ffor^ili:'. Tho ^.tir";" o'' "v ovrr- noctionr. of Pilftl:^ri-- '-It ■'i- 

 lifera ".]iov/6 tliat in l)oth sliapo tind position it iirree^ en- 

 tirel^' v/ith Marnilia and thus v/itli tlie otlier rener:\ nientionr 

 above excopi. Pterir, . 



Tlie tv;o sided apical cell thus formed in Ilarsilia, con- 

 tinues its .'^rov/th and actiiaty, cutting off seT"ients alter- 

 nately towards the dorsal and lateral sides of the stem 

 (Fi.^. 3^), or, since the ventral side of tlie yoim- leaf looks 

 tovrards the apex of the sten (Fir-. 3)>the se^aents are cut 

 off alteniately toward the right and left of tlio leaf itself". 

 Tliis r^oes on until about fifteen or sixteen segments have 

 been formed on eacli side, v;lien th.e apic?,l cell ceases to 

 function as such and apical grovrt-h. ends. Th.e exact fate of 

 tlie apical cell vjas not made out in tlie leaf, but it is 

 probable thiat a periclinal wall is finall^^^ foi-med instead of 

 tlie usual segment wall, as such, a wall ".-as seen several tim.es 

 in the p.pical cell of tlie sporocarp. Sadebeck ('73), Ki-y {'7^ 

 and Bower ('84) ^lave actually observed such, a wall in th.e 

 apical cell of th.e leaf, but they v;ere not able to make out 

 just liov^ many sepjnents v;ere cut off before it appeared. 



In I/Iarsilia th.e great rer-ularity of the division'- in 

 the seg-aents of tlie leaf, as well as tlie fact tliat certain 

 cells remain of th.e full lengt.h of the se-^nent (Fir. 5), 

 mal'e it possible to determine quite exactly t^'e >^umbBr o'" 

 segments cut off (Fi/T. 4). The onlv doubt is in rej^ard to 



