Bulletin Board


Main Page

"Wreck on the Low Grade"

(The Rest of the Story)

By Allen Brougham

-- Published in the January 2005 issue of the Bull Sheet --

Today, January 1, 2005, marks the 48th anniversary of a terrible accident that occurred on the B&O's Low Grade line (No. 4 track) running between Miller and West Cumbo, in the Eastern Panhandle of West Virginia. It was a head-on collision between two freight trains. It left three crew members dead, and six others injured.

I first presented a feature on this accident eight years ago. In that feature [Bull Sheet, January 1997], the report by the Interstate Commerce Commission - with its description, discussion and cause of the accident - was printed verbatim. The blame, regrettably, was upon two operators at Miller Tower for their failure to properly protect the move of a train running against the current of traffic, and by the train dispatcher for overlooking a train order permitting the train to make the affected move.

The report quoted in detail the rules that governed the movement involved, along with a description of mistakes that allowed a train to occupy the track in the opposing direction, resulting in a collision.

The Low Grade line was - and still is - an 11-mile portion of single track taking a circuitous route away from the busier and faster double-track main line in its eastward ascent of what is known as North Mountain. The Low Grade line is about two miles longer than the double-track route, with reduced speed due to its snaking right of way, but with its gentler climb the Low Grade line has typically been the preferred route for heavy trains such as those carrying coal.

Miller Tower was located at the west end of the Low Grade line, at the bottom of the hill, along the bank of the Potomac River. To the west of Miller were three main tracks that assumed a generally riverbank-level route from there to Cumberland, Maryland. To the east of Miller were the two main tracks (numbered 1 and 2 respectively) that ascended North Mountain using the shorter but steeper alignment toward West Cumbo, and the Low Grade line (referred to as No. 4 track) toward the same location. Also to the east of Miller was a joint B&O/Western Maryland yard at Cherry Run, and a connecting track that crossed the Potomac River into Maryland where it joined the main line of the Western Maryland Railway at a place called Big Pool.

The Low Grade line (originally known as the Cherry Run & Potomac Valley Railroad) was completed in 1903, some 61 years after the B&O was opened through the area. Because of its easier ascent for eastward tonnage trains, the Low Grade line was nominally considered an eastbound track, and accordingly was signaled for its eastward direction only. If the occasion developed for a westbound train to use the Low Grade line, rules for Manual Block operation were implemented, and a 'right-of-track' train order was issued. To protect the move at the exiting end for the westbound train, at Miller Tower, the order was first completed to the operator at that location using a 'Form 31,' and the operator was required to block signals and switches to prevent eastward entry onto the track, and to display a red train order signal that could be seen by trains moving in an eastwardly direction.

A red train order signal would require a train to come to a stop, receive, read and understand a train order before proceeding. From a practical standpoint, an eastward train, if routed to No. 4 track, should not even be allowed to advance to the order board in the first place. The interlocking and train order signals worked independently of each other, and if a train were advanced nevertheless, the red order board and the text of the order would prevent further entry.

With protection thus accorded, the train order would get completed at the entry end for the westbound train, at West Cumbo Tower, and delivered along with a clearance card Form A to the crew of the westbound train authorizing the train to proceed.

This was the situation that fateful morning as a 'Extra 6498 West' was given right-of-track on No. 4 track from West Cumbo to Miller.

The report by the Interstate Commerce Commission described the required aspect of the eastward train order signal at Miller Tower as follows:

The report further quoted the train order that had been issued for the affected move, adding that it was repeated and made complete to both offices (Miller and West Cumbo) at 12:01 A.M.:

The Form A reading in part, "Manual Block is clear to Miller. Proceed," was then completed and delivered to the crew of Extra 6498 West at the entry end of the track, West Cumbo.

That train - Extra 6498 West - was, according to the report, originating at that station. There was a yard about a mile west of the tower from which the train was using No. 4 track to double its consist prior to departure thereon.

But in the meantime, the operator at the exiting end of No. 4 track, at Miller Tower, did not block the signals and switches to prevent eastward entry onto the track, and did not display the required red train order signal.

It should be noted, too, that two operators at Miller Tower were involved - not just one. Midnight - by coincidence of timing - was relieving time, and it was the second-shift operator who copied the order while the third-shift operator was entering the office to make his relief. Here is what the ICC report said about it following its post-accident interview with the second-shift operator:

Here, too, is what the ICC report said about it following its post-accident interview with the third-shift operator:

It might well be assumed that the third-shift operator at Miller Tower was relying solely on his ability to 'remember' that he was to protect this movement against the current of traffic, and/or that he felt comfort in the ability of his train dispatcher to maintain the same form of vigilance - and presumably the train dispatcher may have shared the same confidence in his operator. However this may have been, proper protection was not provided, and the presence of a westbound train within that block section known as the Low Grade line was overlooked.

Adding to the situation of having to 'remember' that there was a train needing protection by the operator at Miller Tower was a delay in that train's departure from West Cumbo. More than an hour after the train order was completed, the train affected by that order was still there, at West Cumbo, with air trouble. This delay was crucial, as in the interim the operator at Miller Tower became distracted by the failure of the building's furnace. Remember, this was January, and in an old building with many windows, in the middle of the night, the lack of heat could surely occupy an occupant's mind in a hurry.

Such as it was when, about 1:25 in the morning, an eastbound coal train - Extra 245 East - entered the approach circuit of Miller Tower. According the the Interstate Commerce Commission report, as quoted here:

The eastbound train, with no protection in the aspect of a red train order board to prevent entry, passed Miller Tower onto No. 4 track at 1:33 A.M.

The westbound train, with its crew having no knowledge that an eastbound train was coming, left from West Cumbo on the same track just two minutes later.

It should be remembered that radios were not then in use as a means to communicate with trains. Once the error was discovered, there was no immediate way to alert the crews involved.

The eastward train might have gotten at least some protection by the intermediate wayside signals it passed along No. 4 track had it not been for the location of the westward train at the moment in which the eastward train passed its last signal prior to the collision. In this instance, the last signal it passed informed the crew that there were no trains between there and the next signal it would encounter. But after the eastward train passed that signal, the westward train (which had no signals at all) entered that particular section of track. These intermediate signals functioned automatically; the operators could not change their indications from the towers.

Hurried attempts were made to contact employees living near the Low Grade line - asking them to rush over to the track and flag to a stop any train they should see. But it was too late...

According to the ICC report, the trains collided at 1:42 A.M.

If there could be any optimism that the crews of the trains involved might have been able to see each other in time to avoid colliding, the point of the accident was undeniably the worst place along the line for them to meet. It was within a sharp curve in a cut. Indeed, the headlight beams of both trains shone upon the same embankment as they approached.

From the report of crew members of the westbound train:

From the report of surviving crew members of the eastbound train, who were riding in the caboose:

All of the fatalities were to crew members on the engine of the eastbound train.

According to newspaper accounts of the accident, those killed were B.F. Phillips, engineer; L.R. Holler, fireman; and J.C. Beard, head brakeman. The grizzly task of recovering the last of the bodies took 12 hours.

Injured were James Muir, James Duvall, Lawrence Lowrey, Perry Foltz, and Sylvester Thomas.

Riding in the cabooses of the respective trains and not injured were S.L. Carner, F.A. Prozzo, E.T. Bechtol, and O.C. Markem.

Photos by M. L. Hollar taken at the scene of the accident. (Engine 6500 was the second unit of the westbound train.)

I was first introduced to the aspects of this accident during my qualifying rules examination after I joined the B&O as a student operator in 1970. The late Jack Keefauver, who had been a Cumberland Division tower operator and train dispatcher in his earlier years, was the examiner. Now assigned to the Baltimore Division, he was also the official who had hired me as an employee several weeks earlier. During his discussion, he spent about 15 minutes describing how operators at Miller Tower had slipped up and let a train get by with horrible consequences. He explained how long it had taken to get the bodies out. He spoke from experience; he had been there, at the accident scene, as part of the cleanup effort. He explained, too, that the operators involved might even have been indicted for manslaughter, except that the railroad asked the state not to prosecute. His lecture was a don't-let-this-happen-to-you admonishment.

This, in 1970, was 13 years after the Wreck on the Low Grade. A number of my colleagues had been around during the time of the accident. I can recall numerous discussions about the wreck and how the company had conducted a 'blitz' at interlocking towers throughout the system to fine tune procedures as a result of what had happened. I recall, too, of hearing that there had been some changes to the Book of Operating Rules, but I really did not know what these changes were since the book I was introduced to was a then-updated version. Still, I knew the importance of blocking switches and signals to protect right-of-track moves, and to display train order signals as required. Oh, there were slipups - but mostly because I forgot to clear the signal once it was no longer needed - not (fortunately) the other way around!

Little did I know back in 1970 that there would come a time that I would actually serve at Miller Tower. Twenty-two years later, that's where I was. Quietly nestled within the splendor of the Potomac River Valley, it was a wonderful place to work. But I was acutely aware of the infamous legacy of the Wreck on the Low Grade - surely the tower's darkest hour.

By 1992, train orders were no longer used to divert movements against the current of traffic. Operators were still required to apply blocking devices to protect right-of-track moves, but the train order signal had been eliminated. In fact, the apparatus had been removed from the building.

- - - - -

With the help of a friend who furnished me with a copy of the ICC report on the Wreck on the Low Grade, I printed it in the January 1997 issue of the Bull Sheet. It was then the 40th anniversary of the accident. By printing that report, I felt, it would document the complete story of what happened. That was my intention...

But it raised some questions...

Shortly after that issue was published, I got a letter from a reader saying that there was more to the story. Something important had been overlooked in the ICC version of events.

He (the reader) had been told somewhat earlier by an uncle, who had been a B&O employee at the time of the accident and had since died, that there had been a 'failed signal' at Miller Tower on the night of the accident. Had this signal not failed, he said, the accident could have been prevented.

I did some thinking...

The 'failed' signal, I reasoned, could not have been the 'interlocking' signal (or 'home' signal). That had worked perfectly. The home signal governing movement of the train through the tower's interlocking plant should never have been activated, of course, but the operator admitted that he had committed the error and allowed the train to proceed.

This left the 'train order' signal as the only other signal that could have prevented the accident. But the operator admitted that the train order signal (in this instance, red) had not been displayed.

Or had it?

- - - - -

Refer once again to the description, as given by the ICC, of the required train order signal:

Refer also to the portion describing the operator's failure to display the train order signal:

Perhaps it took someone with some working knowledge of how train order signals were displayed to uncover a possible omission from this characterization, but here it is:

Indeed, at some towers, the bracket for the flag and the switch for the light were in the same place. And very well they might have been (in the same place, away from the desk) at Miller - for all I knew - inasmuch as the apparatus had been removed before I came upon the scene.

But perhaps they weren't.

True, the bracket for the red flag (actually a metal panel) was several steps from the desk, and required opening the window for placement. This I could see by examining the tower's exterior for evidence of a train order signal bracket having at one time been mounted.

But what about the light?

I found my answer by peeling some tape from the tower's model board. Beneath the tape were socket cavities for two light switches, taped over because the lights were no longer in service. Each had been a three-position switch-knob located within easy reach from the desk. Inscriptions on the model board were "ETO" (for Eastward Train Order) on the left side of the board, and "WTO" (for Westward Train Order) on the right side. Each switch-knob could be turned from its neutral position to the left for '19 ' orders (yellow), or to the right for '31' orders (red).

If an operator did not wish to interrupt the dispatcher by taking the time to display the panel, he could at least activate the light - by merely reaching over to the model board, just inches away.

True, display of the light without the panel would have been incomplete. But the display of one without the other - although both were required - would have conveyed the same meaning. And with the red light likely being more visible to a train (at night) than would the panel, this should have offered the needed protection - and the accident could have been prevented.

But was this the way it had been back done in 1957?

Had the light activation feature been introduced to the model board after the accident?

Was it possible that this feature had been removed sometime earlier, and reinstalled later?

Or had the light simply failed?

So I did some checking...

I made contact with several people whose information I greatly value. But for the purposes of this article, my sources shall be nameless. There are still some sensitivities over this accident, and it was felt best to protect the anonymity of those with whom I spoke, and they concurred. Anyway, my sources were folks who had close knowledge of the model board and the train order signals at Miller Tower. It was concluded that the model board - the one I had been using since I arrived on the scene in 1992 - was the same board that was in use back in 1957, and that the train order light could have been activated from it without requiring the operator to leave his seat.

It is curious, then, that the ICC report on the accident mentioned the requirement of both a flag AND a light for the train order signal at night, but then only mentioned the operator's failure to display the flag, not the light.

None of my contacts had ever heard of a train order light failure on the night of the accident. Moreover, if there had been such a failure, surely this would have been suitable fodder for the ICC report. The best conclusion was that neither the flag nor the light had been displayed, and the ICC simply didn't mention the operator's failure to display the light. So, why wasn't the light displayed?

- - - - -

This question might never have been answered (to me) but for a chance meeting last year with a retired gent who had served at Miller Tower during the timeframe of the accident. In fact, he even worked there later that day. (He did not even learn of the accident until after he arrived.) So I asked the question of him. Why had not the train order light been displayed? His answer was unambiguous...

Back in those days, it just wasn't done!

Notwithstanding the point being made in the ICC report that the third-shift operator actually intended to display the train order signal, it was simply not the culture then at Miller Tower - and at many other towers, in fact - to activate train order signals, unless they were needed to notify a specific train that the operator had orders for it to be delivered.

The rationale behind this practice was one of fluidity - and a book of operating rules that effectively brought everything to a standstill if its rules were followed to the letter.

Miller Tower was not unique in this regard, to be sure, but the characteristics of its interlocking and associated trackage did make it a poster child for why the rules were blatantly ignored.

Once again, this tower served as a control point for three tracks to its west, and four tracks and a yard to its east. It was, at times, a very busy place with numerous trains from both directions making use of its interlocking plant.

But the company's book of operating rules was somewhat quirky regarding the red train order signal. Once such a signal was displayed in a particular direction, all trains moving in that direction would be required to stop and receive clearance to pass, even if there were no orders affecting them.

Here is an example: Say there was an eastward red train order signal displayed at Miller Tower to protect a restriction on No. 4 track. But along comes an eastward train - not routed toward No. 4 track, but toward the Western Maryland connection track. It will not be affected by the restriction on No. 4 track in any way, and the operator does not have any orders for delivery to that particular train. Still, the train would be required to stop at the tower, receive a Clearance Form A (saying that there were no orders for delivery), and then try and get moving again. Now let's say that that train is a loaded coal train - which many were - and because it was required to stop, it could lose its momentum and stall once it began its assault upon the grade beginning across the river at Big Pool.

This effectively tied the hands of those whose duties were to keep the trains moving, and a laissez-faire approach was implemented in the towers regarding the train order signal (both red and yellow). It was only used if there was an order to be delivered. (A yellow train order signal - unlike red - did not require the train to stop; the order was simply handed up 'on the fly.')

The Book of Operating Rules had no provision for clearing the train order signal for trains to which no orders were addressed - except (interestingly) passenger trains! This lone exception applied to first-class trains (which scheduled passenger trains were) and extra passenger trains. One could argue that if it was considered safe to summarily clear a train order signal to avoid delaying a passenger train, it ought to be just as safe to clear it for other trains as well. Anyway, that is what they were doing in many of the towers... up until the time of the Wreck on the Low Grade.

- - - - -

The Wreck on the Low Grade had far reaching effects. There was a system-wide crackdown to assure compliance with the rules. In some instances, where towers had multiple tracks and a good deal of traffic, separate train order signals were installed with designations as to which track each applied. Here is what one former dispatcher, now retired, wrote recently:

- - - - -

It is said that the Book of Operating Rules is written in blood. Everything included in the book is there because of something that happened along the way that caused a new item to be added. By the time I came onto the railroad, the rules had been changed. The train order signal was still required to be displayed - by whichever color applied for the most restrictive order - whenever there was an order to be delivered. But there was a procedure in place to clear the signal for any trains (passenger or otherwise) to which no orders were addressed.

This, then, has been the Rest of the Story. Make no mistake about it - the operators at Miller Tower messed up. Not displaying the train order signal could be explained (although not in the words offered by the ICC), but negligence in not applying proper locking devices to their switches and interlocking signals had horrible consequences. Another matter - which likely played a major role - was complacency. There comes a time that somebody can become so proficient at something that it can conceptually be done while asleep. This is where mistakes can pop up.

I believe, too, that the coincidental timing of the change in shifts had a lot to do with things. Had the shift not changed when it did, and the same operator and dispatcher had been on duty throughout the entire episode, the accident probably would never have happened. 'Continuity in involvement' is the term I would apply. Indeed, I always felt that the first minutes - or even an hour - of a new shift are especially vital. That's the time it takes for somebody to get a real 'feel' for what's going on.

But this brings us to some other 'reports' about the accident that were not presented by the ICC , and I will present these as 'Heard on the Street' accounts on the situation. They are:

(1) That the second and third-shift operators had been 'feuding' with each other. So numerous were these reports that I have no reason to discount them. The operators 'said' that both were in the office when train order No. 1 was made complete. Maybe so. But by a number of accounts, the second-shift operator had already completed the order, left the office when the third-shift operator arrived, and the two passed each other on the outside steps without exchanging any words. It has also been reported that the train order was simply left on the desk for the third-shift operator to see - but he did not notice it until after the eastbound train had passed the tower an hour and a half later. It would be hard to imagine that anyone would allow a tiff to interfere with safety - and I'm certain that this was neither of their intentions - but this may have been the consequence nevertheless, as things developed.

(2) That the train order had actually been written and completed somewhat earlier than 12:01 A.M. The explanation given is that the crew of the westbound train - not being next to the tower (West Cumbo) as the train was getting prepared to make its double - needed their order in advance to avoid having to walk the mile or so back to the tower when they were ready to occupy No. 4 track. Could the order have been post-dated? After all, January 1st would not just be a new day, or a new month, but a new year. Could the dispatcher have post-dated the order to incorporate it into the business of January 1st, being that the order would not be fulfilled until then anyway? Not likely, said one source, although he insisted it was his understanding that the order had been issued earlier. No dispatcher would ever issue an order in another dispatcher's name, he said. 'Write it' in advance, perhaps, to get a jump on things for the next dispatcher, but never 'complete' it. How, then, did the ICC come up with the order being No. 1 (the first order of the day) and completed at 12:01 A.M.? This question, however germane it may be, might never be answered.

- - - - -

Here now is a Timeline of Events in the Wreck on the Low Grade, January 1, 1957:

- - - - -

Miller Tower closed in September 2000. I was its final operator. The Low Grade line is now signaled in both directions.