sources; while some give only the income from taxation. Some of the provincial statements are not totalled. The several towns have, however, been classified, the different classes of income have been distinguished as far as possible, and totals have been made in this office for the purpose of comparison.

""The statistics of municipal taxation, taken from the volume on taxation and worked out as above, give the following results for large towns containing over 10,000 souls each, for small towns containing less than 10,000 souls, and for all towns, large and small together.

"The incidence of municipal taxation on all municipal towns, both large and small, may be compared thus

"It will be seen that the general incidence of municipal taxation in Bengal is less than half the average incidence for the rest of India; that the average incidence in the North-Western Provinces is 44 per cent. higher than in Bengal; while in other provinces the incidence is from 72 per cent. to 133 per cent. higher than in Bengal. For towns containing more than 10,000 souls, the incidence of municipal taxation in other provinces is from 76 per cent. to 164 per cent. more than in Bengal. For small towns containing less than 10,000 souls, the incidence of municipal taxation in Bengal is about the same as in most other provinces, except the North-Western Provinces, where the incidence is 24 per cent. smaller, a very large number of petty places being there lightly taxed for police purposes only.'

CHAPTER XXIII.

BIRTHS AND DEATHS.

FOR some years past attempts have been made to collect statistics

Returns of births and deaths from Bengal have always been untrustworthy.

of births and deaths from the whole area of these provinces. The inaccuracy of the results obtained has, however, always been notorious. Exceptional difficulties beset the Sanitary Commissioner for Bengal. His returns have always been incomplete from every district, and absolutely untrustworthy. In 1871 the mortality in Bengal, with a population of 66 millions, was registered at 260,331, which is only a proportion of four in a thousand. The reports of previous years have shown even more unsatisfactory results than this.

In the following table, which the Lieutenant-Governor has taken from the Sanitary Commissioner's report, the districts of Bengal are arranged in order as to efficiency in mortuary registration in 1871:—

Nothing could prove the worthlessness of the returns more clearly than the figures displayed in this statement. The figures for the district of Mymensing, showing a death-rate of 1 per 1,000, would be ridiculously, if they were not deplorably, wrong. Dr. Jackson also remarks that in Burdwan, where so fatal an epidemic was raging, it is probable that not more than one death in several can have been reported.

Dr. Jackson proceeds to show by an interesting analysis that there is a special defect in all districts in the registration of female and of infant mortality. It is in infant mortality that the greatest deficiency appears. In England the deaths of persons under five years of age form 411 per cent. of the total mortality; in Bengal they form only 15.8 per cent. It may be possible that infant mortality may not be so great in India as it is among the poorer classes of the great cities of western countries; but the disproportion cannot approach to what it amounts to in these returns.

The Lieutenant-Governor has, however, already succeeded in effecting

New system of collecting vital statistics from selected areas, from 1st January 1873.

in the present year a tentative reform in the system under which mortuary and vital statistics are acquired. He has freely confessed that the acquisition of accurate statistics is at present impossible over the enormous areas which compose our districts with their vast populations and uneducated agencies; and, while not relinquishing the attempt to do what can be done towards a complete registration, it has been resolved in the meantime to perfect the system on asmaller scale over certain experimental selected areas both urban and rural in every district. These areas have been chosen, as far as was possible, with reference to their geographical situation, so as to be accessible and easily supervised, and it was arranged beforehand that the census should be taken in them with particular care.

At least one town area and one country area have been selected in each district. The town area selected has usually been the headquarters station of the district. The area of the rural registration has been limited to a moderate number of villages, compactly situated, with a population, as a rule, of from 10 to 30,000 inhabitants. In the town areas the duty of collecting these statistics has been generally met and discharged by the Municipal Commissioners, supplemented by a grant from Government. For the rural areas a small special expenditure has been sanctioned. A very wide discretion has in all cases been allowed to the local officers in appointing the agency to collect these returns. It was left to their discretion also whether the registration of births should be included in the first instance.

In order to facilitate these returns, the Lieutenant-Governor has sanctioned the recommendation which Clerk allowed to Civil Surgeons. had been repeatedly urged upon Government by successive Sanitary Commissioners and by district officers, for allowing a sanitary clerk to all Civil Surgeons, to aid them in the compilation of vital statistics, which in these provinces, as elsewhere, is a duty imposed upon the medical officer.

The registration of sanitary statistics from the selected areas in Bengal started fair from most of the Results of the first half-year of 1873. localities in the provinces from the 1st January 1873, and reports have been received from the Sanitary Commissioner on the results which have been collected during the first two quarters of the year. They present undoubtedly a very great improvement on what have hitherto been submitted. In some cases they are no doubt fairly correct, in others very approximately accurate, and in very few are they very widely wrong.

The areas selected are 101 in number, comprising a population of

Mortuary Statistics.

Towns

Rural areas

Combined areas

nearly two millions (1,922,608). The gross mortality of each quarter is returned as follows:

2nd quarter.

7:38 per mille.

5.01

6:05 39

Or as given in the usual form per annum per mille.

That the actual mortality is higher than these ratios, is certain, but it is satisfactory to have made so decided a step as this in advance towards the truth. It is especially satisfactory to note the improved registration evinced during the second quarter, when more experience had been gained.

The mortality according to sex during the two quarters under review compares thus:

The female registration is therefore shown proportionately defec

tive, as might have been expected.

The mortality according to age is returned as follows:

It has already been suggested that infant life may be better in India than in England, but there is suspicion that in great part it is the defective registration of infant mortality in India that makes the death-rate abnormally low. The death of an adult is an event in the family and in the village where it takes place: the death of a child, a few weeks, or months old,-no event at all.

Statements are given in the Appendix which show in detail the selected areas of Bengal and the ratio of their mortality during the

first half year of 1873. The following areas show a mortality at a rate of more than 30 per thousand of the population per annum :

It is probable that these results are approximately correct, and we know that among some of these areas, where the mortality is highest, there have been epidemics of fever and cholera which have raised the death-rates. On the other hand, it is to be remembered that, as a rule, the first half of the year in Bengal is the healthiest half. The areas selected and quoted above are scattered, and not specially unhealthy localities. These figures are not sufficient for us to generalise from, but they seem to shew that the death-rate in Bengal is high. By way of comparison it may be mentioned that the highest annual deathrates registered in England are 40 per thousand, the lowest 17 per thousand, and that the annual average death-rate in England is 22.4 per thousand.

The localities above specified are those in which the selected area system has proved most successful. In some cases it has failed. The town areas of Julpigoree, Dacca, Furreedpore, Sylhet, and Cachar, and the rural tracts of Burdwan, 24-Pergunnahs, Nuddea, Julpigoree, Dacca, Sylhet, Gya, Singbhoom, Maunbhoom, and Seebsaugor, are so far a failure that they all return a death-rate of less than 15 per thousand. As yet the system has been most successful in the divisions of Rajshahye and in Assam. In Assam the special indigenous agencies that exist render the task an easier one than elsewhere. The system has been least successful in the Presidency and Dacca divisions, where unfortunately the indigenous native agencies have been suffered to die out more completely than in other parts of Bengal. In the rural areas failure has most frequently resulted from the area chosen having been too large to manage and superintend properly, but the difficulties which in this respect have been brought to light are now being remedied. The unsatisfactory nature of the Burdwan rural returns in particular is attributable to the excessively large and unmanageable area that had been selected.

A comparative statement is annexed below which shows the rate of mortality in the different months of the year in a few special localities, where it is reported that special pains have been taken to secure correct registration.