CAUSE AND PREVENTION.
In the Accident Bulletin for June, 1910, pages 10 and 11, there are given detailed statistics of twenty-six “prominent train accidents” with the causes of each. They embrace thirteen collisions and thirteen derailments, resulting in sixty-two killed, 306 injured, and a property loss of $261,584. The causes assigned may be grouped under fifteen heads, as follows:
Excessive speed, 5; ran by meeting point, 2; failed to flag, 5; disobeying orders, 1; misunderstanding orders, 1; failure to receive orders, 1; conflicting orders, 1; signal light out and engineman failed to stop, 1; broken rail, 2; explosion of boiler, 1; spreading of rails, 1; washout, 1; trestle failed, 1; insufficient ballast, 1; defective temporary junction of new and old rails, 1. Total, 26.
These fifteen assigned causes may be summarized thus:
Failure of persons, 18; failure of boiler, 1; failure of track and structures, 7. Total, 26.
Of the seven failures of track and structures, the two cases of “broken rails” and one “washout” may be considered unavoidable. The remaining four cases in that group, viz., “spreading of rails,” “trestle failed,” “insufficient ballast” and “defective temporary junction of old and new rails” were preventable, and could have occurred only from neglect of those charged with their care and maintenance.
The one case of “explosion of boiler” may have been due to defective material, or to negligence of the engineman.
We find, therefore, that in this group of accidents, twenty-two were preventable, three unavoidable and one doubtful.
Of the unavoidable, the “washout” may be dismissed as being beyond the control of human agencies, but the “broken rail” calls for further consideration.
Rail failures are generally due to chemical or physical defects, not entirely under control of the manufacturer, and not discoverable by inspection of the finished rails. Under the present practice the manufacture of rails is watched at the mill by the railway company’s inspectors. Specimens from each heat or melt are tested under a weight of 2,000 pounds falling fifteen feet to twenty feet. If the test piece breaks the steel is regarded as too brittle, and the rails from that heat are rejected. If it does not break, but the deflection exceeds the prescribed limit, the steel is too soft, and those rails are accepted as seconds, to be used only in yards and side tracks. All test pieces which do not break under the foregoing drop test are then broken and examined for internal defects. If defects are found, further tests are made, and the heat rejected in whole or in part, on the extent of unsoundness disclosed.