The Glossopteris Flora. 35 
types are unrepresented. It is noteworthy that Lepidodendron and 
Sigillaria occur in association with Glossopteris in South Africa 
and South America and the long strap-shaped leaves which are not 
uncommon in the Glossopteris beds and are usually assigned to 
Neegerathiopsis may he identical with the northern genus Cordaites. 
Making due allowance for the fact that authors are not infrequently 
led to give undue importance to wide geographical separation by 
the use of different names for very closely allied plants, and 
admitting that further research may connect more closely the 
Permo-Carboniferous floras on the two sides of the equator, the 
statement holds good that there is evidence of the existence of 
two more or less well-marked botanical provinces in the latter part 
of the Palaeozoic era. 
Dr. Halle gives an interesting account of the Lafonian glacial 
beds which occur immediately below the plant-bearing strata and 
shows clearly by his description of the rocks, some of which he 
speaks of as tillites (a term instituted by Professor Penck), and 
striated rock-floors that in the Falkland islands, as in other countries, 
where the Glossopteris flora occurs, sheets of ice were spread over 
the land surface of which the Falkland islands are a diminutive 
survival. > 
The question of the widespread ice-action in the southern 
hemisphere and in India in relation to the Glossopteris flora is dealt 
with by M. Paul Bertrand who gives a clear summary of the chief 
geological and botanical facts. After a brief reference to some 
of the hypotheses advanced by way of explanation of the glaciation 
during part of the Palaeozoic period, he remarks that geologists have 
appealed for confirmation of their views to palaeobotanists and to 
meterologists. Unfortunately, he adds, the study of Carboniferous 
and Permian floras instead of facilitating the solution of the prob¬ 
lems raised by the association of the Glossopteris flora and ice- 
formed rocks, raises fresh difficulties. The fact that some of the 
northern genera such as Lepidodendron and Sigillaria, typical 
European Coal-Measure plants, grew in South Africa and South 
America in regions recently emerged from glacial conditions sug¬ 
gests the possibility that the vegetation of the Coal period north of 
the equator may have flourished in a temperate rather than in a 
tropical climate. He proceeds to quote certain conclusions regarded 
by some authors as favourable to this view, adding that if the 
northern flora grew in a temperate climate the lowering of tempera¬ 
ture in the south, sufficient to bring about glacial conditions and to 
effect such a change in the vegetation as is revealed by a comparison 
of the two floras, need not be very great. There are, however, 
serious objections to the acceptance of the view that the nature of 
the Upper Carboniferous and Permian plants of Europe and North 
America is consistent with a temperate climate. M. Bertrand 
points out that it has been frequently asserted that Lepidodendron 
and other Carboniferous plants indicate by their anatomy adaptation 
to an environment physiologically dry ; but it is significant that the 
xylem elements in Lepidodendrese are very much larger than those 
of recent conifers. The diameter of the tracheae in the former 
varies from 100 to 125/x; in the latter from 20 to 25g. In spite of 
the abundance of water which is suggested by the unusual size of 
the conducting channels, M. Bertrand adds, the Lepidodendreae 
endeavoured to limit their transpiration. The stems possessed a 
thick covering of cork ; the leaves were reduced in breadth and 
