inconvenient, and in Bengal it has fallen into such decay, that in some districts the pergunnah boundaries can hardly be ascertained. Detached villages belonging to the same pergunnah may now be found all over the district and in two or more districts. Practically the pergunnah divisions of districts have died out, except for purposes of land revenue payments, in favour of the simpler and more compact arrangements adopted for purposes of police. This arrangement had its origin in Regulation XXII of 1813, by which the Magistrates were directed to divide their districts into police jurisdictions to be named after the places at which the chief police officers were stationed. And in this way the term thannah, which originally meant only the police station, came to be applied to the jurisdiction subordinate to that station. The thannah divisions have been for years growing into greater importance, and are now utilized to a large extent for other than police purposes. The thannah is now the unit of which a sub-division is composed, as the sub-division is of the district. The total number of thannahs in Bengal is 610; the average area of each is 280 square miles. The population

Thannahs.

Outposts.

of thannahs varies from 15,000 to 349,000: some of the large thannahs purposes into outposts or pharrees. which obtains in other parts of India for the collection of revenue and the tehsil

are divided again for police The tehsildari system, Tehsildar divisions lately established. division of territory, is not generally known in these provinces. A system of direct revenue collection and Government management has, however, lately been introduced into several tracts not permanently settled and hitherto let out to farmers or subjected to some such management. The principal tracts affected are the Damin-i-koh of the Sonthal Pergunnahs, Jynteah, the Bhootan Dooars, and Palamow. In Khoordah there is already a tehsildari division. In the province of Assam the land is divided for revenue purposes into mouzahs, or circles of villages, Mouzahs in Assam. the revenue of which is collected by an officer called "mouzahdar," who resides within the circle. There are no limits of villages as distinguished from mouzahs recognized for fiscal purposes. Each mouzah is compact in itself, and there are no revenue divisions intermediate between the mouzah and the individual ryot's holding. A somewhat similar arrangement exists in Chittagong, where estates are very many and very Canoongoe's divisions in Orissa, small. In Orissa also the jurisdiction of canoongoships constitute a civil division for fiscal and executive purposes. The districts are so mapped out that at every twelve or fifteen miles there is a canoongoe's head-quarters. The canoongoes are officials available for every sort of duty between the sub-divisional officers and the people.

The municipal system has been of recent years introduced into these provinces. Calcutta is adminisMunicipalities. tered under a separate Municipality and special laws for the city. There are several Acts for the interior. Twenty-four larger towns are regulated by Act III (B.C.) of 1864; 95 smaller

Townships.

Unions.

townships by Act VI (B.C.) of 1868. Act XX of 1856 enables the 48 villages and small unions in which it exists to pay for their watch and ward and to meet more urgent demands of ordinary conservancy. Act XXVI of 1850 is still in force in one place in Bengal-Jumalpore, in the Monghyr district. The Presidency, Burdwan, and Patna divisions, are those in which the municipal system has most developed itself. There are, however, very few places in Bengal with any pretensions to be called towns, and almost the whole of the vast population of this province consists of persons connected Villages. with agriculture, who live in little rural villages. These villages in Bengal are isolated, detached places, with a number of residents, varying from 200 or 300 to 3,000, living very much among themselves, and clinging tenaciously to their homes. These units are not apt to coalesce into clumps or associations, except in the immediate neighbourhood of the presidency or other such large cities, but each remains with its own houses, and adheres to its own communal servants. Sometimes the Mahomedans live a good deal by themselves or have their own villages apart, and the different castes of Hindoos congregate together in their own quarters. Sometimes the different religions and castes are very much mixed together. This is all a trace of the past, but the old communal institutions by which the village was governed have faded away under the influence of our rule and the zemindary system. The ancient indigenous village system of India still exists in the hilly countries attached to Bengal, but in the plains they have almost disappeared. The chowkidar or watchman now lingers as almost the only vestige of the old municipal commune. The other traces that remain are scanty; some things are in some places regulated by village punchayets or headmen of some sort, but more and more the zemindary agent has supplanted the village_mondul, and the landlord takes the place of the indigenous self-rule. It has been the Lieutenant-Governor's especial object to retrace this order and give the people that measure of selfgovernment and local freedom to which both their old traditions and their modern education alike point, by giving to towns and restoring to villages some sort of municipal or communal form of government. Late inquiries have shown that the materials of old institutions still exist in a greater degree than had been suspected.

DETAILS OF THE LAST CENSUS-TRIBES AND LANGUAGE.

THE census of Bengal, which was effected during the cold weather

THE CENSUS OF 1871-72.

of 1871-72, was the first census of the country that had ever been attempted. Only partial enumerations of particular areas had from time to time been previously. made, and they were either estimates based upon the number of houses in the district incorrectly computed, or they were conclusions drawn from experience and general observation, and entitled to little or no reliance.

The present census computations show the total number of persons counted in the provinces under the Government of Bengal to be 66,856,859.

This aggregate total of population far exceeds that of any previous estimate. With few exceptions every district in the province is more thickly populated than even the most liberal official calculation had allowed for.

The old British territories in Northern India consisted in the last

Previous estimates of the population.

century of the provinces of Bengal and Behar. The area of those territories included the present regulation districts of these provinces and the greater part of the division of Chota Nagpore, but it did not include Assam and Cachar nor Darjeeling, and the Bhutan Dwars, nor until later were Orissa and the adjacent hill tracts annexed by the British Government. The original area was approximately estimated by Mr. James Grant in 1786 in his Analysis of the Finances of Bengal at 97,200 square miles.

The first opinion promulgated concerning the population of Bengal, Behar, and Orissa (including only under the term Orissa what is now comprised in the Midnapore district and part of Hooghly), was that it amounted to ten millions. This was put forth soon after the Company's acquisition to the Dewany, and is the "entire assumed population throughout Mr. Grant's Analysis of the Finances and View of the revenues of Bengal. In the meantime, however, it was found out that this estimate was too low. Sir William Jones, in 1787, forming his judgment from materials to which we have now no access, thought that the population of Bengal, Behar, Orissa, and Benares, amounted to twenty-four millions. In the first year of this century the Bengal Government called for information from the Collectors and

Judges stationed in the districts in the four provinces; but the returns were so imperfect, and, where they were made by these two descriptions of officers, so contradictory, that no general conclusion could be drawn from them. Mr. Colebrooke, in 1802, computed the population to be thirty millions. The Fifth Report of the Select Committee on the Affairs of the East India Company, after consideration of these data, records in 1812 that "if any opinion were now to be offered on a point which has not been subjected to strict investigation, perhaps there would be no danger of exceeding the truth in adopting a medium between the calculations of Sir W. Jones and Mr. Colebrooke, and supposing the population of the four provinces to be not less than twenty-seven millions." Dr. Buchanan had, however, about this period made an estimate of the population of several districts, which he puts very much higher than other authorities.

The semi-official estimate of Mr. Adams in 1835 assumes the population to be thirty-six millions. In 1844, when the territories under the Bengal Government were nearly the same as those now under the Lieutenant-Governor, their population was estimated by Mr. Dampier, the Superintendent of Police in Bengal, at 31,200,000.

Of late years, however, the population of Bengal has been generally accepted at about forty or forty-one millions, and that total has been the recognized basis for legislation and finance. This has always been the accepted estimate since the publication of the Parliamentary Blue Book "on the area and population of India" on the 27th July 1857. The total population of all India is there recorded at 180,884,297; the total population under the administration of the Government of Bengal at 40,852,397. In the administration report of this Government for the year 1870-71, the population is put down after revision at something over 42 millions.

After much discussion instructions were issued by the Government of India that arrangements should be made for a general census of the population in the year 1871, and preparations having been duly made, the work was undertaken in that year. At an early period it was found expedient to abandon the idea of taking the census of the whole of The manner of taking the census, and the vast provinces within the jurisdicthe agency employed. tion of the Lieutenant-Governor on one and on the same day. In many cases indeed an entire subdivision, and even a whole district, was in fact eventually enumerated simultaneously, but it was not found possible or desirable to lay down any precise scheme of operation to be adopted everywhere alike. The time within which, and the agency by which, the enumeration was made, was to a considerable extent entrusted to the experience and discretion of the local authorities. It was directed that each district and division was to be taken as nearly simultaneously as possible with reference to the circumstances of each, provided that the whole must be completed between December 1871 and March 1872. The Bengal census, moreover, did not pretend to deal with the precise status of every individual in the country, man, woman, and child, with rigorous accuracy. That would have been impossible in these provinces. On no previous occasion had so gigantic an enterprise

been undertaken, and the machinery which is available elsewhere, though not wanting in Bengal, had on all sides been suffered to fall into desuetude, and could only be utilized on this occasion to a very

partial extent.

All the information, however, attainable and really necessary for practical purposes was shown in the returns. The names, castes, and professions of all heads of houses and adult males, were recorded; while the women and children in each house were numbered-not named, the caste and profession of the head of the family being sufficient to distinguish the family. Religion also was stated, and adults or grown persons were distinguished from children under twelve. These were the main points shown, while it was also endeavoured to elicit the numbers of schools, of boys and girls who attended schools, of blind persons, deaf persons, insane persons, idiots, lepers, and the like.

Mr. Beverley, the Inspector-General of Assurances, was selected as the most proper officer to supervise the census arrangements from their commencement to their close. To his unwearying devotion to his duties, and the excellence and completeness of his work, the LieutenantGovernor has borne willing testimony.

To facilitate operations, and to prevent any complications which might possibly attend the employment of an unpaid agency, a brief Act (XI of 1871) was passed by the Bengal Legislative Council, authorizing the local officers to appoint enumerators, and providing certain penalties for their misconduct or neglect. The Act invested the enumerator so selected with the power to collect the required information, making refusal to answer their questions on such points a penal offence. It also empowered Magistrates, under certain restrictions, to call for assistance in the matter from landholders or their agents.

The agency employed in the enumeration of the people was chosen as far as possible from among the people. The census was thus virtually effected by the people. In Assam and Behar indeed the enumeration was carried out through the indigenous fiscal establishments of the country; but in Bengal the census was for the most part taken by indigenous agencies or private individuals owning no official allegiance to Government.

Very much was elicited by the census and has been confirmed by recent inquiries regarding the various indigenous agencies still found existing in the country. These agencies were made use of as far as possible. Old institutions, officially supposed to be long ago dead and gone, were still found to survive in many places, and rural agencies, condemned long since, were proved to exist in several parts of the country and afforded much assistance. The reports on the indigenous agencies of the country from the divisional Commissioners, many of which are very full and interesting, have recently been published as a volume of Selections by the Bengal Government.

Many private individuals also accepted office as unpaid enumerators and rendered very useful service. The Government servants of the police and registration departments, schoolmasters and the like, largely contributed to the result as a supervising agency.

N