POSITIVE d:6doitblement. 



99 



that two fertile "spikes" occur in close proximity 

 on the upper surface of the frond, instead at the usual 

 single one ; or the normal " spike " may be divided 

 into three or four branches (fig. 109). In 0. palma- 

 tum there normally occur many " spikes," and, for the 

 most part, on the upper surface, but quite close to the 

 margin (PI. XXXVIII, fig. 7) ; in some cases a 

 " spike " is actually marginal. Roeper and Groebel 

 are, doubtless, correct in stating that the fertile 



Pig. 110. — a, Aneimia. Frond showing the pair of approximated basal 

 fertile pinnae, l, Botrychium ohliquum. Frond showing two fertile 

 " spikes " partly united by their stalks. (After Chrysler.) 



" spikes " of the Ophioglossacese represent lateral 

 pinn« of the frond which have become displaced on to 

 the ventral surface. The basal fertile pinn« of the 

 Schiz^aceous fern Aneimia offer a perfect analogy in 

 support of this view ; these pinii^ converge across the 

 face of the frond, turning their dorsal surfaces towards 

 the axis (fig. 110 a). We can imagine it easy for them 

 to fuse by their outer margins and to come in the 

 course of time to be seated, as a single " spike," on the 

 upper surface of the frond. This was Roeper 's view 

 of the origin of the Ophioglossaceous fertile " spike." 

 It is a very plausible one. The abnormal branching 



