track No. 3 to track No. 5, a commercial track used to allow cars to be placed on the unloading tracks at Westport, which are located on the north side of the main track.

Two of the crossovers are each 191 feet long and the remaining one is 192 feet long, all three being crowded into the restricted distance of 705 feet between the interlocking tower at Westport station and the facing point switch leading from track No. 3 to track No. 5. It does not appear that any physical or operating reason exists which makes necessary the use of such short crossovers at this place. The facing point switch leading from track No. 3 to track No. 5, which limits the extension westward of the crossover leading from track No. 1 to track No. 3, should be removed, independent of any consideration relating to the crossover. Facing point switches should not be used on high-speed tracks where it is possible to avoid them, as they create an element of danger that is eliminated by the use of trailing switches. Track No. 5 should be entered through a trailing switch at some point farther west. There is sufficient tangent between the drawbridge and the point of the curve to make it entirely feasible to replace the No. 10 crossovers leading from tracks Nos. 1 to 3 and 2 to 4 with No. 20 crossovers. There is no danger in the use of a No. 10 crossover between tracks Nos. 1 and 2, as it is used only for back-up movements when it becomes necessary to run trains against the current of traffic. Had No. 20 crossovers been used at this place, being 352.7 feet long and only producing a 1° 42' curve as shown by the New York, New Haven & Hartford standard dimensions of turnouts, it is probable that the derailment of train second No. 53 on October 3 would not have occurred, even though the signal indications were disregarded and the train entered the crossover at unsafe speed.

Time card rule No. 2 of the New York, New Haven & Hartford Railroad reads as follows:

Unless otherwise restricted a speed of 15 miles per hour must not be exceeded over routes governed by dwarf signals.

The signal indications governing this crossover movement were clear, and afforded sufficient protection had they been observed. It is plain, therefore, that the direct cause of this accident was a disregard by the engineman of train second No. 53 of signals and rules provided by the raliroad company to prevent the occurrence of such accidents.

Engineman Clark of second No. 53 was 55 years of age and was the assigned engineman for this run. He entered the service of the New York, New Haven & Hartford Railroad as a fireman in 1883;

he was promoted to be an engineman in 1888 and was dismissed December 20 of that year for responsibility for a collision. He re-entered the service as a fireman in 1893 and was again promoted to be an engineman in 1894, and had been so employed since that date. He was disciplined January 14, 1898, for passing a home signal at danger at New Rochelle, and again July 7, 1905, for passing a home signal at danger at Naugatuck Junction. His habits were good and he had been on duty 5 hours and 29 minutes when the accident occurred, after a rest period of 7 hours and 35 minutes, previous to which he had worked 11 hours and 16 minutes in the 24-hour period from 3:40 a. m., October 1.

In all essential particulars this accident was a duplicate of the accident of the Federal Express on the New York, New Haven & Hartford Railroad on July 11, 1911, which resulted in the death of 14 and the injury of 54 persons. As a result of our investigation of the Bridgeport accident the following recommendations were made:

(1) That in all situations where accidents are likely to occur through the non-observance by enginemen of signals

or

rules calculated to insure safety, automatic train control apparatus should be provided to insure that trains will be brought to a stop in case the signals or rules are not properly observed.

(2) That in the absence of such automatic control apparatus, on tracks where high speed trains are run, switches should not be set to divert a high speed train from one track to another at a crossover which is not safe for high speed until after the train has been brought to a stop.

(3) That at all interlocking plants where trains are operated at high speed over facing switches, approach locking should be provided to prevent the switches being changed from the main or through route to a diverging route after a train has received the signals for the through route.

At the joint hearing in New York City, the vice-president in charge of operation of the New York, New Haven & Hartford Railroad, when his attention was directed to the above recommendations, stated that the only action taken by his company to prevent the recurrence of accidents similar to the Bridgeport accident was "a stiffening up of discipline. No new devices have been installed, because, as he expressed it, "We have not been able to find any that we are satisfied would help." When asked if lengthening the crossovers would not materially decrease the element of danger, this official answered: "No. I would think in the end it would make a worse situa

tion on a railroad than a better one," his reason for this belief being that "It is simply putting a premium on the violation of a rule."

It is hard to understand how the conclusion can be reached, under any contingencies arising in the operation of a railroad, that there would not be greater chance for an accident with a crossover only 190 feet in length, producing a curvature of 6° 48′, than with one 352.7 feet in length, as a No. 20 would be, with a curvature of 1° 42'.

It was also stated by this official that in his opinion a regulation requiring trains to be brought to a stop before switches are set for a crossover movement would not increase safety, his reason for this belief being that when signals were disobeyed, as was the case in both the Bridgeport and Westport accidents, the train would be sent along the straight track with a probability of running into the rear of a train in the block beyond the crossover. Thus, as he expressed it, "you change one accident for another, is all.' The evidence brought out in the investigation of this particular accident, however, shows that had the crossover switch not been set for the diverging route the accident would not have occurred, as Engineman Clark applied the emergency brakes immediately after his engine passed the stop signal, and had his train continued on track No. 1, it would have been brought to a stop about two train lengths beyond the switch.

With regard to the use of automatic stops, the vice-president stated that there was no mechanical appliance that he knew of which he thought would meet the conditions. His attention was called to the following from the report of the Commission's block signal and train control board:

The information obtained from tests, together with knowledge of the general state of development of the art of automatic train control, leads the board to conclude that there are several types of apparatus and methods of application which, if put into use by the railroads, would quickly develop to a degree of efficiency adequate to meet all reasonable demands. Such devices properly installed and maintained would add materially to safety in the operation of trains. In many situations under conditions existing in this country the board is convinced that the use of automatic train stops is necessary to the safe operation of trains.

The board does not wish to be understood as stating that the conditions of entirely acceptable automatic train control, as formulated by it in the ten characteristics published in its report of last year, are fully met by any one of the devices it has thus far examined. On the contrary, the art of automatic train control is still largely in the experimental stage; but it is far enough advanced

to warrant the installation of available devices with a view to their further development to meet the demands of safety in train operation.

Further, the board has no hesitancy in saying that had the railroads directed the same effort toward the development of automatic train-control apparatus that has been devoted to the development of interlocking and block-signaling apparatus we should now have adequate installations of automatic train-control devices which would permit an engineman to handle his train without interference as long as he did it properly, but would intervene to stop his train if he disregarded a stop signal or ran at excessive speed where speed restriction was prescribed.

The railroads have been decidedly lax in developing the automatic stop, and progress has been so slight in this direction that the actual experience which is necessary for the formation of proper legislative judgment is lacking. The board, therefore, does not believe that at the present time legislative compulsion to this end would be wise. It does believe, however, that the railroads should be urged and expected to develop the art of automatic train control so as to provide devices which will meet their operating conditions. This appears to the board to be entirely practicable, and should it not be done with a reasonable degree of expedition steps should be taken by the government to stimulate such action.

When asked what had been done by his company toward the actual development of automatic stops in view of this recommendation, the vice-president in charge of operations said that no definite steps had been taken other than to investigate what was available, and no specific conclusions had been reached as a result of such investigations.

In this connection attention is called to the following extract from the last annual report of the Commission.

While the Commission is not prepared at this time to recommend legislation compelling the adoption of automatic train control upon all lines of railroads, it strongly recommends to the railroads themselves the experimental use of such systems as are now available intended to secure efficient automatic train control.

As previously noted, the roadway and track conditions in the vicinity of the acdident are good, and the construction of the crossover was substantial. All renewals are made with creosoted ties, and on these ties tie plates are used, as well as screw spikes. In some places the chestnut ties are badly worn, due to the base of the rail cutting into them, in some instances to a depth of threefourths of an inch, but a sufficent number of tie-plated creosoted ties are in use to maintain safe track. At the crossover practically all the ties were new and screw spikes were used, and the track conditions at this point were good.

From the evidence taken in this investigation it appeared that the speed limit of 30 miles per hour over the drawbridge

He

at Westport is habitually disregarded. The operator at this bridge testified that the average speed of trains passing his tower (which is located on the bridge) was from 45 to 55 miles per hour when signals were set for a clear track. had never reported a train for exceeding the speed limit over the bridge, and had never been instructed to do so. VicePresident Horn testified that he had no personal knowledge of a violation of this speed limit, but considered it safe to cross the bridge at higher speed than allowed by the rule. He said that he I could never in his own mind determine why a speed of 50 miles per hour over this bridge would not be as safe as the limit fixed by the rule, and in reply to a question as to how he distinguished between the violation of a rule of 30 miles per hour across the bridge and 15 miles per hour on a No. 10 crossover, so far as the safety of passengers or property is concerned, he replied: "One is safe and the other is suicide."

The investigation further developed that there were no speed restrictions governing passenger trains except at specified points which are covered by signals and rules, and passenger trains are expected to make up time when late. Bulletin order No. 291, issued by Superintendent Woodward on September 23, 1912, reads as follows:

BULLETIN NOTICE NO. 291.

Our passenger service has not been running very satisfactorily since the heavy business set in on the 1st of September, although the travel has very much decreased, and there is considerable fault being found on account of so many trains being late.

Every effort should be made to keep the trains on time and conductors are requested to avoid loss of time at stations loading and unloading passengers and baggage, so far as possible, and engineers are requested to do everything they can to make time, except that it is not desired to in any way run trains beyond a safe speed, neither do we wish to have the orders on slow-downs disregarded.

Be very careful when you have delays to make correct reports, and conductors must confer with engineers to make sure that the reports are entirely accurate.

The desire of the company to maintain schedules is shown by the fact that engineers frequently receive letters requesting explanations regarding delays, and are required to explain a loss of one minute in making the run of 40 miles between Stamford and New Haven, as shown by a letter to Engineman Moore, of which the following is a copy:

NEW HAVEN, CONN., Sept. 26, 1912. Mr. C. Moore, Engineer, Springfield, Mass. DEAR SIR: On Sept. 21, train 26 is reported as losing one minute on run engine 1018, Stamford to

(Copy to Mr. F. W. Stanley, foreman.)

I want you to see to it that engineer Moore advises at once in regard to this delay.

When questioned concerning this letter Engineman Moore testified that he did not regard it as a criticism but simply as an effort on the part of the officials to find out what was the cause of delays and to eliminate them as far as possible. While Engineman Moore's view of the purport of such letters is probably correct, it is none the less true that many enginemen would not so regard it. To be called upon for an explanation of a delay of one minute, taken in connection with Bulletin Order 291, would undoubtedly lead many men to give undue prominence to the question of maintaining the schedule, and cause them to take chances in the effort to bring their trains in on time.

Excessive speed as a factor in train accidents deserves serious consideration. The remarkable increase in speed and weight of trains within recent years, and the crowding of tracks and terminals caused by movement of the enormously enlarged volume of railroad traffic, have greatly increased the duties of employees and multiplied the chances of error on their part.

In a recent accident investigation made by our chief inspector of safety appliances it appeared that the general inspector of transportation of one of our largest railroad systems said: "Excessive speed is the cause of 75 to 80 per cent of the catastrophies in the last few years. When asked who was to blame for this, this official said: "The public, in my opinion, because they ride on the railroad that has the fastest train and the railroad that doesn't make that fast time is not patronized. The railroads are forced to do it. They don't want to do it.

We can but repeat what was said in the published report of this accident: "It is true that the public patronizes the railroad that makes the fastest time. But did the public have knowledge that any railroad was operating its trains at such high speed as to make travel upon that road unsafe its patronage would quickly be withdrawn. The great majority of people who ride on railroads desire to travel quickly if it can be done with safety, but their first consideration is safety. They rely upon the judgment of railroad managers, knowing that they are the only ones who have full

knowledge of conditions existing upon their roads, and the assumption always is that the road operating high-speed trains will bring them safely, as well as quickly, to their journey's end."

The cars in train second No. 53 were of wooden construction, and the forward end of the first parlor car was lying against and across the firebox of the overturned engine and immediately caught fire. Three parlor cars were burned. It is obvious that had these cars been of modern steel construction they would not have been destroyed by fire, and it might have been possible to have saved the lives of some of the injured passengers who were caught in the wreckage and cremated. The circumstances surrounding this accident once more demonstrate the inherent weakness of wooden cars and their unsuitability for service in modern high-speed trains, and emphasize the necessity of compliance with the Commission's recommendation in its last annual report relative to the adoption of steel cars.

The volume of passenger and freight traffic on the New York, New Haven & Hartford Railroad is heavy, and trains frequently are required to cross from one track to another. To provide for such movements between New York and Boston, a large number of crossovers are in service. The longest of such crossovers are No. 15; the majority of the crossovers used are No. 10, and some of them are No. 8.

The lowest numbered crossover that can reasonably be expected to provide for the safe passage of a train from one track to another at normal speed is believed to be No. 18 or No. 20. With 12foot track centers and 100-pound_rails, according to "N. Y., N. H. & H. R. R. Standard Dimensions of Turnouts," the radius of curved rail for a No. 18 crossover is 2,804.60 feet, producing a curve of 2° 2' 34"; the radius of curved rail for a No. 20 crossover is 3,369.90 feet, producing a curve of 1°42'. As given on this blueprint, the length of a No. 18 crossover is 306.36 feet, and the length of a No. 20 crossover is 352.70 feet.

Proper measures for the safety of the traveling public have not been provided where crossovers are installed which may be used with safety only at low speed and which are protected simply by signals and rules requiring reduced speed. This fact is emphasized by disasters which have occurred as a result of the disregard of such signals and rules. Longer crossovers than those used by the New York, New Haven & Hartford Railroad should be installed at all points where high-speed passenger trains are to be diverted from one track to another track on which the current of

traffic is in the same direction; then if signals or rules were disobeyed, or if an engineman was incapacitated, the possibility of disaster would be greatly reduced if not entirely eliminated. Until these long crossovers are installed, to provide adequate precautions for the safe movement of trains at any crossover shorter than a No. 20, the switches should be left for a straight route, stop signals should be displayed, and the crossover switches should not be set for the diverging movement until after the train which is to make the movement has come to a stop.

The position was taken by the executive officers of this road that no finding or recommendation with regard to the Bridgeport disaster was ever made by the Interstate Commerce Commission.

It is apparent that this was but a resort to a technicality to evade the point at issue. The records of the Commission show that a report upon the Bridgeport accident, including the recommendations previously quoted herein, was made public by the Commission on August 16, 1911. A copy of this report was mailed to VicePresident Horn, at his request by wire, on August 18, 1911, and his acknowledgment of its receipt dated August 21, 1911, was received at the office of the Commission on August 22. Had the recommendations contained therein been complied with, the Westport accident undoubtedly would not have occurred, and it is therefore apparent that the officials of the New York, New Haven & Hartford Railroad did not take adequate measures to prevent a recurrence of the Bridgeport accident, in view of those recommendations, which they were well informed.

of

As a result of the investigation of the Westport accident, President Mellen, of the New York, New Haven & Hartford Railroad Company, has now wired the Commission that on the 26th day of November instructions were issued to the operating and engineering departments of this road ordering that wherever practicable No. 20 crossovers be used between all parallel main tracks used for trains going in the same direction and through which trains are detoured; and that all trains having a schedule speed of 15 miles an hour, or over, on straight tracks be required to come to a full stop before making a crossover movement. These instructions are not limited to the crossover at Westport, but apply to the whole New Haven system wherever four tracks are used.

It is a deplorable fact, as shown by the following table from the accident statistics published by the Commission, that train accidents have steadily increased in number and magnitude since 1909;

Railroad managers and their employees, as well as State and Federal commissions, must unite in a determined effort to reduce these harrowing railroad disasters to the limits of the unavoidable. This can only be done by ascertaining the real causes of train accidents and taking proper measures to eliminate them.

This Commission is making every effort possible in this direction and has been very ably assisted in the present investigation by the public utilities commission of Connecticut-in a number of other instances by the commissions of other States as contemplated by the accident investigation law of 1910. Railway managers and their employees seem

to have awakened to their responsibilities in this matter. This is shown by the general formation of safety committees, and some railroads have shown a reduction in their casualty list that is worthy of commendation.

But notwithstanding these facts avoidable train accidents continue to occur and exact an increased toll of life and limb. Such accidents occur on large railroads as frequently as on small ones; on wellmanaged roads as well as on roads that are less well managed; on block-signaled roads as well as on roads where the trainorder system is still in force; and by far the greater number of them are due to the same fundamental cause, namely,