i84 MARSILEACEAE [ch. 



At maturity the whole sporocarp spHts like a bivalve shell on access of 

 water: the internal tissues behave as in Pihilaria, but the whole tract of 

 tissue at the ventral margin remains coherent; as it swells it emerges as 

 a worm-like body, and draws out the receptacles and sporangia attached 

 to them, exposing them fully to the water which caused the rupture (Fig. 

 465, B, C, D). 



The fruit of Regnellidium is generally of the type of Marsilea, but there 

 is no worm-like mucilage-ring extended on rupture: the dehiscence is more 

 nearly like that in Pilularia. The megaspores are spherical, and the micro- 

 spores number about 60 in each microsporangium. 



The development of these very peculiar organs has been exhaustively 

 followed by Duncan Johnson, from whose description the following details 

 relating to Marsilea are drawn {Marsilea, Atin. of Bot. xii, p. 119, 1898; 

 Pilularia, Bot. Gaz. xxvi, p. i, 1898). The whole leaf in a very young state 

 has the bifacial initial cell usual for Ferns (Fig. 466, A). From one of its 

 basal segments the sporocarp takes its origin, being, like normal pinnae, 

 lateral {F). It assumes a structure very like a young leaf, but as it grows it 

 curves sharply so that the distal end is directed horizontally. Involutions 

 of its surface then appear on the side facing downwards: these are called 

 the soral canals {sc, C — F), and they are disposed in two alternating rows 

 {H). Each of these is lined on its central face by a tissue designated the 

 "indusium," of which two layers may be recognised {i.inel, a.ind, F, G). 

 The receptacle of the sorus faces this in each canal, and the sporangia arise 

 upon it with a slightly marked acropetal succession {F). The canals close 

 about the time the formation of sporangia begins. The megasporangia are 

 borne distally upon the receptacle, and are followed by microsporangia {ma, 

 mi, Fig. 464, j5). In a strict sense there is thus a gradate sequence, while 

 Campbell states that sporangia may be interpolated at any point in the 

 sorus {I.e. p. 438). The structure of the sporocarp at this stage is revealed 

 by the transverse and median-longitudinal sections {G, H). These facts 

 indicate for Marsilea a soral condition distinctly in advance of the Simplices. 

 Its sori may be described as gradate, with a tendency towards the "mixed" 

 condition. 



The details of development in Pilularia are essentially similar (Goebel, 

 Bot. Zeit. p. 45, 1882). Von Goebel has clearly shown that the origin of the 

 sori is marginal : so that the sporocarp may be referred in origin to a modi- 

 fication of a typical Leptosporangiate sporophyll {Organographie, 191 8, ii, 

 2, p. 1 1 36). 



The development of the sporangia follows the ordinary plan for Lepto- 

 sporangiate Ferns, but Campbell states that it corresponds most nearly to 

 that of the Schizaeaceae. There is no highly elaborated opening mechanism, 

 the sporangia appearing as hyaline sacs. But Campbell has noted in 



