C.—GEOLOGY 75 
for classification. Here and there a plant may have characteristic vegeta- 
tive features, just as minerals may have characteristic colours, but taken 
by and large, the basis is as defective as classifying minerals by their 
colouralone. Modern methods for the determination of cuticular structure 
certainly improve matters as regards leaves and very young shoot tips, but 
they also are deficient, despite the work of many observers. I hope I 
may not be mistaken—I have said flowering plants. The case for 
gymnosperms is rather better; we may place some confidence in their 
determination by vegetative characters, but not flowering plants. For 
this reason we have more confidence in the determination of members of 
the older floras by vegetative characters, though caution is also necessary 
in these cases. 
Unfortunately the very abundance of flowering plants, and their 
absolute preponderance numerically over other types, in the Tertiary 
floras from the very beginning is a hindrance, not a help, to their use in 
geological work. ‘The determination and naming of the specimens is 
not easy, and the protest made by Mrs. Reid and Miss Chandler in their 
recent memoir * is timely in this respect, for workers are far too prone to 
give the name of a living genus to a specimen, and leave one to read down 
the description before one sees that their determination has been made 
on a few scraps that were hardly recognisable. Berry long ago protested 
against the nomina nuda in tertiary floral lists; Reid and Chandler’s 
protest against the use of definite names for indefinite specimens is no 
less deserved ; and Sahni’s statement?’ regarding the Mesozoic flora that 
“it is satisfactory to note that the hostile ranks of the species incerte@ sedis 
have suffered heavy losses’ cannot be taken as a matter of congratulation 
when applied to some lists of Tertiary floras. 
If workers refrain from giving the name of living types to fossils that 
merely look like them, and designate such with some name less committal, 
their floral lists would command more respect, and fewer unsafe deduc- 
tions would result, to the great benefit of geology and of our colleagues 
in other sciences who are coming more and more into the field of geological 
philosophy as we draw nearer to recent times. I refer particularly to the 
geographers, meteorologists, and archzologists. 
Yet there are certain generalisations that may be accepted. (1) The 
earlier Tertiary floras contain angiosperms almost exclusively of arbores- 
cent type—a feature of tropical and sub-tropical vegetation to-day. 
(2) The circum-polar spread, in early Tertiary times, of so many forms 
now living in tropical and sub-tropical lands indicates that these types were 
developed in the colder regions and migrated southwards, and not the 
reverse. (3) There is evidence that some rise in temperature in the 
north temperature zone in N.W. Europe and N.E. America took place 
about middle Eocene times, and that from that ‘ peak’ there has been 
a progressive lowering of temperature, with oscillations during the recent 
Ice Ages. (4) The Tertiary floras of the Northern and Southern hemi- 
sphere are not quite comparable for, while there appear to be considerable 
Variations in the northern hemisphere, there is a greater uniformity in 
46 The London Clay Flora, p. 46 (1933). 
47 Pyoc. Asiatic Soc. Bengal, Presidential Address, 1922. 
