The Evidence

Listen to this chapter, read by the author →

The objects

Three spent cartridge cases were found on the floor of the sniper’s nest. That number is why the official account has three shots. But one had a dent in its lip, of a kind that can happen when a case is used as something other than live ammunition. Investigators have argued it was not fired that day. Three shells is not the same fact as three shots, and the difference has never been fully closed.^{1, 2}

The rifle found on the sixth floor was mail order. A Mannlicher-Carcano, Italian, surplus, bought through a magazine ad under a false name for about twenty dollars including the sight. Adjusted for inflation that is roughly 200 dollars today. Less than a decent suit. Whatever else is true about that day, the weapon at the center of it was the cheapest thing in the room.^{3, 4}

This book argues about that rifle constantly and has never told you what it is, so here it is in one paragraph. A Mannlicher-Carcano short rifle, Italian army issue, the model of 1891 modified in 1938. Bolt action. Six and a half millimeter caliber, feeding from a clip of six rounds that drops out through the bottom of the magazine when the last one is chambered. Serial number C2766.^{5, 6}

It was ordered in March 1963 from a sporting goods company in Chicago, out of the back of a magazine, with a coupon and a money order. Nineteen dollars and ninety-five cents for the rifle with a telescopic sight fitted, a little over twenty-one with the postage. It was sent to a post office box in Dallas rented by Lee Oswald, in the name A. Hidell, which is a name that does not belong to anybody.^{5, 6}

†The telescopic sight is where this gets interesting. It was a cheap four power scope of Japanese manufacture, and when the F.B.I. received the weapon their firearms people found they could not properly align it. To test the rifle at all they had to fit shims under the mount. That is a real problem in the middle of the official account, and both sides have used it for sixty years.^{5, 6}

†Read it one way and it makes the shooting harder to believe: a man firing three shots in six seconds through a sight that could not be trusted. Read it the other way and it makes very little difference, because at that range a shooter can use the iron sights straight through the notch, which is what the rifle was built for and what soldiers did with it. And nobody can say whether the mount was knocked out of true before the shooting, during it, or in the handling afterwards.^{5, 6}

So on the question everybody asks, whether it was a good enough weapon: yes, and the argument that it was not is the weakest one in this whole case. It was surplus, it was cheap, the sight was rubbish and the bolt is slow. It is also a military rifle that fires a full-power cartridge and was carried by an army for fifty years. Marksmen who have tested this actual weapon get consistent hits at that distance. The rifle is not the problem in this case. It never was.^{5, 6}

†A nearly whole bullet was found on a stretcher at Parkland Hospital. Commission Exhibit three ninety-nine is the most argued-over object in this case. The official account has it passing through two men, breaking bone in both. It is flattened slightly and otherwise intact. Ballistics tests have produced similar results, and very different ones. Sixty years on, that bullet is still doing the heaviest lifting in the entire case.^{7, 8, 9}

There is a piece of concrete in the National Archives. A section of curb from Main Street, cut out and taken away in 1964, because there was a fresh mark on it and a man beside it had been cut on the face. That mark is the physical proof that a shot fired in this plaza hit nothing at all. A whole chapter of the official report exists because of a scuff on a piece of sidewalk.^{10, 11, 12}

Police found a long paper bag near the sniper’s nest, made from wrapping paper and tape from the Depository’s own shipping room. A disassembled Carcano fits inside it. It carries a palm print and a fingerprint identified as Oswald’s. It is also, physically, just a bag. Almost everything in this case is an ordinary object that became evidence by being in the wrong room.^{13, 14}

The President’s car itself carried marks. There was a crack on the inside of the windshield, and a dent in the chrome topping of the windshield frame, both consistent with something striking from inside the car rather than outside it. Agents noticed them within hours. The limousine was rebuilt within months, so those marks exist now mainly as photographs and reports. The car was evidence, and then it was a car again.^{15, 16}

†A palm print identified as Oswald’s was found on the underside of the rifle barrel, in a position only reachable with the weapon partly disassembled. It is powerful evidence. It also did not reach the FBI until several days after the rifle did, having been lifted in Dallas and held there. That gap has fed argument ever since. The print is real. The delay is also real. Both things have to sit in the same sentence.^{17, 18, 19}

The sniper’s nest is not a room. It is a stack of book cartons. Someone arranged cases of schoolbooks into a low wall around the corner window, with one box on the sill and two behind, and the whole thing screened from the rest of the floor by a taller stack. It took no tools and about ten minutes. The most photographed crime scene of the century is a man rearranging boxes of children’s textbooks.^{20, 21}

Two pieces of bullet were recovered from the front of the President’s car. They were matched to the rifle from the sixth floor, and between them they weigh less than a coin. Those two scraps of copper and lead do more work in the official account than any witness, because metal does not misremember and does not change its story after thirty years.^{22, 23}

The night of the assassination Dallas police made paraffin casts of the suspect’s hands and cheek. His hands tested positive for nitrates. His right cheek tested negative. Firing a rifle does not reliably leave residue on a cheek, and nitrates are in everything from fertilizer to printer’s ink. The test settles nothing either way. It has been used for sixty years to prove both things. That is a good sign a test is worthless.^{24, 25}

The coupon

Modern readers assume there was something shady about buying that rifle through the post, and there was not. In 1963 you opened a magazine, found an advertisement from a sporting goods firm, cut out the coupon, filled in your name and address, and put a money order in an envelope. Weeks later a rifle arrived in the mail. No permit. No check. No waiting. Millions of guns moved that way every year and nobody thought it remarkable.^{26, 27}

So here is the transaction. In March 1963 a coupon went to a Chicago sporting goods firm, made out to A. Hidell, at a post office box in Dallas rented under his own name the previous fall. Nineteen dollars and ninety-five cents for an Italian surplus rifle with a telescopic sight already fitted, and a dollar fifty for postage. Twenty-one forty-five, by money order. It was shipped by parcel post about a week later. That is the whole purchase.^{28, 29}

And it was not a one-off. In the same period he ordered a thirty-eight caliber revolver by post from a firm in Los Angeles, again to Hidell, again to the same box, on a payment plan with ten dollars down. Two weapons, two states, two mail orders, one false name, one box with his own name on the rental card. The gun that killed the policeman in Oak Cliff arrived in Dallas the same way the rifle did.³⁰

And here is the thing that stops me every time. He was living in Texas. In 1963 he could have walked into any hardware store, pawnshop or sporting goods counter in Dallas, put cash on the glass, and walked out with a rifle. No permit, no record, no name, nothing. It would have taken twenty minutes and left no trace whatsoever. Instead he chose the one method available that generates a paper trail in his own handwriting.³¹

Follow that through, because it decides everything. The strongest single element in the case against him is that chain of paper. The coupon in his hand. The box in his name. The money order with his writing on it. The shipping record. If he had spent twenty minutes and cash in a Dallas shop, none of that exists, the rifle is untraceable, and the case that convicted him in every account ever written simply is not there.³²

†One reading is that a trail like that is exactly what you build if you want a weapon traceable to a particular man. The duller reading is that he was extremely poor, the mail order was cheaper than a shop, he had been using that alias for months already, and in March 1963 nobody had announced any trip to Dallas. He bought it eight months before there was anything to shoot at, and weeks later he apparently used it on somebody else entirely. I lean to the dull one, and not comfortably.³²

†One more thing about the object. The telescopic sight was a cheap Japanese four power unit fitted by the seller, and when the Bureau tested that rifle afterwards the scope was so far out of alignment that they had to shim it before it would group. Read that one way and the weapon was harder to shoot than anybody admits. Read it the other and it tells you nothing, because a scope can be knocked out of true by being thrown between boxes and handled by a dozen people.³³

And there is a consequence to all of this that outlasts everybody in the story. A senator had been trying to restrict mail-order guns since before Dallas and getting nowhere. Then a president was killed with a rifle bought out of a magazine, and the coupon itself went into the hearings as an exhibit. After two more assassinations in 1968, Congress passed a Gun Control Act that ended interstate mail-order firearms sales to private individuals. You cannot buy a rifle by post in America, and the reason is that one.^{34, 35}

The other man in the car

It gets forgotten because of who was sitting behind him, but two men were shot in that car. John Connally, the Governor of Texas, took a bullet through the back that broke a rib, went through his chest, shattered his wrist and lodged in his thigh. He survived it, held office for another five years, and lived until 1993. He is the only person who was hit that day and could talk about it afterward.³⁶

From his first testimony to the end of his life he said the same thing, and he did not say it as an opinion. He said he heard the first shot, knew immediately it was a rifle, and began to turn to look. He said he was still turning, still unhit, when that shot had already struck the President. And then a separate bullet hit him. He put it as knowledge rather than theory: he was there, it happened to his body, and he did not consider it arguable.^{36, 37}

Nellie Connally was beside him in the jump seat and she never moved either. Her account has the President hit, then her husband hit, as two events with a gap between them, and she said she watched the first one happen before the second. She wrote it down, she said it in interviews for forty years, and she was still saying it shortly before she died in 2006. Two people in that car, both certain, both consistent.^{38, 39}

Now the case against him, which is not disrespectful and is not weak. He was hit by a rifle bullet through the chest. People shot through the chest are frequently wrong about the order and timing of the seconds around it, and he himself said he never heard the shot that hit him. A man can be completely honest, completely certain, and completely mistaken about a quarter of a second, and there is a large body of evidence that this is normal rather than unusual.⁴⁰

Here is the part that makes him genuinely valuable, and it is the part both sides leave out. He rejected the single-bullet theory and he also rejected the conspiracy. He said he believed Oswald did it, alone, from that window, and he said that as consistently as he said the other thing. He simply thought the Commission had got one mechanical detail wrong. He is nobody’s witness, which is usually a sign somebody is telling you what they actually remember.³⁷

But follow his account and it does not stay tidy. If a separate bullet hit him, then you need one shot for the President’s throat, one for the Governor, one for the head, and one that missed and injured a man by the underpass. That is four, and four does not fit inside the time available for one man working that rifle. So the most credible eyewitness to the mechanics forces you toward a second shooter he did not believe in. He never resolved that, and neither has anybody else.³²

Who touched what

The rifle found on the sixth floor was described by two of the officers who found it as a German Mauser. It is an Italian Carcano. They corrected it within a day, and a policeman glancing at an unfamiliar bolt action in a dim warehouse is exactly the sort of error people make. It is also the origin of a theory that has run for sixty years, and it was made by the people who were actually there.^{41, 42}

†The nearly undamaged bullet was found at the hospital by an engineer moving a stretcher. Which stretcher it came off has never been settled. His account and the Commission’s reconstruction do not sit comfortably together. The bullet is real and it was tested and it matches the rifle. Where in that corridor it started is a question that was never nailed down at the time, and cannot be now.^{43, 44}

The evidence was handled by a city police department, a federal bureau, and a protective service, in a weekend, with no agreed procedure between them and no single custody record running end to end. Assassinating a president was not a federal crime in 1963. There was no rule about who owned the evidence, because nobody had written one, because it had not come up since 1901.^{42, 45}

Chain of custody exists for one reason: so a defense lawyer can attack it in front of a jury. Every gap above would have been put to a witness, on the record, under oath, by somebody whose job was to break it. Some would have survived and some would not, and we would know which. Instead the objections were made for the first time by authors, years later, with nobody obliged to answer.^{46, 45}

The case against him

I have spent a lot of this book on what does not fit. So here is the opposite, and I am going to do it properly rather than build something easy to knock over. This is the case against Lee Harvey Oswald as a competent prosecutor would put it to a jury, in the order they would put it, with nothing left out because it is inconvenient to me. It is a great deal stronger than most people who argue about this have ever let themselves hear.⁴⁷

Start with the weapon, because it is the strongest thing they have. A rifle was found on that floor. It was ordered by mail from a Chicago sporting goods firm in March 1963, on a coupon filled out in handwriting examiners identified as his, under the name Hidell, and posted to a box in Dallas rented in his name. When he was arrested he was carrying identification in that same false name. Every link in that chain is paper, and paper does not misremember.²⁸

Then the physical traces. His palm print is on the underside of that barrel. His fingerprints are on cartons stacked at the window the shots came from. Fibers caught on the butt of the weapon are consistent with the shirt he was wearing. Somebody will now tell you the print reached the F.B.I. late and I have already given you that objection in full. A jury would be told about the delay. A jury would also be told the print is his.⁴⁸

That morning he broke his routine, went to the house where his wife was staying on a Thursday instead of a Friday, and came back with a long paper package. He told the man driving him it was curtain rods for his room. His room already had curtains, he never mentioned curtain rods to anybody else, and no curtain rods were ever found. A paper bag made from Depository wrapping and tape, long enough for a broken-down rifle, was recovered from that floor with his prints on it.^{49, 50}

Within an hour the building superintendent ran a roll call of his staff, and one name did not answer. Every other employee of that warehouse was accounted for. The man who was missing had walked out of the front door minutes after the shooting, taken a bus, got off, taken a taxi, gone to his room, put on a jacket, picked up a revolver and left again. Innocent people do a great many strange things. That specific sequence is hard to make innocent.⁵¹

And this is the part the argument usually walks around. Three quarters of an hour after the President was shot, a Dallas patrolman was killed in a residential street by a man several people watched do it and afterwards identified. Four cartridge cases were picked up in that road and firearms examiners matched them to the revolver taken off him in the movie theater. Not consistent with. Matched, to the exclusion of every other weapon. That is the hardest evidence in this entire case and it is not about the President at all.⁵²

In the movie theater he did not put his hands up. He stood, struck an officer, drew the revolver, and by the account of the men on top of him tried to fire it. Whatever a man believes about why he is being arrested, that is what somebody does who knows exactly what they are being arrested for.⁵³

†Seven months earlier somebody fired a rifle through a window at a retired general in Dallas, from a garden, at night. His wife testified that he came home that evening and admitted it, and that he had left her a note giving instructions in case he was caught. That note exists. The physical link between that bullet and his rifle was later argued from the lead chemistry, and I have already told you that chemistry has since collapsed. The wife’s account, and the note, remain.^{54, 55}

And through twelve hours of questioning he lied about things that could be tested and were. He denied owning a rifle. He denied ever using the name on the identification in his own pocket. He said the photographs of him holding that weapon were fakes with his head pasted on, before anybody had had time to examine them. Innocent people lie under pressure constantly. They do not usually lie about the specific objects that convict them.^{56, 57}

I want to say this plainly because a lot of people on my side of the argument will not. Put those eight things in front of twelve people in any courtroom in the world and you get a conviction, and you get it quickly. Ownership, opportunity, flight, a second killing with matched shells, violent resistance, a prior attempt and a series of checkable lies. That is not a frame-up. That is a case. Anybody who tells you the evidence against him is thin has not read it.³²

‡So where do I still disagree, having said all that. In one place only. Everything above proves he owned the rifle, handled it, hid it, ran, killed a policeman and lied. Not one item of it proves he was at that window at half past twelve, because there is no witness who saw him there, no photograph of him there, and ninety seconds afterwards he is calm on the second floor. A case can be overwhelming and still be missing the one thing. This one is.³²

The evidence that did not hold

For forty years there was a piece of laboratory science that settled the shot count and almost nobody outside the field knew its name. Bullet lead analysis. The idea is simple. Lead from a given production batch has a distinctive trace chemistry, so if you measure the antimony and silver in two fragments and they match, the fragments came from the same bullet or at least the same batch. It sounds unanswerable, and for decades it was treated that way.⁵⁸

A Congressional committee had the fragments retested with that method in the late seventies. The finding was that everything recovered fell into two groups, meaning two bullets, which supported the single-bullet conclusion and left no room for a third projectile in the car. That result did enormous work. It is quoted in almost every book defending the official account, and it is the reason a lot of careful people stopped arguing about the shot count.⁵⁹

Then metallurgists went and checked the assumption underneath it. They melted and sampled production lead and found that the distinctive chemical signatures are not distinctive at all. Bullets from completely different boxes, different batches and different years can match, and bullets from a single box can differ. The groupings mean far less than anybody thought. A national scientific review said so, and the FBI stopped using the technique entirely in 2005.^{60, 61, 62}

Be precise about what that does. It does not show there were three bullets, or four, or a second rifle. It removes a reason for believing there were only two. The shot count is back where it was before 1977, resting on the film, the shells and the witnesses rather than on chemistry. And most of what is written about this case, on both sides, still cites that analysis as though the last twenty years had not happened.³²

†The morning after, a man walking on the grass south of Elm Street picked up a piece of the President’s skull. He gave it to his uncle, who was a doctor, who showed it to pathologists at a Dallas hospital before it was handed to the federal authorities. At least one of the doctors who examined it said it was occipital bone, from the back of the head. The official record has it as parietal, from the top and side. Those are different places and they carry different implications.⁶³

It was photographed and then it went to Washington, and it is not in the National Archives now. So the one piece of physical material that two sets of qualified people described differently cannot be re-examined by anybody, ever. That is the second time in this book a piece of the President’s body has gone missing from the government’s own collection, and the second time nobody has been able to establish how.⁶⁴

†One more, and it completes something I left half told. The palm print from under that rifle barrel reached the Bureau four days after the rifle did. Here is the part I did not give you. In those four days the Dallas police went to the funeral home where Oswald’s body lay and took a full set of prints from the corpse. The lieutenant always maintained he had lifted the rifle print on the Friday and simply held it. Nobody has ever disproved him. Nobody has ever been able to close it either.^{65, 66}

The recording

The Dallas police ran two radio channels that day and recorded both, one onto a belt of soft plastic and one onto a disc. And on channel one, for about five and a half minutes, somebody’s transmit button jammed open. A microphone on a motorcycle somewhere in that city stayed live, blocking the channel and recording whatever was near it. Nobody thought anything of it for fifteen years. Then somebody wondered whether it had been near enough to hear the shots.⁶⁷

The Congressional committee hired an acoustics firm, and what they did was thorough. They went to Dealey Plaza, put microphones along the route, and fired live rifles from the sixth-floor window and from behind the stockade fence, recording how each shot echoed off the buildings. Every position leaves its own pattern of reflections, like a fingerprint of the space. Then they went looking for those patterns in the noise on that fifteen-year-old belt.⁶⁸

They found impulses. Four of them, and one matched a shot from the knoll. A second team at a New York college refined the work and put the odds at ninety-five per cent or better that somebody had fired from behind that fence. The committee had a draft report saying one gunman. It changed it. That analysis, arriving at the end, is the entire reason the United States government’s second word on this case is the word conspiracy.^{69, 70}

Here is how it came apart, and it is my favorite fact in this whole subject. A men’s magazine printed a flexible plastic record of the recording and gave it away with an issue. A rock drummer in Mansfield, Ohio named Steve Barber bought it and played it over and over, because a drummer listens to a noisy recording differently from an engineer. And underneath the impulses he heard a voice.⁷¹

What he heard, bleeding faintly through from the other channel, was the Dallas county sheriff saying hold everything secure until the homicide and other investigators can get there. That instruction is on the record and it was given about a minute after the shooting, when the motorcade had already gone. Which means the impulses that were supposed to be gunfire were recorded a minute too late. A drummer with a magazine giveaway had just overturned the only official finding of conspiracy in American history.⁷²

The National Academy of Sciences put a panel on it, chaired by a physicist who would later win a Nobel Prize, and confirmed the timing. And a second problem sat alongside it. The motorcycle whose microphone this was supposed to be belonged to an officer who said it was not his, and who was not in the position the analysis needed him to be in. Two independent objections, either of which is fatal on its own.^{72, 73}

†It would be tidy to stop there and it would not be true. In 2001 a scientist published a paper in a forensic journal arguing the Academy panel had made errors of its own and that the impulses do line up with the shots after all. In 2005 several of the original panel, including that physicist, published a rebuttal. Specialists still disagree in print. What is no longer arguable is that the 1979 finding rested on something far shakier than anybody was told at the time.^{74, 75}

Now put this beside the bullet chemistry from a few minutes ago, because the two of them together are the real lesson. The two most technical, most confident, most decisive-sounding pieces of evidence ever produced in this case both collapsed, and neither collapsed because of a conspiracy. One was undone by a drummer hearing a voice. One was undone by metallurgists melting lead. In both cases the science was overturned by more science, in public, and in both cases the books have been slow to notice.³²

What can and cannot be known

The place the President was killed is the place the city was born. In 1841 John Neely Bryan built a cabin on this bank of the Trinity and called it Dallas. A replica of that cabin still stands a few feet from where the motorcade passed. The plaza is named for George Dealey, the newspaper publisher who pushed to build it. Dallas starts here. And it turns here.^{76, 77}

Somebody ate lunch on the sixth floor that day, ten minutes before the shots. Bonnie Ray Williams carried a chicken sandwich and a soft drink up there to watch the motorcade from a window. He saw nobody. He finished, gave up on company, and went down one floor to join Harold Norman and James Jarman. Police found his chicken bones near the window.^{78, 79}

†One witness said a man stopped him at the top of the grassy knoll. Malcolm Summers had thrown himself flat on Elm Street as the shots came. He got up and went up the slope with the crowd. In interviews years later he described a man in a suit, with a gun under his coat, who told them not to come any further. His statement in 1963 says nothing about it.^{80, 81}

People treat conflicting testimony as evidence of a cover-up. Every police force on earth would tell you otherwise. Memory under sudden threat is fragmentary, times get compressed, sounds get relocated, and details are unconsciously filled in afterwards from what people have since read. In Dealey Plaza that effect ran for sixty years. The disagreement is not suspicious. It would be suspicious if they all matched.^{82, 83}

Dealey Plaza is a hard-surfaced bowl with tall buildings on two sides and a railroad bridge across the end. A rifle fired anywhere in it produces a crack and then reflections off concrete and brick. That is why people standing feet apart pointed in different directions in complete honesty. Where a gunshot sounds like it came from, in that specific space, is close to worthless as evidence.^{84, 85}

Here is the whole method of this book in one paragraph. A fact is something with a document, a photograph or an object behind it. A claim is something someone said. Both are worth telling. They are not worth the same. Every claim in this book is labeled. Established, disputed, or reconstructed, because the fastest way to mislead somebody is to tell them something true in the same tone as something unproven.⁸⁶

§This model is built from a survey, from map data, from photographs and from the film. It can settle distances, sightlines and what a given vantage could physically see. It cannot tell you where a shot came from, and it will never pretend to. A three-dimensional recreation is very good at geometry and completely silent on intent. Anyone who tells you their model solved this is selling something.^{87, 88, 89}

Some of the best evidence in this case was damaged by people trying to use it. Films were projected, spliced and copied until frames wore out. The Polaroid was passed hand to hand. The limousine was washed and rebuilt. None of that was concealment. It was 1963, and nobody had a protocol for a crime scene that the entire world wanted to look at at once.^{90, 91}

†Counting the shots should be the easy part. It is the hardest. Three shells were found, but shells are not shots. Witnesses split between two, three and four. One bullet hit nothing and left a mark on a curb. The film has a gap behind a road sign. Every single method available gives a slightly different answer, and none of them can be checked against another. That is the actual state of the evidence.^{10, 92, 93}

A striking number of accounts in this case grew over time. More shots, more men, more detail, told to a camera thirty years later. That does not make them liars. Memory is reconstructive, and a life spent being asked about one afternoon will reshape the afternoon. The single most useful question about any claim here is not whether it is true. Ask instead when it was first said.⁹⁴

The United States government has investigated this twice and reached two different conclusions. In 1964 the Warren Commission found a lone gunman. In 1979 a Congressional committee found a probable conspiracy, based mainly on an acoustic analysis that has since been strongly challenged. So the official position is not one thing. It is two things, fifteen years apart, and the second one is on shakier ground than the first.^{46, 45, 95}

It is worth asking what evidence could still change the picture. Not a new theory, and not another look at a grainy photograph. A document showing someone else’s involvement. A confession with something checkable in it. Physical evidence nobody has tested. Everything else is re-reading a record that has been read a million times. Sixty years in, the shortage is not analysis. It is new facts.^{96, 97}

The point of rebuilding this plaza is not to solve anything. It is that almost everyone arguing about that day has never stood in it, and it is small, and the distances are nothing like what people imagine. Put someone where Mary Moorman stood and a great many arguments simply stop. Not because they are answered. Because you can finally see how little there is to argue about the geometry of.⁹⁸

† disputed   ‡ theory   § reconstructed — everything unmarked is established. Superscript numbers key to the sources below. Corrections

Sources for this chapter

98 sources. Every claim above is keyed to one of them. The complete list for the whole book runs to 1,276.

1. Warren Report ch.3 (CE 543-545)
2. HSCA firearms panel on the dented case
3. Warren Report ch.4 (CE 139, Klein's Sporting Goods order)
4. FBI documents on the purchase
5. Warren Report, chapter IV, the rifle
6. Warren Commission Hearings, FBI firearms testimony
7. Warren Report ch.3 (CE 399)
8. HSCA forensic pathology panel
9. later ballistics literature on both sides
10. Warren Report ch.3
11. CE 3082 (the Main Street curb section)
12. Tague WC testimony (7H552)
13. Warren Report ch.4 (CE 142, CE 626)
14. FBI latent print reports
15. Warren Report ch.3 (windshield and chrome damage)
16. Secret Service reports on SS-100-X
17. Warren Report ch.4
18. FBI latent print correspondence
19. HSCA review of the print chain
20. Warren Report ch.3 (the sniper's nest reconstruction)
21. DPD crime scene photographs
22. Warren Report ch.3 (CE 567, CE 569)
23. FBI firearms comparison reports
24. Warren Report ch.4 (paraffin tests)
25. HSCA review of nitrate testing reliability
26. Federal Firearms Act of 1938 as it stood in 1963
27. contemporaneous mail-order firearms advertising
28. Warren Report, chapter IV, Commission Exhibits 139 and 773
29. Klein's Sporting Goods records
30. Warren Report, chapter IV, Seaport Traders order records
31. Texas firearms law as it stood in 1963
32. the author's assessment
33. Warren Report, chapter III, FBI firearms testing
34. Gun Control Act of 1968
35. Senate hearings on mail-order firearms, 1963 to 1968
36. Warren Commission testimony of Governor John Connally
37. published statements of John Connally, 1964 to 1993
38. Warren Commission testimony of Nellie Connally
39. Nellie Connally, published account, 2003
40. Warren Commission Report, assessment of the Connally testimony
41. Warren Commission testimony of Seymour Weitzman and Eugene Boone
42. Warren Report, chapter IV
43. Warren Commission testimony of Darrell Tomlinson
44. Warren Report, chapter III
45. HSCA final report 1979
46. Warren Report 1964
47. Warren Commission Report, chapter IV
48. Warren Report, chapter IV, fingerprint and fibre evidence
49. Warren Commission testimony of Buell Wesley Frazier and Linnie Mae Randle
50. Warren Report, chapter IV, Commission Exhibit 142
51. Warren Report, chapter IV, on Oswald's movements after the assassination
52. Warren Report, chapter IV, the Tippit killing and firearms identification
53. Warren Report, chapter IV, arrest at the Texas Theatre
54. Warren Commission testimony of Marina Oswald
55. Warren Report, chapter IV, the Walker shooting
56. Captain Will Fritz's notes and later reports
57. Warren Report, chapter IV, interrogation
58. HSCA final report 1979, neutron activation analysis of the Kennedy assassination evidence
59. HSCA final report 1979, volume I, testimony on neutron activation analysis
60. National Research Council, Forensic Analysis: Weighing Bullet Lead Evidence, 2004
61. Randich and Grant, Journal of Forensic Sciences, 2006
62. Spiegelman et al., 2007
63. contemporaneous accounts of the recovery and examination of the Harper fragment, November 1963
64. Assassination Records Review Board findings on missing medical materials, 1998
65. Warren Commission testimony of Lieutenant J. C. Day
66. Dallas Police records of post-mortem fingerprinting, 24 November 1963
67. Dallas Police Department radio recordings, 22 November 1963
68. HSCA final report 1979, acoustics analysis by Bolt Beranek and Newman
69. HSCA final report 1979, findings on acoustic evidence
70. Weiss and Aschkenasy analysis, 1978
71. contemporaneous accounts of Steve Barber's identification of the crosstalk, 1979 to 1980
72. National Research Council, Report of the Committee on Ballistic Acoustics, 1982
73. HSCA and later examination of the motorcycle position
74. D. B. Thomas, Science and Justice, 2001
75. Linsker, Garwin, Chernoff, Horowitz and Ramsey, Science and Justice, 2005
76. John Neely Bryan cabin, Dealey Plaza (Dallas Historical Society)
77. Dealey Plaza named for George Bannerman Dealey, Dallas Morning News
78. Williams WC testimony (3H161)
79. Warren Report ch.3 (chicken remains and bottle, sixth floor)
80. Summers Dallas County affidavit, 22 Nov 1963
81. Summers' later filmed interviews (the added detail)
82. Warren Commission and HSCA witness records
83. established research on eyewitness memory under stress
84. Dealey Plaza acoustics
85. HSCA acoustic testing in the plaza, 1978
86. the confidence scheme set out in this book's introduction
87. CE 884 survey
88. plaza geometry derived from public survey and mapping data
89. the reconstruction's documentation and credits
90. condition history of the Zapruder film and Moorman Polaroid
91. records of the SS-100-X rebuild
92. HSCA findings
93. the Tague curb strike
94. comparison of 1963-64 statements with later interviews across multiple witnesses
95. National Academy of Sciences review, 1982
96. JFK Records Act release history
97. HSCA and Warren Commission evidentiary record
98. the author's statement of method