610 F. ^V. KNOWLTOX MESOZOIC FLORAS OF THE AMERICAS 



and well preserved, the flora as a whole is small and poorly represented. 

 Only eight genera are recognized, and of these several are so poorly pre- 

 served that specific identification was not attempted. A number of the' 

 forms recognized are identical with those described by Wieland^ from the 

 Mixteca Alta, in Oaxaca, from beds also referred to the Lias, This is a 

 comparatively rich flora, comprising 21 genera and about 60 forms. It 

 is especially rich in forms as well as individuals of the peculiar William- 

 sonias. This flora can hardly be accepted at its face value. It seems to 

 me that there must have been either a mixture of horizons or a misiden- 

 tification of generic types. If the genera Noeggerathopsis, Trigonocarpus, 

 Rliabdocarpus, Aletliopieris, Sphenopteris, and, above all, Glossopteris, 

 have been correctly identified, it w^ould certainly argue for a much older 

 position than the Lias, and the Williamsonias and other types of cycads 

 would not be out of place at a higher horizon. 



The entire South American Continent is without a known locality for 

 an undoubted or at least adequate Jurassic flora. As already pointed out, 

 the floras from Chile and Argentina above referred to the Triassic may 

 possibly be referable in whole or in part to the Lias instead of the Ehaetic, 

 but further data must be forthcoming before the matter can be settled. 

 Thus, from Piedra Pintada, on the northern border of Patagonia, Kurtz 

 has described a small collection of plants procured by Roth. They are 

 associated in beds with marine fossils considered to be of Liassic age. 

 Kurtz has compared the plants with the Rajmahal flora of the Upper 

 Gondwanas of India. 



The largest and by all odds the most interesting Jurassic flora is really 

 extralimital. This is the Middle Jurassic flora described by Halle^^ from 

 Hope Bay, Graham Land, 63° 15' south, and just outside the Antarctic 

 Circle. It embraces 61 forms, of which number 21 are definitely identi- 

 fied with previously known forms, and of these IT are found elsewhere in 

 strata believed to be of Middle Jurassic age, although it includes some 

 types that are older and some that might be yomiger. The closest affilia- 

 tion of this flora is shown to be with the well known Jurassic of York- 

 shire, England, there being no less than 9 of the 21 species in common. 

 There are no South American floras of any importance that can be con- 

 sidered contemporaneous with this Graham Land flora. 



Possibly contemporaneous with the Graham Land deposit is a collec- 



* G. R. Wieland : La flora Liasica de la Mixteca Alta. Bol. Inst, geologico de Mexico, 

 no. 31, 1914. 



" T. G. Halle : The Mesozoic flora of Graham Land. Wissen. Ergebnisee d. Schwe- 

 dinschen Sudpolar-Exped. 1901-1903, vol. 3, no. 14, 1913, 



