274 BRYOPHYTA 



the rnany-layered capsular wall (Fig. 132 F, G, H). The dome-shaped 

 archesporial layer divides later into four layers, and every cell undergoes 

 the tetrad-division to form spores. At maturity dehiscence takes place 

 by a transverse rupture, setting free a circular operculum. The foot is 

 considerably enlarged as an haustorium, which is marked off at maturity 

 by the narrow neck of the short seta : and the whole is borne upwards 

 on a more or less elongated pseudopodium developed by the parent 

 gametophyte (Fig. 132 H). 



In the facts thus briefly sketched there is no obvious evidence of 

 sterilisation : it is only when the peculiarly close analogies with the 

 Anthoceroteae are traced that any relation to the theory emerges. The 

 points of similarity with the typical Anthocerotales are seen in the form 

 of the sporogonium, in its absence of apical growth, and in the manner of 

 its primary segmentation : also in the origin of the columella from the 

 whole of the central group of cells, and of the completely dome-shaped 

 archesporium from the primary capsular wall outside it. It differs, how- 

 ever, in the mode of dehiscence, and in the fact that elaters are absent, 

 while the columella is not mechanically functional : it serves no purpose 

 beyond the nutrition of the considerable mass of spores. But as methods 

 of dehiscence, and of distribution -of spores not unfrequently vary within 

 near circles of affinity, this discrepancy does not seem of prime import- 

 ance. Lastly, however, there is, as a point of difference from Anthoceros^ 

 the absence of any functional assimilatory system in the sporogonium of 

 Sphagnum, though it is so well developed in Anthoceros. But, as Haberlandt 

 has shown, 1 functionless stomata, without pores and without intercellular 

 spaces below them, are present in large numbers on the capsules of 

 Sphagnum : from this he concludes " that it is certain not only that the 

 ancestors of the present Bog-Mosses had normal functional stomata on 

 their capsules, but also that the capsules of these ancestors possessed a 

 relatively well developed assimilatory system as well." All these con- 

 siderations taken together point to a close analogy (if nothing more) 

 between the two types. 



On the other hand, Sphagnum has always been ranked as a Moss on 

 such grounds as habit, absence of elaters, and structure of the archegonium : 

 but it differs from all other Mosses in the transverse segmentation of 

 the embryo, and in the absence of an apical cell : also (excepting Andreaea) 

 in the complete dome-shape of the archesporium, and in its origin from 

 the amphithecmm ; these all being features of correspondence with 

 Anthoceros. Such equivocal comparisons, with Liverworts on the one 

 hand, and with Mosses on the other, give Sphagnum itself a special 

 interest : at the same time they serve to link together the two large groups 

 of Bryophytes, and point to the propriety of regarding their sporogonia 

 equally from the point of view of a theory of sterilisation. The columella 

 would thus be held in both cases to be a consequence of sterilisation 

 1 Pringsh. Jahrb,, xvii., p. 474. 



