nc. en]; ■' c l.t •. i c3'.r.^;i.. bul. to a part ol' tlio indusinm 

 as well, tliis name is not Gtrictl;- appropriate. Tliere is, 

 as v/e siuill see, no sinrlc cell after ■^ae two tertiary, 

 marginal colls are fonned v/hicli gives rise to the placenta 

 and sporangia onl;;, nor to these and the v/hole of the 

 indusimn. No true dernic'-itogen v/all is ever formed in this 

 m;^jrgin:il cell, hence the placenta and sporangia are not 

 of strictly epidermal origin. Tliey are formed rather, as 

 v/as foiind in the origin of the sporocar]_3 i:-om tiie lea.f , from 

 a cell that is ca.pahle of fonning at least tv;o meristem 

 layers, as it actually does in the segments of the stalk 

 (Fig. 50). 



In the growth of this part of the capsule that takes 

 place soon after section VI is formed, the marginal cell 

 elongates in a radial direction* At the same time its 

 wall "begins to separate at tlie outer or ventral end from 

 the outer cells of section V (s.c. Fig. 51). Meanv/hilo 

 the marginal cell has grov/n in tlie direction of the lcar;''h 

 of the sporocarp also (Figs. 56, 64). It soon splits by 

 a transverse anticline into ha.lves, of v/hcch the acroscopic 

 one soon comes to be 'lie larger. Then each, of these di- 

 vides by anotlier trtinsverse anticlin- (Fi . r.".) forming 

 thus four cells, of v/liich. the basicopic cell of 1 ii tv/o 

 foiTaed in the acroscopic coll of the first pair soon be- 

 comes the largest (p.m:L - sp.m.c. Fig. 64), and since wo 



