lo Lang, Vasculaj- Anatomy of tJie OpJiioglossacecc. 



this species. Such a distinction is practically impossible 

 in some other species, such as B. virgi?iiajmin, in which 

 cambial activity proceeds at once and all the elements of 

 the xylem are in centrifugally developed radial rows. 

 The leaf-trace of BotrycJmun in endarch, and no adaxial 

 completion of the xylem such as is described above for 

 Heli)ii)ithostachys occurs normally. It may however be 

 pointed out in passing that if all the xylem in the stem of 

 B. virgifiianum is regarded as secondary, this involves 

 regarding the leaf-trace as also wholly composed of 

 secondary xylem. Such an interpretation is not adopted 

 in the analogous cases among Gymnosperms and Dicoty- 

 ledons when secondary thickening supervenes early and 

 makes a sharp distinction of primary and secondary 

 xylem impossible. 



Taking all the facts I have been able to ascertain 

 from the literature and from the study of a number of 

 species of this genus into consideration it seems justifiable 

 to distinguish in the stem of BotrycJnujn centrifugal 

 primary xylem, and, in exceptional cases, centripetal 

 primary xylem as well as secondary xylem. This brings 

 the structure into line with the stem of HelmintliostacJiys, 

 the differences depending on the absence or poor develop- 

 ment of the centripetal xylem and the regular and early 

 supervention of secondary thickening in BotrycJiuun. 



The typeof stelar structure which persists throughout 

 the stem in the simpler species of Ophioglossuvi is readily 

 related to that of specimens of Botrychiiun lunaria in 

 which no secondary growth is evident and the leaf-gaps 

 overlap. In Ophioglossuju the stele consists of a network 

 of separate collateral strands. The xylem of each strand 

 is endarch and normally only centrifugal primary xylem 

 is present. No marked secondary thickening is found. 

 In a species of Op/noglossian collected in Ceylon the 



