LEAF-ARCHITECTURE AS ILLUMINATED BY A STUDY OF PTERIDOPHYTA. 675 



by arched commissures, as in the mature leaf. (5) shows an obvious dichotomy of 

 the blade. Goebel* remarks the absence of an apex (Endtheil) between them ; 

 but this is in itself evidence of the dichotomy. He notes the occasional existence 

 of trifid leaves. These would be explained on the same footing as the ternate leaf 

 of Helminthostachys or Osmunda. Finally, (6) shows four lobes, the two distal 

 beino- clearly the result of dichotomy, since again the apex is absent. The facts, 

 viewed in the light of the juvenile leaves of other Ferns, seem to uphold the in- 

 terpretation of Braun rather than that given by Goebel. They indicate that the 



Fig. 9.— a, base of insertion of the four pinnae in Marsilia polycarpa, showing the vascular connections. 

 6, scheme of their arrangement as result of three dichotomies. 



x 35.) 



quatrefoil leaf is the result of three successive dichotomies developed in a scorpioid 

 sympodium. 



The development of the adult leaf has been traced by Hanstein.I The two 

 lower pinnae arise monopodially, one being inserted lower than the other. \ The 

 leaf is then three-lobed. "Next the apical cell as such ceases its activity, while 

 laterally from it the marginal activity increases, and so the middle lobe gradually 

 divides once more." The interpretation of this, in accordance with experience of 

 other related Ferns, is that the four-lobed leaf results phyletically from three 

 successive dichotomies, though actually the two lower pinnae are produced mono- 

 podially, as is the case with the lower pinnae of so many other Ferns. But the 

 original dichotomy is retained in the origin of the two later pinnae. 



The vascular connections of the mature adult leaf lead to clearer views of these 

 relations. They appear as in fig. 9, a. The insertion of the lower apparent pair is 



L.c, p. 254. 



f Pringsh. Jahrbuch, iv, p. 249. 



\ L.c, pi. xiv, fig. 2. 



