89
which can be taken as the first of the regular censuses, the returns are vitiated by the addition of a percentage to certain items, apparently to compensate for the possible absence of residents from the Colony owing to the unfavourable date on which the Census was taken. From 1891 onwards, however, the Censuses were carried out in much the same manner as they are taken to-day.
Some of these have a special interest worthy of notice. The special object of the 1897 Census was to discover, if possible, in what way the population of Victoria had been affected by the sanitary measures occasioned by the plague in 1894; such as the resumption of Taipingshan, the closing of basements, the removal of cocklofts, and the enforcement of the laws against overcrowding. The Census Officer did not, however, make any comments on this subject in his report, but left it to others to make the proper generalisations from his figures.
In 1901 the Census was extended to include the New Territories, an area which is stated to have had a population at that time of 102,254 persons of whom 17,243 lived south of the Kowloon Range of Hills. No attempt was made, however, to ascertain the occupations of the people living in the New Territories, nor their exact ages, and the figures really represent an estimate only based on the number of houses found in the districts. The enumeration of the Population Afloat was most carefully done on this occasion and showed a total of 40,100 persons.
In 1906 a partial census was taken, confined to the original Colony and New Kowloon, that is, South of the Kowloon Range of Hills. This Census was taken on the application of the Medical Officer of Health who strongly advocated that the intercensal period should not exceed 5 years.
The 1911 Census was on a very much more ambitious scale than had ever before been attempted in the Colony. Full particulars were obtained for all the inhabitants of the New Territories and the Population Afloat of whom in the past only sex and age had been required. Columns recording Married State, Degree of Education, Religion and Infirmities were added to the Census Form and the work correspond- ingly expanded. A better tabulation procedure was also followed, with, no doubt, increased accuracy in the tables produced,
The enumeration of the New Territories was very thoroughly done but it took three months to complete it. If it is true that in those days there was very little movement among the inhabitants of the New Territories, it is probable that thẹ results were almost as accurate as if the enumeration had been completed in one day. From the figures obtained, a complete list of the various villages with the number of their inhabitants was compiled by the District Officer with a view to ascertaining at the next Census (1921) whether the construction of the Railway (then just completed) would have any effect on the distribution and size of the population. Unfortunately the comparison was not made in 1921 and it has been impossible to make it in 1931.
Another interesting aspect of the 1911 Census is that the premature outbreak of revolution in Canton in the Spring of 1911 is stated to have caused a considerable influx of refugees to the Colony and to have inflated the figures of the Census of that year.
The 1921 Census was not so complete as in 1911, questions as to language, religion and infirmities being omitted. The Column for Race was also omitted with the result that it was impossible to classify the cosmopolitan non-Chinese community except by nationality. Thus, Indians and Malays had to be included among British subjects, Annamites and Tonkinese as French subjects and so forth; and there was no means of finding out how many there were of each race. British subjects of Chinese race might also have been included, had instructions not been issued that all Chinese should fill up a form printed in Chinese. One large distinction of race was, therefore, made, but no other. The omission of the column for "industry in which employed" was also a serious one as it affected indirectly the classification of occupations.