2 RHODOSPERME.E. v. 



reduced to a single cell, cast off by the parent plant and capable of continuing an 

 independent existence. If this be the true explanation of the phenomena, we have 

 still to decide (and from analogy only, for evidence of fertilization is wanting) to 

 which of these bodies we shall give the name of -s^wr?, and which we shall call gem- 

 mule. And here different botanists take different views, the positive evideace, as it 

 seems to me, being nearly equally in favour of either. Having no new facts 

 to bring forward, I shall not argue this question here ; but, adopting the names 

 spores and tetraspores for these reproductive cells respectively, shall endeavour 

 briefly to describe their development and modification. 



1 . The spores, then, or gemmidia of J. Agardh, are always congregated in a more 

 or less definite mass, tuft, or cluster, which we shall here call the Sporiferous-nudeus 

 {Nucleus cystocarpii, J. Ag.) Each spore is a cellule, having a gelatino-membrana- 

 ceous, hyaline coat, and containing a dense, subsolid, homogeneous, deeply coloured 

 starchy mass or endochrome, which, on being expressed, breaks up into an impalpable 

 granular dust. The sporiferous nucleus is either wholly naked (as in Wrangelia, 

 Callithamnion); immersed in the substance of the frond (as in Graleloupia, Haly- 

 menia, Dumontia, etc.); lodged in wart-like tubercles (as in Poly ides); or contained 

 within hollow conceptacles (or cystocarpia, Kiitz.) of various forms, which are either 

 dispersed through the frond and partially immersed in it ; or are borne on the ends 

 or sides of the branches, on the midrib of leaves, or on proper fructiferous ramuli 

 developed from some part of the stem, or of the margin or disc of the foliations. 

 In all cases of external conceptacles they appear to be transformations of the ends of 

 the branches, or of lateral ramuli; the transformed branch or ramulus being 

 usually very much shortened, often reduced to a point. And all immersed spori- 

 ferous nuclei are formed from the interior strata of cells, and have generally if not 

 always, a connection with the medullary or central stratum. 



The sporiferous nuclei are of two kinds, distinguished by the manner in which 

 the spores are developed in each. In families of the highest structure (Des. 

 MiosPERMEiE) the nucUus consists of a tuft of articulated, moniliform filaments, or 

 spore-threads, radiating to all sides from a central point, or growing from a pi'oper 

 placenta lodged within a conceptacle. In these nuclei a single spore is formed 

 within one or more of the cells of the filaments ; and the spore-thread at maturity 

 is thus more or less changed into a string of spores. In some, all the cells are 

 changed into spores ; in others, the terminal cell is alone fertile. In the less 

 organised families (Gongyxosperme^) the nucleus is formed either from a single 

 mother-cell, from several detached mother-cells, or from such cells imperfectly 

 joined together in moniliform strings issuing from a central point, or growing 

 from the placenta of a conceptacle. Each mother-cell, whicli is at first filled with a 

 homogeneous endochrome, becomes, by repeated cell-division, converted into a clus- 

 ter of spores, at first retained within its walls ; afterwards, on the bursting of the 

 wall, dispersed. Tims by the evolution of one cell, a favella, or simple globose 

 nucleus containing many angular spores within a hyaline periderm, is formed. By 

 the evolution of several detached but adjacent mother-cells, a favellidium or com- 

 pound favella results. And by the similar evolution of the cells of the moniliform 

 series, the highest form oi favdlidium is produced. In all these cases the general 



