c. The mean age at death has also been used to test the sanitary condition of different classes of persons inhabiting the same town or town district. A very extensive series of , tables, for instance, has been compiled, in which the mean age at death of the gentry, tradesmen, artisans, and paupers of the several parishes of the metropolis is represented, and used as a measure of their sanitary condition.* The parish of St. James's, Westminster, in which the class termed gentry is more likely to be appropriately named than in the poorer and less fashionable parishes, may be con veniently taken as an example. In this parish the average age at death, children included, is 42 for gentry, 26 for tradesmen, 21 for artisans, &c., and 49 for paupers. The differ ences are here so extraordinary, as only to admit of explanation on the supposition of a vast disproportion between the numbers living at the same ages in the several classes. If the element of age in the living is disregarded in comparing the gentry with the artisans out of the workhouse, it must equally be dis regarded in contrasting the gentry with the paupers in the workhouse, who are drawn chiefly from the artisan class. The difference between 49 (the mean age at death of paupers), and 42 (the mean age at death of the gentry) is obviously due to the greater average age of the inmates of the workhouse, as the differ ence between 42 (the mean age at death of the gentry), and 21 (the mean age at death of artisans), is traceable mainly, if not wholly, to the great disparity in the ages of the living members of the two classes. Unfortunately we are not yet in a condition to apply to the mean age at death of these classes of the population, the same correction which, when applied to counties, cities, and city districts, served in so striking a manner to equalize the results. The ages of the living members of the several classes of society is still an impor tant desideratum. The tables under consider ation also give the mean age at death of gentry, tradesmen, artisans, and paupers dying after 21 years of age. In the parish of St. James's, Westminster, the mean ages at death of these classes are 57, 51, 46, and 58 respectively. The differences, though less considerable, are still at total variance with the results of the most accurate inquiries into the value of life of the same classes of society t, and only admit of explanation by supposing a great disparity in the ages of their living members, together with an erroneous method of se lection and classification.
d. The mean age at death has also been employed to test the sanitary state of different climes of society, and of the members of' different professions, without reference to their place of residence. But little objection can be made to this test when applied with ordinary precaution. If the earliest age ad mitted into the tabular abstracts, from which the averages are calculated, is the same in all the classes submitted to comparison ; if during the time over which the observations extend the classes compared with each other have received no rapid accession of numbers ; and if' the deaths are those of the whole body of the profession or trade (or, if of a section only, then of a similar section in every case), the mean age at death will constitute a fair measure of the relative sanitary condition of the several classes in question. But if, on
the other hand, two professions or occupations are compared, in which the age at entry is not the same; or the one is stationary in point of numbers, while the other is rapidly increasing; or if the whole body is taken in the one case, and only the senior or junior members in the other, the results will be quite unworthy of confidence.
Some of the greatest misapprehensions existing with respect to the duration of life of certain classes of the community, are traceable to the selection of the senior members of a class to represent the entire class to which they belong. Nothing is more common, for instance, than to hear royal families, or the members of the aristocracy, or the clergy, or the army, spoken of as long-lived, on the strength of the advanced ages attained by kings, peers, archbishops, or general officers. In making a selection of the more conspicuous members of the class, the significant fact is overlooked that they are also the oldest members, and that they do not attain their exalted rank till a period of life greatly in advance of that at which they entered their several professions. The mean age, at death, for example, of archbishops and bishops of the established church, is upwards of 71 years, but the mean age at death of the whole body of the clergy is about 64 years. A similar disparity would be found to exist be tween the peers and the whole body of the aristocracy, and between general officers in the army, and admirals in the navy, and the whole body of officers in the two branches of the public service.
The tables already referred to as comparing the four classes of gentry, tradesmen, artisans, and paupers in the several districts of the metropolis, supply analogous examples of erroneous selection and classification. In the great majority of the metropolitan parishes, for instance, there is no class of gentry properly so called ; but this class consists, with the exception of a few professional men not im properly mixed up with it, of tradesmen who have retired from business long enough to be entered in the mortuary registers as gentle men. On the other hand, the pauper class is very largely recruited from the ranks of the artisans and labourers, and contains a very considerable proportion of old persons who, being no longer able to earn their hood, have come upon the public for support.
The mean age of the living has been occa sionally resorted to as a test or measure of salubrity. It has been assumed that a low average age of the living members of a class, when compared with the average age of ano ther class, arises, cceteris paribus, from a high mortality leading to a quick addition of young members. This assumption is justified only in those cases in which the addition of young members can be shown not to arise from any other cause, such as an increased demand for members of the class in question. It is there fore a test to be employed with great caution ; and it will also be necessary to show that the age of admission of the classes subject to com parison is the same, or open only to very slight variation.