2o8 GLEICHENIACEAE [CH. 



greatly, and become polynucleate: the cytoplasm and nuclei aggregate 

 closely round the sporogenous cells, and the walls are finally absorbed 

 (Fig. 490,7', k). The definitive sporogenous cell in G. flabellata undergoes 

 successive divisions (Fig. 490, d, j^, but these are continued beyond the 

 limited number usual in Leptosporangiate Ferns, producing a very consider- 

 able mass of spore-mother-cells, of which 45 is a mean number exposed in 

 a single vertical section (Fig. 490,7): their total number is probably about 

 360. If each of these on tetrad-division produced four spores the potential 

 output would be 1440. As tetrad-division approaches the spore-mother- 

 cells separate and round off: the tapetal plasmodium penetrates between 

 them, and is finally absorbed in the usual way. As in the Schizaeaceae 

 the form of the spores is not constant for the Family, being sometimes two- 

 sided {G.pedalis and Stromatopteris), sometimes tetrahedral {Eii-Gl. dicarpa). 

 The estimated potential number of spores is not fully realised: nevertheless 

 the output is very large. Direct countings have given the following results 

 for various species: 



G. flabellata, 794, 695, 838, 634. 



EiL-Gl. circinata, 241, 242. 



Eu-Gl. riipestris, va.r. glaucescens, 220, 232, 244. 



Eu-Gl. hecistophylla, 265, 272. 



G. pectinata, 204, 229, 218, 240 (Williams). 



G. linearis, 251, 319, 120, 128, 194, 228 (Williams). 

 The largest of these figures suggest a typical number of 5 12-1024, which 

 probably applies for G. flabellata, a species regarded on other grounds as 

 phyletically a central type for the genus. All the rest show markedly 

 smaller numbers, mostly falling between such typical numbers as 128 and 

 512. It has been noted that in Eii-Gleichenia the sporangia are large, but 

 the individual spores are also large, a fact that harmonises with their 

 relatively small number. In G. pectinata and linearis, however, the numbers 

 are relatively small and the sporangium is small also. An examination of 

 it in the young state has shown that the number of the spore-mother-cells 

 in a single section is about 15 to 20, and their total number may be com- 

 puted as approximating to 64. This would give a probable number of spores 

 per sporangium approximating to 256, though the actual countings are 

 sometimes much lower (Fig. 491). It thus appears that within the genus 

 there is considerable variety in sporangial spore-output. G. flabellata 

 compares more nearly with the Eusporangiate Ferns than with the Lepto- 

 sporangiates, while Eii-Gleichenia and particularly G. pectinata and linearis, 

 with smaller numbers, correspond to the Osmundaceae and some Schizaea- 

 ceae in approaching the simpler Leptosporangiate state. 



Stromatopteris is somewhat variable in the structure of the sporangium : but it is clearly 

 of a Gleicheniaceous type, though sometimes the annulus is almost horizontal, and the 



