572 DR. J. B. HICKS ON THE GONIDIA AND 



fully increased, if not exposed to too mucli light ; so that in a week it can multiply itself 

 200 or 300 times, whUe the original type has been nearly preserved, the slight alteration 

 being in the elongations of the cells, and a decrease of their breadth. 



Now, although, for the most part, these confervoid filaments generally preserve their 

 peculiar appearance, and are in this manner very readily distinguished from the true 

 root-filaments, whose endoplast is colourless and the cell-walls more or less stained with 

 a brown colour, yet there is a great tendency for each, even after full formation, to pass 

 into the other. This can be frequently seen in their natural state, and can be shown 

 by experiment (PI. LVII. fig. 8). The greater tendency, however, is for the radicles to 

 pass into the confervoid filament than the contrary. The colourless endoplast of the 

 true radicles becomes green and granular ; and ultimately they exhibit all the characters 

 of true confervoid filaments. On the other hand, I placed in the sun a glass full of the 

 filaments wliich had grown, and were then growing, in water. After a week the cell- 

 walls of the older portions had become stained brown, and they had assumed the ap- 

 pearance of those radicles whose contents had assumed the green colom*. 



Some of the filaments which I had grown in water branched in a manner very similar 

 to Braparnaldia tenuis {Stygeoclonium tenue, Kiitzing) ; indeed, had it not been for its 

 known origin, I should have instantly regarded it as such. I have shown one less 

 marked at PI. LVII. fig. 9 (the only one of which I preserved a drawing). In a glass of 

 water, where I had placed Moss, on one occasion I found a very fine specimen of D. tenuis. 

 This is a very unusual place to find this plant ; and though I could not absolutely trace 

 it to a Moss, yet, coupled with the fact that similar growths can be so originated, and 

 also that the radicles produce elongated cilia-like ceUs, it seems to be a poiut worthy of 

 further research, whether or not that genus, or at any rate the above species, may or may 

 not have its origin from Moss in some one of its phases. Nor shoiild this, in our present 

 state of knowledge, be considered a wild speculation ; for we know nothing of the agamic 

 growth of Braparnaldia : we have nothing to militate against its being one mode of 

 vegetative growth of a form considered altogether distinct ; and this is not more extrava- 

 gant than the known fact that these confervoid filaments can produce and spring from 

 Mosses. I again remark, we know so little of the whole possible life-history of these 

 simpler plants, that our want of knoAvledge of a precedent cannot be quoted against it. 



Frequently in the larger filaments, and towards the extremity of those whose growth 

 is not very active, may be seen here and there a considerable separation between the two 

 adjacent ceUs. If this be carefully examined, it will be noticed that this space is filled 

 by a transparent, colourless ceU, which at first sight might be considered to have no 

 contents ; but upon careful examination, and the use of reagents, it will be found that 

 there is colourless homogeneous endoplast, so closely applied to the inner side of the 

 cell-wall as otherwise to escape detection (PI. LVII. fig. 10). Wlien the cells of this 

 filament separate from one another, these transparent cells also become detached, and 

 assume the shape of flattened spheres, or they may become quite globular. Wliether 

 they possess any further history I am unable to tell. It seems rather to be some 

 abnormal condition of cell-formation at the line of the separation of the two portions 

 — a portion detaching itself at the time of division, and forming around it a layer of 



