X THE IIOMOSPOROUS LRPrOSPORANGIATAi 383 



(Trichouiancs) or is divided into two valves (Ilymciiophyl- 

 htui). Many species of the former genus, hov^ever, shov^ an 

 intermediate condition, with the margin of the indusium deeply 

 two-lipped. 



The first sporangia arise at the top of the placenta (Fig. 

 221), hut the apex itself does not usually develop into a spo- 

 rangium. After the first sporangia have formed, new* ones 

 continue to develop. Near the hase of the placenta a zone of 

 meristem is formed, which constantly contributes to its growth, 

 and the young sporangia arise from the surface cells formed 

 from this meristem. The mother cell is very easily distin- 

 guished by its larger size and denser contents. About every 

 third cell seems to develop a sporangium, but this probably is 

 not absolutely uniform. The first wall is usually nearly vertical, 

 and cuts off a narrow segment from one side of the mother cell 

 (Fig. 221, A). This in most cases examined was next fol- 

 lowed by a wall almost at right angles, separating a small basal 

 cell. After these preliminary divisions, which form the very 

 short stalk, the next divisions are exactly as in the Polypodi- 

 acese, and give rise to the central tetrahedral cell with the four 

 peripheral ones. Prantl ( (i), p. 39) states that the first divi- 

 sions of the 'cap cell are also spirally arranged. In T. cyrto- 

 theca (Fig. 221) the tapetum is massive, and composed 

 throughout of two layers. The archesporium divides into 

 eight cells, whose further history is the same as in other Ferns. 

 The annulus in the Hymenophyllacese is large, and situated 

 much as in Gleiclicnia. According to Prantl, it arises in part 

 from the cap cell and partly from numbers one and three of the 

 primary peripheral cells. Where the young sporangium is cut 

 longitudinally (Fig. 221), the annulus cells are at once recog- 

 nised by their larger size, especially upon the dorsal side. 

 Their radial and inner walls become very thick, and a horizontal 

 section (Fig. 221, F) shows that the annulus is not complete, 

 but is interrupted on the inner side where the stomium is formed. 



Apogamy and Afrospory 



Both of these phenomena have been discovered by Bower 

 (8) to occur not infrequently in Trichomanes, and probably 

 further investigations will reveal other instances. Apogamy 

 was common in T, alatum, in which species archegonia were 



