THE GLOSSOPTERIS FLORA i8i 



shows that in fertile fronds the fructification was a marginal 

 one, like the fructification of Pteris'' } On the strength of 

 these discoveries, Feistmantel instituted three groups of 

 Glossopteris founded on the nature of the sori. Carruthers' 

 specimen of G. Browniana from Australia is the type of 

 Group I, and is compared with the recent genus Anthro- 

 pJiyuin ; of Group 2, G. inciica Schimp. is the type, and is 

 compared with Polypodium, while Group 3 is represented 

 by Feistmantel's examples of G. aiigustifolia with the 

 Pteris-like fructification. It has recently been pointed out - 

 that this classification of Indian and Australian forms has been 

 further complicated by the discovery of the Indian type of 

 sori in Australian leaves.'^ The fragment responsible for 

 this supposed complication is imperfectly preserved and 

 shows on its surface three " sub-oval convex impressions," 

 which cannot be accepted as genuine traces of sporangia. 

 In referring to this discovery, Etheridge remarks that the 

 supposed sporangial groups do not convey to him the idea 

 of sori. but adds : " We must however bow to the evidence, 

 such as it is, and it would thus appear that we have in 

 Australia Glossopteris-like leaves possessing two out of 

 three forms of fructification ".* If evidence is unsatisfactory 

 it is hardly a matter of duty to accept it, or to deduce from 

 it far-reaching conclusions I On the whole, it would seem 

 the safer course to admit that as yet no trustworthy example 

 of a fertile Glossopteris frond has been recorded from either 

 India or Australia, and we have certainly no data on which 

 any classification can be leofitimatelv founded. 



Until we possess leaves showing distinct sori and 

 sporangia it is of little use to institute a comparison of the 

 sterile fronds with those of recent ferns. The leaves of 

 species of the polypodiaceous genus Ant/irop/iyum agree 

 very closely in form and venation with Glossopteris ; in 

 some species from Fiji there is little or no midrib ; in others 

 from South America the midrib may extend to the leaf tip. 

 In the fertile fronds the sporangia occur in continuous lines,. 



^ Feistmantel (2), p. 97. - Etheridge (4). 



•^ Mitchell. •* Etheridge (3), p. 237. 



