MARSILIACEAE 551 



It is then specially through Lygodiitm that the nearest connection may 

 be sought with ancestral Fern-types, which should have a protostelic stock, 

 and show dichotomy both in axis and in leaf: large sporangia, with the 

 annulus not necessarily uniseriate, and with a relatively large spore-output. 

 As these characters are less decisive than those seen in the preceding 

 families, the difficulty in locating the Schizaeaceae will be correspondingly 

 greater. 



MARSILIACEAE. 



It is probably in near relation to the Schizaeaceae that the Marsiliaceae 

 find their most natural position. This has been argued by Campbell, 1 and 

 the data relating to the sporangia appear specially convincing. But as these 

 heterosporous plants constitute a peculiarly specialised line, which has 

 probably never advanced further, the discussion of them, however interesting 

 in itself, does not bear directly upon the problem in hand. Accordingly no 

 detailed account will be given of the Marsiliaceae. 



There is, however, one striking feature in their morphology which calls 

 for notice here, since it provides an apparent analogy with the Ophio- 

 glossaceae : viz. the position of the "sporocarp." This curious and complex 

 body may be stalked or sessile and be borne singly attached to the leaf- 

 stalk, or in considerable numbers as in M. pofycarpa : finally it may itself 

 be branched. From its position and structure, as well as from the way in 

 which the sporangia are produced, a foliar character is probable, notwith- 

 standing that the form is far removed from that of any ordinary leaf-segment : 

 and this is the conclusion to which study of the development has clearly led. 



Johnson' 2 found that both in Marsilia and Pilularia the origin of the 

 sporocarp is from a cell of the marginal series of the leaf: he concludes 

 that the capsule is the equivalent of a branch of a leaf in which the 

 marginal cells have been devoted to the formation of sporangia instead of 

 a lamina. Goebel obtained a somewhat similar result from the investigation 

 of M. polycarpa : y here the numerous sporocarps arise in acropetal succession 

 from the margin of the leaf, but from one margin only : they assume upon 

 the leaf-primordium the same position as the sterile pinnae. But the arrange- 

 ment of the cells is different : the sporophylls have a two-sided initial, while 

 the sterile pinnae show from the first a marginal growth. This, however, 

 need not preclude the recognition of the sporocarps as the correlatives of 

 pinnae. 



The analogy of these bodies with the spikes of the Ophioglossaceae is 

 too obvious to escape remark, and some have seen in them and their 

 pinna-character strong evidence that the same must be the nature of the 

 Ophioglossaceous spike. The objections to this facile conclusion are two : 



1 American A'atura/ist, 1904, pp. 761-775. 



'-Ann. of Bot., xii., p. 119; and Bot. Ga:.., xxvi., p. I. 



3 Organography, vol. ii., p. 479. 



