252 MR. W. B. HEMSLEY ON THE 
These totals are equal to about 70 per cent. of the genera, 
and 82°7 per cent, of the species of the whole flora. Similar 
proportions are found in the richest floras, and they therefore 
call for no special remarks. There is no great development 
in Tibet of any order of restricted distribution or of any 
order of peculiar structure or habit. The number and_ per- 
centages of the Composite call for some remark. Some years 
ago the late C. J. Maximowicz drew up a table showing the 
predominating natural orders in seven floral regions, ranging 
from the Caspian through Central Asia, North China, and 
Mandshuria to Japan, and in each of these regions the Composite 
considerably exceed any other natural order; in most instances 
by a third, varying from 15-4 per cent. in the Aralo-Caspian 
region to 7°7 per cent.in Japan. The decline was almost uniform 
from West to East. In Tibet nearly 11 per cent. of the genera 
and 18:7 per cent. of the species belong to this natural order ; 
and it may be added that the preponderance of Composite 
among the high-level plants obtains almost throughout the world. 
Whether this is altogether due to the special meaus for dis- 
persion possessed by members of this order would require 
much investigation to prove, but it may be put forward as 
probable. 
The total absence of certain orders and the poverty of others 
offer, indeed, points of greater interest. One of the most sur- 
prising facts is the extreme rarity of bulbous plants in a country 
where the conditions seem so favourable to the existence and 
propagation of this class, and this more especially, because 
bulbous plants abound in the dry regions of the surrounding 
countries, though at lower levels. Possibly more may depend 
on elevation, and the concomitant temperatures of the soil, 
than is obvious. But the fact remains that petaliferous mono- 
cotyledons are exceedingly rare; and not a single species 
of the Orchidacee occurs in the Tibetan collections dealt with 
in this paper, but Przewalski collected Orchis salina, a neat 
ally of the widely diffused O. latifolia, in the Koko Nor 
region. Considering the extent of saline country, a more 
numerous representation of the Chenopodiacee might have been 
expected. 
