CONFERVOID FILAMENTS OF MOSSES. 583 



marks between many Confervoids is very questionable, and they have doubtless led to 

 much misconception and incorrect assumption. These plants being formed of very simple 

 parts, their characteristics are necessarily ambiguous, and hence the difficulty of fixing 

 their true position ; and this is especially the case in those kinds which are composed of 

 a single cell, whether belonging to those which have been designated by Braun as 

 " true unicellular Algse," or " pseudo-unicellular," or to those which are multicellular, 

 but which spring at some portion of their history from a single vegetative cell (spore, 

 gonidium, &c.). 



3. The third subject to be dwelt upon has already received some attention at the com- 

 mencement of this paper, — namely, that the function and development of the cells 

 separated from the filaments are evidently analogous to those of the gonidia of Lichens 

 and also of many Algae, and therefore the term gonidia* may be rightly applied to 

 them. It is the office of the filaments to produce these gonidia ; or perhaps it should 

 rather be considered that each filament, in its most marked condition, is a series of 

 gonidia developed in a Linear form, each division of which is capable of becoming a 

 gonidium : and this mode of viewing these filaments points out the analogous condition 

 observable in the moniliform filament of gonidia in Collema, each cell of which, although 

 generally segmenting linearly on the binary plan, can at any time assume the other 

 actively segmenting states, as has been pointed out by me in the papers above alluded to. 



The position in which these filaments stand towards the rest of the plant is peculiar. 

 Looking at them as a basis upon wliich the axis producing the organs of the true fruc- 

 tification is supported, they have at first been considered as the homologue of the pro- 

 thallium of Ferns ; but they really represent only a portion of that organ. They cor- 

 respond, indeed, only to the few confervoid cells first generated from the spore ; while 

 the heart-shaped Marchantia-expansion of the prothallium of the Fern is to be con- 

 sidered as corresponding to the stem, leaves, antheridia, and archegonia in the Mosses. 

 They differ also in the respect of producing gonidia, at least so far as is known at 

 present, although it is quite possible (judging from analogy) to suppose that gonidia 

 may be found to be produced from this position of the prothallium of Ferns, under 

 certain conditions. Compared with Fungi, they seem to hold an analogous condition 

 to the mycelium, more particularly apparent in the simpler forms of the Hyphomycetes ; 

 for it is questionable whether the so-called fruit of these Fungi is not rather analogous 

 to the gonidium than to the growth which results from impregnation. It is at pre- 

 sent uncertain whether any antheroids exist in the Fungi, and, if they do, whether the 

 spores take any part in true fecundation. 



It is very necessary to bear in mind that the confervoid filaments play a very im- 

 portant part, if not the most important, in the disseminating of Mosses — much more so 

 than has been generally supposed. Considered quite independently of their powers of 

 producing gonidia, they tend to spread the order in every direction, and cover very much 

 larger surfaces than the species from whence they are derived could do by gamic repro- 

 duction, whereby, simultaneously as it were, large spaces, even many feet in length, are 



* The distinction between "gonidium " and " spore," as marked out by Braun in ' Rejuvenescence in Nature,' is 

 borne out here and in the Lichens much better than in the Algae. 



VOL. XXIII. 4i I 



