Next Selection Previous Selection


Scope of the Inquiry. —The statistics of buildings are based on the descriptions furnished by the local enumerators, who were required when delivering the schedules to enter in their Memorandum Books "the address and description (e.g. , private house, block of flats or block of model dwellings, maisonette, workhouse, hospital, school, hotel, church, chapel, warehouse, stable, etc.) of every dwelling or building." After having completed the collection of the schedules, the enumerator was required to copy into a "Summary Book" provided for that purpose the address and name of the occupier and the description of the kind of building; examples of the manner in which these entries were to be made were provided, and on completion of the Summary Book the work of the several enumerators in each registration sub-district was revised by the Registrar of Births and Deaths.

By this means it was intended to obtain the data from which to derive approximate totals of the most distinct types of dwellings, and also some indication of the number of certain classes of buildings not used as dwellings.

From the nature of the inquiry it would not be possible to claim a very high degree of accuracy for the results; but the replies courteously furnished by the majority of Local Authorities, to whom the figures have been submitted for an expression of opinion, indicate that, on the whole, they may be regarded as reasonably close approximations. In some cases attention was drawn by the Local Authority to considerable errors in classification arising from neglect or misunderstanding of their instructions by enumerators, and the necessary amendments have been made. The most common errors appear to have been misdescriptions of "flats" and of shops, and some of these may have passed undetected.

Statistics of various kinds of buildings have not been compiled at' any previous Census, the general term "house" having hitherto been used in the population tables to denote habitable buildings of all kinds. This practice has now, for the reason stated on page 6, been discontinued.

With a view to comparing the numbers of separate buildings used as dwellings in 1911 with the numbers returned under the general definition "houses" in 1901, the latter are given throughout the tables. It must be observed, however, that the classification of houses in 1911 has resulted in their more accurate enumeration, so that the figures may not in all cases be strictly comparable. In 1901 reference was made to some flagrant cases in which failure on the part of local officers at the previous Census to observe the definition of a house had rendered the figures of very little comparative value; and it is probable that minor inaccuracies occurred in many districts in 1901 and 1911 also.

Classification of Buildings. —Before discussing the figures under the several headings given in the tables it will be desirable to refer briefly to the general principles of the classification. A broad distinction is drawn between "buildings used as dwellings" and "buildings not used as dwellings," the criterion applied being the provision of sleeping accommodation on the premises; thus a shop or office which is structurally part of a dwelling-house is counted along with the latter as a building used as a dwelling, while a lock-up shop is counted as a building not used as a dwelling. In order to account for the total population of each area dealt with, all the buildings used as dwellings are shown, those which are not classifiable under one or other of the definite headings being shown separately as "others"; and persons who did not pass the night of the Census in any "building used as a dwelling" are also distinguished. Under "buildings not used as dwellings" only certain specified headings are given, it being apparent that the returns of other miscellaneous buildings would furnish no useful results and could not be made complete. A stable is a building and a rabbit hutch is not, but at what precise point constructions intermediate between these types cease to be buildings it would be difficult to say.

Buildings used as Dwellings.Ordinary dwelling-houses constitute a very large majority of the buildings used as dwellings; in addition to those buildings described simply as private houses, the heading includes those described as "cottages," "villas," "farm- houses," and generally all buildings designed for the occupation of a single private family. In rural districts and in those towns where but few houses are sub-let to more than one family there is not much probability of error in this number. Where, however, sub-letting is frequent, mistakes may arise from two sources. Perhaps the most important difficulty is the confusion between "flats" and ordinary dwelling-houses.

Some enumerators showed a tendency to describe as "flats" all houses in the occupation of more than one family; while others systematically described as "private houses" those cases in which the houses each consisted of a pair of flats. Thus on the one hand there may be some transference of ordinary dwelling-houses to "flats," and on the other hand some transference, of "flats" to ordinary dwelling-houses. A further possible cause of error, leading to an over-statement of the number of houses, is the return of two or more schedules from a single house which is not described by number or otherwise in such a way as to make it apparent that the schedules relate to the same house.

Flats are distinguished from ordinary dwelling-houses by the fact of their being structurally distinct dwellings within the same building, and comparison with the figures for 1901 is preserved by counting each block of flats, i.e. , all the space within the external and party walls of a building, as one house. The number of separate flats in a block differs very widely according to the type of building. In some localities the large blocks predominate, while in others flats chiefly consist of the upper and lower portions of small houses constructed with separate entrances to the two floors. This kind of "flat" has, as stated above, frequently been confused with ordinary dwelling- houses. Sometimes the method of numbering these flats is an indication of their character, each pair having the same numeral and the upper floor having the letter "a" affixed. But where each flat is numbered separately as an ordinary house, they may in some cases appear under the wrong heading in the tables. These pairs of flats appear most commonly in some of the northern towns and in some parts of London. It may also be noted that the word "flat" is sometimes used for barge, but it is unlikely that the occasional use of the word in this sense has led to any appreciable error. Another and more serious confusion of terms is the use of the word "tenement," which has frequently been applied to flats as well as to portions of sub-let houses, and, where used in the former sense and interpreted in the latter, has probably caused some under-statement of the number of flats. In cases where single flats are built over lock- up shops, each flat is counted as a building used as a dwelling, and appears therefore as a separate block in column 5 of the tables in Volume VI of the Report. The omission on the part of many enumerators to specify the numbers of separate blocks of flats rendered it necessary in such cases to return the summary books to the registrars for completion, and this increased the work of tabulation.

Shops are generally distinguished from ordinary dwelling-houses without much difficulty; but the frequent practice of exposing goods for sale in a room of an ordinary dwelling-house has caused the enumerators some doubt as to the proper description to apply to the building, and the general tendency would appear to be for such doubtful cases to be entered as ordinary dwelling-houses. In a few urban districts it was found on abstracting the figures that no buildings had been described as shops, arid in some other districts the numbers were absurdly small; inquiries in these cases have been sent to the local registrars and the corrections have been made as far as possible. In other cases a shortage in the number of shops has been pointed out by the local authority and has been rectified after inquiry. Workshops such as "blacksmiths' shops," "carpenters' shops," etc., are not included under this heading. Shops classified as "buildings not used as dwellings" comprise all those without sleeping accommodation and therefore include not only lock-up shops such as are built on the ground floor front of blocks of flats or of other large buildings, but also large business establishments in which the staff does not sleep on the premises and there is no resident caretaker.

Hotels, Inns and Public Houses , as shown in the tables, approximate closely to the number of licensed premises; and as the figures have in most cases been passed, by the local councils, they may be regarded as fairly correct. They do not actually coincide with the numbers coming within the cognisance of the licensing authorities owing to the inclusion, on the one hand, of temperance hotels, and to the exclusion, on the other hand, of such places as clubs and railway refreshment rooms and of public houses which are not occupied at night.

Offices, Warehouses, Workshops, Factories which appear as "buildings used as dwellings" are not nearly as numerous as those shown in the table as "buildings not used as dwellings"; those in the former category are mainly occupied at night by resident caretakers, or sometimes (as, for example, banks) by resident managers; those in the latter category are either untenanted or were occupied on Census night only by watchmen, or by other employees who would be enumerated at their own homes on the following morning.

Institutions. —The buildings shown under this heading include all the public institutions given in Table 17 of Volume, I, namely, Workhouse Establishments, Lunatic Asylums, Hospitals, Prisons, Reformatory and Industrial Schools, Institutions certified under the "Poor Law (Certified Schools) Act, 1862," and Certified Inebriate Reformatories and Retreats, as well as the Naval and Military Establishments given in Table 20 of Volume I. In addition to the above, Registered Lodging Houses, Nursing Homes, Boarding Schools, Colleges, Convents, Orphanages and Almshouses are also included. It will be observed that a great variety of buildings is combined in one total and it is not possible to say exactly how far the numbers given in the table relate to separate structures or to separate institutions. At previous Censuses a large institution which was specially enumerated by the chief resident officer was generally counted as a single inhabited house although it may have consisted of more than one separate block of buildings; while a small, institution, such as a row of almshouses, enumerated with the ordinary houses in the district, would be counted as several houses in accordance with the general definition. There has therefore on this account been a slight under-statement of the number of "houses" as hitherto defined, but the numbers are so small in proportion to the total that the effect is almost negligible. The same practice has been followed on the present occasion in order to furnish comparative figures;' and it may therefore be assumed that the number of inhabited institutions does, on the whole, represent the actual number of institutions more nearly than the number of separate institution buildings.

Other Buildings include dwellings over stables, and various buildings, such as clubs, town halls, museums, libraries, schools, theatres, etc., with resident caretakers. Barns, sheds, tents, etc., in which any person slept on Census night are not included under this heading; persons enumerated in such structures are shown in column 11 of the tables, together with those persons who passed the night in the open air and those who were enumerated on board vessels.

Uninhabited Dwellings are all those in which no person passed the night of the Census; an unlet house occupied by a caretaker, or a house left in charge of servants is, therefore, not included under this heading, but appears as an inhabited house. In 1901 an attempt had been made for the first time to divide uninhabited dwellings into two classes, viz., those "in occupation" "and those not in occupation." Uninhabited houses in occupation were those from which the tenants were temporarily absent on Census night, and included such places as dwellings over shops intermittently occupied, ordinary dwelling-houses of which the occupants were absent on visits, seaside and country cottages occupied only during the summer months; uninhabited houses not in occupation were, in fact, empty houses, and included new houses unoccupied on account of their recent completion as well as old houses unoccupied on account of dilapidation; buildings in ruins and others condemned as unfit for habitation were not intended to be counted as dwellings, but it is possible that some few were included. The distinction between the two classes of uninhabited houses was made with a view to showing, as nearly as was practicable from the Census returns, how many houses were liable for rates as occupied premises—inhabited houses and uninhabited houses in occupation being classed together for this purpose. But from the number returned it appeared probable that the large increase over 1891 in the number and proportion of uninhabited houses in some urban areas was due in a great measure to the inclusion of lock-up shops as houses "in occupation." It was therefore considered desirable not to attempt the distinction in 1911.

Dwellings being built are probably returned with a fair degree of accuracy, but a little uncertainty attaches to the figures for shops, offices, etc., and other buildings, which in some cases may have been entered in one or other of these columns even when not intended to be used partly as dwellings.

Buildings not used as Dwellings .—Places of Worship. All churches, chapels, mission halls, Sunday schools, and other places of religious worship as returned by the enumerators are given under this heading, except in those cases where there is a resident caretaker or where the place of worship forms only part of a building, the remainder of which is used for other purposes. For these reasons the numbers given in the table do not correspond exactly with the number of churches and chapels of the Established Church in which marriages could be legally solemnized, and of Places of Meeting certified for Religious Worship. The numbers of such buildings as were on the register at the end of 1910 have been carefully compared with the numbers of places of worship as returned by the enumerators; after allowance on the one hand for buildings belonging to the Established Church in which marriages could not be legally solemnized, and on the other hand for registered places which were not entire buildings or which were inhabited on Census night, inquiries were sent with respect to apparent omissions on the part of enumerators, and the necessary corrections have been made. As a result of these inquiries it was found that in a considerable number of cases the building in question was not, at the date of the Census, used for purposes of religious worship.

Government and Municipal Buildings include all Government and municipal offices not occupied at night, buildings connected with municipal gas, electric, water, tramway, etc., works, police stations, fire brigade stations, public libraries, baths, and school buildings provided by local education authorities. Some errors may have arisen owing to failure to distinguish in all cases buildings, especially schools, under public control from those privately owned, and it is probable that the returns under this heading are consequently somewhat understated. State and Local Government institutions, such as prisons, workhouses and isolation hospitals, are not included here, but under the heading "Institutions."

Shops, Offices, Warehouses, etc. (not used as dwellings) , have already been alluded to, but it may be pointed out that in the numbers included here no distinction is made between occupied and unoccupied shops, etc., and it is probable that some shops with dwellings attached have been inadvertently classed under this heading because the dwellings were uninhabited. It is possible also that some of the offices, warehouses, etc., " shown as uninhabited buildings used as dwellings should have been classed as buildings not used as dwellings.

Theatres and other Places of Amusement. —The numbers under this heading do not show the total number of places of amusement in a given locality, partly because some such places come under the heading "other buildings used as dwellings" on account of having a caretaker or other person resident on the premises, and partly because in many cases halls and rooms occasionally used for concerts, dances, etc., form part of larger buildings, such as hotels or town halls, and are classed accordingly.

The figures for all buildings not used as dwellings include unoccupied buildings, but not those in course of erection. The total number of buildings of all kinds shown in the table does not always represent the number of separate ratings in a district on account of the exclusion, already alluded to, of certain miscellaneous buildings not used as dwellings—for example, stables, coach-houses, cow-sheds, etc.—and also on account of parts of buildings being separately rated.

Number of Inhabited Buildings and Average Number of Occupants .—The number of inhabited buildings or "houses" in 1911 as derived from the returns furnished by the enumerators was 7,141,781, compared with 6,260,852 in 1901; the average number of persons to an inhabited building was 5.05 in 1911, against 5.20 in 1901; in all urban districts (including London and the County Boroughs) the averages were 5.23 and 5.40 respectively, while in the rural districts they were 4.51 and 4.58. The tabulation of the number of buildings does not afford any information as to the relative amount of accommodation in the inhabited buildings returned at the two Censuses. At previous 'Censuses (1891 and 1901) the numbers of separate tenements of 1, 2, 3 and 4 rooms were tabulated, and at the recent Census the tabulation was extended to all tenements of 1 to 9 rooms, but it has not been considered desirable to carry this tabulation further in the case of tenements or to apply it to houses as such. If, however, it may be assumed that between 1901 and 1911 there was, on the whole, no material variation in the size of houses, the averages quoted above would indicate an increase in the amount of house-room per head of population, and the improvement would appear to be greater in the urban than in the rural districts. Among the several administrative counties, the average number of persons to an inhabited building was less in 1911 than in 1901 in all but two English and four Welsh counties.

The exceptions among the English counties are Southampton, where the slightly increased average was due to a large increase in the military population rather than to a diminution of house accommodation, and Monmouthshire, where the abnormally large increase of population due to mining and industrial activity appears to have outstripped the increase of houses; in this connexion it may be noted that the proportion of uninhabited houses in the latter county was low, while the proportion of houses being built was exceptionally high. The four Welsh counties showing an increase in the average number of persons per house are Brecknockshire, Flintshire, Carmarthenshire, and Glamorganshire, in the two latter of which conditions prevailed similar to those referred to above with regard to Monmouthshire.

In the County Boroughs the average number of persons per house in 1911 was 4.99; in all other urban districts (excluding London) it was 4.85; and in the rural districts, 4.51. Comparing one county with another, by far the highest average is found in London, which had 7.89 persons per house; next in order came Middlesex with 5.61, Northumberland and Glamorganshire each with 5.45, Monmouthshire and Durham each with 5.32, The lowest proportions are naturally found among the most rural counties, e.g., Cardiganshire, 3.89; Anglesey, 4.03; Merionethshire and Huntingdonshire, 4.14; Cambridgeshire, 4.15; and Carnarvonshire and Norfolk, 4.18. The following statement shows the large towns having respectively the highest and lowest average number of persons to an inhabited house in 1911; the averages for 1901 are given for comparison. It will be noticed that in both groups of towns the average number in 1911 was generally less than in 1901.


In the Administrative County of London the average number of persons per house fell from 7.93 in 1901 to 7.89 in 1911; in the City of London, however, and in nineteen of the metropolitan boroughs there was an increased average number per Louse in 1911, although in all but four of these there was a decrease of population during the intercensal period. The boroughs of Holborn and Finsbury, for example, decreased in population by 16.9 and 13.3 per cent, respectively, while the average number of persons per house rose by 4.1 and 5.4 per cent.; in both cases the increase in the average per house may be due to an increase in the proportion of the population living in large blocks of flats.

The metropolitan boroughs with the highest and lowest averages, respectively, were as follows:—


The variations in the average numbers per inhabited house in 1911, shown by the large towns and metropolitan boroughs given in the above statements, cannot be regarded as implying similar variations in the amount of house accommodation per person. Apart from the disturbing influence of large institutions and of large blocks of flats, there is undoubtedly considerable variation in the average size of houses as between one locality and another, and consequently a comparison of the average cubic space per person available in the dwellings of the several places cannot be deduced from the above figures. It may be assumed that in the towns with a comparatively high average number of persons per house there was either more sub-letting of ordinary dwelling-houses, or a larger proportion of flats, than in the towns with a low average number of persons per house; but it is at least probable that on the whole any excess in the average which may be due to the latter consideration would be partly compensated for by the greater accom-modation provided in a block of flats than in a single dwelling-house of the same class.

Similarly, in comparing the average number of persons per house in 1901 and 1911 for the same area possible differences in the character of the houses at the two periods must be considered, as well as the slight uncertainty of the actual comparability of the figures for the reason stated on page 193. As regards erection of flats in the intercensal period the Census returns afford no direct indication; towns which in 1911 had but few flats present no difficulty, but for those which had considerable numbers it is not possible to say whether or not they had similarly large numbers in 1901. The effect of public institutions, however, on the average numbers per house is more easily shown. In Epsom Urban District, for instance, out of a total population of 19,156, more than 5,000 were enumerated in asylums which were opened in the intercensal period, and the average number of persons per house rose from 5.86 in 1901 to 7.61 in 1911; a still more striking case is the small Urban District of Thurstonland in Yorkshire West Biding, where a new county asylum contained 1,160 persons out of a total population of 2,041; and the average number of persons per house rose from 4.22 to 9.24. In the rural districts of Amesbury, where the average rose from. 4.96 to 6.08, of Alton, where it rose from 4.42 to 6.02, and of Andover, where it rose from 4.41 to 6.17, the increases were due to the, occupation since 1901 of military barracks or camps.

Quite a different cause of increased averages is apparent in some cases where a rapid growth of population may lead to a temporary shortage of house accommodation; this has already been suggested as the cause of the increased average in Monmouthshire (not of the consistently high average which it has in common with other colliery centres); and in four of the urban districts in that county—Bedwellty, Caerleon, Mynyddislwyn and Rhymney—an abnormally high rate of growth was associated with remarkable increases in the average number of persons per house. It may be observed, however, that the rapid development of a district may actually lead to a decreased average number of persons per house. This is especially true of suburban districts, where building operations on a large scale generally mean the introduction of many comparatively small houses in place of a few comparatively large ones; for example, among the urban districts of Middlesex, the population of Hendon increased by 72.9 per cent., while the average number of persons to a house decreased from 6.64 to 5.78, the population of Southall Norwood increased by 99.4 per cent., while the average number of persons to a house decreased from 6.91 to 5.94,1 and the population of Southgate increased by 124.2 per cent., while the average number of persons to a house decreased from 5.32 to 4.62. In other cases a reduced average number of persons per house is associated with a decline in total population; the following urban districts may be cited as examples:—

Urban District. Average number of persons per
inhabited house.
1901. 1911. Decrease 1911. Decrease
per cent.
Dalton-in-Furness 4.86 4.36 0.50 10,763 17.30
New Windsor 5.66 5.07 0.59 12,681 10.30
Millom 5.60 4.81 0.79 8,612 17.40

A reduced average may also be found in a district in which the growth of population has been arrested shortly before the conclusion of the intercensal period; or in which one part of the district has increased and another part has decreased, as in the Metropolitan Borough of Woolwich.

Proportion of Population in Dwellings of various kinds. —Of the total population enumerated, 86.6 per cent, were returned as in ordinary dwelling- houses, 5.2 per cent, in shops, 3.0 per cent, in institutions, 2.9 per cent, in flats, 1.4 per cent, in hotels, 0.6 per cent, in offices, warehouses and other buildings; the remaining 0.3 per pent, were enumerated on board vessels, or in barns, sheds, tents, etc., or in the open air. As would be expected, the urban districts show a larger proportion than the rural districts in shops, in institutions, and in flats. Comparison of the proportions for the large aggregate areas is afforded by the following statement:—


Number of Uninhabited Dwellings and Dwellings being built .—A large number of empty houses in a district may sometimes be regarded as evidence of a lack of prosperity; but in view of what has already been said (see page 195) the number of houses uninhabited on Census night is by no means a certain indication of the condition of a district. Towns and rural areas vary very greatly as regards the proportions in which the several lands of uninhabited dwelling-houses contribute to the total. A decaying town in which tenants cannot be obtained for a large number of the houses; a suburban district in process of development by enterprising builders, in which a large number of houses are always just about to be occupied; and a seaside town, in which a large number of houses would be occupied if the census were taken three or four months later, may all three show exactly the same proportion of uninhabited dwelling-hmises although the conditions of the three districts are entirely dissimilar.

The number of houses in course of erection is perhaps a better guide than the number of uninhabited houses to the prosperity of a locality; and in view of the interdependence of the numbers it is necessary to consider both together. It has been remarked in the report on a previous Census2 that the proportion of houses building to houses existing is influenced by the average time it takes to build houses and by the average number of years they last. The proportion is affected also by the season of the year by the building operations in the seasons immediately preceding, by the facilities for obtaining money by small builders, "and the price of building materials. Any of these considerations may have considerably affected the numbers at any Census; thus, for example, it has been stated that at the Census of 1891 "the number and proportion of houses in course of erection was exceptionally small, owing, it was presumed, to the severe and prolonged frost that prevailed in the early months of that year."3 For these reasons, therefore, comparisons with previous Censuses must be made with caution. Speaking generally, however, the number of uninhabited houses was less in proportion to the number of inhabited houses at the Census of 1911 than it had been at any of the four preceding Censuses; and the proportion of houses being built was very much less than it had been at any previous Census. The following table shows the numbers of houses— inhabited, uninhabited and being built—at each Census, 1801-1911.


On examining the figures at the last two Censuses for the different classes of area it will be seen that the apparent decline in the proportional number of uninhabited houses is somewhat less in the urban than in the rural districts, and less in the county boroughs than in the smaller urban districts. In all these comparisons, however, the possibility of varying proportions of lock-up shops having been included in the figures for 1901 must be remembered. As regards houses being built, the proportion to the total inhabited is naturally less in 1911, and the proportional decrease greater, in the Administrative County of London than in any other of the groups for which figures are given in the following statement:—


Number of Places of Worship and proportion to population .—The number of places of worship derived from the enumerators' returns in 1911 was 49,970, giving a ratio to population of one to 722; of these 49,970 churches, chapels, etc., 22,797 were returned in the urban districts and 27,173 in the rural districts, the average numbers of persons for each place of worship being 1,235 and 291 respectively; in the county boroughs the average was 1,454, and in the smaller urban districts, 961. No attempt to obtain by means of a Census any statistics of buildings of this kind had been undertaken since 1851. At that Census a detailed inquiry was made embracing not only the number of buildings, but also the denomination of the religious body to which they belonged, the date of erection, the number of sittings provided, the number of worshippers on a certain day (Sunday, 30th March, 1851), and other particulars. The returns were made voluntarily, it having been decided that the terms of the Census Act did not authorize "the inclusion of such particulars among the statutory subjects of inquiry. The results were embodied in a series of detailed tables, prefaced by an elaborate report giving an account of the progress of religious opinions in England and of the doctrines and organization of the various religious bodies. Only the actual number of buildings has been ascertained on the present occasion, and comparison with 1851 is therefore limited to the relation of the number of buildings to population; moreover, the results of the inquiry in 1851 were tabulated for the ancient counties, for many large towns and for registration districts, while the figures for 1911 are given only for administrative areas. In the following statement therefore a few counties have been selected in which the difference between the ancient county and the administrative county with its associated county boroughs is inconsiderable, and for a few large towns of which the boundaries have not been much extended between the two Censuses:—


These figures bear the impress of the obvious fact that, as may be seen also from Table XCIII, a scattered rural population requires more places of worship in proportion to its size than one of urban type. The difference is probably compensated for to a large extent by the larger size of buildings in the towns. As the total population is now much more urban in type than it was in 1851, the proportional number of places of worship required on the scale of accommodation then provided must now be considerably less than in that year, a fact which no doubt largely explains the increase in population per building from 520 to 722. In this connexion it is to be noted that the increases in persons per building shown above for counties are on the whole much larger than those shown for towns. The greatest proportional increase is in Essex, which in 1851 was a county of predominantly rural type, whereas in 1911 only 20 per cent, of its population lived in the rural districts, the large urban communities of West Ham, East Ham, Walthamstow, Leyton, Ilford and Southend having meanwhile come into existence.

As regards other classes of buildings not used as dwellings, no data are available for comparison with any previous Census, and on account of the variety of the buildings necessarily grouped under each heading only the most general comparisons between different localities should be attempted. The total numbers of shops, whether with or without dwellings attached, and the proportion they bear to the populations of the several areas, may reflect to some extent their commercial status, though the mere enumeration of the number of buildings in which retail trade is carried on can form but a very imperfect index of the volume or character of that trade. Similarly the number of separate buildings used entirely as offices or as warehouses, workshops or factories, does not afford-any very definite information as to the mercantile or industrial activity of the place in which they are situated. In the case of Government and municipal buildings and of theatres and other places of amusement no close comparisons can be made without some allowance for the numbers (not ascertained as in the case of shops, etc.) included as "other buildings used as dwellings." The following statement shows the average number of each of the specified kinds of buildings not used as dwellings to 100,000 population in the typical divisions of the country; the average for all occupied shops, etc., inclusive of those used as dwellings is also given:—


The proportional number of shops, offices, etc., is greatest in the county boroughs, and least in the rural districts. The anomaly of the greater proportional number of Government and municipal buildings in the less populous centres may be due partly to the larger number of small schools in those districts, but the cause of the excess of places of amusement in the smaller urban districts as compared with the county boroughs is not so apparent, though it is probably associated with smaller size of the buildings. The greater proportion of places of worship in districts where the population is most scattered has already been noticed and a similar tendency is observable with regard to hotels, inns, etc., though in this case the variations are not so wide.

The fact that this is the first time the Census organisation has been utilised for the collection and tabulation of statistics of various kinds of buildings naturally gives a somewhat tentative character to the results. The problems presented to the very large number of enumerators—especially in towns—in attempting to furnish satisfactory descriptions of all the diverse structures in their districts have probably not all been solved in the best possible manner; and the Registrars of Births and Deaths who were responsible for the revision of the enumerators' work had no previous experience to guide them in this particular matter. It is hoped, however, that the figures now presented, though admittedly imperfect, are not without value. They do not represent the result of any special inquiry, since particulars regarding the buildings in their districts, habitable or otherwise, were in any case required from enumerators as at past Censuses in order to ensure a correct return of population. The question of the repetition in future Census reports of the tabulation of buildings will probably be decided largely by the degree of value found in practice to attach to it.

1 This Urban District contains a large lunatic asylum.

2 Report on the Census of 1861, page 9.

3 Report on the Census of 1901, page 36.

Next Selection Previous Selection