CONFERVOID FILAMENTS OF MOSSES. 571 



By the continuation of this process, chiefly in the terminal cell, and by the growth of 

 the already formed cells, and by the formation of branches from branches continuously, 

 the length of the filaments, and the area they occupy, are extended indefinitely 

 (PI. LVII. figs. 3, 4, 5). The branches are often commenced at an early period of the 

 existence of the cell from whence they arise (PI. LVII. fig. 2). 



The contents of the cells vary much in appearance, according to their age. At first, 

 in the growing end and in the half-formed cell of the branches, there is a tendency to 

 general homogeneousness, although scarcely ever complete. In either case, very shortly 

 after the cell is formed, the contents consist of a transparent, colourless layer (pri- 

 mordial utricle) lining the cell-wall, together with granules of uniform green colour 

 (chlorophyll-granules), in greater or less quantity. In the actively growing cell, the 

 centre contaias a transparent fluid (sap-fluid) ; but in those which have solid, unyielding 

 cell-walls, the endoplast fills up the interior, and the granules crowd each other, 

 sometimes very closely. 



A nucleus can freqviently be observed in the mature cell, though not always. In the 

 contents of the bulging commencement of a branch, no nucleus is to be observed ; nor does 

 the nucleus of the old cell undergo any change in consequence. The nucleus of the new 

 cell must therefore be formed subsequently to the cell, which is certainly not produced 

 by it. This, with many other facts, and with some which will hereafter be given, seem 

 to point out that the nucleus does not necessarily j)lay so important a part in cell-forma- 

 tion as some have supposed. I notice this only incidentally, as the discussion fuUy carried 

 out would be foreign to the purpose of this communication. As I have remarked above, 

 the length of each cell of the filament varies extremely, according to the external con- 

 ditions. Under much moisture and heat, it is very much increased, so that it may be 

 twenty or thirty times longer than wide ; and sometimes the more terminal cells are 

 elongated into delicate hairs, bearing a striking resemblance to the so-called ciha of the 

 Draparnaldice (PI. LVII. fig. 6) (which in some other respects they much imitate), 

 their green protoplasmic contents being drawn out somewhat in the manner ob- 

 servable in that genus. 



The general tendency of the growth of these filaments is towards an arborescent 

 form when growing free and not crowded ; and for a certain time they continue un- 

 altered from the simple plan just described : sometimes they are scarcely branched at all 

 (PI. LVII. fig. 7) ; sometimes the secondary branches are arranged on the plan of a 

 cyme (PI. LVII. fig. 5). 



In these forms the confervoid radicles continue to grow for an indefinite time, ex- 

 ternal circumstances remaining the same ; and in course of time very large surfaces can 

 be covered by them, unless usurped by other plants. They are perfectly capable of 

 independent existence, whether they have arisen from a spore, leaf, stem, or root, when 

 separated from their source ; and hence the erroneous impression of their Algal origin. 

 This is the less to be wondered at, if we notice the growth of one when placed in water. 

 Under these circumstances the activity of its development and linear growth is wonder- 

 cells by division of the whole contents, Niigeli calls the formation of cells by division 'parietal cell-formation.'" 

 — Braun's ' Rejuvenescence in Nature,' Ray Soc. 1853, p. 234. 



