186 



1). H. Scott. 



much-branched petiole and rachis. often perfectly preserved; in the 

 larg-er parts the xylem has a characteristic cruciate transverse section, 

 and bears some resemblance anatomically to Botryopteridian petioles; 

 the tracheides are scalariform. As no leaflets have been found, the 

 branches of the rachis terminating' in the crowded stalks on which 

 the sporan;^ia were borne, it appears that the fossil, as commonly 

 occurring-, represents the fertile frond only. The wall of the sporangium 

 is many cells in thickness; there is no annulus, but a well-marked 

 stomium is present. The existence of germinating spores has been 

 suspected for some time, and quite recently clear proof of this has 



— c 



Fig. 16. Sfauropteris oldhamia. Four germinating spores, from the interior of a 



sporangium. X 335. 



been obtained. In one sporangium the stages of germination are quite 

 as clear as in the original, unassig-ned. specimen, and present the 

 same phenomena, agreeing with those observed in recent Fern-spores 

 when germinating (Fig. 16). Thus we now have good evidence that 

 Stanropteris oldhamia, at any rate, was a true Fern. The sporangium 

 in which germination was first observed, though larger than that of 

 Siauropteris oldhamia, is otherwise verj' similar, and it is probable 

 that this, too, may be referable to a species of Stanropteris. 



The probability is in favour of an affinity between the genus 

 Stanropteris and the Botryopterideae, though the sporangial characters 

 scarcely admit of a definite reference to that family. The discovery 

 of the germinating spores, in the cases just described, much strengthens 

 the conclusion that the Botryopterideae and allied Palaeozoic plants 

 were really members of an ancient race of Ferns. 



Palaeozoic Marattiaceae. 



No conclusion in Palaeobotany has met with more general accep- 

 tance than that of the prevalence of Marattiaceous Ferns in the 



