer layer (pi. 3, fig. 1 ft) ^); but sometimes the 

 increase extends both to the inner and outer part of the membrane, and 

 it is then not broken. -) As pi. 3, fig. 1 b shows, and as is mentioned 



') This circumstance strongly calls to mind the proceeding at the commencement 

 of the formation of branches in the genus Bulhochcete Ag. See Pkingsh. Beitr. z. 

 Älorph. d. Alg. p. 22, pi. 2. 



2) I have not been able to distinguish, previously, two layers in a cell of a 

 Pithophora ready for ramification, neither by optical nor by chemical means. Their 

 existence, at least in some cases, is proved only by the circumstance mentioned 

 above. 



