11.—2

10

should be precisely the same, but clearly this is impracticable from the very dissimilarity of the original facts as regards the occupation of the members, their place of habitation, &c. All that can be done is to take care that the age-groupings compared should as far as practicable be similar, bearing in mind how much the element of age influences the rates of sickness and mortality, and to remember that all such comparisons must be understood to be only of a general character, and in default of more exact knowledge of the precise entities composing the different data. At the same time it may be asserted that, so far as research has shown, there is no reason to believe, in respect of the different investigations, the results of which are set out in the foregoing table, that the constituent elements of any one experience as regards the occupation, place of abode, &c, of the members differ very materially from the others. It will be noticed that the years of lives observed in botli the Odd Fellows and the Foresters inquiries were very large, exceeding in each case a million and a quarter of exposures. Though your experience is considerably smaller, it has the advantage of running over a period of ten years, in lieu of the quinquennial term embraced in the other investigations. The following tabulation presents concisely the broad results of the various data :—

The Rates of Mortality and Sickness according to the Experience of Members of Friendly Societies.

Considerable variation will be observed in both the mortality- and sickness-rates of the respective inquiries ; but, even making allowance for the possible disturbing forces, which I have already alluded to, as regards the units embraced in each experience, the difference between the results is most marked, and worthy of consideration. Eegarding, first, the mortality-rates, the divergency is great, the death-rate among the Eechabites being so remarkably low that I was at once led to examine whether there was any palpable fault in the original returns submitted to me which afforded a solution of this feature. No such explanation was apparent, and, though the years of life observed were not as numerous as I should have desired, they were ample to preclude violent fluctuations in the experience due to paucity of the observations. Whether this low death-rate might not be partially due to the large number of fresh entrants into the society, most of whom would no doubt have passed through some test of examination as to health, next received my attention. Therefore the records in respect of only members who had not been admitted into the society during the period of observation—viz., the ten years 1878-87 —were separately examined, but with the result that their death-rate was found to be almost identical with that for the whole of the members. Indeed, no striking difference was disclosed between the mortality-rates for those admitted into the society before the period of investigation commenced—viz., Ist January, 1878 —and the members who were subsequently enrolled. The low rate of mortality appears to run through the table, and is not confined to one period of life. To test whether there might be anything in the occupations of the members which would throw light upon this subject, though time would not admit of a thorough investigation in this respect into the records for the whole experience, a number of cards taken indiscriminately were examined, but nothing of an exceptional character was disclosed. Thus I was brought back to the fact that, certainly in one respect, all the members were supposed to be on a common basis—viz., total abstainers. It is within my knowledge that in one large assurance company, which maintains a special section for total abstainers, for now twenty years the actual deaths in each quinquennium have only been, in this section, 71 per cent, of the expectation ; but to what period of life this is due, or whether common throughout life, cannot be affirmed, as the facts for groups of ages have never been published. I refer to the results of the experience upon this special subject of this company because it led me to examine what proportion of deaths the actual number in your experience furnishes as compared with the expected number according to the Foresters' experience, and the result is that your death-rate only produces 69 per cent, of the other, or a close approximation to what is stated to be the experience of the assurance company above alluded to. It is not within my reference to argue for or against any connection between total abstinence and longevity; my duty is to state clearly the facts brought out by the present investigation, and to leave my readers to draw their own deductions upon this subject. The only observation I feel called upon to make is that, personally, I am quite impartial upon the matter, and, therefore, no bias can be attributed to my results. Whatever effect total abstinence may have upon health and longevity has hitherto been a matter of speculation, for, until the present inquiry, no investigation of magnitude has been made into the subject as regards rates of sickness, and though the aforementioned assurance company possesses, undoubtedly, observations of a most important character, only the most meagre facts have been made public. Irrespective of total abstinence, it is possible that your members constitute, on the whole, a better class of lives, for the same character which

Rate of Mo: ■tality per Cent. ier Annum. Weeks of Sickness-claim per Me: iber per Annum. Ages. Odd Fellows, 1866-70. Foresters, 1871-75. Reohabites, 1878-87. Odd Fellows, 1866-70. Foresters, 1871-75. Reohabites, 1878-87. Under 25 25 — 35 — 45 — 55 — 65 — 0-632 0-788 1-094 1-647 2-877 5-911 0-753 0-807 1-174 1-802 3-286 6-510 0-603 0-509 0-619 1-119 2-325 5-815 0-742 0-862 1-145 1-905 3-063 8-539 0-831 0-910 1-245 1-914 3-732 9-215 1-057 1-019 1-349 1-866 2-915 8-463