Testing a Mexican Memento Box for the Use of Blood in an Inscription Signed by Frida Kahlo

 

This article describes and evaluates alternative forensic procedures that were used to test the red inscription inside of the lid of an antique wooden memento box that was the property of and was inscribed and signed by artist Frida Kahlo. [1] Tests were performed at the Centro de Investigación Frida Kahlo to determine whether or not the inscription was written using blood or a mixture of blood and other materials. The author presents the results of those tests and makes recommendations for the use of these procedures for screening other works of art for the possible presence of blood. ̶ © 2007 Daniel Friedman

 

The owners of this item and the extensive collection of which it is a part, Carlos Noyola (an antiquarian) and Leticia Fernandez (an art conservator), expressed an interest in evaluating the possibility that artist Frida Kahlo used actual blood, perhaps mixed with other paint pigments, to write the inscription or illustration found inside the lid of this box. Our initial hypothesis was that although Kahlo refers to blood (sangre), and although she used the blood image frequently in her paintings, drawings, and in sketches included in many letters and on pages in her diaries, the inscription on this box was probably comprised of a red paint pigment. We speculated that had actual blood been used, and considering that the box is approximately 60 years old, the inscription should have been browner in color. However because some artists are believed to have on occasion mixed blood or still other body fluids into paint pigments, we wanted to perform a definitive test for evidence of blood on this object.

 

Application and Evaluation of Methods to Screen for Blood on Works of Art

 

With advice from crime scene forensic experts Judy and Don Doje, educators and crime scene forensic suppliers in Ocoee, Florida, we selected several tests that could be performed in the laboratory established by Noyola/Fernandez in the Centro de Investigación Frida Kahlo (CIFK) in San Miguel de Allende, Guanajuato, Mexico.

 

Three different forensic tests for blood were considered: The Hemastix® Reagent Strip[2], The Phenolpthalein Forensic Presumptive Blood Test[3] (known as the Kastle-Meyer test), and the Luminol® test for blood on surfaces[4]. The first two of these were applied, tested and documented in procedures described in this article. We adapted and tested three sample preparation methods using the Hemastix® Reagent Strip and one sample preparation method using the Phenolpthalein Forensic Presumptive Blood Test. Because of the negative results of the first two tests, the Luminol® test for blood on surfaces was held in reserve for future use.

 

Using the Bayer Hemastix® Reagent Strip to Test for the Presence of Blood in or on Works of Art

 

Using Hemastix® reagent strips to test for the presence of blood is considered the most sensitive test available. Hemastix® is a plastic strip to which is affixed a reagent area (Photo 2) that tests for occult blood in urine.[5]The test strip is normally immersed briefly in a urine sample. The reagent surface area is then observed for any color change (which occurs in less than 30 seconds), and then the color of the reagent surface is compared with a color comparison chart provided by the Hemastix® manufacturer. The resulting color visible on the reagent area ranges from orange through green. Very high levels of blood may cause the color development to continue to blue.

 

This test was designed for medical use to screen for minute amounts of blood in human urine and will respond to trace levels of hemoglobin as low as 0.015-0.062 mg/dL of free hemoglobin, with lesser concentrations being detected under some circumstances. The test is equally sensitive to myoglobin as to hemoglobin. Detecting blood at such dilute concentrations is approximately equivalent to detecting 5-20 intact red blood cells per microliter of fluid.[6]

 

 

The test procedure for examining the Frida Kahlo wooden box used a sterile swab to collect pigment or other particles from the surface being examined. The sample-collecting swab is used dry (or if necessary dampened with sterile water), touched to the sample area, and then dampened (if a dry swab was used), and touched to the blood test sensing area of the Hemastix® reagent strip. For the purpose of testing for the presence of blood in or on works of art, such as Frida Kahlo's inscription on the memento box in this study, the sample collection method for use with the Hemastix® reagent strip was tested using three different methods:

 

 

1. Dampened swab method: A swab dampened with sterile water was used to collect pigment and particles from the test surfaces. In each of the damp swab tests, the dampened swab was touched first to the surface to be tested (Photos 6 & 7 below) and second, it was immediately touched to the rectangular surface area of the Hemastix® test strip where the reagent is contained. The reagent test strip color was then read against the color chart (Photo 2).
   It was observed that this method collected an excessive amount of pigment which may have occluded the test result (heavily pigmented swab at right in Photo 3), leading to the decision to continue Hemastix® reagent strip testing using the next two sample preparation methods described below.

 

 

 

 

2. Dilution method: [7] A dampened swab was used to collect pigment and particles from the test surface, then it was immersed in 15ml of sterile water (swab in Photo 4). The Hemastix® reagent strips were then dipped briefly into this liquid and read against the color chart. This method was tried because the damp swab collected excess pigment that may have interfered with a proper reading of the reaction color of the reagent strip. Doje's warns that this method could produce a false negative result so it was used only in the case of excess pigment on an already-collected sample swab.
   
   

 

 

 

 

 

 

 

3. Dry swab method:[8] Because the damp swab sometimes collected an excessive amount of pigment that may have occluded the test result, the procedure was repeated starting with collection of pigment from the Frida Kahlo box inscription test area using a dry sterile swab. The dry swab was then moistened with a single drop of sterile water and the moistened swab was touched to the directly to the reagent strip and read against the color chart (Photo 2 above).

    

   

    

   In the Dry Swab Method (Photo 5), reagent strip #1 was a control that was touched with a swab moistened only with sterile water. Strip #2 was touched with a swab that had been used dry to sample the red-pigmented inscription area on the box, then moistened slightly with sterile water and touched to the reagent area. Strip #3 was touched with a swab that had been used dry to sample our known blood source on a clean wood surface, then moistened slightly with sterile water, and touched to the reagent area. Strips #1 and #2 show no evidence of blood. The red visible on swab #2 is pigment from Kahlo’s box inscription. Reagent strip #3 turned green, confirming the detection of blood from the known-source control.

 

Areas of the Frida Kahlo memento box tested and test controls were selected as follows:

 

1. Pigmented box inscription: a small section of the Kahlo box top interior red-pigmented inscription was selected for sampling. We selected a heavily-pigmented area and one which would not, if disturbed, obscure the original text.

 

 

                                                                                 

 

 

 

 

 

 

2. Non-pigmented box surface: an area on the Kahlo box top interior was selected where there was no apparent inscription and no visible paint or pigment.

 

 

 

 

 

 

 

3.      Known blood reference: a clean wood surface was treated with a known blood-source[9]. This test should produce a positive result in a test for the presence of blood, confirming that the test procedure could detect the presence of blood on a wood surface.

 

 

 

 

 

 

4. A clean, sterile wood surface (a physician's tongue depressor or spatula): a sterile swab, dampened with sterile water, was used to sample the surface of a sterile wooden spatula. This test should produce a negative result in a test for the presence of blood. The sterile water was also tested with Hemastix® reagent strips before each use of that liquid.

Table of Results of Tests for Blood in Frida Kahlo's Signature & Inscription Using the Hemastix® reagent strip

 

The table below summarizes the results of all three variations using the Bayer Using Hemastix® reagent strips to test for the presence of blood. The test results were consistent among all three sample preparation methods, but the "Dry Swab" sampling method was found to be the least invasive and to produce the least difficulty with an overloaded sample while the "Dilute Method" risks a false negative and is generally not recommended.

 

Table I - Blood Presence Test Results Using Hemastix® reagent strips
Test Sample	Dampened Swab	Dilute Method	Dry Swab
Red Pigmented Inscription on wooden box lid	Negative, possibly occluded by excessive pigment	Negative	Negative
Non-pigmented surface on wooden box lid	Faint green, weak positive, possible blood/contaminant[10]	Faint green, weak positive, possible blood/contaminant	Faint green, weak positive, possible blood/contaminant
Known blood reference on wood	Positive	Positive	Positive
Clean, sterile wood surface	Negative	Negative	Negative
Sterile water source	Negative	Negative	Negative

 

False Negative Results using Hemastix® Reagent Strips to Test Artworks

 

Hemastix®  Method #1 (wet swab), #2 (dilute method), and even #3 (dry swab) can all produce a false negative result (indicating that blood is not present when in fact blood is present on the surface), particularly if the sample is over-diluted. Therefore we use the dilute Method #2 only in cases where the sampling swab carries an excessive amount of pigment or debris from the artwork.

Photograph 9 shows four Hemastix® test strips containing various amounts of red-pigmented material from Kahlo’s inscription inside the box lid. Three results showing only red pigment may have obscured a faint positive reaction (causing a false negative conclusion). One test strip (center of three strips at right) contains a faint green (positive for blood) reaction - the same reaction obtained when sampling an area of the box top interior that contained no pigment at all. This could be a false positive reaction and it is discussed next.

 

 

 

 

 

False Positive Results using Hemastix® Reagent Strips to Test Artworks

 

Hemastix® Method #1 (wet swab), #2 (dilute method), and even #3 (dry swab) can all produce a false positive result (indicating that blood is present when in fact it is not), particularly in the presence of a contaminant containing an oxidant.

The sample labeled #2 shown in this photograph was collected from an un-pigmented "plain" wood area of the interior of the box top – an area where a negative result would be expected (no blood detected). But the sample produced a light green reaction on the Hemastix® test strip as shown in this photo. Obtaining a "positive" Hemastix® reaction to the control sample from an un-pigmented (no red inscription) area of wooden surface appeared to represent a false positive. Since the box being tested is at roughly 60 years old, was stored in unknown conditions, and because the artist (or time or storage conditions) could have applied other materials to the work, there may be several explanations including the presence of superficial debris containing other oxidants.

 

Because the Hemastix® method can produce false positives if other oxidates are present, it was important to perform additional blood presence essays using the most-specific Phenolpthalein presumptive blood test method described below.

Test 2 for the Presence of Blood: Using the Phenolpthalein presumptive blood test to test for blood in or on works of art

The Phenolpthalein Forensic Presumptive Blood Test (shown here) is considered the most specific test for the presence of blood on a surface. This bloodstain evidence test kit was designed for forensic use and includes sterile collection media with three reagent chemicals used in this order: ethanol, phenolphthalein reagent, and hydrogen peroxide. The sample turns pink in the presence of blood from humans or other animals. This is a non-destructive test, which is intended to leave the sample intact. The reaction of heme in hemoglobin (in blood) with the hydrogen peroxide is a catalytic reaction, which makes this test very sensitive to very small quantities of blood in the sample. [11] [12] 

 

 

The phenolphthalein presumptive blood test is a highly specific test used to assess potential false positive results of the other blood presence tests on works of art (and in crime scene forensic investigative work). [13] The photograph shows our dry wipe samples of (#6) a non-pigmented area of the Kahlo box top interior and (#7) a pigmented area before treatment of these samples with the reagent chemicals. Using test controls similar to those described for the Hemastix® procedure, the phenolphthalein presumptive blood test involves using an alcohol-dampened sterile swab (using ethanol) or using a dry sterile swab (which will collect less pigment from the test surface) or a sterile filter paper can be used to wipe-sample the surface to be tested. Since in our test of the Kahlo box lid interior there was light pigment visible on the dry folded filter paper, we did not need to use alcohol to moisten the filter paper before applying the three chemicals as catalytic reagents.

 

The dry, folded sterile filter paper was gently rubbed across the surface area to be sampled. Each sample was then treated with in sequence with a drop of ethanol, phenolphthalein [14], and then with hydrogen peroxide [15]. The treated test samples were immediately observed for a color change to pink, which would indicate that blood is present. The samples must be examined closely and rapidly to watch for the color change, and the test sample should be diluted as little as possible.

 

Phenolpthalein presumptive blood test and test controls

 

1.1. Pigmented inscription on art work to be tested for blood: a sterile filter paper was folded into quarters (for convenience) and is then used to collect a small sample of pigmented material from the area to be tested for presence of blood. The photo at left shows the red pigment collected by this sample after the ethanol step. This is pigment from the inscription, not a positive blood-reaction.

 

 

 

 

The phenolphthalein presumptive blood test results showed a negative result on this sample of the pigmented area of the Kahlo memento box, confirming that the pigmented area did not contain blood. The slight pink color at the tip of the folded filter paper shown in the photo at left was slightly diluted but otherwise unchanged from the red pigment collected when the box inscription was dry wiped with the paper.

 

 

 

 

 

Phenolpthalein presumptive blood test controls

1.2. Non-pigmented surface of the same art work: a sterile filter paper was folded into quarters (for convenience) and was then used to collect a small sample of non-pigmented material from the area to be tested. For this project an area of the interior of the wooden box top was selected which was not coated or painted with any visibly detectable paint or pigment.

 

The phenolphthalein presumptive blood test results showed a negative result on the sample of the non-pigmented area of the Kahlo memento box, confirming that the earlier modest positive reaction when using the Hemastix method was a false positive. (Photo 15 shows sample after reagents were applied.)

1.3. Known blood reference: a sterile filter paper used to collect a small sample of material from the reference sample of known human blood which had been previously placed on the surface of a sterile wood spatula and dried. This test should produce a positive result in a test for the presence of blood.

The phenolphthalein presumptive blood test results showed a positive result when performed on a dry-wipe filter paper sample of the known human blood source (a drop of blood placed on a sterile wood spatula and dried), confirming that the chemistry of this test was functioning normally.

 

Photograph 16 shows the characteristic pink stain appearing after the sequential application of ethanol, phenolphthalein, and hydrogen peroxide to this sample.

 

 

 

1.4. Bare filter paper: A sterile filter paper was used with no sample material, to confirm the normal behavior of the reagent chemicals when they are applied in proper sequence.

This test should (and did) produce a negative result in a test for the presence of blood.

[16]

 

 

 

 

Using Luminol® to test for the presence of blood in works of art

 

The Luminol Blood Test is considered a highly-sensitive widely-used screening test for blood on surfaces and is used [usually at a crime scene] to test for the presence of blood on surfaces even when blood is not visible to the naked eye under normal lighting.[17] The Luminol mixture is prepared and sprayed or applied by sterile swab onto an area or surface to be screened for blood. The area is viewed in total darkness. Luminol will luminescence in total darkness if it detects the presence of even minute amounts of blood - in other words, the preparation will glow in the dark. Luminol is thus a highly sensitive screen for presence of blood on surfaces where blood may not be obviously visible in normal lighting.

 

However in ordinary forensic use such as at a crime scene, Luminol is made to be sprayed upon a surface containing dilute or old bloodstains. Obviously one should not spray artwork with this or any other reagent mixture since doing so could damage the work. In the case of Luminol, Don Doje explains that the perborate (or hydrogen peroxide) could damage the artwork, depending on the chemistry of the work's composition.

For the tests discussed in this article, Doje recommended a modified procedure for using Luminol to screen works of art for blood while minimizing the chances of damage to the artwork itself. Used as recommended by Doje, this test procedure should not damage the work. For testing works of art, the Luminol is mixed with water while omitting the perborate until ready for use. [18]

 

 

A sterile swab, dampened with the Luminol/water mixture, is touched to the test area on the surface of the artwork. After sampling the artwork's surface, a catalyzing agent, hydrogen peroxide, is applied directly to the test swab (rather than mixing the Luminol-perborate combination and applying that mixture directly to the artwork).

 

Use of the Luminol Test for Blood when Overcoming Sampling Limitations of the Hemastix® Reagent Strip or Phenolpthalein reagent method:

In examining the Kahlo box top interior inscription, the author did not make initial use of the Luminol blood screening procedure because prior tests using both the Hemastix® reagent test strip for blood and the Phenolpthalein presumptive blood test procedure both indicated that no blood was detected. However we were concerned that the use of a small amount of blood mixed with paint pigments and solvents might have been obscured using the two forensic methnods discu8ssed above.

We found that the Phenolpthalein Forensic Presumptive Blood Test was particularly useful in testing the Frida Kahlo memento box inscription because of the concern that the red pigment collected on our filter paper or cotton swab might itself obscure a positive response indicating the presence of blood. Because this test will cause the presence of blood in a sample to be visible in total darkness as the issue of pigment in the sample when we used the Hemastix® Reagent Strip method is overcome. In follow-up testing we used this test to investigate this possibility further. Those results will be published as an update to this article.

 

Conclusions From Testing the Frida Kahlo Wooden Box for Evidence of Blood

 

1. The Kahlo box inscription does not appear to contain blood. Two forensic procedures indicated that the sampled area of red inscription on the interior of the lid of the Frida Kahlo memento box, while it refers to Kahlo’s blood, does not in fact contain human nor animal blood.

 

2. The Hemastix® reagent strip test method worked well as a sensitive screen of artwork for the presence of blood. Even when a dry swab of a known blood source contained no material visible to the naked eye, the reagent detected blood from the sample with no ambiguity. In three variations of sample preparation, the Hemastix® test produced a negative result (no blood detected) for the pigmented area of the box lid, but it produced a weak positive result (possible blood detected) for an un-pigmented, un-painted area inside the same box top, probably because of non-blood contaminants on the test surface. False positives are possible with this method if there are surface contaminants. False negatives can also occur: weak but valid positive results might be obscured by excessive sample pigment which happens to be the same color as a reaction-color of the reagent strip. Therefore thoughtful sample preparation is essential, obtaining "enough" but not "too much" material. The reagent is so sensitive to blood that very little sample material is needed to produce a reliable result.

 

3. The phenolphthalein presumptive blood test method was sensitive to and easily demonstrated the presence of blood in the control sample. Importantly, this procedure worked well to eliminate an apparent false positive test result with the Hemastix® reagent.[19] Because of the wide variety of possible contaminants that can be present on the surface or within pigmented materials on works of art, especially old objects, it is important to have a false positive screen available, such as the Phenolpthalein presumptive blood test procedure. This test is an important compliment to the Hemastix® test but it must be performed properly and with an understanding of the test’s catalytic chemistry since the phenolphthalein + perborate-wet sample will turn pink when it dries. A novice could mistake this end-reaction as a positive indication of blood.

 

4. It is important that any sampling and tests performed on works of art be conducted so as to protect the integrity of the art work, to avoid damaging it, and at the same time, the test must remain chemically and technically valid. Because it removes the least amount of material, is least likely to affect the surface of a work of art, and because it avoids overloading the swab with pigment or other surface debris, the author prefers the dry-swab Hemastix® method as a starting point for testing works of art for the presence of blood, and dry filter paper wipes when using the phenolphthalein presumptive blood test.

 

5. It is also important that the experimental design of the test procedure include controls of a known blood sample, controls for background contamination, and that other controls are used to identify false positive reactions.

 

Author

Daniel Friedman is a forensic microscopist, paint failure analyst, and building and environmental investigator residing in the United States and Mexico. His education, credentials, and experience can be read at www.inspect-ny.com/danbio.htm and his CV is available at www.inspect-ny.com/DJF-CV.htm .

 

Technical Reviewers

• Judy and Don Doje, Doje's Forensic Supplies, P.O. Box 500, Ocoee, FL 34761
• Ulrik Runeberg, Conservador, Museo de Arte Contemporaneo de Puerto Rico, San Juan, Puerto Rico
• Leticia Fernandez, Conservador, Centro de Investigación Frida Kahlo, San Miguel de Allende, Guanajuato, Mexico. [NOTE: Ms. Fernandez is a proprietor and owner of and provided the item under investigation in this article.]
• Bayer Health Care LLC, Elkhart, IN [Review requested, no response received as of 24 September 2007]

 

References

• Doje's Forensic Supplies, P.O. Box 500, Ocoee, FL 34761 Tel: 407.880.8149 Fax: 407.880.8150 http://www.dojes.com, Judy and Don Doje assisted in specification of the forensic procedures used in this study and were our supplier of forensic supplies for the tests performed.
• "Hemastix® Reagent Strips for Urinalysis, Test for Blood in Urine", Bayer Health Care LLC, Elkhart, IN 46514, product literature AN07016C, Rev. 5/06 USA (C) 1992, 2003 Bayer Health Care LLC

 

FridaKahloPurebadeSangreDF.doc - 29 Sept 2007 - DRAFT -

© Daniel Friedman, all rights reserved

http://www.inspect-ny.com/FridaKahloSangre.htm