680 PROFESSOR F. O. BOWER ON 



sometimes twice, on their course outwards to the margin. No anastomoses are seen. 

 Thus the venation of the segments of the adult leaf is merely an extension of the 

 type shown in the juvenile leaf. But the adult blade as a whole has a rather unusual 

 branching, which is not, however, constant in its details ; but a central type is well 

 shown in Fitch's drawing in Hooker's Second Century of Ferns, pi. xciv. It is, as 

 a matter of fact, a relatively simple development from the type of the ternate coty- 

 ledon. The place of the middle lobe is taken by a scorpioid sympodial series of 

 lobes, equally developed on each side, so that they are arranged in apparent pairs, 

 after the manner of the first pair. But each of the two lateral lobes of the cotyledon 

 is here replaced by a sympodium of the helicoid type. It branches repeatedly, the 

 lower (katadromic) shank developing as a lobe, while the upper (anadromic) shank 

 forks again, and so on. This development is already foreshadowed in the cotyledon. 

 It is a type of branching known as an anadromic helicoid scheme. 



Fig. 13.- — Cotyledon of Botrychium virginianum. (x 4. ) 



A circumstance which may in some degree explain the advanced structure of the 

 cotyledon is that the young plant springs from a bulky saprophytic prothallus. 

 Botrychium also has a large prothallus, and the cotyledon in B. virginianum is a 

 relatively large one. It has already been drawn by Jeffrey # and by Campbell^ 

 The drawing of a cotyledon shown in fig. 13 is from a specimen collected in Jamaica. 

 It is ternate like that of Helminthostacliys, but more deeply cut into secondary lobes, 

 so that a single vein may supply each one of these. The plan of the venation is the 

 same as in Helminthostachys, though simpler. It also is a dichotomy developed as 

 a scorpioid sympodium. From the cotyledon of B. virginianum to the adult leaf 

 the development involves only continued progress along the lines thus initiated, the 

 adult leaf being itself ternate. Fig. 14 shows its apex, but other species, such as 

 B. ternatum, have a more robust habit, with thicker leaves, less finely cut. Fig. 15 

 shows a normal lobe of such a leaf, together with two imperfectly separated segments 

 at its base. A comparison of this with fig. 1 1 of Angicpteris, and fig. 2 of Osmunda, 

 shows a very striking similarity both as regards venation and the relation of the 

 lobes. Notwithstanding the differences noted, the venation is the same as in 



* " Gametophyte of Botrychium virginianum," Trans. Canadian lust., 1896-1897, vol. v, pt. 2, pi. i, figs. 9, 10. 

 t Eusp. Ferns, fig. 7. 



