Levelland

What the file holds

Project Blue Book’s papers on the case make up one unit in the National Archives catalogue, “Vicinity Levelland, Texas, November 1957” (NAID 28972840, 81 images). Among them:

• the record card (image 1) and two undated summaries (images 2 and 35–36);
• the messages of the first days: from the 686th ACWRON at Walker AFB, 4 November (images 7–9); a message in the format of CONAD Manual 55-1 with information from a radar station (images 15–17); pages two and three of a message ending 03/2220Z, which reports the visit of Reese AFB’s provost marshal (images 20–21); and the preliminary report of the 1006th Air Intelligence Service Squadron (AISS), 7 November (images 14 and 19);
• a technical information sheet marked as the sheriff’s statement, given by telephone on 3 November (images 10–13);
• ATIC’s messages to Washington of 8 and 10 November (images 32–34 and 57–58);
• the 1006th AISS’s field report AISS-UFOB-386-57 of 8 November, with six sources, the checks made and the preparing officer’s comments (images 60–78);
• the second page of ATIC’s request for a review by its Air Science Division, an unsigned comment of 5 December 1957 and Capt. G. T. Gregory’s analysis of 3 January 1958 (images 22–25);
• a note of 19 February 1958 by Sgt. Norman P. Barth on the prime source (image 26);
• press clippings (image 27), a card that summarises a published account (image 38) and a magazine article on the Air Force’s statement of 15 November (image 45).

Image 18 belongs to another case: it is page two of a 1006th AISS message about a truck at Kearney, Nebraska (see “The vehicles” below).

Blue Book’s index lists the case on image 176 of NAID 28921348. The FBI’s file 62-HQ-83894 holds a memorandum of 12 November 1957 on reports of stopped engines in the south-western states (Section 9, p. 6). A Defense Intelligence Reference Document of 11 March 2010 on field effects lists the case once, as an item in an appendix: “38. 1957/11/02 TEXAS, nr. LEVELLAND” (DOW-UAP-D128, p. 32).

The night of 2/3 November, in the field report

The 1006th AISS gathered its information on 4–6 November and reported on 8 November (image 60). Its six sources, in the report’s order, with the times it gives in Central Standard Time:

• Part One. A student of 19, driving home from Lubbock at 0005 on 3 November on Texas highway 116: “Approximately four miles west of Smyer, Texas (FJDD2035), the ampmeter on SOURCE’s automobile registered discharge for a moment. Upon the return of the ampmeter to normal, the engine of the automobile began cutting out as if it were not getting any gas. After it had quit running, the lights and car radio went out.” He watched an object, “seventy-five to one hundred feet” long by his estimate, for “four or five minutes”; then “the lights and radio of the car came on and the motor started with no difficulty.” There were “heavy clouds in the area and a light rain falling” (image 62). The preliminary report lists a source “NEWEL WRIGHT” with reliability B (image 14), and the comment of 5 December speaks of “Wright” getting back into his car (image 25); we read him as the source of Part One.
• Part Two. A barber of 30, at 2300 on 2 November, “driving west on Texas Highway 116 about four miles from Levelland”: “he saw a large flame in the west. SOURCE first thought it was caused by lightning. The object, however, moved over his position and his truck engine and lights went out.” Afterwards “SOURCE was able to restart the engine, and after the engine started, the lights of the truck came on.” (image 65) The report notes that he “had no concept of direction and was conflicting in his answers” (image 65).
• Part Three. A man of 43, at 0130 on 3 November, “driving his car looking for objects that had been reported in the area”, saw “just a streak of light one time”; nothing happened to his car (image 68). His occupation is blacked out. That he is the sheriff is our reading: the report gives his schooling as “Officer’s School of Texas” (image 68), and the preliminary report lists “SHERIFF WEIR CLEM” among the sources questioned (image 14).
• Part Four. A Texas highway patrolman, “attempting to search for an unidentified object reported to the Levelland Police Department”: “The flash lasted only a fraction of a second, and was red to orange red in color. This flash occured approximately 1:15 a.m. on the morning of November 3, 1957.” (image 69)
• Part Five. A technical sergeant, at 2335 on 2 November, “about two miles south of Shallowater”, saw a flash for about three seconds: “At this time, the lights and radio on SOURCE’s car went out for approximately one to three seconds. SOURCE and his wife remarked then that it ‘was certainly a strong bolt of lightning to put out our lights and radio.’” (image 70)
• Part Six. A man from Whiteface, at 2045 on 4 November, two days later, seven miles north of Sundown, saw a red object; he stopped his car by the road, and nothing happened to it (image 72).

The preparing officer’s comments (image 78):

“a. Only one observer (Part One) said an object hovered over or on the ground. One observer (Part Two) said an object moved over his head at a low altitude, and one observer (Part Six) saw an object hovering and swinging in a pendulum motion considerable distance above the ground. The other observers saw a light streak across the sky (Parts Three and Four) or a bolt of lightning (Part Five).”

“b. Reference Part Two. This observer’s account of the sighting cannot be relied upon. His statements that the object was going eight hundred miles per hour and that the object was three hundred feet away from him are not compatible with the reported length of time in sight.”

Part Five “is included only as a possible explanation for the other sightings”, and Part Six is “not connected with the other sightings” (image 78). Of the people in the press accounts the report says: “The other witnesses reported in newspaper accounts as having observed an object had either disappeared or returned to their homes, leaving no forwarding addresses.” (image 76)

Engines stopped, then, in Parts One and Two; in Part Five the lights and radio went out for seconds at a flash; Parts Three, Four and Six report no effect on a vehicle. The count is ours.

The first messages

Before the field report, three messages had reached the Air Force:

• The 686th ACWRON, 4 November. “OVAL SHAPED, 200 FEET LONG, 40 FEET WIDE.” and “OBSERVED ON GROUND”; “THREE PEOPLE OBSERVED THE OBJECT FROM AUTOMOBILES AND WHEN THE OBJECT ROSE ABOUT 200 FEET THE ENGINE AND LIGHTS WOULD AGAIN WORK.” It gives the times as “3 NOVEMBER 1957 0430Z AND 0530Z”, which is 2230 and 2330 Central Standard Time on 2 November (our conversion), and the place as “4 TO 5 MILES WEST OF LEVELLAND, TEXAS ON STATE HIGHWAY 16.” Its observers are one person from Kermit, one from Waco and “WEIR CLEM (LEVELLAND COUNTY SHERIFF)” (images 7–8). Neither the Kermit nor the Waco observer is among the six sources of the field report (our reading of the places on image 60).
• A message in the format of CONAD Manual 55-1, on “SEVERAL OBSERVERS IN VICINITY OF LEVELLAND” between “0200Z THRU 0600Z 3 NOV 1957” (image 16), names the same observers from Kermit and Waco and says: “ALL INFORMATION CONTAINED HEREIN WAS OBTAINED FROM AN ADJACENT RADAR STATION (P-51) IN AN ADJACENT AIR DIVISION (34TH)”, and “NONE OF THE PERSONNEL MENTIONED HEREIN WERE QUESTION BY THIS ORGANIZATION” (image 17).
• Pages two and three of a message ending 03/2220Z, whose first page we did not find in the unit, list “CIVILIANS-SHERRIF WIN CLEM” [sic] and others, and report that Reese AFB’s provost marshal, Maj. Daniel R. Kester, “VIITED [sic] THE ALLEGED SCENE IN CONJUNCTION WITH LOCAL CIVIL AUTHORITIES. NEGATIVE RESULTS.” (images 20–21)

On 7 November the 1006th AISS sent its preliminary report: “SOURCES [redacted] CMM [redacted] AND [redacted] WERE ONLY ONES TO REPORT OBJECT LANDED AND THE LNLY ONES WHO SAW FOR OVER A FEW SECONDS” (image 14). ATIC told Washington on 8 November: “CONTRARY TO KEYHOE’S AND WASHINGTON PRESS REPORTS ONLY THREE, NOT NINE PERSONS WITNESSED INCIDENT.” and “OBJECT OBSERVED FOR ONLY FEW SECONDS, AND NOT LENGTHY PERIOD AS IMPLIED BY PRESS.” (image 34)

The sheriff and the deputy. The sheriff’s own statement, given by telephone on 3 November, is a technical information sheet: a light at “0100”, “3 M NW of Levelland”, seen for a “Few Sec” (image 10). The Boston Daily Record of 5 November, quoting the Associated Press, has “Several witnesses, including a sheriff and one of his deputies” (image 27). No paper of the investigation names a deputy. A later card that summarises a published account, “(KEYHOE-1960 P 114)”, has a man whose name is blacked out and “DEPUTY [redacted]” see “A WEIRD GLOW” at “1.30 A.M.” (image 38). The file holds no reports of the Hockley County sheriff’s office; the only paper from the sheriff is the telephone statement.

The vehicles

No vehicle from Levelland was examined in any paper that can be opened. The field report lists its checks in Part Seven: air traffic, balloons, an interview with a professor of mathematics and astronomy at Texas Technological College, the weather, the burning of gas by the oil companies and downed power lines (image 75). None of them concerns a vehicle.

Two summaries do speak of vehicles. The undated summary on image 2: “In one instance a faulty distributor was determined as being the cause for the motor stoppage.” The summary on images 35–36: “In some instances, faulty ignition systems and near freezing temperatures were found during the investigations.” (image 36) Neither names a vehicle or a driver from Levelland.

The only truck the files examine was at Kearney, Nebraska. Image 18 of this unit is page two of a 1006th AISS message ending 09/0226Z; of page one only the last line is on the scan. It reports that a farmer in the “IMMEDIATE AREA HAD EXPERIENCED POWER FAILURE IN HIS TRUCK AT APPROXIMATE TIME OF THIS SIGHTING”, and goes on:

“INVESTIGATION PROVED THAT A ROTOR HAD BEEN CHANGED ON THE FARMERS TRUCK THE PREVIOUS DAY PD ONE PIECE OF THE OLD ROTOR HAD NOT BEEN REMOVED AND HAD WEDGED IN BETWEEN THE POINTS CAUSING THE ELECTRICAL SYSTEM TO BE NOW NON OPERATIVE PD THIS STATEMENT WAS OBTAINED FROM THE MECHANIC WHO HAD REPAIRED THE TRUCK” (image 18)

ATIC’s message of 10 November, marked by hand “LEVELLAND, TEX CASE.”, gives the same finding under “REF. PART I OF PREVIOUS MESG” (image 57), and Part I of ATIC’s message of 8 November is “KEARNEY, NEBRASKA ‘SPACE SHIP’ INCIDENT” (image 32). The 1006th AISS’s own report on Kearney: “A farmer, [redacted], was driving his truck about 2 miles from the scene at approximately the time of the incident when his motor stopped.” And: “The piece of rotor lodged between the points and caused the failure.” (NAID 28972949, image 11) We read the sentences on images 2 and 36 as carried over from Kearney. A magazine article in the unit quotes the Air Force’s statement of 15 November: one report arose from “an automobile mechanic’s mistake” (image 45).

The comment of 5 December 1957 offered water instead: “It is not unlikely that the ignition of an old car could be put out of order by water (blown or splashed onto it). A faulty voltage regulator could account for the discharge of a battery.” (image 24) It examined no car.

The weather

The papers give five versions of the weather that night:

• the 686th ACWRON’s message: “LIGHT MIST”, “UNLIMITED CEILING”, “VISIBILITY 15 MILES”, “4/10 COUD COVER” (image 8);
• the message in CONAD format: “CLOUDY CMM LIGHT DRIZZLE”, a ceiling of “7,000 FEET”, “15 MILES”, “5/10 TO 9/10 OF SKY OBSCURED” (image 17);
• the message ending 03/2220Z: “12,00 FEET” [sic], “15 MILES”, “7/10” (image 21);
• the field report, from Reese AFB and the US Weather Bureau at Lubbock: “A complete overcast existed with a 400 foot observed ceiling and visibility at 3 miles.” and “There was a drizzle or light rain throughout the period.” (image 75);
• the undated summary: “a 400’ overcast, visibility three miles”, with drizzle or light rain, and “the area had been subjected to heavy thunderstorms just prior to the sightings.” (image 2)

The field report’s weather records no lightning or thunderstorm. The lightning comes from the witnesses: the card notes that “all but one observer mentioned lightning or lightning flashes & the existence of rain or mist in the area” (image 1), and Gregory wrote that lightning discharges “were definitely established through the results of numerous investigative reports” (image 23). The files do not say which version of the weather is right.

After 8 November

• 15 November 1957. The Air Force’s statement, known in the unit only from a magazine article, which quotes it as identifying the object as “a natural electric phenomenon called ball lightning or St. Elmo’s fire” and says the same weather might have “soaked the ignition systems of cars” (image 45).
• 5 December 1957. The unsigned comment to Capt. Gregory, in answer to his request for a review by ATIC’s Air Science Division (images 22 and 24–25). It begins: “No two sightings were made at the same time. Therefore no single explanation need be sought for all of them.” (image 24)
• 3 January 1958. Gregory’s analysis, headed “‘Ball Lightning’ -” (image 23).
• 19 February 1958. Sgt. Norman P. Barth of the 1006th AISS on the prime source, a truck driver: “all attempts possible were made to locate” him, and “He was from outside the city and could not be contacted by the sheriff of Hockley County.” (image 26) A summary agrees: “The prime source could not be located for interview and interrogations by USAF investigators.” (image 36)
• March 1958. A handwritten note on Gregory’s analysis refers to a letter from J. Allen Hynek of 5 March, which is not in the unit (our reading of the note, image 23).

The card’s “more than 300 similarly described reports within a 6 day period” (image 1) became, in a statement of 1958 on the investigator’s work, “Approximately six hundred cases were reported in a three week period following this incident.” (image 47)

The FBI

On 12 November 1957 R. R. Roach wrote to A. H. Belmont that “persons in the Southwestern states while driving their cars have allegedly seen UFO’s that caused the engines in their automobiles to stop.” The Bureau’s liaison representative had been told by Air Force Intelligence that “all the cases mentioned above have for all practical purposes been resolved as being nonauthentic or there was no basis for the sightings”, and that the investigators in the field “have not submitted their complete reports but have advised that they are all more or less negative.” (FBI file 62-HQ-83894, Section 9, p. 6) The memorandum names no town. Four days earlier the 1006th AISS’s report on Levelland had ended “forwarded as unresolved” (image 78); setting the two side by side is our comparison.

What the file says about its label

The record card marks “Other” and writes “Ball lightning”. Its summary begins: “This case triggered off more than 300 similarly described reports within a 6 day period because of the nation-wide publicity and sensationalism given by the press, radio and TV.” and notes that “all but one observer mentioned lightning or lightning flashes & the existence of rain or mist in the area.” Its comment (image 1):

“After very extensive checks and detailed investigations by the Air Force and with complete collaboration with both Air Force & non-governmental scientists it was concluded that the sighting was due to a very rare phenomena, ball lightning.”

Blue Book’s index: “5120 2-3 Levelland, Texas (Folder) (*) Multi Other (BALL LIGHTNING)” (NAID 28921348, image 176).

What the same files say:

• The field report. It lists four possibilities, St. Elmo’s fire, ball lightning, “Reflection of excess burning gas from a very low cloud cover” and “any combination” of them, and resolves none: “It is felt that insufficient knowledge of the above phenomena exists at this organization to completely resolve this sighting. Therefore, this sighting is forwarded as unresolved.” (image 78)
• The comment of 5 December 1957. “(In spite of occasional reference to ball lightning in supposedly reputable publications, it is not felt that ball lightning is sufficiently respectable scientifically to be brought into use in explaining any of these sightings, particularly where ordinary lightning appears to furnish an adequate basis for explanations).” (image 24)
• ATIC, 8 November. Ball lightning was among “THE FOLLOWING POSSIBILITIES, REPEAT POSSIBILITIES, BEING CHECKED” (image 34).
• Gregory, 3 January 1958. “In the opinion of the undersigned, after careful search, study, and consideration of all data available, the phenomenon was undoubtedly related to the meteorological conditions that existed in the area at that time”. Of the cars: “The close proximity of ball lightning or lightning discharges is capable of ionizing the air that may, in turn, effect moisture-laden ignition components of a motor vehicle.” (image 23)
• The summary. “No definite conclusion has been reached regarding the possible effects of ball lightning on ignition systems.” (image 36)

The label rests on possibilities that the field report listed and did not resolve, and on an analyst’s opinion; the status of the case stays unresolved. The files hold no measurement or physical trace of the phenomenon, no vehicle examined, no photograph and no instrument record, so the mark ◆ does not apply.

Where it was

The field report places the sightings on roads around Levelland: Texas highway 116, “Approximately four miles west of Smyer” (Part One) and “about four miles from Levelland” (Part Two); “the unmarked roadway known as the Oklahoma Flats Highway” (Part Four); “about two miles south of Shallowater” (Part Five); and, two days later, “seven miles north of Sundown” (Part Six) (images 62–72). The 686th ACWRON’s message has “4 TO 5 MILES WEST OF LEVELLAND, TEXAS ON STATE HIGHWAY 16.” (image 8), and the sheriff’s statement “3 M NW of Levelland” (image 10). No document gives coordinates. The pin is OpenStreetMap’s point for Levelland (relation 115231, 33.5871° N, 102.3777° W), rounded to 33.59° N, 102.38° W. It marks the town, not a position of any sighting.

Why this scores where it does

Witnesses: 3. Six sources questioned within days by an Air Force squadron, among them a highway patrolman and, by our reading, the sheriff. Not higher: of the two whose engines stopped, one is an account the preparing officer said “cannot be relied upon” (image 78); the patrolman and the man we read as the sheriff saw only a flash or a streak (images 68–69); and the prime source was never found (image 26).

Radar: 0. None.

Optics: 0. No photograph; the card marks “PHOTOS” as “No” (image 1).

Physical trace: 0. No trace and no measurement. The only vehicle examined in the files is the truck at Kearney (NAID 28972949, image 11).

Multi-channel: 2. Accounts only, of a light and of engines and lights that failed. The weather data describe the conditions, not the phenomenon.

Documentation: 4. The card, the field report with six interviews, the messages, ATIC’s messages, the comment of 5 December, Gregory’s analysis and Barth’s note can all be opened. Not 5: the prime source was never interviewed, the papers give five versions of the weather, and one page in the unit belongs to another case.

Anomaly: 3. Part One stays unusual: an object on the road, an engine that stopped and restarted, told by a source the preparing officer found reliable (image 78). Parts Three and Four are streaks, and in Part Five the witness himself took the flash for lightning.

Data today: 2. The report’s times and map references allow the sequence and the places to be reconstructed; the effect on the vehicles can no longer be examined.

The overall is 1.6: class D, testimony only.

What would settle it

• An interview with the prime source, the truck driver the investigators could not find (image 26).
• J. Allen Hynek’s letter of 5 March 1958, referred to by hand on image 23 and not in the unit.
• The first page of the message whose second page is image 18.
• A record of a vehicle from Levelland examined after the event; there is none in the files.