HYMENOPHYLLACEAE 



581 



h 



than in ffymenopkyllum, while on the other hand the receptacle is usually 

 longer, and has more continued intercalary growth, the question arises whether 

 the larger number of sporangia will approximately compensate for their lower 

 individual output. A computation was made of the output per sorus in 

 H. 7'unbridgense and dilatatum, and compared with a similar computation 

 in T. rfiiifonne and radicans* with the result that, notwithstanding the great 

 variations in spores per sporangium, the output per sorus appears approxi- 

 mately uniform for the cases quoted. Thus the increased length of the 

 receptacle and higher number of the sporangia tends to compensate the 

 smaller output per sporangium which is seen 

 in its extreme form in some species of 

 Trichomanes. 



It has been seen that in G/eichenia, where 

 there is a median dehiscence of the sporangium, 

 elbow room is required for the process of 

 mechanical ejection of the spores, and that this 

 is only possible where the sporangia are loosely 

 arranged. In a crowded sorus such as that of 

 the Hymenophyllaceae the necessary space is 

 not available. In Loxsoma this difficulty is met 

 by the incomplete development of the annulus, 

 which then merely opens the distal end of the 

 sporangium and allows the spores to be shaken 

 out. But in the Hymenophyllaceae the whole 

 mechanism is altered by the adoption of the 

 oblique lateral dehiscence : as this is usual in 

 the basipetal sori of the Gradatae it demands 

 the greater attention. The sporangia are placed 

 relatively to one another as seen in the diagram 

 (Fig. 327), in which the cells of the annulus 

 traversed in the median section of the sporangium are indicated by 

 heavier lines. It is plain that on dehiscence taking place laterally, the 

 distal side of each annulus has freedom to alter its form independently 

 of the adjoining sporangia: 1 the oblique position of the annulus thus 

 finds a practical explanation, and is even a necessity where the sporangia 

 are short-stalked and imbricate. The dehiscence is, however, aided by its 

 occurring usually in strict basipetal sequence in the Hymenophyllace;i< .- 

 after the lateral slit is formed, the annulus is first of all reflexed, and then 

 recovers with a sudden jerk, which often dislodges the whole sporangium, 

 and scatters the spores. The result in Tric/n>mancs is that the distal 

 end of the receptacle may remain bare of sporangia, an indication of the 

 extent of the intercalary growth, while young sporangia may still be found 

 around its base. 



FIG. 327. 



Diagram illustrating the relative 

 position of the sporangia on the 

 receptacle in the Hymenophyl- 

 laceae. It was constructed from 

 Prantl's section of a mature spor- 

 angium of Triclioina lies Speciositm. 



1 As a matter of fact the freedom is greater than is shown in the diagram, for the s 

 alternate, and are not disposed in closely consecutive orthostichies. 



