This article describes a series of experiments that show the result of stepping in a blood-like liquid, and then stepping on another surface, so as to better understand the indications/lack of indications of blood stains at the Bundy crime scene.
There are several specific sub-issues to the general subject of recognizing tracks in blood. These include…
* Recognizing imprinted patterns (as by fabric or shoe soles)
* Recognizing shape: overall outlines or repeated shapes
* Smearing or wiping up stains, to obscure the source
* Reconstituting a stepped-in pool.
The first of these topics is discussed herein; the others will be the subject of later articles.
LATEX PAINT: During the criminal trial an issue was introduced by Dr. Henry Lee as to whether a wavy lined pattern on the end of the crime scene envelope was caused by the shoe of an otherwise unknown person at the crime scene.
(FIGURE 1, Crime Scene Envelope Detail. [H_LEE1.JPG].) FBI Special Agent Doug Deedrick conducted experiments that led him to believe the imprint was made from Goldman’s jeans. Though Deedrick’s experiments were not described in detail, we see some of the FBI’s procedures for studying imprints in the November 20, 1996 civil trial testimony of FBI Special Agent William Bodziak concerning his identification of the Bruno Magli shoes…
On the bottom [of a photograph exhibit he is describing] are three
impressions I made wearing Exhibit 395, the Bruno Magli shoe, the
right shoe, and those were made with latex paint where I actually
stood in a tray of latex paint, took a step out of the tray onto a piece
of paper.
So, it appears that the FBI is satisfied to simulate blood with latex paint for the purpose of making impressions. If it is good enough for the FBI, it is good enough for me. In the following, 100% latex paint was used as the transfer medium for pattern recognition, except in one case (Trial #4) where a 40% mixture of paint with water was used, and showed results that seemed too washed out and indistinct to be consistent with blood stains. (A full red color was used, except on the terra cotta Quarry Tile, for which pink was used for contrast.) FIGURE 2, Red Latex Paint Imprinted onto Arizona Flagstone. [TB_02.JPG]
PAVED SURFACE: The surface onto which the foot steps to leave an imprint is the “paved surface.” At the Bundy crime scene that surface was a cast concrete slab into which had been impressed a form to give the appearance of “grout lines” of a tile-paved walk. In fact, there was no tile in this area; all surfaces were concrete. Furthermore, the surfaces showed some roughness and were patterned during the finishing, as was attested to in the criminal trial by the rebuttal to Dr. Lee’s belief that he had seen footprints of a stranger on the walk. FBI specialists testified that what Lee saw were artifacts in the finished concrete surface which included both trowel marks and the footprint of a concrete worker who was doing the finishing. (See photographs 15.11 and 15.12 in Bodziak’s “Footwear Impression Evidence,” 2nd. Ed., 2000.) FIGURE 3, Closeup of Nicole’s Walk, Showing Surface Irregularities. [TB_03.JPG]
I have a cast concrete walkway in my back yard, but it is flat and does not have the patterned effect of Nicole’s walk. Nonetheless it could make an approximation for the imprint of bloodstains seen at Bundy, but there were a couple of practical drawbacks to using it. One problem was cleanup. Latex paint makes a fairly tenacious bond to concrete, and it would be a lot of labor to clean up extensive stains. More important, the only way to document the result is to take a photograph, in which there is a question of getting the scale right, the focus is imperfect, and it is difficult to make the photograph directly downward without creating shadows that appear in the image. It is more useful to make the imprint on a piece of thin paving tile (typically a foot square and ¼ inch thick) which can be let to dry and then moved to the bed of a scanner to make a contact scan. So, a few trials were made onto cast concrete surfaces for calibration and thoroughness, but most of the impressions were made onto portable paving surfaces. FIGURE 4, Cast Concrete Back Walk. [TB_04.JPG]
The following surfaces were tried…
Concrete Stepping Stone – Bought at a local brick yard for this experiment. It is extremely coarse-grained, but has a flat surface. FIGURE 5, Concrete Stepping Stone Imprinted with Work Sock. [TB_05.JPG]
Cast Concrete Walk – The aforementioned backyard walkway; flat and of medium roughness.
Float Finish Concrete – My front porch; flat and almost as smooth as glass, is an example of this. This was used for economy houses fifty years ago with the idea that the smooth finish did not require further (and expensive) covering. However, the surface is slippery (and somewhat unsafe) when wet, and it is my impression that this style of finishing has not been allowed in Los Angeles for exterior walking surfaces in several decades.
Glazed Tile – This and the other samples came from the floor covering department of Home Depot. The glazed tile is flat and very smooth – like the page of a glossy magazine; there is a slight colored random pattern below the glazed surface. 16” square, ¼” thick, the only sample I used that was bigger than 12” square. FIGURE 6, Imprints on Glazed Tile. [TB_06.JPG]
Semi-Glazed Tile – Glaze is applied to the surface of the tile with a mesh – similar to “silk screening,” but using a metal screen. This gives a small amount of surface roughness, and in some places the pattern of the screen can be discerned. The tile itself has some thick/thin surface pattern, and overall the surface varies in height by at least 1/16 of an inch, sometimes abruptly. (UPC# 014373 200181)
Unglazed Tile – This is flat, with a very absorptive surface. The “Maui” style from Home Depot was used.
Mexican Paving Tile – Appears to be unfired; a hard, flat, porous surface of light terra cotta color. One inch thick.
Quarry Tile – The only sample used that was smaller than 12” square, this was 6”. Flat, porous, terra cotta colored, this has a slight surface roughness, sand-grain sized; it is 3/8” thick. In order to give contrast between the imprinted paint image and the background, white paint with a small admixture of red (to also show up on white paper) was used. Both components were 100% latex paint and gave a pink result. FIGURE 7, Imprints on Quarry Tile. [TB_07.JPG]
Arizona Flagstone – Also from a brick yard, this comes in 12” squares, ¼ to 1 inch thick. An individual tile has a completely flat surface, but is rough, as though with fine grains of sand.
IMPRINT OF DENIM: The original objective of these experiments was to examine Deedrick’s conclusion that the imprint on the envelope came from Goldman’s jeans. It had occurred to me that the envelope imprint could have been the result of some other fabric, such as socks that covered the killer’s feet. I imprinted from three different pairs of jeans onto paper and several paving surfaces in 5 different trials of as many as a dozen impressions each.
I had earlier analyzed Figure 1 and determined that the line spacing was 15.4 lines per inch (lpi.) and this compared closely with 16 lpi that Bodziak claimed. Since Bodziak had better equipment and access to the original photographs, I consider that my work confirms his findings, and the line spacing on the Bundy envelope was 16 lpi. (Also, based on my own survey of shoe stores, I believe Bodziak’s claim that 16 lpi is too fine a pitch to be found on any shoe.) My own experiments in impressing a denim pattern gave a range of line spacing from 17 to 21 lpi. Considering that variation, it is reasonable that another sample of denim could give 16 lpi, and I believe that Deedrick’s 16 lpi is consistent with denim as a source. FIGURE 8, Image Imprinted from Denim Onto Paper. [TB_08.JPG]
As far as the shape of the impressed image (I have called it a “squiggle” shape), both on the envelope and in my experiments… They are indistinguishable to my eye. So, I am satisfied that the mark on the envelope probably came from a secondary transfer with a denim intermediary. However, Deedrick’s claim that this denim was the fabric of Goldman’s jeans seems unsupportable to me. From looking at images from three different samples of denim, I can not see a difference, and I think that all Deedrick can really say is that the envelope image came from denim, not necessarily Goldman’s jeans.
FOOT COVERING FABRIC: Having satisfied myself that the envelope image could have come from denim, I then wondered if it could not also have come from a sock, and so extended the experiments to imprint latex paint on various paving surfaces while wearing different socks. The very first experiments were conducted with the sock on the bare foot, then I discovered that clearer images resulted if the space between the sock and the foot was stuffed with an absorbent material, such as another sock; after trial #15 I standardized on a stuffing of four layers of Kleenex (doubled). The results of trials #8 and up use some procedure for stuffing the sock. (There was no impediment to mobility or surefootedness with any of these for sock stuffing techniques.) FIGURE 9, Image Imprinted from Work Sock Onto Paper.[TB_09.JPG]
Also, in preliminary experiments I resorted to old socks in my sock drawer, but quickly realized that the findings would not be reproducible, since they were the result of a variable and unknown amount of wear. So, all of the results reported here (except Trials #3, #9, #10, #19, and #23) used new socks. I bought my first lot at an inexpensive source (K-mart) and my second lot in a more expensive source (the Robinson’s May Co. department store). All of those were men’s work socks, dress socks, or casual socks, of one description or another. One of the socks I used, a casual sock that did not go high enough to cover the ankle (a “foot sock”) had different surfaces inside and out. As normally worn, the sock has a typical recognizable weave on the outside, but has an indistinct furry surface on the inside. A number of trials were made wearing this inside/out to see if the amorphous surface would be unrecognizable as a fabric (sometimes it was).
I also noticed as I progressed that the fabric pattern could be completely concealed on some surfaces with a finer line count in the fabric. So, I bought and used for some of the trials, women’s knee-high hose, which has an extremely fine fabric of very fine threads. FIGURE 10, Image Imprinted from Knee-Hi Hose onto Paper. [TB_10.JPG]
The short result of these experiments is to discover that all of the socks I studied produced images (where there was a fabric image, and not just a blob or a haze) that were very distinctive and conspicuously of a different shape than the squiggles of denim. (Except some of the dress socks produced a squiggle pattern, but so much finer it would not be confused with denim for that reason.) So, I can rule out a sock as the transfer medium for the image on the crime scene envelope. At that point, I became convinced that the envelope image came either from Goldman, or the killer’s helper, who may have been wearing denim jeans, and knelt on the corner of the envelope during the second phase of Goldman’s murder (he was squatting during the first phase). (The killer, himself, was wearing the sweatsuit with the blue-black fibers, you may recall.)
But, I was at that point set up to explore another question that has been contentious in AFOJS, and that is the claim that since no other footprints than Simpson’s were discovered by the police, there could not have been any other person there after the blood started to flow. The remainder of my work with socks was devoted to exploring the conditions under which blood could be tracked without leaving a print that was recognizable as having been left by a sock. (The issue here is not the concealment of a full footprint. In the article, “Finite Spreading Rate of the Blood Pool,” I showed that the main blood pool from Nicole’s slashed throat would not be of a sufficient extent by the time the killers fled – whoever that was – to step in with a full foot. However, there would be splatters, dribbles, and globs of blood that could be tracked around, and under many conditions a sock- or shoe-covered foot would reveal the fact by the detail in the partial imprint.) The representative imprint studied here is an inch high and three inches long, as caused by stepping in a pattern of paint that is a quarter of an inch by three inches in extent. FIGURE 11, Image Imprinted from Dress Sock onto Cast Concrete. [TB_11.JPG]
METHOD: The experimental method was extremely simple and straight forward. A teaspoon of paint was taken from the paint can and dribbled on an old newspaper. The sock-covered foot was placed in that, and then used to step on a double thickness of paper (to simulate the envelope). Then, I stepped on the paved surface, then back to the paper. The procedure was continued until the images had nearly faded into oblivion. Typically this took a dozen steps, but with some combinations the paint persisted for more than 20 steps. (This can not be extrapolated to the persistence of liquid on shoes, since the sock/Kleenex combination is more absorbent.) Conditions of the experiment were marked on both the paper and the paving beforehand; the step numbers were labeled after the trial. By this method, the odd-numbered steps were on paper and the even numbered steps were on the paved surface. When the paint was dry, the portable paved surfaces were subjected to a contact scan and the image saved. Non-portable surfaces were photographed; paper images were retained in file folders. FIGURE 12, Experimental Setup. [TB_12.JPG]
SOCK RESULTS: The typical sequence for the experiment is to progress through three phases:
1. The paint is so thick that the transferred image is an unrecognizable blob.
2. Beginning at the outside, and in later steps the whole image, may reveal the woven pattern of the sock, sometimes in almost photographic detail.
3. The image breaks up in the final steps, and eventually falls into invisibility.
However, there are important variations on this general pattern. In cases where the paved surface is rough and/or the weave of the fabric is fine, it never happens that fabric weave is imprinted on the paving in a recognizable way. In fact, I have formed the hypothesis that an “effective surface roughness” can be defined for a paved surface which is a roughness at which fabrics with a greater thread count will not reveal themselves. For example, Glazed Tile shows all imprints, even the finest fabric, Knee-Hi hose with a thread count of about 35 lpi. From that, I conclude that the effective surface roughness of Glazed Tile is greater than 35. On the other hand, Semi-Glazed Tile does not show the Knee-Hi pattern, but does show men’s dress socks at about 25 lpi. As a result, I conclude that the effective surface roughness of Semi-Glazed tile is between 25 and 35 lpi.
It is not a perfect model, but it is useful in predicting the general behavior of different fabrics on different surfaces.
SURFACE ROUGHNESS: The critical characteristic determining whether a pattern will reveal itself is the roughness of the paved surface. My own experiments demonstrated that almost any sock will not betray its pattern on my own back walk, or on the concrete stepping stone that I tried. These are both rough concrete surfaces – the stepping stone rougher than the walk. Now, Nicole’s front walk has been characterized as likewise being a rough concrete surface, and if it is rougher than my back walk, then it is warranted to believe that almost any sock will not have revealed its pattern at the Bundy crime scene.
In Figure 3, I showed Bodziak’s photograph of Nicole’s walk, obliquely lit to highlight surface irregularities. I made a similar oblique photograph of my own walk (yellowish color because outdoor film was used with an incandescent light source). I have scaled the two photographs to be the same size and show them side by side in FIGURE 13, Comparison of Surface Roughness. [TB_13.JPG]
It is a somewhat subjective call to judge roughness from these pictures, but I would say that Nicole’s walk is somewhat (maybe 50%) rougher than mine. However, I also think that Nicole’s walk is not so rough as the concrete stepping stones I used.
FOOTPRINTS/SHOE PRINTS: A couple of trials were also made with a full bare foot, and with a full imprint of a flat soled (no pattern) shoe. This was done on both glazed tile and the back walk. FIGURE 14, Full Prints on Glazed Tile. [TB_14.JPG]
Although the outline of the object was obvious as being a shoe or a foot, the fine detail pattern could be discerned on the tile, but not on the concrete.
CONCLUSION: Since a wide variety of socks were tried that did not show their pattern when Latex paint stains were impressed on cast concrete here, I believe they would not have revealed themselves at the crime scene either. If the killers at Bundy wore almost any kind of socks -- or were barefooted -- while they worked, the weave of the socks, or the fine lines on the bottom of the bare foot, would not be recognized by the police later.
FABRIC PRINT EXPERIMENTS
Standard on Crime Scene Envelope -- Squiggle at 16 lpi.
FEBRUARY 11, 2002, 100% Latex Paint, 80°F. Concrete Walk, Existing Socks
NO. TIME SOURCE PAPER PATTERN TILE LPI
1 3:00 p Levi Leg Squiggle, 2-4 haze 18
2 3:10 p New Work Sock Herringbone, 1-9 haze 20
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 14, 2002, 100% Latex Paint, 74°F. Glazed 16”x1/4” Tile, Existing Socks
3 4:30 p Old Dress Sock Squiggle to 12+ recogz 32
FEBRUARY 15, 2002 (4) 2:50 PM, 40% Latex Paint: Smudged, Indistinct Levis pattern on paper.
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 15, 2002. 100% Latex Paint, 74°F, Semi-Glazed 12”x1/4” Tile, K-Mart Socks
(014373 200181)
5 3:00 p Levi Knee Squiggle to 11+ recogz 19 ½
6 3:30 p Small Lycra Sock Chevrons, weak, 3/12 recogz 20
7 3:40 p M. Casual (foot) sock Arch, weak, 7/13 recogz 11
8 3:50 p Cotton Crew Sock over 2 Double rows, 3/11 recogz 18 ½
9 4:00 p Old Dress Sock over 3 Squiggle, 2/6 recogz 28
10 4:10 p Old Brwn Sock over 2 Waffle, squiggle @7 recogz 15
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 16, 2002, 100% Latex Paint, 64°F, Raw Mexican 12”x3/4” Tile, MayCo Socks
11 3:10 p Dribble on Levi Leg Squiggle, 5/8 recogz 21
12 3:25 p Khaki Dress Sock +2 Chevron, 5/11 recogz 9 (x3)
13 3:35 p Peds over 2 socks Double Squiggles, 5/13 recogz 23
14 3:45 p Casual (foot) sock I/O Waffle, 9/23 recogz 15
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 17, 2002, 100% Latex Paint, 58°F, Semi-Glazed, 12”x1/4” Tile)
15 4:00 p Worn Levi Pocket Poor image, recognizable recogz 19 ½
16 4:30 p Knee-Hi over 4 Kleenex Square, 95% unrecogz haze 23
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 18, 2002, 100% Latex Paint, 64° Arizona Flagstone
17 2:15 Gray Work Sock + 4Kx Chevron recogz 18
18 2:40 Knee-Hi over 4 Kleenex Square Grid haze 40
19 3:10 Old Dress Sock + 4 Kx Squiggle 5% 31
FEBRUARY 18, 2002 Concrete Stepping Stone
20 2:20 Gray Work Sock + 4 Kx Chevron haze 18
FEBRUARY 18, 2002 “Maui” Tile
21 2:25 Gray Work Sock + 4Kx Chevron recogz 20
22 2:35 Knee-Hi + 4 Kleenex Square Grid 15% 35
FEBRUARY 18, 2002 Mexican Raw Paving Tile
23 3:00 Old Dress Sock + 4Kx Squiggle recogz 21
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 19, 2002, 100% Latex Paint, 65° Concrete Walk (back)
24 2:40 Nylon Dress Sock + 4K Squiggle haze 24
25 3:25 Cotton Dress Sock+4K Various, weak haze 23
FEBRUARY 19, 2002 Arizona Flagstone
26 2:45 Nylon Dress Sock + 4K Squiggle recogz 26
27 3:20 Cotton Dress Sock+4K Chevron recogz 23
FEBRUARY 19, 2002 Semi-Glazed Tile
28 2:50 Nylon Dress Sock + 4K Squiggle recogz 25
29 3:10 Cotton Dress Sock+4K Squiggle recogz 24
FEBRUARY 19, 2002 Mexican Raw Paving Tile
30 Nylon Dress Sock + 4K Squiggle recogz 26
31 3:15 Cotton Dress Sock+4K Chevron recogz 24
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 20, 2002, 100% Latex Paint (white tinted with red), 71°F, Quarry Tile, terra cotta
32 12:35 Knee-Hi + 4 Kleenex Mesh recogz 34
35 12:45 Foot Sock, In/Out Vague recogz 11
38 1:00 Gray Work Sock Herringbone recogz 21
39 1:30 Cotton Dress Sock Herringbone recogz 19
FEBRUARY 20, 2002, 100% Red Latex Paint, Cast Concrete Walk
33 12:40 Knee-Hi +4 Kleenex Lines haze 35
34 12:50 New Levis Squiggle haze 17
37 ? Foot Sock, In/Out Vague haze ?
FEBRUARY 20, 2002; 100% Latex Paint, Semi-Glazed Tile
36 1:45 Foot Sock, In/Out Vague haze 17
NO. TIME SOURCE PAPER PATTERN TILE LPI
FEBRUARY 21, 2002, 100% Latex Paint, 87°F., Float Finished Concrete Porch.
40 1:30 Knee-Hi + 4 Kleenex Mesh recogz 29
41 1:35 Foot Sock, Inside/Out Cobblestone recogz 15
42 1:40 Nylon Dress Sock Chain recogz 25
43 1:45 Gray Work Sock Chevron recogz 11 (x3)
FEBRUARY 24, 2002, 100% Latex Paint, 62°F., Glazed Tile
NO. TIME SOURCE PAPER PATTERN TILE LPI
44 10:15 Full Shoe, Plain Sole Recognizable recogz --
45 10:30 Full Bare Foot Recognizable recogz --
46 10:45 Partial Bare Foot SMEAR -- variable --
FEBRUARY 25, 2002, 100% Latex Paint, Cast Concrete Walk
47 2:00 Full Bare Foot, SMEAR
48 2:15 Bare Foot, DRY WIPE
49 2:30 Bare Foot, DAMP WIPE
MARCH 1, 2002, 64°F. - 68°F. Cast Concrete Stepping Stone
NO. TIME SOURCE FINAL IMAGE
50 10:30 Size 12 M “Venturini” Shoe in 67% Latex @ 2 Min. ABORTED
51 11:15 Shoe in Pint of 100% Latex @ 2 Minutes Ebbs 90%/50% in sole/heel voids
MARCH 2, 2002, 66°F, Cast Concrete Stepping Stone
NO. TIME SOURCE FINAL IMAGE
52 10:30 Shoe in 67% Latex @ 22 Minutes No Recognizable Imprint
MARCH 2, 2002, 70°F. Raw Concrete Stepping Stone
53 11:40 Shoe in 100% Latex @ 22 Minutes Ebbs 30%/10% in sole/heel voids
FIGURE 15: Sequence of Impressions on Glazed Tile. [TB_15.JPG]
SOCKS RESULTS MATRIX
(visibility of stuffed foot covering on various surfaces)
SURFACE
A
B
C
D
E
F
G
Concrete Stepping Stone
0
0
0
0
0
0
0
Concrete, Cast, no Texture
0
0
0
0
0
Concrete, Float Finished
100%
100%
100%
100%
Tile, Glazed
100%
100%
100%
100%
100%
100%
100%
Tile, Semi-Glazed
100%
0
100%
100%
100%
0
Tile, “Maui,” Unglazed
100%
15%
Mexican Paving Tile, Raw
100%
100%
100%
100%
100%
Quarry Tile, Terra Cotta
100%
100%
100%
100%
Arizona Flagstone
100%
5%
0
100%
100%
INFERRED
A: Work Sock (gray)
B: Dress Sock (old, black)
C: Knee-Hi Women’s Hose (Pay-Less)
D: Casual Foot Sock
E: Dress Sock (new, black, nylon)
F: Dress Sock (new, black, cotton)
G: Casual Foot Sock, inside-out
LEVIS RESULTS MATRIX
SOURCE
SURFACE
LPI
TILE RESULT
Worn, 2/17
Semi-Glazed Tile
19 ½
visible
Current, 2/11
Cast Concrete Walk
18
haze
Current, 2/15
Semi-Glazed Tile
19 ½
visible
Current, 2/16
Raw Mexican Paving Tile
21
visible
New, 2/20
Cast Concrete Walk
