who were not insured is much smaller is explained by the large number of fatal accidents chargeable to steam railroads which are not insured under the act. (Tables I. and II.)

It should be stated at this point that here as well as throughout the chapter the figures covering non-fatal accidents are the ones upon which to establish standards, or the prevailing tendencies related to industrial accidents. Fatal accident figures are more likely to show fluctuations, because of the smaller number of cases involved, and because of the fact that this type of accident runs to the extreme limits, and is affected by many variable factors.

The next point to be considered is the number of non-fatal accidents that happened in certain groups of industries. The division into one hundred and nineteen distinct trades or occupations, shown by Table III., has been condensed here into twenty-four main groups of industries.

The frequency distribution of accidents in these industries is shown graphically by means of the figure in Diagram 1. On the basis of the classification employed, there appears a range of accident occurrence from 10 to 13,568.

The ten industries showing the largest number of accidents are iron and steel, 13,568; textiles, 13,562; road, street and bridge transportation, 11,398; fifteen other industries (broom and brush makers, charcoal and coke burners, cigars, electric light and power companies, electrical supplies, gas works, ore works, rubber factories, straw workers, tobacco, turpentine distillers, gas and electric companies, "not specified" manufacturing and mechanical industries, and other miscellaneous industries and occupations), 10,698; trade, 8,107; building trades, 7,310; leather and its finished products, 5,713; metal and metal products, 4,299; lumber and its manufacture, 2,727; and food and kindred products, 2,198.

The incidence of accidents shown by these figures should be interpreted largely as so many separate facts, based on the experience in Massachusetts. In comparing these figures with those of other States due weight should be given to whatever differences may exist in methods of collection and classification of accident data. Also, in comparing the accident frequency in one industry with that of another, allowance should be made

for variable conditions based on the grade of labor and the number of employees.

In a large way the accident incidence, as shown above, is an indication of the general hazard of the industry. With these figures serving as an index, the work of finer analysis of conditions should be greatly facilitated. (Table III.)

The distribution of fatal accidents by industries is summarized in the following table:

From these figures it will be seen that, although the ten industries showing the largest number of fatalities are practically the same as in non-fatal accidents, the order of incidence has shifted. Road, street and bridge transportation, and the building trades had the largest number of fatal accidents. Together these two groups of industries are to be charged with more than 50 per cent. of all the fatalities. (Table IV.)

If the figures were available, it would be interesting and valuable to show for each industry the relation between the number of employees and the number of accidents. It is possible, however, to get accurate figures only in certain classes of industry. Table V. in the appendix to the report gives a cross-section view of this relation for twenty-five selected branches of industry in Massachusetts. The table shown below makes a summary of this study.

Number of Accidents per 1,000 Employees for Twenty-five selected Branches of Industry.

General average for group, 91 accidents per 1,000 employees.

15

Following the analysis of accidents by industries comes the study of the causes of accidents. On the basis of the reports of accidents there have been tabulated one hundred and twentytwo primary causes. The figures for non-fatal accidents have been summarized as shown in Diagram 2, under twenty-eight

main classifications. Twenty-nine thousand seven hundred and thirty-seven non-fatal accidents, or 33 per cent. of the total number, were caused by hand labor; 11,375, or 12 per cent., were caused by machinery peculiar to special industries; 8,417, or 9 per cent., were the result of falls of various kinds; 4,331, or 4 per cent., were eye injuries; 104, or 10 of 1 per cent., were occupational diseases.

The large number of accidents caused by hand labor, almost one-third of all the causes combined, is obviously a type of injury difficult to prevent.

By regrouping the figures into accidents caused by hand labor, as before, machinery accidents, and all others, it is found that 39 per cent. were due to miscellaneous causes, 33 per cent. were the result of hand-labor accidents, and 28 per cent. were directly related to machinery. (Table III.)

The following table shows the causes of fatal accidents arranged in the order of their magnitude: