INTRODUCTION xvii 



tidal streams^ including the Mangrove formatioHj are quite peculiar, the 

 extent of these littoral forests is considerablej the habits and structure of 

 the principal species have been thoroughly studied ; * this, however, is not the 

 place to say more regarding them. 



The reader will I trust not misunderstand me. The nature of the soil, and 

 consequently the nature of the rock, by the disintegration of which the soil 

 has been formed, is a most important factor in forestry. It will gradually 

 be more fully understood that Teak forests on trap require a different treat- 

 ment from those on gneiss, on limestone, on sandstone and calcareous shales or 

 on alluvial soil ; and doubtless Indian foresters will eventually find that 

 many trees thrive better on certain classes of soil than on others. The large 

 area covered with basaltic rocks in Western India affords opportunities for 

 promising studies in this respect. The Sal tree on the Satpura range has its 

 western limit on the sandstone of the Pachmarhis. It does not thrive on trap, 

 and hence it is not found further west in that region. A carefully prepared 

 list of species found in the trap cou.ntry of the Peninsula and of those which 

 are wanting, though they are found on other formations in the same region, 

 may prove instru.ctive in many respects. 



At the British Association assembled at Brighton in 1872, 1 had the honour 

 to explain before the geographical section my views regarding the distribution 

 of forests in India. My paper was published in Ocean HigMcays, 1872, p. 200, 

 and in the Transactions of the Royal Scottish Arhoriciiltural Society, Yll, 88 

 (1873). It was illustrated by a map, on which I had, on the ground of the imper- 

 fect meteorological data then available, entered the regions of greater or less 

 moisture, as indicated by the annual rainfall. I laid stress upon a subject 

 which is now universally recognized, the intimate relation between moisture 

 and forest growth in India. Since then the Indian Meteorological Department 

 has been established, and the maps published by that Department are more 

 accurate and much more detailed than the map which T ventured to compile in 

 1872. Still the great regions then established remain, and with the addition 

 of the dry region in the Irawadi valley, not known to me in 1872, they are : 



First : I. The arid region with a scanty and irregular rainfall, never 

 exceeding 15 inches annually, and an atmosphere dry nearly throughout the 

 year, comprising Sind, the southern Punjab and western Eaiputana. 



Secondrthe 4ree dry regions, annual' rainfall betweenlf and 30 inches,t 

 viz. : II. The northern dry region, adjoining the arid region on its north 

 and east sides, comprising the greater portion of the plains of the Punjab to 

 within a short distance from the foot of the Himalaya, Eastern Rajputana 

 and a large part of what are now called the United Provinces, extending 

 eastwards along the Ganges beyond Cawnpore. III. The Peninsular dry 

 region, comprising the greater part of the Deccan and Mysore. IV. The 

 Eastern dry region in the Irawadi valley, comprising the following districts. 

 West of the river : the southern portion of Shwebo, the eastern half of the 

 Lower Ghindwin, Sagaing, Pakokku and Minbu. East of the river: the 

 western portion of Mandalay and Kyaukse, Myingyan, Meiktila and Magwe. 



Third, the two moist regions, rainfall 75 inches and upwards, viz : V. The 

 "Western moist region, extending from the Gulf of Cambay to Travancore and 

 comprising the country from the coast to some distance east of the crest of the 

 Ghats. VI. The Eastern moist region, comprising the eastern Himalaya with 

 a narrow strip along the outer ranges, extending north-west as far as the 

 Eavi, the greater part of Assam, Eastern Bengal^the northern part of Upper 

 and nearly the whole of Lower Burma. 



* Schimper, Indo-Malayisohe Strand ]?lora (1891), Plant- Geography, p. 90, 287 (1903). 

 Prain, Plora of the Siindrihan in Becords Bot. Survey India II. 240. Hooker, Sketch 

 of the Plora of British India, 26. 



f In the excellent large Rainfall Map of India^ published by the late Henry 

 Blanford in 1883, and in later maps published by the Meteorological Department, 

 the regions are limited by lines of 10-20, 20-30, etc., inches of the annual rainfall, 



b 



