on Polysiphonia iirceolata. Later I have found, on Flustra foliacea, some discs which 

 I think must be referred to the same species; they differed in their sliglitly larger 

 dimensions and in the margin being partly continuous, the filaments being united 

 to the extremities. These discs were thus still more similar to ErythropeUis, but 

 the filaments had always partly free endings. The filaments were 3,5 — 5 /j. thick, 

 narrowest ■at the border, broadest in the middle of the frond. The spores were 

 4 — 5 fj. in diameter. 



Localities, Sk: Off Hirshals (XO and YK), 11,5 to l,i meters, August. 



Goniotrichum Kiitz. 

 1. Goniotrichum elegans (Chauv.) Le Jolis. 



Le Jolis, Alg. mar. Cherb. p. 103; J. Agardh (1883) p. 13; Berthold (1882) p. 26; Hauck, Meeresalg. p. 518. 

 Bangia elegans Chauvin, Mem. Soc. Linn. Norm. t. 6 (not seen) ; Rech. sur Torg. d. plus. genr. d'Algues, 



Caen 1842 p. 33 ; Harvey Phyc. Brit. pi. 246. 

 Ceramium ceramicola Fl. Dan. tab. 2207 fig. 2 (? not the description). 



This plant attains a length of at least 5 mm. in the Danish waters. The fila- 

 ments are below up to 50 tj. thick, above they become gradually thinner and are 

 at the summit only 15 rj. thick. The increase in thickness below is usually due 

 only to the thickening of the gelatinous outer wall, the diameter of the cells not 

 increasing, and the cells forming usually a single row. There may be, however, 

 more than one cell at the same level. This was caused, in the cases examined by 

 me, not by longitudinal division of the cells but by displacement of the cells, so 

 that the growing axes became inclined, the cells dividing then as usually by walls 

 perpendicular to the growing axis and becoming arranged in two irregular longitu- 

 dinal rows, or even more than tw^o cells may occur at the same level (fig. 15 

 The outer wall is usually uniform, limited outwards by a fine line. Sometimes, 

 however, the cells are provided with a denser, special membrane. In the plant 

 represented in fig. 15 £ a rather thick, dense cuticular-like outer-wall was visible 

 in the lower part of the plant; the cells were here also provided with dense spe- 

 cial membranes, and between these and the outer membrane a stratification was 

 often visible. 



The ramification takes place in a manner reminding one of the so-called false 

 branching of the Scytonematacece. The branches rise at a great distance from the 

 end of the filament, a cell growing outward through the gelatinous wall, dividing 

 by a wall perpendicular to the new direction of growth (fig. 15 B, C). The further 

 growth results in the branch coming to form a direct continuation of the principal 

 filament and often takes nearly the same direction as this, the upper part of the 

 principal filament being more or less pushed aside and taking the appearance of 

 a branch (fig. 16). The cell lying at the origin of the branches is divided by a 

 transverse wall as well as all the other cells. New branches very often arise below 

 the older; even in old filaments new branches may arise (fig. 15 A). 



10* 



