XC X-^^^'Sil^^^ INTERSTATE COMMERCE COMMISSION REPORT OF THE DIRECTOR BUREAU OF SAFETY ACCIDENT ON THE SOUTHERN PACIFIC RAILROAD TORTUGA, CALIFORNIA SEPTEMBER 20, 1938 INVESTIGATION NO. 2294 UNIV. OF FL Lli. DOCUyENTS DEPT. ',/^^^. i 1- LimV?^/^ U.S. DEPO»ITO?'> te UNITED STATES GOVERNMENT PRINTING OFFICE WASHINGTON: 1939 SUMMARY Inv-2294 Railroad Southern Pacific. Date September 20, 1938. Location Tortiiga, Calif. Kind of accident Head-end collision. Trains involved Passenger Passenger. Train Nos. 44 5. Engine Nos 4304 4362. Consist 11 cars 14 cars. Speed Standing 30-40 m. p. h. .Operation Timetable, train orders, and automatic block-signal system. Track Single; tangent; 0.44 percent descending westward. Weather Clear. Time 1:36 a. m. Casualties 11 killed; 139 injured. Cause Switch opened directly in front of approach- ing train. • 2 To THE Commission: December 2, 1938. On September 20, 1938, there was a head-end collision between two passenger trains on the Southern Pacific Railroad at Tortuga, Cahf., which resulted in the death of 8 passengers and 3 employees and the injury of 132 passengers, 3 railway mail clerks, 3 Pullman em- ployees and 1 train-service employee. The investigation of this acci- dent was made in conjunction with a representative of the Railroad Commission of California. LOCATION AND METHOD OF OPERATION This accident occurred on that part of the Los Angeles Division designated as the Salton Subdivision which extends between Indio, Cahf., and Yuma, Ariz., a distance of 121.8 miles. This is a single- track line over which trains are operated by timetable, train orders, and an automatic block-signal system. At Tortuga, a siding, 3,642 feet in length, parallels the main track on the north; the accident oc- curred on this siding at a point 232 feet west of the east switch. En- trance to the east end of the siding is made through a No. 10 turnout with a maximum curvature of 7° 22' ; the clearance point is 191 feet from the switch points. Approaching from the east the track is tangent more than 8,000 feet to the point of accident and a distance of more than 4,000 feet beyond. The grade is 0.44 percent descending westward more than 8,000 feet to the point of accident and a distance of about 1,000 feet beyond. Westward home signal 6789 and westward distant signal 6795 are located 68.5 and 2,069 feet, respectively, east of the switch involved, with approach-lighting circuits for both signals ex- tending 11,648 feet east of the home signal. Eastward home signal 6790 is located on the south side of the track opposite the westward home signal; these signals display red aspects when the block east of signal 6790 is occupied and the switch-points of the east switch are opened one-fourth inch or more. 3 INTERSTATE COMMERCE COMMISSION 1' -t; ' Sig.679S- -Sig.6789 c Yuma^ Arizona 6. 6 m/. 1 > Araz. iTci^Ca/if. 28. Omf. 1 G/omJs /3.5m/. < 1 Amos 6. 1 m!. ) : Tor/uga CP o/a.) 4-1 mi. i > Iris 6. 9 m/. ' Ni/and 56 6 m/. \> Inc/io,Ca/if. Po/n/ ofoccfden^ — '■ - \ \ -■ Sig.6790 / V \ ^ \>- *0 ACCIDENT ON SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 5 The automatic block signals involved are of the one-arm, lower quadrant, semaphore, approach-lighted type and display the following aspects: Home signal: Red Stop. Green Proceed. Distant signal: Yellow Proceed, prepared to stop at next home signal. Green Proceed. The switch stand at the east switch is a Southern Pacific standard switch stand and is located on the north side of the main track with the center of the switch shaft 7 feet 2 inches from the north rail. The operating table has two lever-slots 45 degrees apart on the quadrant, in which to secure the operating lever. There is a hasp on the lever corresponding to one under each slot for inserting the lock. The lever is 21 inches in length and is fastened to the shaft 2 feet 3% inches above the switch tie on the east side of the stand and its normal position is vertical. To set the switch for the siding the lever is disengaged from the slot, raised to horizontal position, then pushed 45 degrees to the north slot where it is pushed downward to engage the slot. The switch shaft is 6 feet 2 inches in height and at a point 5 feet above the head-block there is a banner 18 inches in diameter which displays red aspects to the east and the west when the switch is set for the siding. Except within yard limits switches within auto- matic-block territory are not equipped with lights; the point of accident was not within yard limits and there was no light on the switch stand. Rule 104 (C) reads in part as follows: Trainmen and other employees must not unlock derails or main track s'nitches to enter main track until the trains which are to be met or passed have cleared the switch, and must not stand within twenty feet of the switch stand while a train is closely approaching or moving over the s'nitch. When practicable, they must take position on the opposite side of track from sv\itch stand while train is passing. When a s'nitch is thrown, the employee using it must see that both points have moved to proper position. A switch must be fastened as soon as thrown either way, and, when locked, the chain pulled to insure that lock is securely fastened. The maximum authorized speed for passenger trains is 65 miles per hour. Westward trains are superior to trains of the same class in the opposite direction. The weather was clear at the time of the accident, which occurred at 1:36 a. m. 6 INTERSTATE COMMERCE COMMISSION DESCRIPTION No. 44, an east-bound passenger train known as "The Californian," consisted of two baggage cars, four chair cars, one diner, three Pullman tourist sleeping cars and one lounge car, in the order named, of standard all-steel construction, except the third, fourth and fifth cars, which were constructed of Cor-Ten and stainless steel and were so-called lightweight streamlined cars, hauled by engine 4304, and was in charge of Conductor Rankin and Engineman Mason. At Indio, 67.6 miles west of Tortuga, the crew received train order No. 473 which conferred on No. 44 right over No. 5 from Indio to Araz Junction, 47.6 miles east of Tortuga. At Niland, 11 miles west of Tortuga, train order No. 405 was received, reading as follows: No. 5 meet No. 44 at Tortuga and No. 4 and No. 6 at Iris No. 44 take siding This train left Niland at 1:12 a. m., according to the train sheet, 4 minutes late, entered the west switch of the siding at Tortuga and about 1:32 or 1:33 a. m. stopped with the head end of the engine 232 feet west of the east switch; it was struck by No. 5 about 3 or 4 minutes later. No. 5, a west-bound passenger train known as "The Argonaut," consisted of one combination mail and baggage car, one baggage car, one coach, two chair cars, one Pullman tourist sleeping car, one diner, one lounge car, five Pullman sleeping cars, and one business car, in the order nai:iied, of standard all-steel construction, except the fourth car which was a so-called lightweight streamlined car con- structed of Cor-Ten and stainless steel, hauled by engine 4362, and was in charge of Conductor Cantrell and Engineman Richardson. At Yuma, 54.2 miles east of Tortuga, the crew received train order No. 473 and at Glamis, 19.6 miles east of Tortuga, train order No. 405 was received. This train left Glamis at 1:12 a. m., according to the train sheet, 3 minutes late, passed Amos, 6.1 miles east of Tortuga, at 1:30 a. m., on time, according to the statements of the crew, and while traveling at a speed estimated to have been between 30 and 40 miles per hour collided with No. 44 on the siding at Tortuga. No, 44 was shoved westward approximately 88 feet by the impact; the locomotive was practically demolished and stopped on its left side with its front end 320 feet from the siding switch and about 20 feet north of the north rail; the tender stopped on its left side, badly damaged, attached to and behmd the locomotive. The first and second cars, coupled together, were derailed and stopped upright on the siding; the front end of the first car was telescoped a distance of about 3 feet by the rear of the tender and it stopped at a point 20 feet west thereof; the rear end of the second car stopped on the transom of the third car. S. P. 2418, the third car, and the front car ACCIDENT OK SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 7 of an articulated two-car unit, was derailed and stopped upright and in line with the siding; the front end was telescoped a distance of 18 feet by the rear end of the second car; the superstructure of the front end containing 16 seats was demolished; all the passengers killed were in this car. S. P. 2419, the rear car of this unit, and the other seven cars were neither derailed nor damaged. The locomotive of No. 5 was practically demolished and stopped on its right side on the siding at an angle of about 45 degrees thereto, with its rear end fouling the main track and the front end about 10 feet from the front end of the locomotive of No. 44; the tender and the first car stopped on their right and left sides, respectively, north of the track, next to and parallel with the locomotive; both were badly damaged. The second car, T. N. O. 608, passed the first car and the tender and stopped upright on the siding with its front end telescoped a distance of 18 feet and against the overturned locomotive. The rear end of the third car and the front end of the fourth car, S. P. 2438, buckled to the south and stopped on their left sides at an angle of about 45 degrees to the main track, with their north ends adjacent to the rear end of the second car and their greater portions lying on the south side of the roadbed; the third car was badly damaged and the fourth car was demolished. The head end of the fifth car was forced to the north, and stopped upright adjacent to and parallel with the fourth car; it was badly damaged. Both trucks of the sixth car and the leading truck of the seventh car were derailed ; these cars stopped in upright positions and remained coupled to the remaining seven cars which were neither derailed nor damaged. The employees killed were the engineman and the fireman of No. 5 and the fireman of No. 44; the employee injured was the head brakeman of No. 5. SUMMARY OF EVIDENCE Engineman Mason, of No. 44, stated that Head Brakeman Jacobson opened the west switch and derail at Tortuga and No. 44 entered the siding. The locomotive stopped at a point about 200 feet from the east switch which was properly set for che main track. The westward and eastward home signals were in proceed position at that time. The engineman extinguished the headlight, released the automatic brakes on the train, left the independent brake applied, got off the right side and proceeded to oil and inspect the locomotive. He reentered the engine cab on the left side as No. 5 approached and looked ahead through the right front window for identification; when the locomotive was \vithin 200 or 300 feet of the switch the illumination from the approaching headlight revealed that the switch points were open; simultaneously the engineman of No. 5 sounded two blasts of the whistle and Engineman Mason shouted a warning to his fireman, who 8 INTERSTATE COMMERCE COMMISSION was behind him, and jumped out the left gangway; the accident oc- curred 2 or 3 seconds thereafter. No. 5 was not in the block when No. 44 stopped and he estimated that No. 44 had been standing about Z}i minutes when the collision occurred. Subsequent to the accident Brakeman Jacobson stated to the engineman that he had thrown the switch. He said that he is personally acquainted with Brakeman Jacobson and has considered him to be very dependable; he briefly conversed with him after stopping on the siding and Brakeman Jacobson appeared normal in every respect at that time; no reference was made to the switch during the conversation. The weather was clear and the visibility was good at the time of the accident. Head Brakeman Jacobson, of No. 44, stated that he read and fully understood the train order requiring No. 44 to take siding at Tortuga; he opened the switch and derail for the entrance and rode on the pilot of the engine to the point where the train stopped near the east end of the siding; he then conversed briefly with Engineman Mason but no mention was made of the switch. About 30 seconds after the train stopped, as he started toward the east switch, he saw the beam of the headlight of No. 5 come around the curve, located about 8,000 feet east of the switch. As he proceeded toward the switch he had no thought of heading No. 5 into the siding. Nearing the switch he observed that the eastward home signal indicated that the train was in the block which is about 3 miles in length. Somewhere between the locomotive and the switch, it slipped his mind that No. 5 was to hold the main track, and about 15 seconds after his arrival at the switch he leisurely unlocked the switch, lined it for the siding, hung the lock in the hasp, in accordance with his usual practice when meeting No. 5 at this point, and took a position on the opposite side of the track about 4 or 5 feet from the south rail. About 15 seconds later, when No. 5 was closely approaching, he suddenly realized his error, hurried to the switch and tried to close it, but the locomotive passed him before he could remove the lock from the hasp. He said that immediately thereafter he left to assist the injured, leaving the switch lined for the siding. He did not know whether No. 5 was past the distant signal when he opened the switch and due to the glare of the headlight he could not estimate how far distant the train was at that time nor at the time he hurried to close it. On being informed that the switch was found lined and locked for the main track after the accident occurred, he stated that to the best of his knowledge he did not close it and he did not return to it after the collision occurred. During the 9 months he had been acting as brakeman on No. 44 this train usually had met No. 5 at Tortuga; in nearly all cases No. 5 took the siding and he has opened tlie switch to head that train in; he said that without a doubt this custom was the cause of his error. Had there been a light on the switch" stand it probably would not have helped him discover ACCIDENT ON SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 9 his error earlier, but had it been so equipped the indication might have been seen by the engmeman or the fireman of No. 44 who could have warned him by shouting or sounding the whistle. Brakeman Jacob son stated that he is 63 years of age and has been employed as passenger brakeman on the division involved for a period of 18 years. He was off duty a period of 51 hours before reporting for the trip on which the accident occurred; he had ample rest during this time, having slept undisturbed about 9K hours the night previous, and K to 1 hour during the day, before he reported for duty at 8:05 p. m. He had no worries of any kind, took alcoholic drinks very seldom, used no nar- cotics, and was physically and mentally in normal condition. Conductor Rankin, of No. 44, was on the rear platform of the rear car as his train stopped near the east end of the siding about 1:32 or 1:33 a. m. He could see the reflection of the headlight of No. 5 about 3 miles distant, and he watched the approaching train until the east- ward home signal displayed a red aspect indicating to him that the locomotive had passed it, and shortly thereafter he heard tln^ee short blasts of the engine whistle, following which the collision occurred. He estimated that from 8 to 15 seconds elapsed between the time of the display of the red aspect of the eastward signal and the time of the collision. As No. 44 stopped in the siding, he saw Brakeman Jacobson with his lighted lantern near the locomotive, at which time No. 5 was rounding the curve about IK miles distant, but he did not again see him before the collision. No. 44 had been standing about 3 or 4 minutes when the accident occurred. He could not see the position of the banner on the east switch stand because of the darkness; had the switch been equipped with a light, he could have observed its position at any time. He had worked with Brakeman Jacobson during the past 4 years and considered him to be safe and dependable. He stated that the brakeman was in good spirits during the trip on which the accident occurred, was alert in the discharge of his duties, and appeared normal in every respect. Rear Brakeman Hall, of No. 44, stated that he was examining a new journal bearing on the north side of the rear car immediately prior to the collision; the train was shoved backward about one car length by the impact. Conductor Cantrell, of No. 5, stated that the air brakes were tested at Yuma, a running test was made when departing, and they func- tioned properly en route. His train passed Amos, 6.1 miles east of Tortuga, at 1:30 a. m., on time. When the whistle was sounded for the station at Tortuga, at a point about 1 mile distant, he gave the engineman the communicating signal for the meeting point; it was acknowledged. The brakes were applied to reduce the speed of the train from about 60 to between 30 and 40 miles per hour when the 10 INTERSTATE COMMERCE COMMISSION locomotive was near the distant signal, apparently for the purpose of identifying No. 44, after which they were released. While looking out the open door at the rear end of the third car when approaching Tortuga, the visibility was good, and at a point between the distant signal and the home signal he observed that the home signal was displaying a green aspect, which indicated that the switch was lined for the main track. The headlight of No. 5 was burning brightly and that of No. 44 was extinguished. About 30 seconds after the release of the brakes the engineman made an emergency application near the east switch, and the collision occurred immediately thereafter. About 35 or 40 minutes after the collision he examined the switch and found it undamaged and lined and locked for the main track. Rear Brakeman Soutar, of No. 5, stated that approaching Tortuga the brakes were applied when they were near the mile board, reducing the speed from about 60 to about 35 miles per hour. Preparing to identify No. 44, he looked out the open door in the rear vestibule of the thirteenth car at the time of the brake application and observed that both the distant and the home signals displayed green aspects. Signal Supervisor Bliss stated that after the accident he inspected the signals involved and found that they were operating properly. In his opinion, a light on the switch stand involved would be of no additional advantage from a safety standpoint, as the westward signal is very close to the east switch. Trainmaster Cantrell stated that he first observed the westward signal at the east end of the siding about 6 a.m., and it was functioning properly. According to records furnished by Southern Pacific officials, when No. 5 departed from Yuma on the date of the accident there were no steam leaks visible on locomotive 4362 which would obstruct the view of the engineman or the fireman. During the period from May 1 to the date on which the accident occurred. Head Brakeman Jacobson had been on duty on No. 44 at 50 meeting points with No. 5; at 46 of these meeting points No. 5 took siding; 34 meeting points were at Tortuga, of which No. 5 took siding in 32 instances. Engineer of Car Construction D alley, of the Southern Pacific Co., stated that he was fairly famiUar with the construction of the light- weight, streamlined equipment involved in this accident, having been at the car works when these cars were built. He said that the chief engineer of the Pullman-Standard Car Manufacturing Co. said they met all the strength requirements of the Post Office Department speci- fications, the underframe being designed for a buffuig strength of 400,000 pounds. He stated there was no denying the fact that the lightweight cars suffered greater damage than the regular equipment in the accident but he could offer no explanation for it as they have been operating for years in the same service without any trouble, but had ACCIDEXT OX SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. H never had an accident test of this kind before. He thought the Pullman Co. made a test of a section of an underframe at its plant about November and that they may have a record of it, showing a test of 1,000,000 pounds applied to the center sills, wliich seemed to pro- vide a good factor of safety. He stated the underframe stood the accident very well although it cracked, but the superstructure did not stand it so well. His opinion was that R. I. car 4120 jumped up above the coupling when its center- plate rivets sheared so that it telescoped the Hghtweight car 2418. The car body, the roof, underframe and the truss construction are of Cor-Ten steel; the hghtweight equipment suffered more damage in both trains than did the conventional equipment and hghtweight car 2438 in train No. 5 was the only car damaged beyond repair; all the passengers killed were in lightweight car 2418, and he thought that if the superstructure had stood the shock as well as the center sills, there would have been considerably less damage and probably no fatalities. It was his opinion that had car 2418 been of open-hearth steel, standard construction, it would have telescoped just as far and as badly and the loss of life would have been as great. However, he also stated that m his opinion Cor-Ten steel being much harder than open-hearth steel, its shearing value would probably be less than open- hearth, and stated that this was demonstrated by some of the other cars where the open-hearth end posts bent back but did not break, while on car 2418 the Cor-Ten end posts were sheared off. It was his opinion that end posts of open-hearth steel would have better resisted shearing. He stated that he intended to give further consideration to light- weight cars and discuss their construction with his superiors; and also that the chief engineer of the Pullman Co. had advised him he was going to give these matters further study. He said the use of open- hearth steel for center sills, and also for end posts to comply with the Post Office Department specifications, is bein^ seriously considered, as well as the possibility of applying a plate from the corner post to the bolster, from the side sills to the belt rails, and welding this plate to the frame construction. He said he was still of the opinion that it is safe to operate lightweight cars in association with conventional steel cars and that he did not think it necessary to have them placed at the rear of conventional equipment. Master Car Repairer Peek stated that he heard the statements of the engineer of car construction and agreed with them from a practical standpoint. According to the carrier's specifications of these so-called lightweight cars, the car structure in general, including underframes, side framing, end and roof framing, etc., were fabricated of high tensile alloy steel of self-supporting, welded construction as developed by the Pullman 12 INTERSTATE COMMERCE COMMISSION Co., the yield point specified for this steel being a minimum of 50,000 pomids per square inch and the ultimate strength a minimum of 75,000 pounds per square inch; the cars were sheathed with corrugated stain- less steel. The specifications provide for buffing stresses of 400,000 pounds at the draft gear without the use of buffers. The weigh ts shown in records for articulated cars 2418 and 2419 are 86,770 pounds and 86,150 pounds, respectively, and for unit cars Nos. 24.36 and 2438, having two trucks each, they are 104,500 pounds and 104,700 pounds, respectively. Following is a statement of damage to equipment as reported by the carrier: West-bound No. 5 (moving— 14 cars) East-bound No. 44 (standing— 11 cars) Position in train No. Estimated damage Position in train No. Estimated damage Engine 4362 5126 608 401 2438 2534 $50,000 8,000 4,500 5,000 60,000 1,500 Engine 4304 4129 4120 2418 2419 2436 $50,000 First car B&M-SP First car Bag-RI-., 2,000 Second car Bag-T&NO Second car Bag-RI Third car (Art. Ch.) ... . 3,500 Third car Chair-T&NO. 40,000 Fourth car streamline chair (Lt. Fourth car (Lt. Wt.) Wt.). Fifth cai (Lt. Wt.) Fifth car Chair-SP . No damage back of third car. No damage back of fifth car. Total, $224,500. 129,000 95,500 Southern Pacific lightweight cars 2418 and 2438 were built in May, 1937, by the Pullman-Standard Car and Manufacturing Co. Records secured from this company show that in October, 1936, tests were made of a section of a Cor-Ten steel underframe which was duplicate of that used on the Southern Pacific streamlined cars involved in this accident. The section was 8 feet }& inch in length, which represents the greatest unsupported span which occurs at the center of the car, the cross-sectional area being 18.883 square inches. The test was made in a wheel press, extensometers being applied at several points on the center-sill section. Actual extensometer stress readings were made up to 900,000 pounds, at which point the instru- ment readings were discontinued, but calculated stresses were applied up to 1,000,000 pounds pressure, at which point it was apparent that the yield point had been reached. The same section of underframe was subjected to a test made by the Pennsylvania Railroad, at Wil- mington, Del., about February, 1937, and withstood a pressure of 920,000 pounds before any sign of failure or distortion appeared. Chief Engineer Parke, of the Pullman Co., also referred to another test of this same type of car made by the Pennsylvania Railroad in connection with the construction of cars of the same type for their ACCIDENT OX SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 13 service. For this purpose the Pullman Co. furnished a complete skeleton car frame, without interior finish, windows, etc., and the test was made in February, 1938, at Altoona shops of the Pennsylvania Railroad. The actual, full-size car frame was subjected to a longi- tudmal compression test and the report covering this test shows that the center sills began to spread just back of the draft casting when a pressure of 395,000 pounds was indicated by the hydrauUc gage, or at a pressure of 370,000 pounds as indicated by Brinell. The test was stopped at this point and a plate jU by 22)^ by 25 ^{e inches was applied to the bottom of the sill, after which the test was continued up to a pressure of 909,000 pounds on the hydraulic gage. The report shows that the draft casting failed at 882,000 pounds and the center sill cracked at 909,000 pounds. DISCUSSION According to the record in this investigation, the trains involved in this accident regularly met at Tortuga or some other nearby point. No. 5 was superior by direction but usually it was instructed by train order to take siding in meeting No. 44. In this instance No. 44 was given right over No. 5 but subsequently an order stipulating a meet between these trains at Tortuga designated that No. 44 take siding. The head brakeman of No. 44 had frequently occupied the same position when his train met No. 5. On the night of the accident he had read the meet order and understood its contents; he opened the west switch to permit his train to enter the siding, and then rode on the engine to the east end of the siding so as to be in position to open the east switch after No. 5 had passed. According to his statement, he was walking from the engine toward the east switch when he saw the headlight of No. 5 rounding the curve at a point about 8,000 feet distant. At that time he stated he had no thought of heading No. 5 into the siding; however, after arriving at the switch, instead of waiting until No. 5 passed, he waited only about 15 seconds, then opened the switch, hung the lock in the hasp and took a position on the opposite side of the track in accordance with his usual practice when meeting No. 5; he had stood at that point about 15 seconds when he realized his error; he then rushed to the switchstand but before he could restore the switch to normal position No. 5 passed that point and struck No. 44. The evidence indicates No. 44 had been standing on the siding 3 or 4 minutes when the collision occurred. The statements of various witnesses corroborate the head brakeman's statement up to and including the time he opened the switch. From the evidence it was impossible to determine just where No. 5 was located when the switch was opened. \\Tien the conductor of No. 5 was between the distant and home signals he looked out from the 14 INTERSTATE COMMERCE COMMISSION rear door of the third car and saw that the home-signal aspect was green, at which time the speed of his train had been reduced to 30 or 40 miles per hour. The distance from the distant signal to the home signal is approximately 2,000 feet; placing the conductor of No. 5 approximately 340 feet behind the front of the locomotive it follows that the locomotive was within 1,660 feet of the switch before it was opened. However, it is probable that it was much closer. The con- ductor of No. 5 said his train was near the east switch when the emergency application of the brakes was made. The conductor of No. 44 was looking directly toward No. 5 as it approached and he was observing the eastward home signal at the instant the red aspect was displayed; according to his statement the collision occurred between 8 and 18 seconds thereafter; assuming that the speed of No. 5 was 40 miles per hour, the locomotive would then have been between 470 and 880 feet distant from the switch when it was opened. It is probable, therefore, that not as much time elapsed between the time the switch was opened and the time the head brakeman returned to it as he indicated, because a red aspect would have been displayed by the westward home signal the moment the switch was opened. Out of a total of 34 days on which No. 5 and No. 44 met at Tortuga from May 1 to the time of the accident, when the head brakeman involved in this accident was on No. 44, No. 5 took siding 32 times; according to his statement the fact that No. 5 was to hold the main track in this instance contributed to his confusion. The evidence indicates he was in normal condition and was apparently alert at all times in the discharge of his duties prior to the time he left the engine to go to the switch. Under rule 104 (C), he should not have stood within 20 feet of the switch as No. 5 was approaching, and should not have unlocked the switch until that train had passed. The evidence is to the effect that a service application and an emergency application of the brakes were made and that the whistle on No. 5 was sounded just prior to the accident, M'hich is conclusive that the engineman of that train was not incapacitated. Had the switch involved been equipped with a light it is possible that the red aspect might have attracted the attention of the engine- man, the fireman, or the conductor of No. 44 in time to enable one of them to call the attention of the brakeman to his error. The investigation disclosed that all of the fatalities to passengers occurred in the tliird car of No. 44, which car was the forward section of a two-car articulated unit of lightweight, streamlined design. The car was telescoped a distance of 18 feet, or slightly more than one- fourth of its total length, sustaining far more serious damage than either of the two cars ahead of it. In No. 5 the car which sustained ACCIDENT ON SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 15 the greatest damage was the fourth car, which was practically de- molished and which was likewise a lightweight, streamlined car. In both of the trains involved in this accident these lightweight cars were being operated in association with heavy, all-steel, standard equipment, the lightweight cars being the third, fourth, and fifth cars in the 11-car east-bound train, and the fourth car in the west-bound 14-car train. All other cars in each of the two trains were of heavy, all-steel, standard type, thus placing the light-weight cars between the standard cars and ahead of the heavier diners, lounge cars, and sleepers. Obviously, in the collision, the most violent impact occurred be- tween the two locomotives, which were practically demolished and a great amount of the destructive shock was thus dissipated. The destructive shock then progressed backward, carrying to the fifth car in No. 5 and to the third car in No. 44, there being no damage to equipment in either train back of these points. There was consid- erable damage to the forward standard cars in each train, further dissipating the destructive force, but the first lightweight car in each train suffered the greatest damage. It is evident that the collapse of these cars further absorbed the destructive shock to such an extent that but little damage occurred beyond them. The statement of the damage to equipment discloses that of the total of $124,500 estimated damage to cars, $100,000 or 80 percent of the total amount was on lightweight cars which constituted only 16 percent of the total number of cars involved. The 21 standard cars, constituting 84 percent of the total cars involved, suffered only $24,500 damage, or 20 percent of the total amount. Considering only the eight cars within the zone of actual damage at the forward end of each train, the six standard cars in this zone suffered $24,500 damage or an average of $6,083 per car, while the two lightweight cars in same zone averaged $50,000 damage per car. The evidence indicates that no tests were made by the Southern Pacific as to the strength and stability of these lightweight cars as compared with the heavy steel cars of standard designs with which they were associated in trains. Apparently the cars were accepted from the manufacturer with the understanding that they met the requirements of Post Office Department specifications for railway mail cars, and that the underframes were designed to withstand buffing stresses of 400,000 pounds. The Postal Department specifications require a safety factor of two in the calculation of buffing stresses, fixing the minimum for actual failure due to buffing shocks at 800,000 pounds. It appears from the records that the manufacturer relied upon calculations and the result of two tests made with an 8-foot section of the underframe in the matter of buffing strength rather than on any test with a complete car structure. A few months subsequent to the 16 INTERSTATE COMMERCE COMMISSION construction of the Southern Pacific cars involved in tliis accident, tests were made by another carrier of the same type of car, using a complete underframe with skeleton car body, furnished for the purpose by the same manufacturer, and in these tests the center sills began to spread at 395,000 pounds pressure; a reinforcmg plate was then appUed, and pressure was then increased to approximately 900,000 pounds before failure occurred. It does not appear that a similar reinforcing plate was applied to the Southern Pacific cars. During the development in recent years of the so-called lightweight, streamlined equipment for passenger-train service, the Bureau of Safety has on various occasions stressed the necessity for maintain- ing that high degree of safety, which has been developed by years of experience in steel passenger-car construction. In the annual report for the fiscal year ended June 30, 1936, in referring to streamlined, lightweight cars, the following statement was made: Equipment of this type must be considered as being still in the experimental stage. As yet there is no case on record where a train of this type has been involved in a serious collision or derailment of such character as to furnish data relative to its ability to withstand impact shocks and tearing and shearing stresses The accident here under investigation constitutes the first demon- stration of the ability of such equipment to withstand impact shocks of a major collision; what would have occurred had either train been made up entirely of standard equipment or lightweight equip- ment is a matter of mere conjecture or speculation. The results of this accident raise a grave doubt whether the operation of such light- weight cars between or ahead of heavy, all-steel cars of standard design should be permitted until their ability to withstand shocks with equal safety has been proven by appropriate tests; furthermore, the effect of operating cars without buffers in association with cars having buffers also should be given careful consideration. CONCLUSION This accident was caused by a switch being opened directly in front of an approaching train. RECOMMENDATION It is recommended that railroad officials give serious consideration to discontinuance of operation of so-called lightweight cars between or ahead of standard cars unless and until the strength of construction has been determined by suitable tests to be substantially the same as that of other cars with which they are associated. Respectfully submitted. W. J. Patterson, Director. ACCIDEXT OX SOnHKRX PACIFIC RAILROAD, TORTUGA, CALIF. 17 117445—39 2 18 INTERSTATE COMMERCE COMMTSSIOX TJ^fffj^h ACCIDENT OX SOUTHERN PACIFIC RAILROAD, TORTUGA, CALIF. 19 20 INTERSTATE COMMERCE COMMISSION accidp:xt ox southern pacific railroad, tortuga, calif. 21 22 INTERSTATE CO:\rMERCE COMMISSIOX Figure 7.— View showing damaged condition of baggage car C. R. I. & P. 41l'9, tlie fir.n ear in No. 41. FiGURK 8. — View sliowing dainak'ed cinidition of baggage car C'. K. 1. A; F. 4r-'n, the second car in Xo. 44. ACCIDKXT OX SOTTHERX PACIFIC RAILROAD TORTUGA, CALIF. 23 24 INTERSTATE COMMERCE COMMISSIOX I o UNIVERSITY OF FLORIDA 3 1262 08855 9165