This article is extracted from Statistician's Report 1911.

1. General. In most of the British Possessions and Dependencies the census methods of England and Wales are followed more or less closely, the population being enumerated by means of householders' schedules. In India and Ceylon, however, the conditions are greatly different to those of other countries, and consequently it has been found necessary to adopt different methods of enumeration. The enormous population to be enumerated, the general illiteracy of the masses, the varieties of race, and the distinctions of caste and religions, all combine to make the taking of an Indian census a task of considerable difficulty. In Canada the census has been modelled upon that of the United States rather than on that of England and Wales, and in that Dominion the schedules are not left with the householders, but are filled up by the enumerator himself. Another important difference in Canada is that the population de jure is enumerated.

2. Census Methods. As a rule in all British Possessions and Dependencies each country is divided for census purposes into districts placed under a superior officer, often named the " Supervisor" or "Enumerator," who is under the immediate control of the Superintendent of the Census; these districts are again parcelled out into convenient areas for purposes of collection. For each of these latter districts a collector is appointed. The time allowed for the collection of the schedules varies in different countries; in a great number of cases it is found that the collector has to fill in the schedules himself, and the time taken in collection is thus prolonged.

Magistrates, registrars, and civil officers are usually appointed as supervisors or enumerators for their respective districts. The supervisors frequently arrange the division of their districts into collectors' sub-districts, and subject to the approval of the superintendent select their own collectors, and instruct them in the performance of their duties. They arrange for the proper distribution and collection of census forms and books, and audit and examine the collectors' books. The method of payment of the collectors varies considerably; some countries have adopted the system of paying by results, i.e., so much for every person recorded; in other countries a fixed sum is allotted to each official, while in others the collectors are paid by the day. The method of payment by results has been objected to as tending to bring about an exaggeration of the number of persons enumerated. In order to give as much publicity as possible to the fact that a census is about to be taken, advertisements and announcements are usually inserted in the press, and the co-operation of persons of local prominence --- such as teachers, and other officials --- in each district throughout the country is secured(1).

1. Cape of Good Hope.

From the commencement of the colonisation of the Settlement at the Cape of Good Hope in 1652, it was the practice to have annual enumerations of the population, live-stock, and agricultural produce for the purpose of communicating the ascertained results to the Netherlands East India Company. The earliest known of these returns, however, is that for the year 1687, from which date, with a few casual exceptions, they run down to 1785 . These annual returns were also made under the English administration from 1823 to 1856, when they were discontinued, owing to the untrustworthiness of the results. The first complete census of the Colony in the proper acceptation of that term, was taken in 1865, when the total white population numbered 20,638 persons. The second census was taken in 1875, and the third in 1891. It was originally intended to take the fourth census in 1901 in conjunction with the rest of the British Empire; the abnormal conditions created by the military operations in that year, however, compelled the abandonment of the project. Now that it was too late to combine with the rest of the Empire, the chief object to be served was to effect a simultaneous census throughout British South Africa, and it was found that this could not be carried out until 1904. Before the final arrangements were made for the census of that year an Inter-colonial Census Conference was organised at Pretoria to discuss the mode of procedure to be adopted in order to secure uniformity in the more essential points in all the different Territories. The Colonies represented were as follows: Cape of Good Hope, Natal, Transvaal, Orange River Colony, Bechuanaland Protectorate, and Basutoland. As the Cape Government had, prior to the date of this conference, completed the preliminary arrangements for the census in that Colony, the schedules and forms adopted by the other South African Colonies followed with minor modifications the lines which the experience of the Cape had proved useful and practical. The information required to be collected on the schedule at the census of 1904, according to the plan agreed upon by the Conference of Delegates, embraced the following categories: (a) Name. (b) Sex. (c) Age. (d) Conjugal Condition. (e) Relation to head of household. (f) Profession or occupation. (g) Sickness or infirmities. (h) Race or nationality. (i) Country where born and length of residence in Colony. (j) Religious denomination. (k) Education. (l) Description of dwelling and (m) Live-stock kept.

(i.) Census of 1904. The Census Act of 1903 provided for the taking of an account of the number of persons and the number of each kind of live-stock within the Colony, and also for the collection of agricultural, educational, ecclesiastical, industrial, friendly societies' and other statistics by the census supervisors and enumerators. The Colony was divided for census purposes into 108 districts, corresponding to the Fiscal Divisions of the Colony proper, and the Magisterial Districts in the Transkeian Territories. In the former of these the Civil Commissioners and in the latter the Resident Magistrates were appointed ex officio supervisors. The basis finally adopted for fixing the enumerators' sub-districts was to parcel out the country, so that an urban area might be completely traversed by the enumerator in from three to five days, and a rural area in up to nine or ten days. In some instances, however, over a fortnight, and in two cases as much as twenty days, had to be allowed for the collection of the schedules. The total number of the collectors' districts was 2394. Special arrangements were made for the Native (2) census.

(ii.) Tabulation. In addition to the householder's schedule, special forms (viz., Agricultural, Education, Ecclesiastical, Industrial, Fisheries, and Friendly Societies) were used at the census of 1904. In the work of tabulation two distinct methods were employed. The particulars regarding individual inhabitants, viz., Race, Sex, Birthplace, Religion, Age, Degree of Education; Conjugal Condition, Profession or Occupation, and Sickness or Infirmity, were entered from the schedules on to cards; while, on the other hand, the particulars as to buildings and dwellings and those contained in the special forms were transcribed to the tabulating sheets direct.

2. Natal

The first occasion in which a census was taken in this country was in 1891, but no attempt was made in that year to enumerate the natives. The second census was taken in 1904 in conjunction with the census for the whole of British South Africa. The arrangements, as well as the nature of the particulars collected, at the census of 1904 were very similar to those in the Cape of Good Hope (3).

3. The Orange Free State.

The first census of the Orange Free State was taken in 1880, the second in 1890. The third census was taken in the year 1904, and was synchronous with the censuses of the other South African Colonies(4). The Magisterial Districts of the Colony (24 in number) were adopted as census districts, to each of which was appointed an enumerator. There were 405 sub-districts, in which 424 collectors and 39 interpreters were employed. The arrangements for the census and the nature of the information collected were similar to those adopted in the Cape of Good Hope, as adopted at the Pretoria Conference. Three householders' schedules were used; one in English, one in Dutch, and the third for natives. Eight special schedules were also used. The cost of the census of 1904 was £6146. For the purposes of tabulation, cards, similar to those used in the Cape of Good Hope, were adopted at the Orange Free State Census of 1904.

4. The Transvaal and Swaziland.

The first census of the Transvaal was taken in 1890; no details are available as regards the number of enumerators, and the general carrying out of the work. In 1896 a census of the municipality of Johannesburg, with very full details, was taken. At the census of 1904, taken in conjunction with the censuses of the other South African Colonies, the whole of the Transvaal and Swaziland was divided into 21 supervisors' districts, and 779 collectors' districts. These included 28 districts for the railways, and 6 for the military establishment. The total (5) number of supervisors was 21, collectors 2115 (including 1120 native assistants, and 31 interpreters).

5. Mauritius.

The first census of this island appears to have been taken in 1846. The second census was taken in 1851, and from that year onward a decennial census has been taken, usually on or about the date of the census of the United Kingdom. At the census of 1901 the staff employed comprised 15 superintendents, 98 enumerators, and 239 collectors.

6. Seychelles Archipelago.

Though the Seychelles Islands are dependencies of Mauritius, separate censuses have been taken decennially since 1851.

7. Protectorates and Minor Possessions.

In most of the British Protectorates and minor Possessions in Africa, censuses, more or less complete, were taken in 1901, under the authority of special Ordinances.


1. India. In the Indian Empire the social system and the ethnic distribution are the most complicated, the most varied, and the most elaborate that the world presents, and the work of taking a census is thereby rendered exceptionally costly and laborious. Further, the general illiteracy and narrow horizon of an agricultural population not only render the supply of enumerating agency less abundant in India, but also increases the work of enumerating, since it puts out of question the practice of leaving the responsibility of filling in the schedule to the householder(6). In the work of census-taking, the official administration is used as far as possible; the ordinary administrative business of the country necessitates the presence in each village community of an official who is generally acquainted with every house and family in the place. It is thus possible to largely decentralise the census work. Furthermore, in India there is practically no overlapping of areas.

(i.) Historical. In certain provinces such as Madras and the Punjab, the custom of making periodic estimates of the population, founded on more or less accurate data, is of very old standing, but the first systematic attempt to obtain a census for the whole of the Empire of India was made in the years 1871 and 1872, and even then resulted in an incomplete enumeration. Earlier provincial censuses had been taken in the North-West Provinces in 1853 and 1865, in Oude in 1869, in the Punjab in 1868, in the Hyderabad Assigned Districts in 1867, and in the Central Provinces in 1866. In Madras quinquennial returns had been prepared since 1851. Mention is made of a census having been taken in Bombay in 1716, while unsuccessful attempts are recorded for 1833- 4, 1849,1851 (twice), and 1861, the first really satisfactory census appearing to have been that of 1864(7). The enumeration mentioned above as having been made in 1871- 2 was, in addition to being incomplete, also non-synchronous. The first synchronous census of all India was that taken in 1881, and this has since been followed by similar enumerations in 1891, 1901, and 1911. The earliest attempts to enumerate the people of any portion of British India. were probably those made in 1820(8).

(ii.) Census Methods. The organisation of the census of India is placed in the hands of a Commissioner, under whom Superintendents are appointed in charge of each province, large agency, or more than one small agency. Under the Superintendents are Charge-superintendents, each of whom has charge of a group of "circles." The smallest sub-division for census purposes is a "block" consisting of from 30 to 50 houses in charge of an enumerator. Above the "block" comes the "circle", a compact group of from 10 to 15 blocks, or about 500 houses under a supervisor, who is responsible for the work of all the enumerators in his circle. At the census of 1901 there were in the whole Indian Empire in round numbers 9800 Charge-superintendents, 122,000 Supervisors, and 1,325,000 enumerators(9), in all about 1,456,800. A special feature of the Indian Census is the preliminary enumeration, which is made some few weeks prior to the actual census day. At this enumeration all the required particulars are duly entered for every person then resident in each enumerator's district, so that on census day it is only necessary to bring up to date these returns. This is effected by striking out the particulars concerning those who have died or departed in the meantime, and adding others for those who have since arrived or been born.

(iii.) Method of Tabulation and Cost. At the censuses of 1881 and 1891 the information contained in the schedules was extracted on "abstraction sheets", but at the census of 1901 this method was abandoned in favour of what is known as the slip or card system. The total actual cost of the 1901 census was £137,674 or £146,265 including charges not debited against the census budget in the public accounts(10).

(iv.) The Schedules. In India it is specially laid down that the schedules must be filled in by the enumerators, but this does not apply to Europeans or to natives of high rank. The schedule for the census of 1871 contained space for the name, sex, age (under and over 12 only), caste or tribe, religion, occupation, education and infirmity. The census of 1881, the first synchronous enumeration of all India, added inquiries as to the individual ages, conjugal condition and language, and at the census of 1891 inquiries as to birthplace and nationality were included. In 1901 the scope of the schedule was the same as in 1891.

2. Ceylon. The census of Ceylon, both in scope and machinery, bears a strong resemblance to that of India, and this resemblance was at the census of 1901 increased by the adoption of the Indian method of preliminary enumeration, and subsequent revision. The Government Agents and the Chairmen of Municipalities and Local Boards are, under the Census Ordinance, ex officio Commissioners of Census for their respective areas. A supervisor is the officer in charge of a "circle", i.e. a group of enumeration "blocks", which contains from 50 to 100 houses and for each of which an enumerator, in rural districts the village headman, was appointed.

(i.) Historical Under the native rulers of Ceylon there does not appear to have been anything in the nature of a census, though there is evidence, both historical and physical, to show that the population must have been very much larger than at present. The decline in the population was probably due to the wars and internecine strife which preceded the establishment of the British Dominion. The earliest enumeration of which there is any record is that taken by the Dutch East India Company in 1789. This census, although apparently incomplete, seems to have been more successful than the first attempt in 1814, under British rule. In 1824 the first complete census of the island was commenced; the returns were published in 1827. At this census particulars were obtained as to sex, race and age, but the age details related only to the numbers above and below the age of puberty. In 1868 the first legislative provision for a census was made, and three years later the first census, in the proper acceptation of that term, was undertaken. Since 1871 a decennial census has been taken.

(ii.) Census Methods. At the census of 1901 great importance was attached to the "drilling" of the enumerators. Each supervisor was required to put his enumerators through a practical course of training by making them visit a number of houses and fill in schedules for the residents. These trial forms were examined by the supervisor and mistakes in them were pointed out to and corrected by, the enumerators. The schedules used were of five kinds :(a) form to be filled by householders, (b) form to be filled by enumerators, (c) form for the enumeration of the estate population, (d) form for sea-going population, and (e) form for the outdoor population. The form (a) was issued in English only, the rest were issued in each of the three languages, English, Singhalese, and Tamil. In the 1901 schedule a column was added for conjugal condition, and various amendments were made in regard to the particulars asked for concerning nationality, religion, occupation and English literacy. Particulars of caste are not asked for in Ceylon. For the tabulation of results the "Slip System" was used, on the same lines as the system used in the census of India(11).

3. Hong Kong. The first regular census was taken in Hong Kong in the year 1881; it was, however, in certain respects, incomplete. Later censuses were taken in 1891, 1897 and 1901, but were not satisfactory in regard to certain of the particulars (12) asked for. The census is taken by police officers, and the "double block" system has been adopted. Under that system each census block is worked by two enumerators, instead of one, the size of the blocks being correspondingly increased. The two enumerators work together, and it is found that a great deal of time is thus saved, as one man can interrogate the members of the house, while the other takes down in writing the information thus obtained. A special feature of the Hong Kong census is the enumeration of the Chinese living in various kinds of water craft.

1. Commonwealth of Australia. The historical development of census-taking in each State of the Commonwealth is discussed in some detail in the succeeding Section of this Report(13). It has, however, been thought desirable, in order to preserve the continuity of this Section, to here give a brief account of the past censuses of Australia.

(i.) The Creation of the Colonies. Although the shores of Australia had been visited from time to time during the 16th and 17th centuries by various Spanish and Portuguese expeditions, it was not until the 23rd August, 1770, that the history of Australia was brought into political connection with western civilisation. It was on that date that Captain Cook took possession "of the whole eastern coast, from lat.38⁰ to this place, lat.10.5⁰ to S., in right of His Majesty King George the Third". Cook, however, proclaimed British sovereignty only over what are now the eastern parts of New South Wales and Queensland, and formal possession, on behalf of the British Crown, of the whole of the eastern part of the Australian Continent and Tasmania was not taken until the 26th January, 1788. It was on this last date that Captain Phillip's commission, first issued to him on the 12th October, 1786, and amplified on the 2nd April, 1787, was read to the people whom he had brought with him in the "First Fleet". <A full historical account of the period referred to may be found in the" Historical Records of New South Wales," vol. 1., parts 1 and 2>.

The western part of Australia was annexed on the 2nd May, 1829, when Captain Fremantle, who had been despatched from England in H.M.S. Challenger in charge of a party of intending settlers, hoisted the British flag on the south head of the Swan River, and took possession of "all that part of New Holland which is not included within the territory of New South Wales". Tasmania was constituted a separate colony in 1825, and in 1836 South Australia was created a "province". In 1841 the separation of New Zealand from New South Wales was proclaimed, while Victoria and Queensland were separated in 1851 and 1859 respectively. Though the boundaries of some of the colonies were changed after the year 1859, no further colonies were created subsequent to that date. The following table shows the dates of annexation, first settlement and creation of the several colonies which now form the Commonwealth, as well as the date of the first regular census taken in each individual colony:

Colony	Date of Annexation	Date of First Permanent Settlement	Date of creation as Separate Colony	Date of first Census
New South Wales Tasmania South Australia Victoria Queensland Western Australia	1770 1788 1788 1770 1770 1829	1788 1803 1836 1434 1824 1829	1786 1825 1834 1851 1859 1829	1828 1841(14) 1844 1854 (14) 1861 (14) 1848

(14) Previously included with New South Wales.

A more complete account of the colonisation of Australia may be found in "The Official Year Book of the Commonwealth of Australia" No. IV., pp.12 to 27.

(ii.) Early Musters. Though regular Censuses were not instituted in the several Colonies until the years specified in the preceding table, population returns in one form or another have existed from a very early period in the history of Australia. The earliest enumerations were known as "musters" and although the actual results of very few of them have been preserved, it is probable that during the early days of Colonisation they were of frequent occurrence. The first official "muster" was taken in 1788, soon after the new settlement at Sydney Cove was formed, and in 1803 the first "muster" of Convicts in Van Diemen's Land was conducted. In 1813 Liverpool was included among the "mustering" stations; and in 1815 vital statistics were given for the towns of Castlereagh and Richmond, thus showing the spread of settlement along
the Nepean River. In the reports of subsequent "musters" the growth of the infant colony may be clearly traced. Thus in 1822 appear such names as Evan District in County Cumberland, Wilberforce on the further bank of the Hawkesbury, Illawarra, and Argyleshire. The statement of 1823 gives the new "mustering" stations of Bringelly, Campbelltown, and Cawdor; in 1824 Moreton Bay, consequent upon Oxley's discovery of the Brisbane River, and Melville Island, indicating the settlement of a military post on the northern coast of Australia, first appear. The growth of the metropolitan district is also shown in the returns; thus in 1825 the Field of Mars is shown as a distinct area, while such additional districts or parishes as Sackville Reach, Pitt Town, Kelso, and Christ's Church also appear.

(ili.) The Development of the Census. The first regular Australasian Census was taken in New South Wales in November, 1828, and included the population at Moreton Bay (now Queensland), and Van Diemen's Land (now Tasmania). Particulars were asked as to the names, ages, and civil conditions of the inhabitants. The next census was taken in 1833, and was followed by another in 1836, when arrangements were made for the enumeration of the population of the newly established settlement at Port Phillip (Victoria). These two censuses were similar in their scope and provisions to that of 1828. In 1841 separate censuses were taken in New South Wales (including the Moreton Bay and Port Phillip districts) and Tasmania. New Zealand, which was first settled in 1839, was expressly exempted from the provisions of the New South Wales Census Act 1841; but it was not until May, 1841, that the islands composing the group were proclaimed an independent colony. Up to that date they continued a dependency of New South Wales. The first regular census was taken in South Australia in 1844, and in Western Australia in 1848. The dates of succeeding censuses are shown in the tabular statement given in Chapter (15) IV hereinafter. All the early censuses of the Australian colonies were based in their main features upon the methods adopted in England, and do not at present call for further notice.

(iv.) Recent Censuses and Census Conferences. While the fact that the census methods adopted in the several Australian colonies were based on a common prototype had the beneficent result of conducing towards general uniformity, it also gave rise to the reproduction of certain defects in English methods and to want of conformity to local circumstances. In course of time it appeared evident that the methods of inquiry and tabulation were in certain respects ill adapted to Australasian conditions, and in 1890 a Conference of Statisticians was held at Hobart for the purpose of preparing a scheme by means of which the census information could be collected and tabulated throughout Australasia in a uniform and effective manner. Though neither Queensland nor Western Australia was represented at this Conference, the decisions of the Conference were followed in the collection and compilation of the censuses in these colonies. This conference resulted in considerable improvement in the degree of uniformity attained in the censuses of 1891 and in the increased fertility of the inquiries.

In 1900 another Conference of Australasian Statisticians was held in Sydney. The colonies represented were New South Wales, Victoria, Queensland, South Australia, Western Australia, Tasmania, and New Zealand. The object of the Conference was to agree to such measures as would lead to uniformity in regard to (a) the date of the census; (b) the subjects of inquiry; and (c) the methods of compilation and tabulation. Although practical uniformity was attained in some respects in the census methods of the Australasian Colonies in 1901, yet the facts that the modes of presentation of the reports and tabular matter were not uniform, that the results of all the inquiries were not completely tabulated for all the States, and that differences arose as to interpretation of terms, rendered it difficult and in some cases impossible to obtain totals for the whole Commonwealth.

2. New Zealand. Reference has already been made to the facts that New Zealand, which was first permanently settled in 1839, was separated from New South Wales in 1841, and was expressly exempted from the provisions of the New South Wales Census Act of that year. In 1852 representative government was granted in New Zealand, with a Legislative Council for the whole Colony, and six provincial district Councils, presided over by an elective superintendent. The provincial governments were abolished in 1876.

(i.) The First Census, 1851. So far as can be ascertained, the first general census of New Zealand was taken in 1851 under authority of an Ordinance (No. 8) of the Legislative Council. The different settlements constituted Census Districts. Resident Magistrates promulgated all information for filling in schedules, etc., and collectors were appointed by the Governor-in-Chief at a rate not exceeding ten shillings per day. No information is available as to scope of inquiry, methods of compilation, or cost of census.

In 1853 the Constitution Act came into operation, and an effort was then made to present the General Statistics of the entire Colony in one comprehensive and authorised compilation. But after considerable delay and some embarrassing difficulties it was found that the work was incomplete in several particulars. The census abstracts had their own peculiar difficulties, arising from such circumstances as the census having been taken at different times in the several Provinces, and even in the same Province in different months, on successive occasions. There was also want of uniformity in the schedules, not merely as to absence of details but also as to important branches of inquiry. Tables which could be completed for parts of the Colony, could not be completed even approximately for the whole.

(ii.) Census of 1858. The census taken in 1858 under authority of the "Census Act 1858", was the first to be taken throughout the whole of New Zealand on a uniform system. The machinery for the collection of the census information in each Province was virtually organised by its own Provincial Government. The divisions of Provinces under the Constitution Act were adopted as census districts. The number of districts was seven. The appointments of enumerators for the several Provinces were all made by the Governor, on the recommendation of the several superintendents. Sub-enumerators were appointed by enumerators. Masters or keepers of asylums, gaols, etc., were appointed sub-enumerators.

(a) Scope of Inquiry. The following were the subjects of inquiry in 1858:
(1) Domestic and social condition.
(2) Number and descriptions of houses and buildings.
(3) Names and surnames of the people.
(4) Relation to heads of the households.
(5) Their condition as married, single, etc.
(6) Age and sex, profession or occupation.
(7) Extraction and place of birth.
(8) Religious denominations.
(9) The date of entry into Colony: whether British subjects or aliens.
(10) The number of persons able to read or write.
(11) The number of acres under crops, sown grasses, etc., and the number of acres fenced.

Information was also requested concerning churches, chapels, etc., their situation, the number of persons each would contain, and the number generally attending, and as to hospitals, asylums, dispensaries, etc., savings banks, friendly and benevolent societies, mechanics' institutes, and other literary or scientific institutions, manufactories, mills, works, mines and quarries. The returns showing this information were, however, incomplete, and of little value.

(b) Compilation. The returns were compiled by enumerators from the schedules, and transmitted to the Registrar-General.

(iii.) Census of 1861. The census of 1861 was taken under authority of the "Census Act 1858", and the "Census Act Amendment 1860". The person responsible for the carrying out of the census was the Registrar-General. The machinery for collection of the census information in each Province was virtually the same as at the preceding census. Arrangements were made to obtain the required information, not only as formerly for each Province of the Colony, but also for each electoral district. Electoral districts as defined by the Representation Act 1860 were constituted census districts, of which there were 43. The scope of inquiry was similar to that of 1858.

(iv.) Census of 1864. The census of 1864 was taken on the same lines as that of 1861, the number of census districts being 45. The military settlers and defence forces in the North Island were a new addition to the population, and were included in the numbers for the Provinces, but not in those for the Electoral Districts. Officers of Regiments were appointed enumerators.

(v.) Censuses of 1867, 1871, 1874, and 1878. The census of 1867 was taken under authority of the Census Act 1867. There were 64 census districts. Special inquiries concerning uninhabited houses and houses building were introduced this year.

Schedules Ill. and IV. of the Census Act, 1858, were repealed by the Census Act of 1867, and the following subjects of inquiry were substituted: Extent of holding, freehold or leasehold. Extent of land fenced. Extent and description of fencing. Extent of land broken up but not under crop. Extent of land in crop in detail. Horses, cattle, sheep, etc. Annual production of butter and cheese. All kinds of machinery and description of manufactories.

The methods adopted at the censuses of 1871 and 1874 were substantially the same as in previous years. The census of 1874 was the last of the triennial censuses which commenced in 1858. In 1878 it was thought that the Provinces were too large to admit of proper supervision and control as enumerators' districts, and the colony was accordingly subdivided into 25 districts. As the Census Act 1877 did not apply to Maoris, the census taken of them by the Native Department was quite distinct in character from that taken of the rest of the population. Since this Act was passed statistics of manufacturing industries have also been collected at each of the New Zealand censuses.

(vi.) Censuses of 1881, 1886, 1891, and 1896. The census of 1881 was the first of the quinquennial censuses which have been taken in New Zealand since that year; the system adopted in 1881 was similar to that at the previous census. The New Zealand householder's schedule contained very extensive inquiries, as, in addition to the particulars of persons information had to be given thereon as to lands occupied and unoccupied, the nature of tenure, the amount of fencing, the numbers and description of live-stock and poultry, the annual produce of butter and cheese, and the number of threshing and reaping machines, steam ploughs, and harrows.

In 1886 a more elaborate census was taken of the Maoris than at any previous census, the tabulation being effected through the "Native Officers" according to numbers, sex, ages, principal tribes and counties.

At the censuses of 1891 and 1896 the inquiries made embraced the same subjects as before, with the exception that in the latter year no information was required as to land cultivation and live-stock, for the reason that these matters were dealt with annually under the provisions of the Agricultural and Pastoral Statistics Act 1895. In 1896 the inquiry as to the domestic and social condition of the people covered the following matters: (a) Number and description of dwellings. (b) Names. (c) Relation to head of household. (d) Conjugal condition. (e) Sex and age. (f) Occupation. (g) Nationality and birthplace. (h) Religion. (i) Naturalisation. (j) Sickness, accident and infirmity; and (k) Education. Manufacturing statistics were also obtained on special schedules, and returns were collected under the fourth schedule of the Act of 1877, as to the number of churches, their accommodation and the attendance thereat, and as to land and building societies, mechanics' institutes, and other literary or scientific institutes. In 1896 the card system of compilation was adopted for the first time.

(vii.) Censuses of 1901 and 1906. As a result of the deliberations of the Australasian Statistical Conference held in February and March, 1900, arrangements were made for a uniform householder's schedule, and for general agreement in methods of compilation, so as to render the results for the several Colonies more fully comparable. As regards New Zealand, the most important alteration in the schedule was by way of introducing an inquiry as to length of residence of all persons not born in the colony. In 1906 the schedule was practically identical with that of 1901. In 1906 the Maori population was not enumerated by the officers who took the general census. It had been found impossible to take an account of the Maoris for one night, and the required particulars were collected during the month of April by "native agents" and magistrates of "native districts". A third count was made on special lines, viz., that of the population of the Cook and other South Sea islands, which now form part of New Zealand.

(viii.) Population and Cost of Censuses. The following table gives particulars of the population enumerated and the cost (exclusive of printing) of each census from 1851 to 1911, inclusive:

Census	Population		Cost of Collection			Cost of Compilation, etc. (£)	Total Cost (£)	Cost per Head d.
	European No.	Maori No.	European Census (£)	Maori Census (£)	Total (£)
1851 1858 1861 1864 1867 1871 1874 (16) 1878 1881 1886 1891 1896 1901 1906 1911	26,707 61,224 106,315 184,131 220,123 256,393 299,514 414,412 489,933 578,482 626,658 703,360 772,719 888,578 1,008,468	n.a n.a n.a n.a n.a n.a 45,470 43,595 44,097 41,969 41,993 39,854 43,143 47,731 49,844	n.a n.a n.a n.a n.a n.a n.a n.a n.a 10,119 9,734 10,216 12,202 14,310 18,058	.. .. .. .. .. .. n.a 106 278 573 789 819 962 1,378 1,419	n.a n.a n.a n.a n.a n.a n.a n.a n.a 10,692 10,523 11,035 13,164 15,688 19,477	n.a n.a n.a n.a n.a n.a n.a n.a n.a 3,959 5,608 5,317 6,237 8,229 11,123	n.a 2,018 3,296 6,245 6,787 9,649 8,741 10,084 12,250 (17) 14,651 16,131 16,352 19,401 23,917 30,600	n.a 8 7.5 8.25 7.5 9 6.25 5.25 5.5 5.75 5.75 5.25 5.75 6.25 7

(16) First Maori Census
(17) Approximate

1. Canada. The Department of Agriculture, which has charge of the Canadian Census operations, has shown itself far more ambitious than the Governments of other parts of the British Empire, and has modelled the census upon that of the United States. An elaborate investigation is made into the components of the general wealth and economic condition of the country. In 1901 the enumeration proper was conducted with the help of 17 chief or supervising officers for the Provinces or Territories, of 351 directing and revising commissioners and assistant commissioners for the census districts, and of 8800 enumerators for the polling divisions.

(i.) Historical. The first Canadian Census, and probably the first regular census of modern times, was that of the province of Quebec, then known as La Nouvelle France. This census was taken in the year 1665. From that year onward until 1754 enumerations were of frequent occurrence, no fewer than fifteen regular censuses (in addition to numerous "statements of population") being taken during that period. Censuses of "Canada" were taken in 1765 and 1784, while in "Upper Canada" --- or what is now practically the province of Ontario --- the census was annual from 1824 till 1842. Censuses in the different districts continued to be frequent until 1861, after which date no official enumeration took place until 1869-71. The census of 1871, the first after the passing of the British North American Act in 1867, included Ontario, Quebec, New Brunswick, Nova Scotia, and Prince Edward Island, other parts of the Colony having enumerated their populations just previously. Since 1871 the census of Canada has been decennial, but an intermediate census was taken in the North West Territories in 1885, and in Manitoba in 1886.

(ii.) Census Methods. There are two important points to be noticed in connection with the Canadian Census, viz. (a) that the de jure population (18) is enumerated, and (b) that the schedules are filled up by the enumerators and not by the householders. In 1901 there were 206 census districts, and 3204 sub-districts. The enumeration in the field was required to be completed within 30 days, including time for the revision of all schedules by the Commissioners before transmission to the Census Office.

(iii.) Tabulation and Cost. In tabulating the returns, the Hollerith (19) electrical tabulating machine, introduced in the United States, was employed at the 1891 and subsequent Canadian Censuses. The general object and working of this machine is described elsewhere. Particulars as to the cost of the census of 1901 are not available. The cost of the census of 1891 (when the population enumerated was 4,833,239) was £108,000, of which £85,000 were for the enumeration (local expenses), £20,000 for compilation and tabulation (office expenses) and the remainder represented printing and incidental expenses.

As regards the scope of the inquiry, the first census (1665) enumerated the population de jure by families, age, sex, conjugal condition, and professions and trades (not occupations of the whole people). The scope of the inquiry has varied in succeeding censuses; the "professions and trades" were soon dropped, but agricultural statistics and particulars of buildings were, on the other hand, included. The scope of the censuses taken during the 18th century was not generally so wide as in the preceding censuses. During the 19th century the inquiries differed materially in their scope. The ages at first consisted of two groups only --- persons under and over 16 --- but the number of groups was gradually increased. Birthplaces appear for the first time in 1817 (in a census of Nova Scotia), occupations reappear in 1827, in which year also the earliest reliable statistics of religion, births, marriages, deaths, education, and industries are given; infirmities were first recorded in 1842 (20).

During more recent censuses, owing to the large amount of information which is sought to be obtained, a number of different schedules, each comprising a considerable number of questions and printed in both English and French, are used. The census of 1891 was taken with 9 schedules, comprising in all 216 questions, and the census of 1901 with 11 schedules, comprising 561 questions, as shown in the following tabular statement:

Canada schedules and enquires at 1901 Census (21)
No. of Schedule	Subject	No. of Inquiries
I.	Living persons	34
II.	Buildings, lands, churches and school	35
III.	Deaths	15
IV.	Farmlands, fruits and plantations	37
V.	Field products	37
VI.	Live-stock and animal products	26
VII.	Agricultural values	25
VIII.	Manufacturers	67
IX.	Forest products and furs	51
X.	Fisheries	130
XI.	Mines	104

Besides the foregoing, there were used at the 1901 census of Canada special forms for obtaining records of (a) persons temporarily absent, (b) persons such as boarders and lodgers, not present when the enumerator called and of whom particulars could not be given by the head of the household, (c) cheese and butter products, and (d) manufactures of clay products . The schedules contained inquiries to ascertain for every trade, occupation, and profession the time each person was employed and the amount of his earnings, and for all children of school age the time at school in the year.

2. Newfoundland and Labrador. In this Colony the first complete census appears to have been taken in 1857. Later censuses were taken in 1869, 1874, 1884, and 1891, and since the last of these dates the census has been taken decennially.

In many respects the census of Newfoundland resembles that of Canada, and deals not only with persons living, but also with births, marriages, and deaths (for the year preceding that in which the census is taken), fisheries, buildings, ships and boats, primary production, mines and minerals, and mills and factories.

The census of Labrador is not simultaneous. A large number of the population can be found at their homes only during the winter season --- a time when it is impossible to take the census owing to difficulties of communication. In the summer these people are engaged in the fisheries off the coast --- but the enumerator counts them as resident in their place of abode during the winter. Special provision has to be made to avoid duplication of entry, and in most cases the schedules are retained until December, when most of the people have returned home, and the sheets can be checked (22). The number enumerated, therefore, appears to be a compromise between the de jure and de facto population.

1. General. In the West Indies (with which may be grouped British Guiana, British Honduras, and the Bermudas) the particulars collected and the arrangement made for the censuses are similar to those in other British Possessions. The principal feature of note is the inquiry which is usually made as to colour and native population.

2. Bahama Islands. In these islands there has been a decennial census since the year 1891. Particulars are collected with regard to sex, age, conjugal condition, profession or occupation, nationality, infirmities, and education, and a special inquiry is made as to the number of people vaccinated (23).

3. British Honduras (24). In this Colony one quarter of the schedules were printed in Spanish at both the 1891 and 1901 censuses. As regards the degree of education, the return is more clearly defined than usual with reference to writing, persons being required to state whether they could write their names.

4. Other British Possessions (25). In other British possessions (including Bermudas, Barbados, Jamaica, the Leeward Islands, Trinidad, and British Guiana) censuses were taken in 1891 and 1901. The particulars collected and the arrangements made were much the same as elsewhere, but in the smaller colonies the returns comprise only a few sheets. The principal points calling for remark are the inquiries as to colour and Indian population.


01 See" Modes of Census-taking in the British Dominions," R. H. Hooker, M.A. Journal of the Royal Statistical Society, Vol. LVII., p. 327 <back>
02 See Census Report, Cape of Good Hope, 1904, p. i <back>
03 See Census Report, Natal, 1904, p. 22 <back>
04 See Census Report, Orange River Colony, 1904 <back>
05 See Census Report of the Transvaal, 1904, Vols. I and II <back>
06 See "On Census-taking and its Limitations." Journal of the Royal Statistical Society, J. A. Baines, C.S.I. (now SirJ. Athelstone Baines), Vol. LXIII., p .. 44 <back>
07 See Census of the Island of Bombay, 1864, p. 1 <back>
08 See Geographical, Statistical, and Historical Description of Hindustan, by W. Hamilton, 1820, and Reports on the State of Education in Bengal, by W. Adam, 1835-38 <back>
09 See Census Report of India for 1901. The supervisors of India were generally subordinate revenue officers, village accountants, and where non-officials were appointed, as in towns, the position was ordinarily looked upon as an honour. Many enumerators were volunteers, and received special certificates where the work was well done. Minor officials were also largely employed for this purpose, and the number of paid enumerators was relatively small <back>
10 See General Report, Census of India, 1901, Part I., p. xv <back>
11 See The Census of Ceylon, 1901, Vol. I., p. 36 <back>
12 See Census Report of Hong Kong, 1901, p. 1 <back>
13 See Chapter IV, p. 36, Statistician's Report, 1911 <back>
14 Previously included with New South Wales <back>
15 See Chapter IV, p. 36, Statistician's Report, 1911 <back>
16 First Maori Census <back>
17 Approximation <back>
18 See Chapter II, p. 11, Statistician's Report, 1911 <back>
19 See Chapter II, p. 15-16, Statistician's Report, 1911 <back>
20 See Modes of Census taking. Hooker, J.R.S.S., Vol. LVII., p. 337 <back>
21 See Census report, Canada, 1901, Vol I, p.vii <back>
22 See Census Report, Newfoundland, 1901, p. v <back>
23 See Census Report of Bahama Islands, 1901, p. 7
24 See Census Report, British Honduras, 1901, p. 11
25 See Hooker, " Modes of Census-taking in the British Dominions." J.R.S.S., Vol LVII., pp. 341-2