OUTSIDE CARS.

Outside car accidents caused 470 deaths in coal-mining operations in North America during the period under consideration, or 2.6 per cent of the fatalities due to all causes. In proportion to the number employed the standard or average fatality rate due to this cause was 0.86 per 10,000. Since practically all of these accidents occur to outside employees they should perhaps have been calculated with reference to outside employments only, but the necessary information as to the number of outside employees is not available for the North American coal fields as a whole. The statistics for Pennsylvania prove conclusively that most of the fatalities due to outside cars have occurred to outside employees. As brought out in the subsequent discussion of the fatal accident rate by occupations, the rate of outside car accidents to underground employees in Pennsylvania was 6.3 per 10,000 for the anthracite coal field and 3.9 per 10,000 for the bituminous coal field. In the anthracite coal fields 38.2 per cent of outside accidents were due to outside cars, and in the bituminous coal field 45.9 per cent.

The term "outside cars," as used in mining statistics, is, however, of a rather indefinite meaning, at least for some of the coal areas, and there are strong reasons to believe that in some of the States this class of accidents is combined with "mine car" accidents, which have been previously discussed. Most of the outside car accidents in the consolidated table have been reported from Pennsylvania and the eastcentral section, and it is, therefore, very doubtful whether the rates for other coal fields and coal-mining States can be accepted as accurate. For some of the States no outside accidents at all have been separately returned, although it is a practical certainty that such accidents have occurred. For this reason a discussion of the fatality rates by States would obviously be of no practical value. It may be emphasized here, however, that the risk factor due to outside cars is, undoubtedly, one which demands serious consideration.

MOTORS.

Motors caused a total of 30 fatal accidents in the coal fields of North America during the period under consideration, or 0.2 per cent of the casualties due to all causes. The standard or average fatality rate due to motors was 0.05 per 10,000 employed, the rate having been highest in the western section, where it reached 0.14 per 10,000, and zero in the west central section, for which no fatalities of this kind were officially reported. It is very doubtful whether all the fatal accidents due to motors used in mines have been properly classified as such, and it may reasonably be supposed, considering the rapid increase in the use of electric motor power in mining, that more deaths have occurred than have been officially reported, or properly

classified. Granting probable defects in the returns, it is surprising to find that the fatality rate due to motors should be as low as 0.06 per 10,000 in the eastern coal field, which comprehends the States of Maryland, Pennsylvania, and Ohio. In Illinois, for illustration, out of 1,391 fatal accidents only 4, or 0.08 per 10,000 employed, were officially ascribed to motors, while in Ohio 19 out of 1,027, or 0.47 per 10,000 employed, were attributed to this cause. No deaths due to motors were officially reported as having occurred in the coal mines of Pennsylvania, which, no doubt, is partly due to errors in classification. For these reasons, it would serve no practical purpose to discuss the fatality rates due to motors in the different coal areas. and coal mining States, but the self-evident defect in coal mining statistics clearly brings out the necessity of a uniform classification. of causes to be agreed upon by the mining officials of the different States.

EXPLOSIONS OF GAS OR DUST.

Statistical defects are inherent in all classification, but they can be materially reduced by a critical examination of individual returns. Of no group of causes is this more true than of gas and dust explosions, explosions of dynamite and powder, blasts, etc. There are practical difficulties to be overcome which at times will baffle even the most skilled, for cause and effect will often be confused, and it will occasionally occur that the true but insignificant cause of a coalmining accident is overshadowed by the disastrous effect. A can of powder explodes and results in a subsequent gas and dust explosion with a considerable loss of life. Should such an explosion be classified under explosives or explosions? Or, a naked light ignites fire damp and causes a small gas explosion, which results in the explosion of an underground powder magazine, with disastrous results to life. Should such an accident be classified as gas explosion, or as a powder explosion? A compromise is necessary in such cases, but it would be of great value if definite rules governed in the statistical classification. It is due to these differences in the determination of the true cause that so many apparent errors and contradictions occur in coal-mining accident statistics and without a definite understanding among the different mine officials, a decided improvement can not be expected for many years to come.

As classified in Table XXIII of the appendix, there have been 2,571 fatal accidents due to gas and dust explosions in the coal fields of North America during the period under consideration, or 14 per cent of the fatalities due to all causes. The standard or average rate of accident frequency was 4.71 per 10,000, against 4.04 for mine cars and 1.77 for deaths due to explosions of powder or dynamite. Gas and dust explosions, therefore, are among the most important causes of mining

fatalities and their geographical distribution is a matter of particular importance.

The variations in rates in the different coal fields are shown in the following table:

FATAL-ACCIDENT RATES IN COAL MINING IN NORTH AMERICA DUE TO GAS OR DUST EXPLOSION, FOR THE PERIOD 1899 TO 1908.

Rate per 10,000 employees.

The variation in rates ranges from a minimum of 1.25 per 10,000 for the east-central section to 17.17 for the southern section. The wide range is due primarily and largely to fundamental differences in the character and depth of the coal beds in the several coal areas, which are brought out in a striking manner by reference to the tables for the different coal-mining States. Gas and dust explosions have been combined, because of the fact that dust explosions per se are probably extremely rare in coal mines, while dust explosions caused by previous gas explosions are quite common, and particularly so in some of the States of the southern and Pacific coast fields. Coal dust, as a factor in mine explosions, has only been regarded as such within very recent years, after a number of particularly serious disasters, in which there could be no doubt that what had been a gas explosion in its inception had subsequently become an explosion of coal dust. There is probably no more serious question to-day in many of our mining States than the increasing danger of dust explosions, due possibly to the use of coal-mining machinery and to other causes which, as yet, are not fully understood. The subject of dust explosion is too complex to be discussed in this place,(") but it may be pointed out that measures of prevention in

17.17

11.85

6.22

4.71

3.48

2.73

1.92

1.25

a The most important recent contribution to the subject is Bulletin 425 of the United States Geological Survey, on "The Explosibility of Coal-dust," Washington, D. C., 1910. This bulletin includes a very complete bibliography. Reference should also be had to Bulletin 383 of the United States Geological Survey, on Notes on Explosive Mine Gases and Dust," by R. T. Chamberlin, 1909. Among other important recent contributions reference should be had to an article on "The Dust Problem in Coal Mines," by Joseph Virgin, Engineering and Mining Journal, October 9, 1909; Coal-dust as a Factor in Mine Explosions," by Henry M. Penn, Engineering and Mining Journal, July 4, 1908; "Equipment for the Prevention of Mine Explosions, with Special Reference to

be better understood when it is stated that heretofore the United States Geological Survey has given the details of causes of fatal accidents in only 3 specific groups, while in some of the tables in the appendix to this study the facts are given in detail in 21 groups. The practical value of the tabular analysis of coal-mining accidents, provided the facts by causes are sufficiently numerous, is, of course, in exact proportion to the detailed grouping of individual but welldefined specific causes responsible for coal-mining casualties. Much would be gained by uniformity in the method of tabulation by causes, but efforts in this direction should insist rather upon a comprehensive tabular analysis than upon condensation. (")

The importance of details is best illustrated by specific causes of modern significance, such as deaths due to mining machinery, electricity, boiler explosions, etc. (For further details of the causes of accidents by States, see Table XXIII of the appendix.)

THE FATAL-ACCIDENT RATE DUE TO PRINCIPAL CAUSES.

The rate of fatal accidents by principal causes determines with scientific accuracy the degree of risk exposure to particular hazard in mining experience. The following table is identical with the table on page 454, previously discussed, except that for each cause the rate. per 10,000 exposed to risk has been calculated, instead of the percentage distribution of causes, as in the former table. Since the number of some of the causes is small, it has seemed best to use 10,000 employees exposed to risk one year, instead of the usual basis of 1,000. The table is based upon a risk exposure of 5,459,436 mine workers for one year and 18,346 fatal accidents occurring in the coal fields of North America during the decade ending with 1908, or a part thereof, since the returns for some of the States are not complete. The rates for individual coal-mining States and the details for the different coal areas, by number of casualties and the rate per 10,000, will be found in Table XXIII of the appendix.

For suggestions for improving coal-mining accident statistics, see Engineering and Mining Journal, June 2, 1900, and subsequent issues. Among the more important works on the causes of coal mining accidents are Mining Accidents and their Prevention, by Sir Frederick Augustus Abel, New York, 1889; Explosions in Coal Mines, by W. N. and J. B. Atkinson, London, 1880; Essays on the Prevention of Explosions and Accidents in Coal Mines, by Creswick, Galloway, and Hopton, London, 1874; Elements of Mining and Quarrying, by C. Le Neve Foster, London, 1903; and Practical Coal Mining, by T. H. Cockin, New York,

FATAL-ACCIDENT RATE IN THE COAL MINES OF NORTH AMERICA DUring a TEN-YEAR PERIOD, BY CAUSES.

The average fatality rate for the North American coal field, according to this table, was 33.6 per 10,000, or 3.36 per 1,000 of persons employed one year.

FALLS OF COAL OR ROOF.

The average fatality rates by principal causes during the ten-year period 1899 to 1908, as given in the above table, may be considered the standard by which the relative frequency of accident occurrence in the different coal fields and coal-mining States can be measured. The standard rate of accident occurrence due to fall of coal and roof, slate, etc., combined was 15.67 per 10,000, but, as brought out by the following comparison, the degree of risk varies considerably in the different coal fields.

FATAL-ACCIDENT RATES IN COAL MINING DUE TO FALLS OF COAL, ROOF, SLATE, ETC., DURING A TEN-YEAR PERIOD.

This comparison is of very considerable practical significance. For the first time, the true rate of risk from a particular cause in mining