BY T. W. E. DAVID. 249 



9. Eliminating the habit of growth, of which we know nothing 

 definite in Gangamopteris, it is questionable how far the leaves 

 of the two genera Glossopteris and Gangamopteris can be separated, 

 except in extreme cases. 



10. It is evident that those leaves of Sagenopteris, possessing a 

 moderately developed mid-rib, such as S. elongata, Br., are with 

 difficulty separated from those of transitional species of Glossop- 

 teris, but once such leaves referred to Sagenopteris are seen attached 

 to their leaf-stalk, or it is permitted to observe the fructification, 

 the separation is possible. 



11. We have no leaves in Australian rocks exactly analogous 

 to Anthrophyopsis, Nath. 



12. In all probability it will be necessary to separate Feist- 

 mantel's S. (?) polyphylla from Sagenopteris, and restrict Dacty- 

 lopteris for its reception. It presents the additional peculiarity 

 of possessing the secondary veins all of one order. 



10. Stratigraphical Distribution of Glossopteris in 

 Australia. 



Glossopteris is known to occur in Queensland, N. S. Wales, and 

 Tasmania. It has also been recorded from New Zealand, but 

 hitherto its occurrence in that country cannot be considered to be 

 an established fact. As regards the occurrence of Glossopteris 

 there. Sir James Hector says* — " At the base of the Kaihiku 

 Series are the Glossopteris beds of Mt. Potts." To these 

 Kaihiku beds he assigns a Permian age, but the range of the 

 genus Glossopteris is stated to extend into the higher Wairoa 

 Series of Triassic age. On this Sir James remarksf — " In some 

 districts the Wairoa Series is divided into two horizons, yielding 

 marine fossils, separated by sandstones containing fossil plants 



* Indian and Col. Exhib. N. Zealand Court, Geol. Cat. and Guide, 1886, 

 p. 77. 



t Ilnd. p. 74. 



