of this again, extending through Burma from north to south, we find broken highlands and plateau, traversed by no definite mountain ranges, but forming one continuous chain of rugged tableland, stretching from the Kachin Hills on the north, through the northern and southern Shan states to the Karenni country on the south. This tableland is intersected by the trough of the Salween River. Beyond the Shan states is China in the north and Siam in the south. But the province of Burma does not end with the Shan states. There is a long strip of coast land, averaging perhaps 20 miles in width, but occasionally narrowing to 10, which extends down the western edge of the Malay peninsula, and includes the districts of Martaban and Tenasserim. This is also part of the province under the administration of the Lieutenant-Governor of Burma. It includes the Mergui archipelago, and is chiefly remarkable for the long broken coast-line, extending through 16 degrees of latitude, flanked by hundreds of islands which once formed part of the peninsula. The total length of Burmese coast-line from North Arakan to South Tenasserim is not much less than 550 miles. The total area of the province is 171,500 square miles, and its population (in 1891) was 7,600,000.

This brief outline of the configuration of the continent of India will suffice to make clear the allusions to the different provinces and forestry divisions of the country which follow hereafter.

The Island of Ceylon, though not included in the Indian Empire, must, from the geographical point of view, be regarded as a part of India. The latitude of its southern extremity is only six degrees north of the Equator. The extreme north latitude of the north-western corner of Kashmir is thirty-seven degrees north, and the altitude of the land surface of India varies from flats a few inches above sealevel to peaks 28,000 feet above it.

With the great variation in the elevation of the surface and the considerable distance involved from south to north, great variations in climate in the different parts of the country are a natural outcome. And, as a matter of fact, every variety of climate is to be found in the country and every condition of physical extremes from the Equatorial to the Arctic. And occasionally these sharp differences are found within close proximity the one to the other. As will be shown later the great changes in elevation and climate have a very marked

influence on the distribution of the numerous species of trees to be found in the forests of the country, varying with the different zones-hot moist, hot dry, arid, temperate and

arctic.

Some account of the rainfall of the continent will be necessary here, since it exerts such a marked influence on the distribution and nature of the forest growth and general flora. As elsewhere in the tropics the rainfall in India has marked characteristics, the feature of importance being the periodic rainfall or monsoons. The distribution and character of the various types of Indian forests are primarily influenced by the monsoon rainfall. The monsoons prevail at two seasons of the year. The first or south-west monsoon rains brings to an end the hot weather season, falling between June and September. They first strike and give rain to the whole of the lower western coast of the peninsula and the western coast of Bengal and Burma. The second or north-east monsoon falls between October and December, being chiefly confined to the eastern coast of the peninsula. Some parts of the country receive rain from one or other of the monsoons, other regions being affected by both. The southwest monsoon rains from the Arabian Gulf fall with their full force on the western coast districts from the Tapti southwards and the Ghát Range behind them; as also on the coast lands and outer hill ranges of Tenasserim, Pegu, Arakan and Chittagong; the plains of Bengal, the outer slopes of the Eastern Himalaya and the Khasi and Tipperah Hills. This monsoon also reaches in a lesser degree the western slopes of the Himalaya, the whole length of the outer southern parts of this great mountain system also receiving a monsoon rainfall at this period from the rain clouds travelling up from the Bay of Bengal. The north-east monsoon rains, on the other hand, provide the chief rainfall of the coast of the Carnatic, from the Kistna River southwards and inland to the outer ranges of the hills of the Eastern Gháts; of the Mysore tableland, the Javadi, etc., and even reach the edge of the Western Gháts. This region receives only a scanty and often failing supply of rain from the south-west monsoon. The regions which receive both monsoons are confined to the peninsula, the inner areas of the Deccan and the Carnatic. In addition to the monsoons, varying amounts of rainfall due to local precipitations are received in different parts of the country, as for instance the so-called "Christmas "

rains in the United Provinces and elsewhere, the showers known as the Mango showers, and the local rainfall in Assam. These are often of very considerable local importance.

Omitting the part of the country in the south which is subject to two monsoons, the cold weather season commences in October or November (the former in the northern provinces) and lasts till the end of February or end of March respectively. This cold season is followed by the hot weather commencing in early March (or early April, the latter in the north) and lasting till the monsoon breaks in early June or July. It is during the hot season that the forests suffer so severely from fires; though the modern fire conservancy arrangements introduced by the Forest Department have done much to mitigate this evil. This season is, however, a somewhat trying one for the Forest Officer.

To complete these brief preliminary remarks a glance must be taken at the geology of this remarkable continent. It is in itself a fascinating study. To Oldham, of the Geological Survey of India, in his Evolution of Indian Geography (Vol. III, R.G.S.), we are indebted for an extremely able account of India's geological origin. It reads like a romance, whilst at the same time affording the Forester data which enable him to understand several important factors relative to tree distribution.

"Measured by the vast ages of geological existence, the peninsular area of India (the region of southern tableland) is by far the oldest. On the north-western borders of this area, stretching across the plains of Rajputana, are the remnants of a very ancient range of mountains called the Aravalli. To the south of these mountains the peninsula of India, as we know it now, has been a land area since the close of the palæozoic era. Across the extra peninsula regions to the northwest of the Aravalli Hills the sea has repeatedly flowed even to the commencement of tertiary ages; and between the two regions thus separated by the Aravallis there exists most striking differences both in structure and in conformation. The present shape of the peninsula-itself but a remnant of a far more widely extended continent-has only been assumed since the occurrence of the vast series of earth movements which resulted in the creation of the region of depression-the alluvial basins of the Indus and of the Ganges. Over the substratum of granite and gneiss which forms the 'bedrock' of peninsular geologic structure, and eastward from the Aravallis, stretch the wide red sandstone deposits which are

known as the Vindhyan system, and which (even when buried beneath the Deccan trap) preserve a generally horizontal position. Geologists maintain that these widespread unfossiliferous beds are but the detritus washed down from the peaks and valleys of the inconceivably ancient mountain range which is now represented by the comparatively low and degraded Aravallis. Almost coeval with the Aravallis (and possibly at one period connected with them) is the much broken and ragged formation known as the 'Eastern Gháts,' overlooking the Bay of Bengal. So ancient is this eastern buttress of the peninsular tableland that since the close of the paleozoic era the waters of the bay have never washed westward, and the coast of Madras was the eastern coast-line of that pre-Indian continent of which India is now the muchdiminished representative. But whilst the Aravallis were clearly the north-western limit of this prehistoric continent, it is not quite so clear what formed the boundary on the northeast. There was no Gangetic basin in those days, and it was probable that the Rajmahal Hills and the hills of Assam continued the land area to the Himalaya east of Sikkim ; for it is certain that the Eastern Himalaya are vastly older than the western, and that the Burmese mountains are comparatively young. Next followed a long period of repose and of the silent process of alluvial deposit by river action, during which the wide central beds called Gondwana were formed. Here we must note the existence of ice-worn boulders and the evidence of former glaciers in Rajputana; and at this point we are faced with the almost indisputable fact that the India of the Aravallis and of the Rajmahal Hills was but an extension from South Africa. The evidence which has been collected to prove this ancient connection seems to be conclusive. Plants of Indian and African coal measures are identical, and not only plants, but the fauna of that period claim a similar affinity. Near the coast of South Africa a series of beds occur which is similar in all respects to an existing Rajmahal series; whilst the distribution of marine fossils proves that to the north-west and to the south-east respectively of a land barrier, which must have included the Maldives, Laccadives and Madagascar, were two distinct seas. This land connection must have existed at the commencement of cretaceous times. There are no marine sedimentary beds in the Eastern Himalaya analogous to those of Burma, of the North-West Himalaya and of the mountains west of the

Indus. These wide highlands, together with the great plateau of Tibet, were then under sea, subject (so far as we can tell) to no great earth movements in very early ages, but undergoing quiet and placid intervals of subsidence and upheaval, of alternate existence as open lacustrine land surface or sea bottom. At the close of the jurassic period not only were the North-Western Himalaya non-existent, but the very rocks of which the ranges to the west of the Indus are formed were still uncreated. Only the Aravalli peaks stood lonely and silent, overlooking a vast north-western sea. Not till the close of the cretaceous period was India shaken by a series of eruptive cataclysms into something of its present shape. A succession of volcanic eruptions, exceeding in force and grandeur anything that the world has ever seen (except, indeed, it be in South America), covered 200,000 square miles of India with lava and tuffs to a depth of thousands of feet. India must have been for a time one vast volcanic furnace. This was but the prelude to the mountain building. Then, at the commencement of the tertiary period, set in that long succession of earth movements which, culminating in intensity about the pliocene period, are still in perceptible activity. The sea was driven back; rocks were crushed and forced upwards until marine limestones were upheaved to 20,000 feet above sea-level. Then were the sea-formed rocks of the trans-Indus hills ranged and folded and set in order. It was the period of the creation of our Indian frontier.

"Geologists have decided that the fossils of Burma and of the western frontier alike place the formation of these regions in the eocene, or latter part of the tertiary era. But the great bulk of the North-Western Himalaya must have been a formidable mountain barrier in times previous to the eocene period, and, moreover, even in those early times, the river systems of the Himalaya must have been traced very much as they are at present. At the foot of the Himalaya there existed for geologic ages a long series of river deposits which have been compressed and upheaved in very recent times to form what is known as the Siwalik Range, an entirely subsidiary and secondary range of hills which flanks the Himalaya on its southern face, forming an elevated longitudinal valley, in which Dehra Dun is situated, between itself and the foothills of the main system. In the neighbourhood of those rivers which issue from Himalayan valleys and cross the elevated valley and the bordering Siwalik Hills there have