Wednesday, 22 March 2017 22:24

Alexandra Zapruder, Twenty-Six Seconds: A Personal History of the Zapruder Film (Part 2)

Written by

In the second installment of this book review/essay, Jeff Carter focuses on questions of authenticity, alteration, and the NPIC analyses which occurred over the week-end of the assassination but which the CIA later tried to deflect and all but make disappear from the record.

Part 1 of this essay

What the Zapruder Film Is (and Isn’t)

The Zapruder film is (most probably) an intact and authentic 8mm motion picture sequence. Information appearing in the film corresponds with common segments of other amateur films taken in Dealey Plaza during the assassination event, as well as existing still images. The extant images match the general description provided by Abraham Zapruder, the man who filmed the images, during his live televised appearance at WFAA studios in Dallas approximately two hours after the shooting. Later suspicions Zapruder film frames may have been removed or altered, after the film was processed and initial copies printed, gradually gained momentum in the late 1970s/early 1980s as a previously unacknowledged analysis of the film was revealed which challenged the established chain of custody with the film’s possession. Suspicions increased after the Assassination Records Review Board took specific interest in authenticating the film in the late 1990s. Although there is not currently any hard evidence that tampering took place, the presence of a Zapruder film (original or copies) at the CIA’s National Photographic Interpretation Center (NPIC) on the weekend of the assassination has been effectively established, even as official records of this event have inexplicably failed to appear.

Limits to Fakery

NPIC analysts at work
during Cuban Missile Crisis

The most precise description of a possible how and when pertaining to alteration of the Zapruder film was developed by Doug Horne, who had worked as Chief Analyst for Military Records for the Assassination Records Review Board (ARRB) in the 1990s. Horne assisted in the joint efforts between the ARRB and Kodak to preserve and assess the authenticity of the Zapruder film. During this process, as former employees of NPIC added detail to events on the weekend of the assassination, Horne came to realize two things: two separate teams developed distinct sets of briefing boards from selected frames of the film; and, from recollection (albeit many years after the fact), each team believed they were handling the original Zapruder film—one group working from an 8mm film reel, and the other from an unslit 16mm reel.1 Horne postulated that, in the approximately twelve-hour period between the work of the two teams, the original film could have been sent to a top-secret CIA film facility attached to a Kodak plant in Rochester, NY (Hawkeye Works) and there revised over the course of the day on an optical printer. A freshly altered “original” film was then presumedly returned to NPIC for a new set of briefing boards, and the existing prints of the original film were swapped out.2.

Information pointing to two separate briefing boards, and two different film formats used to create them, should not be dismissed. Official clarification may yet be discovered, perhaps in the still missing official history of the Zapruder film’s presence at the NPIC written by Dino Brugioni. As speculation has otherwise filled the vacuum, it’s worth considering what was, and was not, possible to do manipulating film images in 1963. Evidence of an 8mm reel of film on one night, and an unslit 16mm reel the next does not automatically or logically lead to an alteration hypothesis.3

The alteration argument vis-à-vis the Zapruder film has been prone to a certain illiteracy regarding the mechanics and science of special-effects filmmaking, specifically the use of the optical printer, which ranges from mildly informed to wildly uninformed, even as the whole of the argument requires intervention of such machines. Roland Zavada, a retired Kodak specialist hired by the ARRB to authenticate the Zapruder film, explained technical issues mitigating against alteration in a patient, if somewhat exasperated, response to Doug Horne’s theories and criticism published in the fourth volume of Horne’s Inside the Assassination Record Review Board.4 The substance of Zavada’s response can be, and is, supported by relevant professional technical and descriptive texts, as well as, if sought, personal affidavit from technicians experienced in practical application of optical printers for celluloid-based motion pictures (a skill set largely displaced since the advent of digital technologies). The notion that elements within the Zapruder film’s frames could be removed or rearranged at will, let alone done so without evident and obvious trace, is completely mistaken. Such sorcery was not possible with the available optical printer technology, and, for what was possible, the relatively short time period available in Horne’s hypothesis would not allow for anything but very limited—very limited—activity.

An Oxberry 1600 aerial optical printer,
a common commercial model

In an article titled “The Cinemagic of the Optical Printer”5, Linwood Dunn lists the variety of visual effects achievable on the optical printer: creating transitions such as dissolves and wipes of varying complexity; changing image size and position on screen; frame modification such as speeding up or slowing down a sequence, or “freezing” a select frame; optical zooms; superimpositions; split screens; adding motion, e.g., creating a rocking effect for a scene set in a boat or aircraft. He then describes “special categories” of effects work: travelling mattes “used to matte a foreground action into a background film made at another time”; blow-ups and reductions used to convert formats, e.g., 16mm film converted to 35mm; anamorphic conversion to change aspect ratio; and “doctoring and salvaging” which includes salvaging unusable scenes due to mechanical or human error on set or adding elements to previously filmed scenes.6.

Claims of Zapruder film alteration usually cite changing image size, frame modification, superimposition, travelling mattes, and doctoring. Where these claims tend to fail is by misunderstanding necessary limitations in the use of these techniques. A common claim is that an altered Zapruder film has removed or repositioned bystanders along the visible motorcade route through doctoring and superimposition combined with a travelling matte of the Presidential and Secret Service limousines. What is not understood while making such claims is, prior to the introduction of digital workspaces, mattes and superimpositions found seamless effect by utilizing hard vertical and horizontal lines within the frame to join separate elements, or by adding images to a flat uniform background. Consistent vertical or horizontal separation points or uniform backgrounds within the Zapruder film are virtually nonexistent because a) the sequence is always in motion as Zapruder panned with the motorcade, b) the motorcade varies in size within the frame as it approaches and passes Zapruder’s zoomed-in lens, and c) the shaky hand-held filming is inconsistent (i.e., this is not a steady locked-off pan performed with a tripod).7

Any element within the frame said to have been removed from the Zapruder film would require an equal consistent element to replace it; for instance, removing a bystander from the Dealey Plaza lawn would require additonal lawn in place for the requisite number of frames, just as a replaced bystander closer to Elm Street would require a replacement background consistent with what already is visible (portions of road, sidewalk, landscaping and other persons). These replacement elements must also adjust plausibly in perspective as Zapruder’s camera drifts and pans, and blur when the camera is unsteady. Again, this is long before digital technologies, and the workspace of each individual celluloid frame was 8mm in diameter. Theoretical radical alteration of the Zapruder film would require exacting work in multiple areas of each frame, for many dozens of frames, which would require many weeks, at least, to accomplish.8 At the end of such a process, it would be necessary for the results to appear as a seamless element of the original, an impossible task to conceive. Any removal of persons, geographic features, or even splatter from a large exit wound, should be obvious through inconsistencies produced by attempting to replace the lost information. If the Zapruder film was in fact somehow radically altered, appearing as it appears today, then it would stand as the single greatest trick shot in cinema history, even as the technique developed by these magicians would never be exploited for any other purpose, or even rumor of such incredible feat leaked as the magicians never sought credit.

Another important consideration for determining what is possible with an optical printer is the requirement for precise testing related to exposure and color temperature, to maintain consistency as film stocks have varying exposure indexes and grain structure. Print stocks used with optical printers are different from those used in the field, and production of an intermediate internegative with these stocks is a necessary part of the process,9 adding generational loss. Alteration of the Zapruder film would then require not only seamless work within the frames, but also assuring the resulting altered film’s colour, exposure, and grain is consistent with the original 8mm film stock, a feat with no known precedent.10 Discussing this, Roland Zavada determined that the minimum time to evaluate these factors, including filming, processing, and viewing the necessary tests, would have been more than seven hours,11 which factors poorly in considering an alteration scenario limited to Sunday November 24.

Z-313: a painted blob and debris removal?

Incredibly, although Zavada’s peer-supported professional opinion mitigating against alteration to the Zapruder film should have largely diminished the controversy, the notion of alteration has since hardened, and a substantial number of persons have somehow become convinced that radical alteration is a proven fact. In truth, time constraints and technical limitations make plain that if alteration was in fact engaged in that Sunday, it would necessarily be limited to, for example, a “blob” added to a frame or a black mask added to a few frames. However, even this work appears unlikely due to the difficulties in returning the altered product to an undetectable plausible 8mm “original”.12.

Aside from the technical reasons mitigating against Zapruder film alteration and substitution, a set of other considerations was articulated by Josiah Thompson in his 1998 article “Why The Zapruder Film Is Authentic.” 13 Thompson notes, from the officially vetted timeline, the original Zapruder film was in the possession of either Abraham Zapruder or representatives from LIFE Magazine that entire weekend. This notion is no longer assured. Even so, Thompson makes the point there was no means to ensure additional copies of the original intact version were not created before the film could be presumedly delivered to Hawkeye for alteration. For example, an extra copy could have been printed surreptitiously at the facilities in Dallas on the first day, or a copy perhaps made by the FBI from a borrowed Secret Service print, as discussed in memos from Saturday November 23.14 Thompson also notes that there are numerous films and photographs depicting the same sequence (or portions thereof) which potentially could require alteration as well (some thirty-eight persons had cameras in use during this sequence), and, as important, on the weekend of the assassination it could not be known if all photos and film had been accounted for—that is, a then unknown film or photograph could appear later to reveal the forgery.

Finally, other than a painted “blob” or black mask to hide wounds, it is unclear what exactly it is believed the alleged alteration is concealing. In the numerous films and still photographs which feature portions of the exact sequence captured by Zapruder, and in sequences taken before and after Zapruder was filming, there is nothing to suggest a person or event which would require excision, such as during the limousine turn which does not appear in the Zapruder film (although Abe Zapruder suggested he had filmed it during his Warren Commission testimony). One frequently cited presumed alteration is the slowing down and near complete stop of the Presidential limousine in the moments ahead of the fatal (Z312) shot which, it is claimed, was removed from the film. This is not true, but can appear that way because Zapruder is panning his camera to follow the passing vehicle; the camera itself in motion assumes a certain pace even as the vehicle slows within the frame. The slowing of the limousine becomes apparent if the viewer is able to identify Zapruder’s panning motion as a separate element from the motion of the vehicle, and follow as the pan in turn slows to keep the limousine relatively centered in frame. The camera pan actually gets ahead of the vehicle, highlighting its decrease in speed. That the limousine had come near to a complete halt can be observed in the person of Secret Service Agent Hill who rapidly gains on the static chassis. The acceleration of the vehicle is also obvious, and is even more so in the Nix film.

The Zapruder Film Is Not A Precise Clock

According to Dino Brugioni, one of the NPIC staff interviewed in the late 1990s and 2000s, representatives from the Secret Service were at NPIC on Saturday evening November 23, 1963 and were “vitally interested in timing how many seconds occurred between various frames.” Brugioni’s colleague Ralph Pearse informed these men that the Zapruder Bell & Howell Zoomatic 414PD was “a spring-wound camera, with a constantly varying operating speed”, and attempts to determine precise timing would be “unscientific” and could lead to false conclusions.15 The Secret Service agents insisted, and Pearse apparently used a stopwatch to gauge time between “various frames of interest.” Later testing by the FBI would determine that the Zapruder camera ran at an average speed of 18.3 frames per second, and, with that established, it was claimed that a count of frames between significant events appearing in the Zapruder film, divided by 18.3, could produce a precise reading of the time between which these events occurred, particularly the timing between presumed shots.

This formula unfortunately bypasses the important qualifier “average”, as it became commonly reported that the camera’s film speed was 18.3 frames per second, and thereby it was claimed the Zapruder film could serve as a precise clock for the assassination sequence.16 This is not the case, due to the spring-wound mechanism of Zapruder’s camera which, as Ralph Pearse noted, had a “constantly varying operating speed.” This factor is apparent in the results of the tests done by the FBI’s Lyndal Shaneyfelt, “focusing the camera on a clock with a large sweeping second band”, later counting frames from the developed film to ascertain the number of frames per second as determined by the sweeping second band. A “sync” motion picture camera, with a crystal sync oscillator maintaining consistent operating speed, would indeed produce repeatedly the exact same number of frames per second, but a spring-wound camera would vary.17 This spring-wound effect is reflected in the FBI report:

“This study has been made by checking the film speed of the Zapruder camera at ten second intervals throughout the full running time of a fully wound camera. Several checks were made on a full roll of film and it was found that the film speed of the camera when fully wound runs at an average speed of from 18.0 to 18.1 frames per second (fps) for the first ten seconds. It gradually increases to 18.3 to 18.5 fps for the next 20 seconds, then gradually decreases slightly to 18.1 fps for ten seconds before the final twenty seconds that run at an average speed of 17.6 to 17.9 frames per second. Mr. Zapruder has stated that the camera was fully wound when he started filming the President’s motorcade.”18.

According to the above calculation, the Zapruder film, once the Presidential car comes into view (the 132 frames of the head of the motorcade accounts for approximately 7.3 seconds) was exposed at 18 to 18.1 fps for about three seconds, and then “gradually” increased to 18.3 to 18.5 fps for its duration. The 353 frames, according to the FBI’s calculation, occurred over somewhere between 19.138 seconds to 19.332 seconds (without accounting for the “gradual transition from 18/18.1 to 18.3/18.5). The shooting sequence (LIFE 12/6/63 frame Z-190 to Z-312) occurred from somewhere between 6.595 seconds and 6.666 seconds (again not accounting for the “gradual” transition), a difference of between one and two frames. So, while not demonstrating extreme variation, the FBI’s work, at least as described, demonstrates that, giving or taking even two frames in a short span, the Zapruder film cannot be considered an exact clock. Other tests on similar cameras noted even greater disparity between individual “checks” than a few tenths of seconds.19 Such disparity is more in keeping with the advice of NPIC’s Ralph Pearse that a spring-wound camera’s operating speed was constantly varying and that attempts to measure precise timing could lead to false conclusions. In fact, the FBI’s “average” speed seems unusual for these cameras in that the results inferred suggest comparatively minute differences.

Might the FBI have dropped a high frame count pass and a low frame count pass recorded by the Zapruder camera during their speed tests, in the interest of arriving at a more precise statistical average? This statistical method is known as a “truncated mean.”20 An odd reference to frames-per-second appears in a chart presented to the Warren Commission in January 1964, presenting timing scenarios for the Presidential limousine’s approach to Dealey Plaza, based on measurements which identify a high and low miles-per-hour determination (15 mph and 12 mph) with a similar constant frames-per-second count (“22 fps” and “17.6 fps”).21 It is very tempting to speculate that these numbers—22 fps and 17.6 fps—might represent the high and low markers of the FBI’s speed tests with the Zapruder camera. Shaneyfelt told the Warren Commission “we ran through several tests of film … and averages were taken.” (WCH Vol. 5, p. 160)


In 1967 CBS time-tested five same-model cameras and got varying results


If so, the presumed “average speed” of 18.3 frames-per-second is, as Pearse told the Secret Service, meaningless in context of the assassination as there is no possibility or means to determine the frame rate when Zapruder’s camera actually ran on November 22. In theory, the “constantly varying operating speed” of the spring-wound camera would mean the frame rate varied across the duration of any filmed sequence. Although Pearse articulated this, and Brugioni apparently attached this information to the first set of prepared briefing boards, the insistence of the Secret Service agents suggests determining a time sequence for the assassination was an investigative priority. This insistence would create for the developing lone assassin narrative a series of problems.

How Did LIFE Magazine Know The Camera Ran At 18fps?

Before the FBI ran their speed tests with the Zapruder camera, LIFE Magazine’s article “An End To Nagging Rumors” (December 6, 1963) already states: “from the movie camera’s known speed of 18 frames a second—two frames a second faster than it should have run—it is possible to reconstruct the precise timing …” Zapruder’s Bell & Howell camera, according to its operating manual, was supposed to run at 16 frames per second in its RUN setting. That it actually ran some two frames faster could only be determined through tests similar to what the FBI would later do—filming a clock with the original camera. The LIFE Magazine article does not directly state that LIFE itself conducted tests and determined the speed, it says only the speed is “known”.

Although there is nothing in the record about testing Zapruder’s camera before the FBI took possession of it on December 4, 1963, it appears highly likely that a test to determine the speed of that camera was undertaken as part of an official investigation, connected with the Secret Service and CIA, sometime during the week following assassination. Information derived from this test was subsequently shared with LIFE Magazine. 22 Philip Melanson’s 1984 essay “Hidden Exposure: Cover-Up and Intrigue in the CIA’s Secret Possession of the Zapruder Film” first noticed a brief aside in an December 4, 1963 FBI memorandum discussing the possession of the camera: “(Zapruder) advised this camera had been in the hands of the United States Secret Service Agents on December 3rd, 1963 as they claimed they wanted to do some checking of it.”23 If the Secret Service were in possession of Zapruder’s camera on December 3rd, they may well have been in possession of the camera before that date. The memorandum certainly does not clarify.

When the Secret Service visited NPIC on the evening of November 23, 1963, “vitally interested in timing how many seconds occurred between various frames” Dino Brugioni recalled: “Ralph Pearse informed them, to their surprise and dismay, that this would be a useless procedure because the Bell and Howell movie camera (that they told him had taken the movie) was a spring-wound camera, with a constantly varying operating speed.” A 1975 CIA description of the same NPIC event states that since “the film had been taken in a spring-powered movie camera, it was not possible to determine precise time between shots without access to the camera to time the rate of spring run-down.”24 Access to the camera was necessary to determine the information the Secret Service was intent on establishing. That the Zapruder camera, and even the Zapruder film original, may have been, or probably were, examined at NPIC shortly after the assassination should have been an expected procedure. The Secret Service considered themselves holding “primary jurisdiction in a case of this nature”,25 and, as Philip Melanson notes, “the Secret Service of the 1960s and early 1970s had some sort of technical dependence upon the CIA.”26.

An FBI memorandum dated November 29, 1963, generated by Dallas field agents, discusses a meeting with Secret Service Special Agent John Howlett, in which Howlett described an ability to determine the distance from the alleged sniper’s nest to the Presidential limousine at the time of shots striking the President, ascertained from 8mm movies of the assassination.27 Howlett places the first shot, “where the President was struck the first time in the neck”, at “approximately 170 feet”. Paul Mandel’s LIFE article also places the first shot at 170 feet ( “The first shot strikes the President, 170 feet away…”, also identified as Zapruder frame 190 since 122 frames are then counted to the third shot which “over a distance of 260 feet, hits the President’s head.”). Howlett would inform the FBI the fatal shot was at “approximately 260 feet”. As Howlett was meeting with the FBI men, LIFE’s issue with Mandel’s article was being readied for the printers. It is hard not to believe that Special Agent Howlett and LIFE Magazine’s Paul Mandel received their information from the same or similar sources, derived from analysis conducted at NPIC.

A later chart created by the Secret Service, listing distances which differed slightly from Howlett’s,28 and associating these distances with Zapruder film frames (CE884), would situate the given distance of the first shot as equivalent to Zapruder frames 200 or 201, shortly before JFK disappears behind the Stemmons Freeway sign in the film. A certain flexibility in determining position and frame number has been introduced as early as Howlett telling the FBI men on November 29 that the Secret Service “using the 8 millimeter film have been unable to ascertain the exact location where Governor JOHN B. CONNALLY had been struck.” This uncertainty reflects the difficulties for the developing official story, as the FBI’s Robert Frazier had determined on November 27 that the bolt-action rifle in evidence required at least 2.8 seconds to operate between shots at moving target, the equivalent to approximately fifty Zapruder frames. Determining that Connally was not struck until somewhere around Z-250 (in relation to a first hit on JFK at frame 200) is not supported by the Zapruder film, where it appears the strike occurred at least 20 frames earlier.29 Differing from Howlett, Mandel in the LIFE article, provides a precise frame for a shot striking Connally (Z-264):

“The first shot strikes the President, 170 feet away, in the throat; 74 frames later the second fells Governor Connally; 48 frames after that the third, over a distance of 260 feet, hits the President’s head. From first to second shot 4.1 seconds elapse; from second to third, 2.7 seconds. Altogether, the three shots take 6.8 seconds—time enough for a trained sharpshooter, even through the bobbing field of a telescopic sight.“ (Paul Mandel, “End To Nagging Rumors: The Six Critical Seconds”, LIFE Magazine, December 6, 1963)30

In her book, Alexandra Zapruder ponders the irony that her grandfather’s film had displaced the view from the purported sniper’s nest; standing in, so to speak, for “seeing the assassination through Oswald’s eyes”. In actuality, the true irony is that, by insisting on establishing exact timing and ignoring Ralph Pearse’s advice, federal investigators wrapped themselves into a straightjacket trying to explain the visible shooting sequence, and the “exact” timing of the film, against the self-imposed limitation of three shots and one bolt-action rifle. Ultimately the Warren Commission had to go with both the single bullet theory and the claim that it could not determine when the first shot was fired. For its part, the HSCA’s photographic panel seemed to determine that the President was struck before disappearing behind the freeway sign in the film and also endorsed the single bullet theory, which are mutually exclusive.

What Happened At The NPIC November 23-25, 1963?

Dino Brugioni in 1962

It appears that two sets of “briefing boards” were independently created—one through the Saturday evening into Sunday morning and one through Sunday evening into Monday morning—both using frame blow-ups derived from a copy of or the original of the Zapruder film. Dino Brugioni was involved with the Saturday night event, and Homer McMahon the Sunday evening event, as developed by Doug Horne. Brugioni’s recollections are corroborated by a CIA submission to the Rockefeller Commission made in May 1975.31 This document, describing an analysis of the Zapruder film at NPIC, matches Brugioni’s account of the presence of the Secret Service, that establishing elapsed times between rifle shots was of primary concern, and the subsequent production of briefing boards. The document states the Secret Service “were present during the process of analysis” and took away one set of briefing boards, while CIA Director McCone retained another. The briefing board set “was controlled carefully; very few people saw it.” Notably, the document does not date the event, instead choosing to vaguely locate it in “late 1963.” Results of the analysis are deflected: “We assume the Secret Service informed the Warren Commission about anything of value resulting from our analysis of the film, but we have no direct knowledge that they did so.”

On the day following this first disclosure of a Zapruder film analysis at NPIC, the Rockefeller Commission requested “memoranda or other textual information provided to the Secret Service by CIA after NPIC’s analysis of the Zapruder film.” The CIA responded a week later, claiming they “had no indication in our records that any such written material was provided to the Secret Service. Attached are copies of the only textual matter in our files pertaining to the NPIC’s analysis of the Zapruder film.”32 Xerox copies of six “written or typed papers” were attached, described as the total existing documentation of an analysis process which spread over a thirty-six hour period and featured the production of two separate briefing board sets. That the May 7 CIA Addendum included information about the “spring-powered camera” which appears directly derived from Brugioni’s briefing board notes attached, but no such notes are among the sparse released documentation on May 14, does not inspire confidence that the CIA is on the level here.33.

Among the six papers provided to the Rockefeller Commission is a typed page which features an undated columned list featuring four “panels” with Zapruder frame numbers listed below each panel. Each frame number has a corresponding “print” number, totalling 28 prints. This appears to be for a set of briefing boards presumably created the weekend of the assassination, perhaps the second session, as Brugioni said his boards consisted of less than twenty prints. Handwritten notes on another page calculate time needed to “shoot internegs”, process, test, and make three prints. During interviews in the 1990s, Homer McMahon and Ben Hunter recognized their handwriting on this document, and also on portions of another handwritten document recreating the previously described typed briefing board chart.34 Three more handwritten pages are included, author unknown, which appear to have been created at a later date than the November 23-25 analysis as these pages feature charts and calculations which refer directly to information appearing in LIFE ’s December 6 article “An End To Nagging Rumors.”


These relatively unsophisticated charts were presented as artifacts of the 1963 NPIC analysis,
even though they were clearly drawn up later.


In fact, these pages seem to have been drawn up by a person completely unaware of the first weekend briefing boards, or that the Secret Service had already possessed the information that appeared in LIFE. The hand drawn charts feature phrases from the Mandel article in quotation marks: “74 frames later”; “48 frames after that”; “2 FPS than it should have been run”. A question is written out: “how do they know frames of first and second shot?” Timing calculations cluster the page, with division tables setting scenarios of 18fps (attributed to LIFE) and 16 fps (the camera’s speed according to its operating manual). Alternative shooting scenarios, most of which feature Zapruder frame 242 as a second shot, appear next to the LIFE attributed shooting sequence of Z-190—Z-264—Z-312. Whatever is going on with these unsophisticated charts, the impression left by the CIA’s 1975 presentation on the NPIC analysis—from lack of documentation to the sketchy attribution of “late 1963”—is of a conscious decision not to admit analysis occurred on the weekend of the assassination. Making it appear the NPIC, the premiere image analysis lab anywhere at the time, relied on timings and frame numbers printed in LIFE Magazine served to deflect attention from the actual analysis done, as did the diversion of highlighting the Secret Service’s supposed sole responsibility to share “anything of value resulting from our analysis.” The NPIC analysis event had been effectively disappeared from the record.


The typed frame chart produced as part of NPIC’s records. This may be from the second analysis event, Nov. 24-25, 1963.

The briefing panels in the record seem derived from the above typed chart.
Dino Brugioni was certain these were not the charts he had created during the first analysis event Nov. 23-24, 1963.

This motion sequence features the selected frames from the above chart.
That the panning of Zapruder’s camera gets ahead of the slowing vehicle is apparent.


For its part, the Secret Service had nothing to add, claiming that by 1979 all documents relating to the assassination had been passed to the National Archives. Nothing directly attributed to an NPIC analysis appears. The Warren Commission—which sponsored two conferences in April 1964 at which the Zapruder film was closely analyzed in the presence of Bethesda and Parkland doctors, ballistics experts from Edgewood Arsenal, FBI agents, Commission attorneys, and even John and Nellie Connally—did not receive any information regarding the November 1963 NPIC analysis.

In her book, Alexandra Zapruder asks about the NPIC event: “Who cares when it happened?” That is not the appropriate question. More appropriately: Why was the NPIC analysis hidden from the official record and the official investigation, and then, when uncovered in 1975, its “when” was obscured and its documentation was obviously incomplete?

A reason for this may be the NPIC analysis clearly demonstrated that a lone gunman conclusion was not viable; that something like the “flurry” of shots described by Secret Service Agent Roy Kellerman—seated in the passenger seat of the Presidential limousine—was more apparent. Homer McMahon, during his 1990s interviews, said it was his impression that “he saw JFK reacting to 6 to 8 shots fired from at least three directions.”35 Robert Kennedy would tell Arthur Schlesinger Jr., on December 9, 1963, that CIA Director John McCone, who received the NPIC’s first briefing boards, had indicated to him “there were two people involved in the shooting.”36 A few hours after McCone’s briefing on Sunday November 24, LIFE Magazine’s publisher C.D. Jackson sent instructions to Dallas to negotiate the remaining rights to the Zapruder film which had been explicitly left out of the contract signed the previous day. An internal LIFE memo would note that “C.D. Jackson bought the copyright to Zapruder's film to keep it from being shown in motion.”

The Zapruder Film Proves Conspiracy

A week after the assassination, the Secret Service was continuing its investigation utilizing a shooting sequence which commenced with a first hit at either Zapruder frame 190 or frame 200. At the same time, LIFE Magazine was preparing its December 6 issue featuring an article which placed the first shot at Zapruder frame 190. Years later, a House Select Committee on Assassinations photographic panel systematically analyzed the Zapruder film in a manner similar, if not more extensively, to that done previously by the NPIC.37 The HSCA panel would report: “At approximately Zapruder frame 200, Kennedy’s movements suddenly freeze; his right hand abruptly stops in the midst of a waving motion and his head moves rapidly from his right to his left in the direction of his wife. Based on these movements, it appears that by the time the President goes behind the sign at frame 207 he is evidencing some kind of reaction to a severe external stimulus.”38


Zapruder frames 190, 200, and 207. Analysis determined Kennedy began to react to a "severe external stimulus" at this point.


The Warren Commission Report would claim “it is not necessary to any essential findings of the Commission to determine just which shot hit the Governor.”39 This is not true, as essential findings of the Commission included the determination that only three shots were fired, all from a particular bolt-action rifle found on the sixth floor of the Texas School Book Depository. If the President was reacting to a “severe external stimulus” (i.e. a shot) before disappearing behind the Stemmons freeway sign, as seen in the Zapruder film and as determined by both expert panels in 1963 and 1978, there was not enough time to operate the rifle’s bolt and fire a second shot to strike Connally consistent with his observed reaction (struck approximately Z224-230). The Commission’s Single Bullet Theory proposes that Kennedy and Connally react to the same bullet as they come into view at Zapruder frame 222-223, although in the film it appears obvious the President is already reacting to external stimulus while Connally is not. It has been suggested that Connally’s reaction is somehow delayed, although the smashing of his rib bone by the passing bullet would initiate an immediate involuntary reflexive response.

Since the time of the HSCA, independent researchers have been successful in aligning close analysis of the Zapruder film with eyewitness testimony and with other photographic evidence.40 With this work, the determination advanced by the analysis in 1963 and 1978 that the President was struck by a shot at a point between Zapruder frames 190-200, before disappearing behind the Stemmons Freeway sign as seen in the film, has been corroborated by the accounts in the official record of at least a dozen witnesses, and their interlocking observations are further supported by the photographic record apart from the Zapruder film.

The testimony of Jacqueline Kennedy exemplifies this support for a first shot circa Z-190. She told the Warren Commission that she turned in her seat to directly face her husband as the result of a commotion, a noise, which can be identified as this first strike (which probably hit in the back, as witnesses located behind the Presidential vehicle described his reaction as a slump to his left). Mrs. Kennedy can be observed in the Zapruder film as turning just ahead of the disappearance behind the sign, and afterwards her hat remains largely visible holding this position, looking directly at her husband. Proponents of the single bullet theory are suggesting that a shot from a high-powered rifle blasted through Kennedy's neck and struck Connally, while Mrs Kennedy looked directly on, closely positioned, and she didn’t realize what had just happened. What is observable in the Zapruder film is that Jackie Kennedy, looking directly at her husband in the moments before the devastating shot at Z-312, is bewildered as to the source of her husband’s distress.


Mrs. Kennedy turned to look at her husband as the result of an audible commotion,
generally conceded as the strike of a first shot. She is doing so before the vehicle disappears behind the sign.


Dino Brugioni, during his 2009 interviews, recalled that the Secret Service agents who arrived with the Zapruder film at NPIC on November 23, 1963, and who directed the analysis of the film “in individual stop frames”, paid particular attention to the portion of the film which showed the Presidential limousine just ahead of the Stemmons sign, its subsequent disappearance behind the sign, and then the frames after it reappeared. The Zapruder film is unique in the photographic record as capturing this portion of the assassination sequence, and what it shows cannot be reconciled with the official conclusion of a lone assassin—as the Secret Service, and its CIA partner, surely realized less than forty-eight hours after the event.


1 The 8mm film in Zapruder’s camera was actually a spool of 16mm film, exposed along one side and then flipped and exposed on the other. After processing the film would be slit down the middle, the two halves spliced together to make one continuous roll of developed 8mm film.

2 For an overview of the National Photographic Interpretation Center and excerpts from Horne’s work, see Bill Kelly, "Washington Navy Yard NPIC", JFK Countercoup blog

3 That the Saturday 8mm reel is assumed to be the Zapruder original relies on Dino Brugioni’s recollection that there was film information between the sprocket holes. Brugioni’s memory appears fairly solid, and is corroborated on crucial points by the available sparse official documentation, but the Zapruder film possession timeline is tight because LIFE Magazine did its own work with the film at some point over the first weekend. If Brugioni is mistaken on this detail, then he was working from a Secret Service first generation copy of the film. Brugioni remembers an 8mm projector was used to view the film, but it is hard to believe NPIC employees projecting the actual original due to risk of damaging the film. It could also be that the Zapruder original was retrieved from LIFE on Sunday, possibly delivered to Hawkeye to create additional copies, and then sent to NPIC for creation of a second briefing board. Roland Zavada determined in his authenticity report that the Zapruder original initially remained as an unslit 16mm reel, as seen at NPIC Sunday night. The compartmentalization of the two briefing board sessions may reflect that the first was an “in-house” analysis, and the second featured a differing set of impressions.

4 Zavada’s open letter can be read here: It is a response to Chapter 14: "The Zapruder Film Mystery", Douglas P. Horne, Inside the Assassination Record Review Board, Volume Four.

5 Linwood G Dunn, ASC., "Cinemagic of the Optical Printer", American Cinematographer Manual, Fifth Edition, 1980. The Fifth Edition features a unique section on special effects cinematography. Dunn’s company Film Effects of Hollywood was established in 1946, and Dunn was a pioneer in optical printer technology. The American Cinematographer Manual has served as an essential professional reference book since its first edition was published in 1935. The latest Tenth Edition appeared in 2013. These volumes are compiled and published by the American Society of Cinematographers.

6 The specific examples for this final category are much simpler than might be inferred by the term “doctoring”. In the film It’s A Mad, Mad, Mad, Mad World, a gag was to feature a truck bumping into a wooden shack which subsequently collapses. During filming, the breakaway shack was pulled before the truck had backed up far enough for the gag to work. Using the optical printer, the frame was split vertically between the truck and the shack, and the frame portion of the intact shack was held (frozen) until the other frame portion saw the truck reversed to the position that would sell the intended gag. Note that the ability to achieve this effect depended on a lack of moving elements in the portion of the frame featuring the shack, as can be seen in the movie itself. A second example was of using split screens, trick cuts, and superimpositions to create close explosions and artillery fire near a group of actors playing refugees for a film titled One Minute To Zero (the desired effect was unsafe to attempt on the set.)

7 ”A Hollywood or other film production requiring postproduction optical effects is a product of a carefully planned and executed script in advance. The key subject matter, foreground and background scene content, camera image focus, depths of field, masks or mattes, etc., are carefully executed ahead of time and incorporated into the camera film that becomes the optical master…(the Zapruder film) was handheld, unsteady, panned to follow the limousine causing bystanders and background to be blurred and Zapruder jerked as reflex reaction to rifle shot reports or other stimuli.” Zavada, p. 19.

8 Consider the time required to produce relatively simple shots of the USS Enterprise against a black space background, as described in an online article ( discussing Film Effects of Hollywood's association with the first Star Trek television show. This indicates the time-consuming and sometimes imperfect results using optical printers. The effects seen in the original Star Trek program are nothing compared to claims of Zapruder film alteration.

9 “Preparation of an internegative which closely simulates the characteristic of the original has always been the goal of optical houses throughout the industry. In spite of the superb quality frequently achieved in internegatives, it seems virtually impossible to attain characteristics identical to those of the original negative in the duplicate generations for the following reasons: 1) The non-linear response of photographic film limits the range over which the following generations can duplicate an original. The internegative is one or two generations away from the original, depending on the stock used. 2) Many variable elements are introduced during the processing of the internegative. 3) The exposure characteristics of the optical printer may vary from time to time.” Mehrdad Azarmi, "Exposure Control of Optical Printers", American Cinematographer Manual, Fifth Edition, 1980.

10 “There is no known film production history that would provide a technology reference for the use of an 8mm KODACHROME II camera film as a printing master to allow subsequent significant optical special effects into selected scenes and then reconstitute the adjusted images on to an 8mm KODACHROME II daylight film ‘indistinguishable’ from the camera original.” Zavada, p. 18.

11 Zavada, pp. 30-32.

12 One text cited as “proof” that altering the Zapruder film was plausible has been Techniques of Special Effects Cinematography by Raymond Fielding. When excerpts of alteration arguments were shared with Fielding by Zavada in 2006, Fielding’s response included: “in my judgment there is no way in which manipulation of these images could have been achieved satisfactorily in 1963 with the technology then available … if such an attempt at image manipulation of the footage had occurred in 1963, the results could not possibly have survived professional scrutiny … challenges regarding the authenticity of the NARA footage and assertions of image manipulation … are technically naïve.” Zavada, p. 18.

13 The article is derived from a presentation made at a conference in Dallas November 20, 1998. (

14 DeLoach to Mohr, "8 Millimeter Color Film Taken At Scene of Assassination"

15 Horne, Inside the Assassination Records Review Board, Volume Four, p. 1233. This fascinating and important information, derived from an interview conducted by Peter Janney, is worthwhile considering in full: “… He also said that the Secret Service was vitally interested in timing how many seconds occurred between various frames, and that Ralph Pearse informed them, to their surprise and dismay, that this would be a useless procedure because the Bell and Howell movie camera (that they told him had taken the movie) was a spring-wound camera, with a constantly varying operating speed, and that while he could certainly time the number of seconds between various frames if they so desired, that in his view it was an unscientific and useless procedure which would provide bad data, and lead to false conclusions, or words to that effect. Nevertheless, at the request of the two Secret Service agents, Ralph Pearse dutifully used a stopwatch to time the number of seconds between various frames of interest to their Secret Service customers. Dino Brugioni said that he placed a strong caveat about the limited, or suspect, usefulness of this timing data in the briefing notes he prepared for Art Lundahl.”

16 The HSCA’s photographic panel did note in its report “only the average, and not the precise, running speeds for the camera are known.” Despite this, the panel would go ahead and calculate time between frames anyway. HSCA Report Appendix, Volume VI, p. 31.

17 “In crystal drive systems, a crystal oscillator of extremely high accuracy at, or in, the recorder, provides the sync pulse. The camera, in turn, is driven by a specially designed D.C. motor and control circuit which is capable of operating in exact synchronism with a self-contained crystal oscillator of comparable accuracy…both camera and recorder reference to self-contained crystal oscillators which are so accurate the effect is the same as if they had been tied together.” Edmund M. Di Giulio, "Crystal Controlled Cordless Camera Drive System", American Cinematographers Manual, Fifth Edition, pp. 469-472.

18 FBI Memorandum, Griffith to Conrad, January 31, 1964.

19 CBS did their own tests for their 1967 news special on the Warren Report. Using five cameras, the same model as the Zapruder camera (not the actual camera), their tests filming a clock with a sweeping hand resulted in a fair amount of disparity. Roughly matching the timing of the shooting sequence, the common exposed frames came in at 6.16, 6.70, 6.90, 7.30, and 8.35 seconds. Pat Speer: “IN 1967, CBS PURCHASED FIVE IDENTICAL CAMERAS AND FOUND THAT THEY RAN 15.45, 17.7, 18.7, 19.25, AND 20.95 FRAMES PER SECOND, A SIMILAR RANGE WITH A SIMILAR AVERAGE OF 18.4 FPS.” A New Perspective On the Kennedy Assassination, Chapter 2B

20 “It involves the calculation of the mean after discarding given parts of a probability distribution or sample at the high and low end, and typically discarding an equal amount of both.”

21 CD 298, p. 59 ( and CD 298, p. 62 ( It should be noted that 17.6 frames per second is cited in the FBI’s January 31, 1964 memorandum in reference to average running speed during the final twenty seconds of the Zapruder camera’s wind. This does not explain how “22 fps” entered the record. Further discussion is found in Pat Speer, A New Perspective On the Kennedy Assassination, Chapter 2B

22 LIFE’s publication schedule was such that editions were assembled a week ahead of publication date. So the December 6 edition would have been largely prepared by the weekend of November 29-Dec 1, and on the newsstands by mid-week.

23 FBI 105-82555 Oswald HQ File, Section 16, pp. 30-31 This report also states the “camera was set to take normal speed movie film or 24 frames per second.” This is incorrect: the Bell & Howell camera’s normal run speed, as noted in its operating manual, was 16 frames per second. The camera had no setting to reproduce 24 frames per second.

24 This comment was most likely derived from Brugioni's briefing board notes.

25 Memorandum 11/25/63, CD 87, p. 91.

26 Philip H. Melanson, "Hidden Exposure: Cover Up and Intrigue in the CIA's Secret Possession of the Zapruder Film", The Third Decade, Vol. 1, Issue 1, November 1984.

27 Barrett/Lee, Dallas, 11/29/63.

28 Howlett’s measurement for the fatal shot is “approximately 260 feet”, whereas the Secret Service chart (CE 884) notes the distance as 265.3 feet.

29 The FBI’s Frazier would tell the Warren Commission that Connally’s wounds could not have occurred past Z-231, if the shot was fired from the designated TSBD 6th floor window. A week after Howlett shared information with the FBI, the Secret Service would promote a different set of measurements, extending the shooting sequence to the equivalent of Z-217, Z-283, and Z-343 (CE 585). A Visual Aid Guide presented in January 1964 by the FBI to the Warren Commission (CD 298) would include a similar extended measurement whereas the first shot strikes at “167 feet”, the second at “262 feet”, and a third at “307 feet”—a full 45 feet beyond the location of the headshot seen in the Zapruder film. This Visual Aid Guide is therefore saying the fatal shot at Z312 is the second shot in the sequence. See Pat Speer, A New Perspective On the Kennedy Assassination, Chapter 2B ( for more discussion.

30 Mandel goes on to describe a sharpshooter test, using the “director of the National Rifle Association”, firing “an identical-make rifle with an identical sight against a moving target over similar ranges for LIFE last week. He got three hits in 6.2 seconds.” Later, at the request of the Warren Commission, the FBI investigated this sharpshooter test. It was determined that the sharpshooter used by LIFE was not “the director” of the NRA, and the test had no connection to the NRA. The test target was approximately fifty yards away and moved “from right to left and back, running for a distance of thirty-three feet in one direction.” (CD 1309) This test may not have been directly related to the Zapruder camera speed test results, as numerous media outlets, including LIFE, were interested in timing tests with a similar rifle very soon after the assassination, even in the absence of any published exact time for the shooting sequence. The “nagging rumor”—that there wasn’t enough time for three shots—probably derived from observation of the bolt action mechanism of the purported assassination weapon. Five decades later, well-founded skepticism remains.

31 "Addendum To Comment On Zapruder Film"

32 "NPIC Analysis of Zapruder Filming of John F. Kennedy Assassination"

33 A handwritten note written by then NPIC Director John Hicks, Brugioni’s boss in 1975, attests that these are “the only known” documents available. In a 2009 interview, Brugioni recalled discovering one of his briefing boards from 1963 during the 1975 review, and that Hicks was distressed about this.

34 Douglas P. Horne, Inside the Assassination Record Review Board, Volume Four, p. 1230.

35 Douglas P. Horne, Inside the Assassination Record Review Board, Volume Four, p. 1224.

36 For discussion of this see Bill Kelly, "CIA Director Told RFK Two People Shooting at JFK"

37 “The Zapruder film was viewed by this group on a frame-by-frame basis and at various speeds approximately 100 times.” HSCA Report Appendix, Volume VI, p 16.

38 HSCA Report Appendix, Volume VI, p. 17.

39 Report of the President’s Commission on the Assassination of President Kennedy, p. 19.

40 see, for example, Pat Speer, A New Perspective on the Kennedy Assassination, Chapter 12 Barb Junkkarinen, "First Shot/First Hit Circa Z-190", Kennedy Assassination Chronicles, Volume Five, Issue Two, 1999

Last modified on Thursday, 23 March 2017 13:11
Jeff Carter

Jeff Carter is a filmmaker and audio technician based in Vancouver, Canada. Along with Len Osanic, he produced the web series 50 Reasons for 50 Years in 2013.

Find Us On ...


Please publish modules in offcanvas position.