TABLE E.-Illustrating the Adjustment of the Age-Returns.

THE CHAIRMAN said he was pleased to announce that the Society had received a bequest of 100l. to the Building Fund from the late Hon. John Eliot Sanford, of Massachusetts, who had been an Honorary Fellow since 1870. It was a very gratifying incident, and almost if not quite unique in the history of the Society, that one of its Fellows in the United States should remember them when arranging his affairs. He thought it would be only right that they should convey to his executors the Society's deep appreciation of the circumstance.

Sir ATHELSTANE BAINES then read his paper.

Mr. T. A. WELTON said that in the census of Ireland of 1851 there was a diagram showing the numbers returned under successive

ages, and the inequalities were stupendous; the number at the age of 51 was about one-twentieth of the number at the age of 50; and there were other differences of a similar kind.

Mr. A. H. BAILEY said he was not qualified to criticise this paper. If they could get returns for about 1,000 or 10,000 lives accurately, they might deduce life-tables from them, and these would be valuable; but he could not deal with the millions of Sir A. Baines's paper, which appeared so uncertain as to the facts behind them.

At

Mr. NOEL A. HUMPHREYS said there could be no question about the extreme interest of the paper, especially of that part which attempted to gauge the effects of famine and plague on the ageconstitution of the Indian population and upon its death-rate. the same time it seemed somewhat difficult to accept the results arrived at as conclusive, bearing in mind the uncertainties and difficulties connected with taking a census of an Oriental population, and in the absence of complete and accurate registration of births and deaths. He thought the choice of the word "Peradventures as applied to an Indian life-table was a very happy one; but some of the conclusions drawn in the paper suggested that this word might be applied generally to some of the other life-tables which were referred to in Table I of the paper, the deductions from which were certainly difficult to accept. That table correctly showed that in England, of the sanitary progress in which we were justly proud, the increase in the duration of life during the twenty-five years 1875-1900 was about 10 per cent., judged by the official life-tables; but it also showed that the increase in the duration of life in the second half of the last century was equal to 27 per cent. both in Sweden and in Holland, and to 22 per cent. in the last twenty years of the century in Italy. These conclusions suggest the desirability of definite and accurate information as to the precise methods adopted in the construction of those foreign life-tables. The sanitary progress of England during the last thirty years was so unquestionable and so encouraging that it seemed most difficult to believe that the mean duration of life had increased more rapidly in several other European countries than in England and Wales. He had just read with a great deal of interest the Memorandum of Mr. G. F. Hardy based on the Census Returns of India, and it was impossible not to admire the ingenuity of the tables which he has based on these Census Returns. At the same time, one could not forget the definition given by one of the most eminent of their Presidents, Dr. Farr, of what constituted a true life-table. He said "a true life-table was not deduced from population alone nor from deaths alone, but from the ratio the one bears to the other at successive ages." Now Mr. Hardy had not, it is clear, in his possession materials which would enable him to make a true life-table, but his so-called life-tables were both ingenious and interesting. It is a fact that so-called life-tables were formerly constructed from deaths alone, but Dr. Farr had shown, by a

comparison with the results of his true life-tables, the precise extent of the fallacy of that false method. Mr. Hardy had attempted to construct a life-table from population alone, which was a most interesting experiment, and he should like to see his method tested, which could be done without much difficulty. If his method were applied to the English census figures, and a table constructed on a basis similar to that adopted for the Indian tables, the result might be compared with that derived from the correctly constructed life-tables issued from time to time by the Registrar-General. The difficulty Mr. Hardy experienced in constructing a life-table from population alone naturally beset him also when he attempted to estimate the rates of mortality in the Indian population from the Census Returns. Some of his results are unquestionably startling. For example, he deduces from the census figures that in the province of Bombay the average annual death-rate rose from 21°2 in 1881-91 to 33 2 ten years later-in 1891-1901. This signifies that the increase of the Bombay death-rate in the later ten years was equal to 50 per cent., and implies 3,000,000 additional deaths during those ten years. In view of the extreme difficulty which surrounded this computation, it seemed reasonable to hope that by the method adopted the effect of plague and famine in the province of Bombay during the ten years 1891-1901 has been over-estimated by Mr. Hardy.

Sir SHIRLEY MURPHY congratulated the author upon this important paper, and all the more considering the enormous difficulty which attended the compilation of the statistics of population.

The CHAIRMAN said Sir Athelstane Baines possessed, perhaps more than any other man, the faculty of clothing the dry bones of statistics with humour, so as to make everything he wrote on that subject delightful to listen to; and certainly this Paper was no exception. He had profited by the great store of experience he had laid up during his service in India, so that he was always able to produce an interesting paper when one was required; and they were grateful to him for this one, which would be a valuable addition to the Journal.

He took it that the circumstances and the conditions under which Indian statistics were collected comprehended many varied considerations, and were modified by the peculiar religious beliefs and observances of the people. For instance, there was a practice which, he was told, prevailed in parts of India of marrying a child to an adult young woman, and then for the father of the bridegroom to take the bride to his home, and to provide for the relief of his son and himself from the fate which attached to childless persons by producing children for the benefit of his son until the son had a sufficient age to become in effect as well as in fact the husband of the lady. He noticed, in reading Mr. Thurston's Ethnographical Notes on Southern India, that that custom was undergoing, in certain places, a change in what was described as the direction of decency-that

some tribes did not like the father of the bridegroom taking this function upon himself, and they transferred it to another gentleman selected by the bride. When one had to deal with the circumstances of a population affected by a custom so foreign to our own, one could understand there would be difficulties in the way of obtaining accurate statistics.

He begged to formally move a vote of thanks to Sir Athelstane Baines.

Sir ATHELSTANE BAINES, in acknowledging the vote of thanks, said that his paper was avowedly no more than a stopgap, which covered but a portion of the wide field which would repay investigation. The differences in the European life-values to which Mr. Humphreys had referred had been introduced into the subject with the intention of showing that nowhere could these computations be considered to have more than a temporary currency, even when worked out by statisticians of authority equal to that of any of our English workers in this line, with the exception, perhaps, of Dr. Farr. He was not prepared to conclude that in the hands of such skilled manipulators the differences were attributable to paucity of data, to the exclusion of the possibility of real changes in the agecomposition of the people. In India, then, the greater variability of the conditions would justify still greater differences in the calculated values than in Europe, whilst the value of lifetime, as a whole, is not unlikely to be far lower. The remarkable difference between the death-rate in Bombay in 1881-91 and in 1891-1901, on which Mr. Humphreys had commented, was explicable by the prevalence of plague from 1896, and of famine which subsisted until about nine months before the census was taken. He saw nothing incredible in the variation. The main object in dealing with such topics was to obtain a mean, or average, population. The best foundations available were the two communities of the north and south respectively, one carefully observed, the other unaffected by famine, which could be contrasted with populations elsewhere, known to have been under different influences. In regard to the peculiar family relations mentioned by the Vice-President, people must not run away with the idea that such customs prevailed over India at large. They were, in fact, confined to a small and backward community in the south-west. Moreover, a similar custom is to be found in the reputedly more civilised population of the lower Volga, if the evidence of the novelist Maxim Gorki is to be trusted. Prolificity in India was not stimulated by domestic custom, but the abundance of children was due, as he had pointed out, to the almost universality of marriage in those of productive age.

FOOD TAXATION in the UNITED KINGDOM, FRANCE, GERMANY, and the UNITED STATES.

By S. ROSENBAUM, M.Sc.

[Read before the Royal Statistical Society, 19th May, 1908. The Rt. Hon. Sir CHARLES W. DILKE, Bart., M.P., President, in the Chair.]

I.-Introductory.

So far as I am aware the subject of food taxation has never before been brought before the members of this Society. Sir Robert Giffen and the late Mr. Stephen Bourne have, in papers read before this and other learned societies, and in written contributions to various journals and reviews dealt with the subjects of food and taxation; but I cannot discover that they have ever attempted to demonstrate the connection and relation between them. It would be unwise and unnecessary to suggest any explanation for this omission, but it has occurred to the writer that in view of the prominent part which the subject occupies in the leading controversial political question of the day, this omission should be rectified, and the present paper is a contribution towards that end.

A little reflection at once suggests that if the facts as to food taxation are to be made known, they should, as far as possible, be made to relate to the leading industrial countries of the world. Nostra nos sine comparatione delectant. We are prone to derive too great pleasure from the contemplation of our own economic position; some of the conceit is only taken out of us when our position is compared with that of other countries and with other times. The importance of this warning has, it is hoped, been amply heeded by the presentation in this paper of comparable information covering a long period of years of the food taxation of the United Kingdom, and her three largest industrial rivals, France, Germany, and the United States.

The question of food taxation may be regarded and discussed from two separate and distinct points of view. There is, firstly, the point of view of the Exchequer; how much is raised annually from the taxes levied on foodstuffs. There is, secondly, the question of the effect of these taxes on the nation, and whether food taxes bear ultimately more hardly on the masses of the people than taxes of the same amount raised by any other means. The discussion of each of these questions would more than suffice for a paper of