XIV 



GAMETANGIA AND SPORANGIA 



291 



of the plants that bear them, modified it may be in some measure by the 

 biological conditions under which they live. The comparison is best drawn 

 between antheridia and homosporous sporangia, both of which, as already 

 shown, are conservative in their character. A rough but by no means exact 

 numerical estimate may be made by counting the spermatocytes traversed 

 in the median vertical section of an antheridium, and comparing it with the 

 number of spore-mother-cells in a similar section of the corresponding sporan- 

 gia(Fig.28i). Much depends upon thesectionbeingreallymedian.atangential 

 section being liable to give too low a figure. Other sources of inaccuracy of 

 the comparison may arise from the shape, and the degree of standardisation 

 of the parts. In most of the advanced Ferns the sporangium is strictly 



Fig. 281. Antheridia of various Ferns, to show differences in number of sper- 

 matocytes. a = Gleichenia pectiriata, after Campbell (231, Fig. 209, B), 

 46 cells in section; b=Osmiinda finiramoinea, after Campbell (231, 

 Fig. 195, D), 33 cells; c -Hymciiophyllum, after Campbell (231, 

 Fig. 217), 26 cells; d = Trie ho manes vcnosuin, after von Goebel (Organo- 

 graphie. Fig. 9r2, v), 18 cells; e-Nephrodiiim Filix-inas, after Kny 

 (Wandtafeln), 19 cells; f=Ceratopteris thalictroides, after Kny {Parkeri- 

 aceen, Taf. XVHI, Fig. 11), 7 cells; g~Schizaea piisilla, after Britton and 

 Taylor (PI. 3, Fig. 52), 7 cells. 



Standardised in size and shape, and the spore-output can therefore be fairh' 

 arrived at from the number of spore-mother-cells traversed in median section. 

 In the Eusporangiate Ferns they are more apt to be variable even in the 

 same species or individual, as is seen in Danaea {Land Flora, Fig. 286), or 

 Mai'attia {I.e. Fig. 285). The antheridia may also be variable in the species 

 or even in the individual, as is clearly shown by Campbell's drawings of 

 Gleichenia laevigata {Ann. Jar d. Bot. Buit. Viii, PI. xi), which give sperma- 

 tocyte-counts as divergent as 95, ^6, 35. Notwithstanding such sources of 

 error the following table shows a substantial parallelism of the figures for 

 the spermatocytes and the spore-mother-cells as seen in vertical sections of 

 the antheridia and the sporangia of the Ferns named. In the fourth column 

 the spore-output is given, as estimated in the larger sporangia from sections, 



.19—2 



