Dedication

TO HIS MOST GRACIOUS MAJESTY KING EDWARD VII EMPEROR OF INDIA

THIS WORK IS BY HIS ROYAL PERMISSION DEDICATED

General Preface

THE first edition of The Imperial Gazetteer of India was published in nine volumes in 1881. A second edition, augmented to fourteen volumes, was issued in the years 1885–7. A revised form of the article on India, greatly enlarged and with statistics brought up to date, appeared as an independent volume in 1893, under the title of ‘The Indian Empire: Its Peoples, History, and Products.’ All of these were edited by the late Sir William Wilson Hunter, K.C.S.I., who formed the original plan of the work as far back as 1869, when he was first entrusted with the duty of organizing a statistical survey of the country, and who wrote most of ‘The Indian Empire’ in its final form with his own hand. His untimely death in 1900 has deprived the present edition of the advantages of his ripe experience and literary skill.

The second edition having for some time passed out of print, it was resolved by the Government of India that a new edition should be prepared in connexion with the Census of 1901. The changes, however, that have been introduced, both in the general scheme and in the methods of compilation, may justify its being considered as a new work rather than a new edition. The bulk will be raised from fourteen to twenty-six volumes, including a companion Atlas; and the single volume of ‘The Indian Empire’ has been expanded into four volumes, entitled respectively ‘Descriptive,’ ‘Historical,’ ‘Economic,’ and ‘Administrative.’ Moreover, while the main work will be arranged as before in alphabetical order, it is intended to issue in India, for official and local use, a parallel series in which the several articles relating to each Province or large group of States will be collected together in separate volumes. No less considerable are the changes that have been made in methods of compilation. Apart from the Historical volume and a few other chapters of ‘The Indian Empire,’ the whole of the work has been written by officials in India under orders of the Indian Government; and every page has been submitted to the criticism of the several Administrations or Departments concerned.

The task of editorial supervision has been shared between India and England. In India the first outlines were drawn up in consultation with Sir Herbert Risley, K.C.I.E., at the time when he was Census Commissioner. On his nomination to be Secretary in the Home Department, Mr. W. S. Meyer, C.I.E., was appointed editor for India; and to him are due the detailed regulations under which the greater part of the work has been executed, the general scheme of the Atlas, and the primary revision of most of the chapters dealing with India as a whole. When he, in turn, was promoted to be Secretary in the Finance Department, he was succeeded by Mr. R. Burn, who has carried out the primary revision of most of the other volumes, and otherwise completed the work that had to be done in India. [^1] The editor in England throughout has been Mr. J. S. Cotton, who was closely associated with Sir W. W. Hunter in both the former editions. On him, in subordination to a committee appointed by the Secretary of State, has rested the responsibility for the final form of the work, and the duty of seeing the whole through the press.

The volumes of ‘The Indian Empire’ have been entirely rewritten, with the exception of the history of the British period, where the personal impress of Sir W. W. Hunter’s knowledge and style is preserved. The principle adopted was to entrust each subject to an author best qualified to deal with it either by special study or official experience. The names of the authors are usually appended to their chapters; but in other cases the text, as finally approved, can be regarded only as a composite production based largely upon their contributions, and their names are then recorded elsewhere. This course has been found necessary in order to maintain the character of official authority that attaches to the work.

The articles that make up the body of the Gazetteer have likewise been rewritten, for the most part by officials who had already acquired local experience as Census Superintendents of their Provinces or States in 1901. A list of their names will be found prefixed to the first volume of the Gazetteer proper. As in the previous editions, the articles are based on materials collected by District officers and officials of Native States, supplemented by special contributions from experts. More detailed acknowledgements will be found in the prefaces to the volumes of the Provincial Gazetteers. It may here be mentioned that, since the last edition was compiled, Upper Burma and Baluchistan have both been included within the Empire, and two new Provinces have been constituted, while much more accurate information is now available for the Native States generally.

To come to details. Comparative statistics are usually given for the three Census years, 1881, 1891, and 1901; but in most cases later figures for 1903-4 have been added. The most recent administrative changes have been incorporated or referred to in footnotes, so far as possible. No novelties have been introduced in the way of spelling, the usage now generally adopted in India having been followed, even where this usage cannot be considered satisfactory or consistent. The only important change is the substitution of the mark of length (–) for the accent (ˊ) over long vowels. Some hints for guidance in pronunciation are given in an Introductory Note, prefixed to each volume. Another Introductory Note supplies a brief explanation of the monetary system and the weights and measures used in India, for it has not been found practicable to convert these uniformly into their English equivalents.

In the present volume, being the Descriptive volume of ‘The Indian Empire,’ chapters i, ii, iv, v, vii, viii, ix, and x have been written by the authors whose names are subscribed to them. Dr. Blanford died before he was able to revise his proofs. Of the other chapters, that on Meteorology is based upon materials contributed by Sir John Eliot, K.C.I.E., and that on Ethnology and Caste has been abridged from Sir H. Risley’s chapter in the Report on the Census of India for 1901. The Index has been compiled by Mr. J. W. Browne.

INTRODUCTORY NOTES

Notes on Transliteration

Vowel-Sounds

a has the sound of a in ‘woman.’

ā has the sound of a in ‘father.’

e has the vowel-sound in ‘grey.’

i has the sound of i in ‘pin.’

ī has the sound of i in ‘police.’

o has the sound of o in ‘bone.’

u has the sound of u in ‘bull.’

ū has the sound of u in ‘flute.’

ai has the vowel-sound in ‘mine.’

au has the vowel-sound in ‘house.’

It should be stated that no attempt has been made to distinguish between the long and short sounds of e and o in the Dravidian languages, which possess the vowel-sounds in ‘bet’ and ‘hot’ in addition to those given above. Nor has it been thought necessary to mark vowels as long in cases where mistakes in pronunciation were not likely to be made.

Consonants

Most Indian languages have different forms for a number of consonants, such as d, t, r, &c., marked in scientific works by the use of dots or italics. As the European ear distinguishes these with difficulty in ordinary pronunciation, it has been considered undesirable to embarrass the reader with them; and only two notes are required. In the first place, the Arabic k, a strong guttural, has been represented by k instead of q, which is often used. Secondly, it should be remarked that aspirated consonants are common; and, in particular, dh and th (except in Burma) never have the sound of th in ’this’ or ’thin,’ but should be pronounced as in ‘woodhouse’ and ‘boathook.’

Burmese Words

Burmese and some of the languages on the frontier of China have the following special sounds:—

aw has the vowel-sound in ’law.’

ö and ü are pronounced as in German.

gy is pronounced almost like j in ‘jewel.’

ky is pronounced almost like ch in ‘church.’

th is pronounced in some cases as in ’this,’ in some cases as in ’thin.’

w after a consonant has the force of uw. Thus, ywa and pwe are disyllables, pronounced as if written yuwa and puwe.

It should also be noted that, whereas in Indian words the accent or stress is distributed almost equally on each syllable, in Burmese there is a tendency to throw special stress on the last syllable.

General

The names of some places—e.g. Calcutta, Bombay, Lucknow, Cawnpore—have obtained a popular fixity of spelling, while special forms have been officially prescribed for others. Names of persons are often spelt and pronounced differently in different parts of India; but the variations have been made as few as possible by assimilating forms almost alike, especially where a particular spelling has been generally adopted in English books.

Notes on Money, Prices, Weights and Measures

As the currency of India is based upon the rupee, all statements with regard to money throughout the Gazetteer have necessarily been expressed in rupees, nor has it been found possible to add generally a conversion into sterling. Down to about 1873 the gold value of the rupee (containing 165 grains of pure silver) was approximately equal to 2s. 6d., or one-tenth of a £; and for that period it is easy to convert rupees into sterling by striking off the final cipher (Rs. 1,000 = £100). But after 1873, owing to the depreciation of silver as compared with gold throughout the world, there came a serious and progressive fall in the exchange, until at one time the gold value of the rupee dropped as low as 1s. In order to provide a remedy for the heavy loss caused to the Government of India in respect of its gold payments to be made in England, and also to relieve foreign trade and finance from the inconvenience due to constant and unforeseen fluctuations in exchange, it was resolved in 1893 to close the mints to the free coinage of silver, and thus force up the value of the rupee by restricting the circulation. The intention was to raise the exchange value of the rupee to 1s. 4d., and then introduce a gold standard (though not necessarily a gold currency) at the rate of Rs. 15 = £1. This policy has been completely successful. From 1899 onwards the value of the rupee has been maintained, with insignificant fluctuations, at the proposed rate of 1s. 4d.; and consequently since that date three rupees have been equivalent to two rupees before 1873. For the intermediate period, between 1873 and 1899, it is manifestly impossible to adopt any fixed sterling value for a constantly changing rupee. But since 1899, if it is desired to convert rupees into sterling, not only must the final cipher be struck off (as before 1873), but also one-third must be subtracted from the result. Thus Rs. 1,000 = £100 – ⅓ = (about) £67.

Another matter in connexion with the expression of money statements in terms of rupees requires to be explained. The method of numerical notation in India differs from that which prevails throughout Europe. Large numbers are not punctuated in hundreds of thousands and millions, but in lakhs and crores. A lakh is one hundred thousand (written out as 1,00,000), and a crore is one hundred lakhs or ten millions (written out as 1,00,000,000). Consequently, according to the exchange value of the rupee, a lakh of rupees (Rs. 1,00,000) may be read as the equivalent of £10,000 before 1873, and as the equivalent of (about) £6,667 after 1899; while a crore of rupees (Rs. 1,00,000,000) may similarly be read as the equivalent of £1,000,000 before 1873, and as the equivalent of (about) £666,667 after 1899.

Finally, it should be mentioned that the rupee is divided into 16 annas, a fraction commonly used for many purposes by both natives and Europeans. The anna was formerly reckoned as 1½d.; it may now be considered as exactly corresponding to 1d. The anna is again subdivided into 12 pies.

The various systems of weights used in India combine uniformity of scale with immense variations in the weight of units. The scale used generally throughout Northern India, and less commonly in Madras and Bombay, may be thus expressed: one maund = 40 seers; one seer = 16 chittaks or 80 tolas. The actual weight of a seer varies greatly from District to District, and even from village to village; but in the standard system the tola is 180 grains Troy (the exact weight of the rupee), and the seer thus weighs 2·057 lb., and the maund 82·28 lb. This standard is used in official reports and throughout the Gazetteer.

For calculating retail prices, the universal custom in India is to express them in terms of seers to the rupee. Thus, when prices change, what varies is not the amount of money to be paid for the same quantity, but the quantity to be obtained for the same amount of money. In other words, prices in India are quantity prices, not money prices. When the figure of quantity goes up, this of course means that the price has gone down, which is at first sight perplexing to an English reader. It may, however, be mentioned that quantity prices are not altogether unknown in England, especially at small shops, where pennyworths of many groceries can be bought. Eggs, likewise, are commonly sold at a varying number for the shilling. If it be desired to convert quantity prices from Indian into English denominations without having recourse to money prices (which would often be misleading), the following scale may be adopted—based upon the assumptions that a seer is exactly 2 lb., and that the value of the rupee remains constant at 1s. 4d.: 1 seer per rupee = (about) 3 lb. for 2s.; 2 seers per rupee = (about) 6 lb. for 2s.; and so on.

The name of the unit for square measurement in India generally is the bigha, which varies greatly in different parts of the country. But areas have always been expressed throughout the Gazetteer either in square miles or in acres.