282 RESEARCH IN CHINA. 
and Podozamites emmonsi is, I believe, more or less distinctly for a Rhetic 
correlation; and this reference seems to find support, not only in the 
general affinity of the flora to that of the Rhetic in eastern Asia and Europe, 
but also in the absence of species distinctly characteristic of the Jurassic. 
It may well be urged that a more satisfactory collection containing 
additional species might bring to hand forms indicating an earlier or a 
later age. The possibility, which most paleobotanists will admit, that 
the discovery of additional species might reveal the presence of types 
characteristic of the Lower Gondwanas, such as would indicate a refer- 
ence of the beds to the Permian, in accordance with the fossil mollusks 
and the stratigraphy, is the chief cause for this discussion. For if it is 
permissible even tentatively to regard the flora as falling within the upper 
limits of the Paleozoic, it then will not be necessary to conclude that 
the K’ui-chéu series transgresses from the Permian into the Mesozoic. 
Even in the small amount of material available the representatives 
of Nilsonia, Podozamites, and Angiopteridium are in themselves sufficient 
to preclude a consideration of the flora as possibly belonging to that of 
any horizon of the latest Carboniferous or Permian in Europe or North 
America. The alternative that they may be fragments of the Lower 
(Paleozoic) Gondwana flora, near whose province the K’ui-chdéu basin lies, 
would at first hand seem to be favored by the slightly Mesozoic aspect of 
the other Gondwana floras and the appearance of Secondary types in the 
flora of the Middle Gondwana. 
Viewed from the Paleozoic Gondwana point the evidence is largely, 
but far from wholly, negative. Not only have none of the types charac- 
teristic of the Lower Gondwanas yet been found in the K’ui-chéu basin, 
but the species from the latter include nothing, with the possible though 
doubtful exception of Cladophlebis petruschinensis, that, so far as I am 
aware, has yet been found at any point in the Lower Gondwanas. Fructi- 
fications of the group represented by Angiopteridiwm richthofeni are not 
yet known below the upper Trias; nor is the genus Nilsonia of an earlier 
date. Podozamites lanceolatus occurs in the Upper Gondwanas at Jabalpur 
and in the Trias at Ipswich in Queensland; but the presence of the genus 
Podozamites itself at K’ui-ch6u argues against a Permian age. Further- 
more, it is important to note that Cycads appear to dominate in the 
K’ui-chéu flora, while in the Paleozoic Gondwanas they are very rare or 
almost wanting and are practically restricted to the genus Plerophyllum. 
Remains of Cycads are perhaps as infrequent in the Permian rocks of the 
Gondwana province as in the contemporaneous strata of Europe. The 
same is true of Cladophlebis. Regarding Czekanowskia rigida, whose 
distribution is Mesozoic, it must be said that the identification by both 
Schenk and Schmalhausen is questionable. 
