under 70; 101 were 70 or under 75 years of age; 70 were between 75 and 85, and 3 were above the latter age.

The following table shows the distribution of deceased members of the Steam Engine Makers' Society according to the length of time elapsing between their superannuation and their decease:

DECEASED MEMBERS OF THE STEAM ENGINE MAKERS' SOCIETY WHO RECEIVED SUPERANNUATION BENEFITS, BY LENGTH OF TIME SUPERANNUATED, 1856 TO

1902.

[Source: The Seventy-ninth Annual Report of the Steam Engine Makers' Society, 1903.]

The next table shows the record of membership, deaths, and superannuation of the Typographical Association for the period 1881 to

1903.

STATISTICS OF MORTALITY AND SUPERANNUATION OF MEMBERS OF THE TYPOGRAPHICAL ASSOCIATION, 1881 TO 1903.

[Source: Typographical Association, 109th Half Yearly Report, 1904.]

a Calculated half yearly and the results added to obtain rate for year. Not including special grade members. The average age on joining of the 81 special grade members who have been placed on the fund since 1891 was 55.12 and the average age at death of the 35 who died prior to December 26, 1903, was 57.40.

These data are of course only suggestive, but coming down to recent years they indicate the present trend of rates of death and superannuation and the corresponding costs. The actuarial facts affecting the operations of trade unions were presented in extensive reports made in 1867 to 1869 by a parliamentary committee appointed to inquire into the organization and rules of trade unions in the United Kingdom. From what was said in connection with friendly societies, and the showing of the tables just given, as well as from the recent interest in and growth of the movement for providing superannuation, it is clear that data reflecting the conditions of forty years ago can be of only secondary importance at this time. While the fact involves an element of unsteadiness, it is also true that the power of the trade union to modify its system, or such details as may require change, with greater facility than is the case with institutions of a more stable form, makes for an adaptability that serves to enable them to avoid actual insolvency or disruption by effecting the changes shown to be necessary by current conditions. Another advantage possessed by trade unions is that the risks are composed of a class of men of more than average standing physically. As a writer on the subject has said: "The unions offer their advantages not indiscriminately, but to every young, healthy, skilled workman of good character; his election is not a matter of course." It is of especial importance to unions paying liberal benefits that they should restrict admission on the basis of both age and health. Thus the rules of the Steam Engine Makers' Society provide that "no person shall be admitted a member of the society who is over 35 years of age, unless he has been a member of the society previously for at least seven years consecutively, in which case such candidate can be admitted up to the age of 40 years. All candidates joining the organization at an age above 30 are obliged to provide a certificate of age at their own cost, or some satisfactory proof." In this union the initiation fee increases rapidly with the age of the member, rising from 8s. ($1.95) at 21 years of age to 80s. ($19.47), or ten times as much, at 40 years of age. The dues are also higher for members admitted above the age of 30 years, being 1s. (24 cents) per year for each year of age above 30 in addition to the ordinary dues of 1s. (24 cents) per week. Thus a member entering at 35 years of age pays 5s. ($1.22) per annum in addition to the 52s. ($12.65) ordinary dues. Candidates, however, may pay the extra dues in a lump sum of 10s. ($2.43) for each year they exceed the age of 30, if they prefer.

The date of organization of the majority of existing unions is so recent that a standard of superannuation has not yet been attained, the more so as unionism shows as yet a constant and even a rapid growth. As offset to this is the fact that the growth and expense of superannuation benefits are likewise increasing at a rapid rate, so

that the question of a balance of income and expenditure is as yet. an open one. Since, however, there is no legal, binding force in the agreement of a trade union to insure members at a rate named, it is not bankruptcy in the eyes of the law for a union to either reduce or even repudiate entirely the payments on account of superannuation, death, or accident benefits. The agreement to pay benefits at all is at best a conditional one, and the refusal of young men to pay increased levies for the benefit of older members is obviously a matter that the union can not penalize beyond the suspension of such recalcitrants, which can not be regarded as a remedy, since it leaves the older members, for whose approaching superannuation provision must be made, the sole contributors to the funds from which they hope to derive the support of their declining years.

The process of what may be called an actuarial education is a slow one, and the efforts of trade unions, as of friendly societies, to get on a correct basis have resulted in secessions and the formation of new societies which, on account of insufficient charges, must of necessity repeat the experience of readjustment within a few years or go to pieces. A certain percentage of the officers and members are impressed with the importance of a safe working margin in the form of accumulated funds, not only because it affords the actual protection needed, but because it unifies the men, making them less likely to abandon the union in some temporary stress or dissatisfaction, thus giving the union a degree of stability and capacity to meet the employer in questions involving the employees' interests that is regarded as making the occasions of dispute much less frequent. On the other hand are those members who, in their anxiety for cheap insurance, insist on keeping the contributions at a low rate, hoping that by some good fortune they will be able to realize the estimated benefits, even though it is most unlikely from the standpoint of the actuary.

In spite of all this, however, and of the actual scaling down of benefits or the increase of the levy, there is still the argument in favor of the union that its chief purpose is not insurance, but the improvement of trade conditions; and if the union is maintained and holds together in times of trade dispute, it may often be felt that it has justified its existence and that its members have reaped greater actual benefit, personally, as well as for their fellow-craftsmen, than if insurance alone had been sought and secured in a form apart from the trade union.

FEMALE MEMBERS.

The employment of women, especially numerous in the textile industries, raises the question as to their connection with trade unions, the fact being speedily disclosed that they form but a small percentage of the total membership of such organizations. So far as the

subject of benefits is concerned, they are not even affected in proportion to their actual numbers, since they are for the most part members of unions that pay small benefits. The table given below shows the number of females found in trade unions for each of the ten years 1898 to 1907, by groups of trades:

NUMBER OF FEMALE MEMBERS OF TRADE UNIONS IN THE UNITED KINGDOM, BY GROUPS OF TRADES, 1898 TO 1907.

[Source: Board of Trade Report on Trade Unions, 1905-1907.]

Total..

117,184 122, 096 124, 082 121, 119 123, 419 120,615 126, 342 137,082 163,887 201,709

The number of trade unions having female members in 1907 was 182, as against 149 in 1896-the first year for which data as to female membership are available—and 154 in 1897. This latter number was not again equaled until 1905, though there has been an increase in both numbers and membership each year since 1903. The total female membership in 1907 was 8.4 per cent of the total membership of trade unions, as compared with 6.7 per cent in 1904 and 6.9 per cent in 1898. The following table shows more in detail the number and distribution of female members of trade unions in 1907 and the percentage of female trade-unionists found in each trade or group:

NUMBER AND PER CENT OF FEMALE MEMBERS OF TRADE UNIONS IN GREAT BRITAIN, BY TRADE GROUPS, 1907.

[Source: Board of Trade Report on Trade Unions, 1909.]

a In unions which admit both male and female members the exact numbers of each sex are not always known, but the numbers stated in the table are approximately correct.

In some cases the number of females is an estimate rather than an exact report, but enough is positive to show that females form but a small part of the membership in most instances, though in the unions representing the textile trades, in which more than 85 per cent of all female trade-unionists are found, nearly one-half of the members are females.

STATISTICS OF ACCIDENTS.

A body of accident statistics is accumulating as a result of the laws relating to industrial inspection and of various laws requiring accidents to be reported, particularly the notice of accidents act, 1894. The last-named law requires employers in certain industries to notify the board of trade of any accident which causes to any person employed in the industry either loss of life or bodily injury, preventing him on any one of the three working days next after the occurrence of the accident from being employed for five hours at his ordinary work. These reports come short of furnishing adequate data from an insurance standpoint by failing to require a report of the period of resultant disability. This leaves the experience of friendly societies and compensation schemes as the more valuable material on which to base insurance estimates. However, these facts do not eliminate pure accident data from consideration in a study of the question of providing protection for workmen from the consequences of their employment, though the value of the data now in hand is lessened by the fact that the standards of reporting nonfatal accidents, as fixed by the various acts, have been changed at comparatively recent dates, and further by the fact that there is said to be an increasing tendency to observe the provisions of the factory and workshops acts in the matter of making reports.

The labor department of the board of trade furnishes a table showing the number of accidents reported in the United Kingdom under the various acts for the years 1898 to 1907, distinguishing fatal from nonfatal accidents. The latter are not comparable industry by industry, owing to differences in the laws governing their report. The better observance of the laws in recent years is said to explain in part the marked increase in the number of accidents reported under the factory acts. The table follows.