The XB-70 Valkyrie
The advanced XB-70 Valkyrie was initially created in the 1950s as a high-altitude nuclear strike bomber capable of flying at Mach 3 (three times the speed of sound)—any adversary would have found it impossible to counter such a bomber.
By the early 1960s, the introduction of new Surface-to-Air Missiles (SAMs) jeopardized the survival of high-speed, high-altitude bombers. Additionally, more affordable nuclear-armed ICBMs (Intercontinental Ballistic Missiles) were starting to be deployed. Consequently, in 1961, the costly B-70 bomber initiative was terminated before any Valkyries had been produced or flown.

Even so, the USAF bought two XB-70As to test aerodynamics, propulsion, and other characteristics of large supersonic aircraft. The first XB-70A, on display at the National Museum of the US Air Force, flew in September 1964, and it achieved Mach 3 flight in October 1965. The second Valkyrie first flew in July 1965, but in June 1966, it was destroyed following an accidental mid-air collision.
A shot for publicity and advertising
As told by John Fredrickson in the book North American Aviation In the Jet Age, The California Years 1945-1997, the fragmentation order for the XB-70 flight of Jun. 8, 1966, called for an airspeed calibration run followed by a sonic boom run to measure the intensity of the boom at a given Mach number and altitude.

The final item on the test card was a rendezvous with a Lockheed F-104 Starfighter, a T-38 Talon, a McDonnell F-4 Phantom II, and a Northrop F-5 Freedom Fighter for a group photograph. General Electric built the engines on all five aircraft types and wanted a shot at publicity and advertising. Clay Lacy was at the controls of the Learjet photo chase plane.
It would be an XB-70 indoctrination flight for Maj. Carl Cross. Experienced NAA chief pilot Al White was asked to fill in for Fitz Fulton, who had a scheduling conflict. The preflight briefing started at 6:00 a.m. With the first two objectives accomplished, the XB-70 slowed and descended to 30,000 feet to allow the other airplanes to form up. The formation was flying eastbound between Mojave and Barstow.
XB-70 pilot Al White recounted the event in the March 1996 NAA Retiree Bulletin:

‘[W]e heard a loud bang followed immediately by shouts of “Midair! Midair! Midair!” over the radio. Neither Carl nor I said a word, probably because we heard rather than felt the bang and were both listening intently. The chatter on the radio was coming fast. I heard the words “You’ve lost your stabilizer but you’re doing okay.”‘ I was sitting there transfixed, trying to figure out who had the problem I knew it was close because I had heard it.
Flat spin
‘After about ten seconds the XB-70 started to roll slightly to the right. When I tried to correct the movement. I got an immediate clue as to who had the problem. The airplane yawed violently and continued to roll. It felt like the giant airplane was trying to snap roll. “‘Bail out! Bail out!” the radio crackled.
‘My immediate fear was that the forward fuselage would break off and that we would not be able to eject due to the loads… Fortunately the XB-70 fuselage held together, but during the second roll, the wing came off. From that point on I would be hard pressed to describe the maneuvers the airplane went through. Just before I ejected, I realized we were in a flat spin.

‘When Joe Cotton [in the rear seat of the T-38] commanded us to bail out, I was trying to apply engine power on the right side to counteract the yaw, but it was already obvious that the plane was out of control. [AI White struggled with the encapsulated B-70 emergency egress system because his arm was trapped. White was unable either to assist or even communicate with copilot Carl Cross. White could see Carl’s head moving but did not observe him activating his escape capsule.]
‘As the airplane gyrated down though the overcast, I wrestled with my arm problem and became less aware of what was going on around me. Everyone has limitations under pressure, and by then I’d reached the limit of my ability to act.’
He continues;
‘When I finally got my arm loose, I had no idea how far we had fallen. I only knew I had to get out now. I squeezed the trigger and fired out of the airplane with the capsule doors open.
F-104 collided with the vertical stabilizer of the XB-70 Valkyrie

‘Suddenly it was very quiet. Then there was a big whoosh as the airplane went by just below me in a spin. I thought the next time around it might get me, but my parachute opened, and the airplane dropped away. Several seconds later I heard a terrible WHOMP as it hit the ground.
‘The capsule clamshell door had a small window at the top. The next thing I saw through it was the cactus on a hill about 100 yards away. When I hit, it sounded like a metal garbage can full of bricks landing on a concrete sidewalk after a three-story drop. It felt worse than anything I’ve ever felt. I got the top door open and was trying to crawl out when the T-38 went by. I waved as best I could. It was great to see that they knew where I was, and that I was alive.
‘I got out, stood up, and pulled my jacket out, but had trouble getting it on. Even though it was a late spring morning in the desert I was getting very cold and stiff, so stiff that I couldn’t sit back down pulled the parachute around me for warmth. [AI White was flown by helicopter to the hospital.]’
White concludes;

‘We later learned that the horizontal tail of Joe Walker’s F-104 hit the turned down wingtip of the XB-70. The collision pitched his airplane up and rolled it across to top of the XB-70’s fuselage. The F-104 crashed into the Valkyrie’s twin vertical tails and exploded. Walker was killed instantly. The XB-70 was left with no directional control or stability. For reasons unknown. Carl did not eject.’
Why the F-104 collided with the XB-70
As usual after a headline-grabbing accident, news commentators hyperventilate, investigations are launched simulations are run, and fingers are pointed. Unfortunately, the man who was responsible, Joe Walker, perished in midair. So, the Air Force had to find someone else to blame. Four Air Force officers were charged with causing the crash.
They were Colonel Joseph F. Cotton, the XB-70 test force director; and Colonel Albert M. Cates, director of systems test at Air Force Flight Test Center; Lt. Col. James G. Smith the director of information (soon to be public affairs); and Lt. Bill Campbell, the chief of media relations at Edwards.

According to Campbell, at the urging of a Texas congressman and head of military appropriations, the Air Force convened a “collateral board”` to find those responsible.
The concept of wake vortex was little understood in 1966, but there was an invisible tornado constantly emanating from each of the XB-70 wings anytime it was aloft. It is likely that the F-104 got too close, was sucked into the powerful vortex, and was involuntarily hurled into the XB-70 vertical stabilizer.
The photos in this post feature the deadly photoshoot that caused the mid-air collision between XB-70 AV2 and F-104N N813NA.
North American Aviation In the Jet Age, The California Years 1945-1997 is published by Schiffer Publishing and is available to order here.
Photo by U.S. Air Force