226 THE JOURNAL OF BOTANY 
provinces, for the determination of which I have, after long de- 
liberation, resorted to the number of species of the ten largest 
natural orders in each er as the eet —_ of their 
botanical differences. The nine province 
Tue Kastern Himaraya, ti aa ie Sikkim to the Mishmi 
cia ag ae in Upper Assam. 
ii, Toe Western sputtigad al extending from Kumaun to Chitral. 
iii. Toe Inpus Paty, including the Punjab, Sind, and Rajpu- 
tana west of the Aravallirange and ‘Jumna river, Cutch ‘and Gujarat. 
Tue Ganextic Puan from the Aravalli hills and Jumna 
ae to Bengal, the Sundarbans, the plain of Assam, and the low 
untry of Orissa north of the Mahanadi river. This province is 
divisible into a subprovinces—an upper dry, lower humid, and 
the oe ban 
. Matapar th a very extended sense—the humid belt of hilly 
or diounudiont country Anas 2 along the western side of the 
bev peninsula, from the mouth of the Tapti river to Cape 
Co omor It includes the Sisrikeain’ Kanara, Malabar proper, Cochin, 
Travancore and the seen Islands. 
Deccan in a very broad sense—that is, the whole c 
sdicalivaly oF elevated table. land of Fills east of Malabar and oxith 
of the Gangetic and Indus plains, together with, as co a 
the low- “ieee strip of coast land ex xtending from Oxinas to Tin 
velly, known as the Coromandel coast 
vii. Ceynon and the Maldive Islands. 
viii. Burma, bounded on the N. and N.E. by the flanking moun 
tains on the south of the ergs valley and China, on the east by 
China and Siam, on the west by Bengal aad the Indian Ocean, and 
on the south by the State of Keda in the Malay Ponitisuia. The 
ee sige. and possibly the Wiecbar, belong to the Burmese 
rovin 
Matay Peninsuxa, from Keda to Singapore, including 
the British protected States in this peninsula. ‘The British pro- 
vinces proper are Wellesley, the island of Penang, eanee and 
crip err The Nicobar Islands may belong to this province. 
A glance at the map of India shows that, in this atten to 
delimit thea heey provinces geographically, large areas are in 
some cases difficult to apportion, as, for example, Gujarat, of = 
the N.W. half is probably referable botanically to Sind, the 8.E. t 
crossing it, carrying with them types of the Malabar flora. The 
flora of the trans-Indus mountains bounding the Indus Plain 
* The independent kingdom of - extending for five hundred miles 
between the ores and — _ Himalaya, is here left out of account, from 
ignorance of its flora. Exc very limite d collection made in the valley of 
Khatmandu by Wallich in aaa, ies flora of Nepal is all but unknown. 
= are the tween the floras of and Kumaun, the two meet 
n Nepal, as indicated by Wallich’s Sole which further contain a con- 
didotable number of endemic specie 
