These returns show no change from the last century to the present, for in the forty-one years from 1754 to 1795 the deaths were 1 in 48, while from 1795 to 1815 they were 1 in 46, taking the whole population over 15. If the numbers between the ages of 20 and 50 be considered, the ratio for the first period is 1 in 84, and for the second, 1 in 84. It would seem, therefore, objectionable to introduce any observations before 1750, but since that time no restriction appears to be necessary. We have given in the last number the mortality for Carlisle. This being a large town, has been thought to be well suited to give the average mortality for an insurance company. In the larger cities the chance of dying is greater; in the country districts smaller; so that this affords a fair average for the whole country. The liabilities to mistakes and errors are supposed also to be smaller than for a whole nation. But in an old country, where the government is strong and respected, if a system of registration is carefully devised, and continued for a long period, the returns for a whole nation would deserve more confidence than for a single city. The wide extent of country, and the long continuance of the observations, increase the probability of a fair average. We shall introduce into our average the observations of Sweden, Norway, Prussia, Hanover, Saxony, and England, with much confidence in our results. Of these the greatest weight should be given to England, because so many of our people are sprung from this stock, and race is supposed to have some influence on longevity. The Swedish observations seem, however, to be well suited for obtaining a reliable table of mortality. The country is not so far north as to be injuriously affected by cold; it is free from the malarious diseases of southern latitudes; it is inhabited principally by a rural population, with only one considerable city; the people are industrious, religious, and intelligent; the census is taken frequently, and the reports for each province scrutinized most carefully for errors; the returns of the deaths have been kept up for more than a hundred years; they have been made with great care and labor; the population is large; they have been exposed to every variety of seasons, of epidemics, of war and peace, of famine and abundance; surely such returns are entitled to much confidence. The expectation of life in Milne's Swedish table is, however, nearly two years less than at Carlisle from 15 to 50, and continues below it to the end of life. It is more than a year below Dr. Farr's. The Swedish table of Dr. Price is still lower. But this is no good reason for excluding these observations, for we do not know beforehand whether the American mortality will conform most to the English or the Swedish experience. The probabilities are, that the deaths here will be greater than in either country. But whether this is true or not, the Swedish observations deserve much esteem by our life offices, and we shall not hesitate to allow them considerable weight in our average. Dr. Price's is so old, and so near the limit where we have thought proper to exclude the observations, that we shall allow it less weight than the others, but we shall not feel at liberty to exclude it entirely. In column second at the end of this article is Dr. Price's Swedish table, and in the third column is the adjusted rate of mortality. The influence of adjustment is very slight, as the large numbers observed and the quinquennial intervals of ages of the living and the dying have prevented any serious anomalies. In column fourth is Mr. Milne's Swedish table, and in the next column its adjusted mortality. Both adjustments are made in the same way, by taking the geometrical mean of five successive rates as the true amount for each age. In the next two columns are the rates of mortality for the next two periods of twenty years, obtained from the observations published by Dr. Farr in the sixth volume of the registrar general's reports. The mode of construction which we have adopted is that proposed by Dr. Farr, but in adjusting we have taken the mean of seven consecutive rates of mortality instead of five, because the observations being given for every five years, the adjustment by fives was not so satisfactory. The next table we will introduce (column eighth,) is founded on the observations in Norway, published by Dr. Farr in the same volume of his reports. These were officially communicated, and seem to be made with care. They extend from 1800 to 1840, but the ages of the living are only given for the last census in 1840. The table we have constructed is, therefore, only for the last ten years, from 1830 to 1840. The mode of construction we have employed is the same we have used for the Carlisle observations, which, though more laborious, is more reliable than any of the methods proposed. The actual number of the living at each age being approximately obtained, the rates of mortality that will give the observed deaths for these numbers of the living are more likely to be correct than the rates that give the proper number of deaths in a stationary population. The numbers of the living and the dying being smaller than for Sweden, and for ten years only, we shall give less weight to this Norway table than to the Swedish. The rates of mortality are, however, less, and nearer to Farr's. The next table is derived from the observations of Mr. Finlaison, actuary of the National Debt Office in Great Britain. His report to Parlia ment was made in 1829, and gives the particulars of the deaths and ages of the government annuitants and of the nominees under the several government tontines, beginning as early as 1693, and ending in 1826. The facts are numerous, perfectly reliable, continued through a long series of years, and very carefully collected and arranged. Mr. Finlaison's results have not, however, been very much esteemed. They differ considerably from our best tables, and contain anomalies not found elsewhere. This arises, we think, from the selection of lives which would exert a greater influence in a tontine than in an insurance office. When a policy is first issued the insurer is known to be in perfect health. His physycian and the company's medical examiner, as well as the insured himself and his friend, unite in testifying to this fact. The rate of mortality at any particular age will therefore be very different among recent in surers and among those who have been long insured. The difference is still greater in the tontines, since many of these persons were chosen because of their vigorous health and their promise of long life. The mortality among such persons soon after admission would be very different from the average rate for persons of the same age taken at random or among the members of an insurance company. And this is sufficient to account for the irregularities in these tables. Besides, Mr. Finlaison did not use all the materials he had collected and published, but only a portion of them which he thought most worthy of confidence. His "tables of annuities, computed for the government," were founded only on the Irish Tontine," the Tontine of 1789," and that "of the Sinking Fund as observed to the end of the year 1822." 66 In the table we have inserted at the end of this article we have used all the observations of Mr. Finlaison, omitting only the first set, because founded on observations which were made before the middle of the last century; and the result is free from many of the anomalies of his tables, and everyway more worthy of confidence. We have adjusted it precisely in the same manner he did for his tables. Thus, we have added together his two summaries, Nos. 8 and 15, and subtracted No. 1, comprising in our result 21,350 lives, of whom 12,275 yet survived in January, 1826. We have then obtained the ratio of the living and the dying at each age, and adjusted these ratios by taking the geometrical mean of five consecutive terms, and then of each three of these results, following precisely the same method as Mr. Finlaison. To this mean two corrections were applied to get the rate of mortality at 15, 16, etc., because it is the ratio of the living and the dying at the average age of 15, 16, etc. The rates finally obtained are to be found in column ninth at the end of this article. At the earlier ages when the influence of selection is felt, the table is quite irregular. But for the older ages, when this objection disappears, the large number of lives, the exactness and accuracy of the observations, and the absence of all withdrawals, make the figures worthy of more confidence. We have inserted in column tenth the mean of Finlaison's two tables, but we do not think them worthy of much weight in the proposed combination. In column eleventh is to be found Farr's Northampton, which is very different from Dr. Price's, partly because it is founded on recent observations, but mainly because it has been properly constructed from the numbers of the living and the dying. Dr. Price, with great ingenuity, supplied the numbers of the living from the ages of the dying; but his bypothesis, though better than those made by his contemporaries, was not an approximation to the truth for the younger ages. Farr's is deserving of every confidence. As it embraces the mortality for seven years, from 1838 to 1844, among a population nearly as large as Carlisle; as the facts on which it has been based have been observed with care, and the table constructed on correct principles, it is worth nearly as much as the Carlisle table. Price's Mortality Milne's Milne's Sweden, Sweden, Norway, Finlaison, Finl'son, Farr's Age. Sweden. adjust'd, Sweden, adjust'd. 1795-1815, 1815-35. 1825-35. 1745-1826, male, fe. North'n. 5899 .0064 6098 .0061 .0066 .0052 .0055 .0071 .0064 .0056 15.. Price's Mortality Milne's Milne's Sweden, Sweden, Norway, Finlaison, Finl'son, Farr's Age. Sweden. adjust'd. Sweden. adjust'd. 1795-1815. 1815-35. 1825-35. 1745-1826. male, fe. North'n. 20.. 5697 79 5903 74 80 101 65 Price's Mortality Milne's Milne's Sweden, Sweden, Norway, Finlaison, Finl'son, Farr's Age. 81.. 468 THE importance of the crime of forgery, and the confusion which it is capable of creating in the transaction of both public and private affairs, have in all time engaged the serious attention of the ministers of justice. Notwithstanding this, however, the arts of the forger have never received that attention, in an educational point of view, which the importance of the subject demands. In the progress of civilization cupidity has very nearly attained the dignity of a science, and how to detect and how to avoid the arts of the counterfeiter has well nigh become a necessary part of commercial education. Forgery, in law, may be defined to be the fraudulent making or alteration of any record, deed, writing, instrument, register, stamp, etc., to the prejudice of another man's rights. This broad field of operation is open to a great variety of means with the freest use of scientific principles. If we consult the records of this species of crime, we discover the arts of the forger to be contemporaneous with the advance of science. Indeed, the propagation of the truths of the science of chemistry, among all classes of society, seems not only to have facilitated the arts of the falsifier, but in some cases to have been available for obliterating the evidence of murder. The application of chemical processes in the perpetration of crime have, in some instances, demonstrated the greatest triumphs in that science; and had they been used for scientific purposes alone, they would have clothed their discoverers with imperishable honor. Photography-one of the brightest gems of modern chemistry-bas achieved some of its greatest exploits in efforts to elude the " bank-note detector." But in the adaptation of the latest truths of science, the forger never forgets the ruder methods of his art, now reduced to an unprecedented degree of perfection. The smooth erasure, the over careful preserva |