The "Lapel Flip" as Reflected Light

The Handhold Specular Goes Dark in 224

			2.0X 1.5X 1.0X 0.5X 0.25X

The handhold specular disappears in frame 224. This change is coincident with JBC's chest going dark. With a clothing movement explanation, this is just a coincidence. This is not just a case of two things happening at the same time.  The events I am discussing are occurring in one frame and both are  caused by less light reaching the camera from those areas which could be explained by less light reaching those areas. It seems obvious that JBC's clothes moving have nothing to do with what is happening on the handhold so maybe clothing movement has nothing to do with what is happening on JBC's chest, either.  One thing can be said for certain about the handhold specular. If the reflection is from the vertical part of the handhold, then it can't be coming directly from the sun (explained below) so the light striking the handhold would have to be reflected from somewhere.   If the reflection was coming from the right front of the car and it traveled near the side of the car then the same light source could be illuminating both the handhold and JBC's chest. 

The Basics of Reflections

 In order to follow the arguments presented, some basic knowledge of the properties of reflections is required, but not much. Since the handhold is a cylinder, the way light is reflected off cylinders is of particular importance and will be discussed in more detail. If you learned anything about reflections in school, then you learned how specular reflections work even if they weren't called that. A mirror is an example of an extremely specular reflective surface.  The light reflects "true" in that it does not spread so that one can actually see an image.  Light that is parallel remains parallel after the reflection if the surface is flat.  A surface that spreads the light some but not a lot will not produce clear images but still can produce bright, directional specular reflections.  The direction of the reflected light depends on the direction of the incident light and that of the surface normal.  The surface normal at a point is the vector (direction) that is perpendicular to the surface at that point.  (Diagram)  The surface normal direction is the same everywhere on a flat surface but not on a curved surface. It is different at different points.  Whether a surface is flat or curved, at a given point the incident ray, reflected ray, and surface normal all lie in a plane with the surface normal bisecting the angle subtended by the incident and reflected rays.  This is the law of reflection; angle of incidence = angle of reflection.  The diagram says it better.  For simplicity, I will discuss reflections off vertical cylinders, i.e., ones with a vertical longitudinal axis.  The cylinder tends to spread light around horizontally so that being able to see a specular reflection from it and the position of the specular on the reflecting object is not very sensitive to the horizontal location of the observer.  The visibility of the specular is very sensitive to the vertical location of the observer, however.  If the incident light is parallel such as sunlight, then the specular can only be seen in a vertical band whose height is the length of the cylinder.  (More on reflections off cylinders)  A reflection from the rounded corners can be seen over a wide range of directions because the surface normal points in a wide range of directions.  The location of the reflection does not change much with changing illumination angles or changes with observer position because the points on the corner where the surface normal is right for two different observer positions  are not far apart.  More simply, the corner is not very big so the reflection could not move very much and still be on the corner.

The Handhold Specular Revisited

There are three parts of the handhold on which the reflection might lie which are he top, the side,  and the corner..  A specular has already run across the top by frame 222 and the angles are wrong for a reflection from there.  The car would have to turn quite a bit to the right to make it come back and the road curves left.  The surface normals on the corner sweep out a large volume of space so that a reflection from there is seen over a wide range of viewing angles if,sunlight is striking it.  This also means that the reflection will not be particular bright when viewed from anywhere and it will be small.   Zapruder is looking down at the handhold at about 5.5 degrees.  The sun is about 37 degrees above the horizon.  That means Zapruder was not nearly in the right position to see the specular from the side if it is directly illuminated by the sun.  He would need to be about 37 degrees below the car.  The road slopes down at about 3 degrees but that can't change the direction of any light reflected off the car by more than 6 degrees.  The slope of the road actually makes the reflection a bit lower in the horizontal direction of Zapruder.  Any light that is reflected as a specular back to Zapruder off the handhold would have to be moving in a nearly horizontal direction.  It should in fact be traveling upwards at about 5.5 degrees plus a bit more for the slope of the road or about 8 degrees in world coordinates if the slope of the road is 3 degrees.  That would translate to 5 degrees in limo coordinates.  As we have seen, we can't locate the horizontal position of the source of the reflection because the reflection can be seen over a wide horizontal angular range.  We can determine its approximate vertical position, however, or more precisely, the vertical angle of the incident light.

What Part of the Handhold is the Reflection Coming From?

222	223	248

The images on the left show the handhold in 222, 223, and later in 248.  The car was closer in 248 so the image was scaled for comparison.  The reflection disappears in 224 and when we can first clearly see this area again in 230, our view of the handhold is blocked by the hat.  It appears obvious in 222 and 223 that the reflection is not from the top.  In 222 the reflection looks as though it is below the corner but in 223, not so clear.  Generally, they appear too big and bright for a reflection off the corner in those two frames.  The surface normal changes a lot over small distances on the corner.  As a result the corner spreads light over a wide range making the reflection visible over a wide range but, for the same reason, it doesn't send much in any one direction.  There is no specular visible from the handhold from the time the hat moves enough to see it in about  233 until 247 when a smaller, higher specular appears.  It remains more or less constant until the view is blocked by the sign on the lamppost in frame 268.  It reappears on the other side of the sign in frame 269.  In the next frame it disappears for two frames and then reappears in frame 272. This can all be seen in this clip of Zapruder frames 263-280 .   There is plenty enough change in the vertical and horizontal angles to make the specular appear to move if it was coming from the horizontal or vertical parts and the illumination is coming from the sun.    Roy Kellerman who is sitting directly in front of JBC is leaning back when the handhold specular disappears in 270.  Kellerman's head (if the light is coming directly from the sun) or body could be blocking the sunlight for the next two frames.   Kellerman's shadow must be falling on the corner of the handhold until 247.  The road turns to the left so the car probably does as well.  This would move Kellerman's shadow to the left in the car.  By 247, it has turned enough to move his shadow off the corner of the handhold.  The reflection seen in 222 and 223 is rom the side of the handhold.  The horizontal direction (azimuth) of the source of this reflection can't be determined but if it is coming from in front of JBC then it is headed toward his chest.

The Sharp Left Edge

The brighter blue area on JBC's chest in 222 and 223  has a sharp boundary on the viewer's left in 223. Since the shutter of Zapruder's camera was open nearly half the time, half of all movement will occur during the exposures. Any clothing movement rapid enough to cause as much change as seen between 222 and 223 and then 223 and 224 would leave a blurry boundary. Frame 222 has a fairly wide intensity (brightness) gradient at the left edge of the bright area.  In 223, there is a narrow gradient (sharp boundary) at the left edge.  (Intensity graphs)  The intensity graphs reveal that the two bright spots on the (viewers') left of the boundary of the bright are in 223 have very sharp boundaries but there is no edge between them.  The area between the spots is nearly  as dark as the dark area to the left.  People have a tendency to think of the frames as instantaneous snapshots.  In fact, the shutter was open nearly half the time the camera was running so half of the motion that occurs happens during the exposures.  The stripes are too sharp for the amount of vertical blurring in 223.  The intensity graphs show that there are narrow, fairly sharp edges between the stripes.  One explanation is that the stripes are actually bright spots that are moving during the exposure.  This results in streaks (or stripes).  There isn't much blurring because the spots are not over any one point on the film very long.  The motions of the spots and the blurring determine the direction of the streaks.  I currently don't know in which direction the streaks form but am assuming that they formed left to right for this discussion. The  stripes in 223 are bright spots moving across the rear window frame and the bright spots on JBC result from the top and bottom stripe moving a  little past the edge of the window frame and spilling off onto his jacket. As can be seen in the images on the left, the window frame has a curve at the bottom.  The intensity graph of 223 also shows a curvature along the left boundaries of the spots. The curved left boundary is the edge of the shadow of the window frame.  The curvature is exaggerated on JBC's chest because he is not perpendicular to the illumination.  If he is rotated by 45 degrees then the horizontal axis on JBC's chest is stretched by 40%.  If he is rotated by 60% then the stretching is 100%.  The left edges of the spots on the jacket are sharp because the spot is on the edge of the window frame very briefly producing a strobe-like effect. The image above and to the left has lines drawn from the spots on the jacket to the top and bottom stripes.  This method is inherently inaccurate and the top line is drawn too high but it shows that the windshield chrome is at least in the ballpark as the source of the illumination.

JBC's Left Side Gets Darker, Too

When the bright area goes dark in 224, JBC's left side gets darker as well.. I compared the intensities of some areas away from JBC in 223 and 224 as controls. The area sampled on JBC's right side is to his right of the bright area, i.e. the dark part, in 223.

Sample location        (avg intensity 224)/(avg intensity 223)

Connally's left side             0.86

Connally's right side            1.04

Roy Kellerman                    1.01

Jackie's shadow                  0.97

It seems the left side of JBC's jacket should be unaffected by the right side opening up or bulging out.  For some reason, there is more illumination on JBC's  left side in 223 than there is in 224 but the difference is not nearly as much as in the "lapel flip" area.  This might just be due to JBC turning more toward the sun.   

The Hat Blocks the Reflection (maybe)