^{page 139 note *} Whether the mean duration of life on the Continent, and especially in Germany, is less than in England, cannot, for want of sufficient observations, be ascertained with certainty. It appears, however, that in England more people attain to a higher age than upon the Continent. According to the accounts of the census published in the year 1841, there were in England and Wales, out of 15,927,867 persons, 1,147,108, or 7·2 per cent., above 60 years of age, and amongst them not less than 249 were of 100 years of age and upwards. By the census in Belgium in 1831, out of 4,000,000 there were only 16 of 100 years of age and upwards. In Wurtemburg, in 1832, out of 1½ million there was only one of that age; and by the last census in Prussia, at the end of 1849, the number exceeding 60 years of age was scarcely 6 per cent. of the total population. The actual experience of the Gotha Bank, although it is not yet sufficiently extended to give a solution to this question on a positive basis, has up to the present time confirmed the above hypothesis. Notwithstanding that the table of mortality used by the Bank shows, for the higher classes of age above 55 years, an essentially greater mortality than all the approved English tables, still, according to the tenor of the reports of the Bank, the actual mortality at these ages is up to the present time, on an average, greater than was expected. (See Report for 1848, § 16.) It may, indeed, be a question, to decide what influence the frequent voluntary retirement of healthy members has on the increase of the mortality of the members remaining in the Company, and whether it is not equal to the profits which the outgoing members yield to the funds of the Company.

^{page 141 note *} The increased premium is so computed that at the time of payment of the sum assured the reserve shall be equal to the full value paid. If, for instance, a person thirty years old wishes to receive the sum assured at sixty years of age, in case it should not have become a claim by death at an earlier period, he will have to pay, with his original premium, a further annual premium of ·854 per cent. If this party has been already five years assured, when he first desires at the age of thirty to modify his assurance in the way described, the yearly premium will then only be ·810 per cent.

^{page 143 note *} Whether the Company also values its reserve by the Northampton Table is not known. Such a proceeding is not to be recommended to them; on the contrary, those Companies which with the view of safety have computed their premiums according to one of the tables showing too high a rate of mortality, such as the Northampton or Sussmilch, should calculate the values of their policies by one of the tables which show the real mortality already experienced in the Company, or which may be expected for it. The tables most suitable for the computation of premiums for Life Assurance Companies are indisputably the excellent tables of Brune, deduced from the observations of the Prussian “Widows' Provident Fund, and those constructed from the observations of the seventeen English Life Assurance Companies. The following comparison will show how the values will come out according to these and according to the Northampton. It will be seen that the value reserved, according to the latter table, at corresponding ages and periods of assurance (some few eases excepted), is too small. At the division of the surplus, a portion of the profits included in the premiums, and which belongs to the future, to cover unfavourable fluctuations of mortality, will by this process of calculation be anticipated out of the actual value of the risk.