Microscopic Hair Comparison

History 

The use of micro­scop­ic hair com­par­i­son in crim­i­nal cas­es can be traced back to the mid-19th cen­tu­ry, with one of the ear­li­est exam­i­na­tions occur­ring in 1847 dur­ing the mur­der inves­ti­ga­tion of Duchesse de Praslin in France. The the­o­ret­i­cal foun­da­tion for micro­scop­ic hair analy­sis was estab­lished by French crim­i­nol­o­gist Edmond Locard, who proposed that every con­tact leaves a trace,” sug­gest­ing that peo­ple con­tin­u­ous­ly trans­fer par­ti­cles of hair and oth­er evi­dence with­out know­ing. In the United States, micro­scop­ic hair analy­sis was first con­sid­ered in 1882 in Knoll v. State, where an expert tes­ti­fied that hair found in a sus­pec­t’s wheel­bar­row matched the vic­tim’s hair in length, mag­ni­tude, col­or, and in every oth­er respect.” However, this case was reversed on appeal, with the court warn­ing that such evidence was of a most dangerous character.” 

Despite these ear­ly con­cerns over its valid­i­ty, micro­scop­ic hair com­par­i­son analy­sis devel­oped into a com­mon foren­sic tech­nique. The process gen­er­al­ly involved exam­in­ing the mor­phol­o­gy of hair strands, includ­ing the scale type of the cuti­cle, pat­tern of the medul­la, and pig­men­ta­tion in the cor­tex. Analysts used pat­tern recog­ni­tion to com­pare unknown hairs to sam­ple hairs, typ­i­cal­ly reach­ing the con­clu­sion that the hairs were con­sis­tent with” each other or micro­scop­i­cal­ly indis­tin­guish­able.” The FBI, which first start­ed using micro­scop­ic hair com­par­i­son in the 1950s, main­tained that prop­er­ly trained tech­ni­cians fol­low­ing the cor­rect pro­ce­dures should reach the same conclusions. 

Problems with micro­scop­ic hair analy­sis began to emerge as DNA test­ing became avail­able. Despite the advance­ment in DNA test­ing, hair com­par­i­son con­tin­ued to be used when DNA test­ing was deemed too time-con­­sum­ing, expen­sive, or oth­er­wise unavail­able. In 1994, the U.S. Department of Justice estab­lished a task force to review cas­es han­dled by an overzeal­ous gov­ern­ment wit­ness. By 1996, the DOJ report­ed that 28 peo­ple, many con­vict­ed based on micro­scop­ic hair com­par­i­son, had been exon­er­at­ed through DNA test­ing. A 2002 study com­par­ing micro­scop­ic hair com­par­i­son with mito­chon­dr­i­al DNA sequenc­ing con­clu­sive­ly demon­strat­ed that DNA typ­ing was far more accurate. 

The unre­li­a­bil­i­ty of micro­scop­ic hair com­par­i­son had seri­ous con­se­quences. Roger Coleman was exe­cut­ed in 1992, and hair com­par­i­son played a sig­nif­i­cant role in his con­vic­tion, despite lin­ger­ing ques­tions about his inno­cence. While results of hair com­par­i­son are sub­jec­tive, the pros­e­cu­tor in Mr. Coleman’s case told his jury that based on sci­en­tif­ic analy­sis of the hair, It would be extreme­ly unlike­ly that any­one else would have hair that would be con­sis­tent with this hair.” In an inter­view, the tri­al judge lat­er acknowl­edged that the expert wit­ness at the tri­al nev­er com­pared the hair found on the vic­tim in Mr. Coleman’s case to any­one oth­er than Mr. Coleman. 

In April 2015, the FBI announced that a review of their micro­scop­ic hair com­par­i­son analy­sis tes­ti­mo­ny in 268 cas­es revealed FBI expert wit­ness­es gave erro­neous tes­ti­mo­ny in 96% of cas­es. Eventually the FBI flagged near­ly 3,000 files where ana­lysts may have pro­vid­ed improp­er hair com­par­i­son evi­dence. A 2018 update from the FBI revealed that erro­neous tes­ti­mo­ny was found in 93% of 484 reviewed cas­es. Nine peo­ple whose con­vic­tions par­tial­ly relied on improp­er hair com­par­i­son had already been exe­cut­ed, and five more had died in prison.

Standard of Care

Macroscopic char­ac­ter­is­tics of hair, those that can be observed with­out a micro­scope, include length, col­or, and tex­ture. However, these char­ac­ter­is­tics alone are not enough to deter­mine a hair’s ori­gin. The first steps in the hair com­par­i­son process involve deter­min­ing whether the unknown hair sam­ple is of human or ani­mal ori­gin. Microscopic char­ac­ter­is­tics of hair include the medul­la, pig­men­ta­tion of the cor­tex, and the types of scales present in the cuti­cle. These char­ac­ter­is­tics dif­fer in human and animal hair.

Labeled dia­gram of a hair follicle.

OpenStax College, CC BY 3.0 <https://​cre​ativecom​mons​.org/​l​i​c​e​n​s​e​s​/​b​y/3.0>, via Wikimedia Commons

The medul­la, or inner­most lay­er of the hair shaft, if present, is made up of pig­ment. The medul­la may be a con­tin­u­ous, unbro­ken line of pig­men­ta­tion, or may be inter­rupt­ed, frag­ment­ed, sol­id, or not exist at all. An inter­rupt­ed medul­la includes pig­men­ta­tion bro­ken up in reg­u­lar inter­vals, while a frag­ment­ed medul­la has pig­men­ta­tion bro­ken up in uneven incre­ments. A hair strand with a sol­id medul­la means that pig­men­ta­tion is found through­out the entire hair, both in the medul­la and the cor­tex (largest part of a hair shaft, lay­er sur­round­ing the medul­la). Generally, the cor­tex of human hair con­tains pig­ment gran­ules, or melanin, that give the hair col­or. Pigment dis­tri­b­u­tion varies from indi­vid­ual to indi­vid­ual, as well as the size of the pig­ment gran­ules found in the cortex.

While pig­men­ta­tion in human hair tends to be denser towards the out­side of the cor­tex, pig­men­ta­tion in ani­mal hair tends to be denser towards the medul­la. Human hairs have gen­er­al­ly con­sis­tent col­or and pig­men­ta­tion through­out a strand, where­as ani­mal hair shows vary­ing col­or changes through­out a strand. Another promi­nent dif­fer­ence appears in the medul­la: In humans, the medul­la is thin and takes up no more than 1/​3 of the hair’s total width. In ani­mals, the medul­la tends to be much wider (half or more than half the hair’s total width), and more well-defined. The cuti­cle, or out­er­most lay­er of a strand of hair, is made up of over­lap­ping scales that pro­tect the inner lay­ers of the hair. Different mam­mals have dif­fer­ent hair cuti­cle pat­terns: rodents have coro­nal cuti­cles that look like a stack of crowns, while cats have spin­ous scales that resem­ble petals. The scales of human hair are flat and nar­row, called imbricate. 

A human black hair surface in light microscope. Parts of the surface layer can be seen

Human hair under a microscope

Juan de Vojníkov, CC BY-SA 4.0 <https://​cre​ativecom​mons​.org/​l​i​c​e​n​s​e​s​/​b​y​— s​a/​4.0>, via Wikimedia Commons

For ani­mal hairs, exam­in­ers can deter­mine the spe­cif­ic species, though they can­not defin­i­tive­ly deter­mine the sin­gle ani­mal the hair came from. Through analy­sis, ani­mal hairs recov­ered from evi­dence might estab­lish con­nec­tions, poten­tial­ly tying an indi­vid­ual to a crime scene, vehi­cle, or loca­tion of a vic­tim. When exam­in­ing human hairs, ana­lysts must look at the arrange­ment of scales, dis­tinc­tive appear­ance, and dis­tri­b­u­tion of hair char­ac­ter­is­tics across dif­fer­ent hair regions. 

When com­par­ing a known hair to an unknown hair, an exam­in­er must use a com­par­i­son micro­scope, which is made up of two com­pound light micro­scopes con­nect­ed by an opti­cal bridge. This con­fig­u­ra­tion allows an exam­in­er to look at a known and unknown hair simul­ta­ne­ous­ly. There are typ­i­cal­ly two con­clu­sions an exam­in­er can reach: (1) the hair samples are not con­sis­tent with” each oth­er or (2) micro­scop­i­cal­ly indis­tin­guish­able.” This micro­scop­ic com­par­i­son can only deter­mine whether the unknown hair sam­ple exhibits sim­i­lar char­ac­ter­is­tics to the known sam­ple and whether it could have orig­i­nat­ed from the same source. 

If these hairs are deemed to be con­sis­tent with one anoth­er based on micro­scop­ic find­ings, exam­in­ers may use DNA and blood pro­tein test­ing to deter­mine if they came from the same indi­vid­ual. Microscopic assess­ment is per­formed first because it is more cost effec­tive and faster than DNA and blood pro­tein test­ing. Individual strands of hair can only be attrib­uted to an indi­vid­ual if the fol­li­cle is attached to it. If a hair is forcibly removed, the fol­li­cle may stay intact, allow­ing ana­lysts to exam­ine the fol­lic­u­lar tag, where blood and tis­sue from the indi­vid­ual may be present. DNA test­ing of this mate­r­i­al pro­vides iden­ti­fi­ca­tion to a high lev­el of con­fi­dence, where­as sim­ple hair com­par­i­son can­not pro­vide any defin­i­tive iden­ti­fi­ca­tion of an individual.

Concerns with Microscopic Hair Comparison 

Microscopic hair com­par­i­son analy­sis suf­fers from sig­nif­i­cant sci­en­tif­ic lim­i­ta­tions, as the analy­sis relies on sub­jec­tive assess­ment from exam­in­ers rather than objec­tive, quan­tifi­able mea­sure­ments, allow­ing for incon­sis­ten­cy between ana­lysts. This tech­nique can only indi­cate whether a ques­tioned hair is con­sis­tent with” a known sam­ple, cre­at­ing a poten­tial class of con­trib­u­tors, rather than iden­ti­fy­ing an individual source. 

Forensic ana­lysts have his­tor­i­cal­ly over­stat­ed the sig­nif­i­cance of their deter­mi­na­tions in tes­ti­mo­ny, imply­ing lev­els of cer­tain­ty that sci­ence sim­ply does not sup­port. As found in the National Registry of Exonerations’ (NRE) December 2023 report, Microscopic Hair Comparison Analysis and Convicting the Innocent and the FBI’s review of micro­scop­ic hair com­par­i­son analy­sis cas­es, FBI experts rou­tine­ly used erro­neous state­ments to make the evi­dence sound stronger than it had sound­ed in writ­ten reports.” Testimony from these experts repeat­ed­ly exceed­ed the lim­its of the sci­ence in ques­tion. NRE’s review of exon­er­a­tions involv­ing micro­scop­ic hair com­par­i­son analy­sis deter­mined that at least 129 indi­vid­u­als had been false­ly con­vict­ed based at least in part on this evi­dence, with 15 of these exonerees hav­ing been sen­tenced to death.

Even when the tes­ti­mo­ny was care­ful­ly phrased, it con­tributed to the con­vic­tion of inno­cent people…I sus­pect that this is because the appear­ance on the stand of an expert wit­ness, who said they were using sci­ence and the evi­dence impli­cat­ed the defen­dant, led the jury to think that the evi­dence must have been pow­er­ful evi­dence of guilt, even if the expert avoid­ed say­ing it explicitly.”

Simon Cole, Director of the National Registry of Exonerations, in an inter­view with Mother Jones.

Case Studies 

Gary Nelson 

Gary Nelson spent 11 years on Georgia’s death row after being con­vict­ed of the 1978 rape and mur­der of six-year-old Valerie Armstrong. 

The case began in February 1978, when Valerie and her eight-year-old friend S.W. had been play­ing togeth­er, and they decid­ed to walk to the store. When Valerie did­n’t have mon­ey for can­dy, she told S.W. she would ask Uncle Al” or Alphonso Swinton, for mon­ey. At this time, Mr. Swinton shared a house with Gary Nelson. According to S.W., when they arrived at Uncle Al’s house, they part­ed, and the last time she saw Valerie, she was walk­ing toward the door while a man worked on a car in the dri­ve­way. Valerie’s body was dis­cov­ered the next day in the woods behind her house, with a bro­ken knife found nearby. 

Mr. Nelson, who had pri­or minor run-ins with the police and lived in the area, became an imme­di­ate sus­pect. Although he vol­un­tar­i­ly met with police a week after Valerie’s body was found and allowed his house to be searched, he was arrest­ed in May 1978. Mr. Nelson was indict­ed by a grand jury in October 1979. The state had no direct evi­dence link­ing Mr. Nelson to the crime — no eye­wit­ness­es, fin­ger­prints, or con­fes­sion — only cir­cum­stan­tial evi­dence based on hair analysis. 

The hair evi­dence was cen­tral to the pros­e­cu­tion’s case against Mr. Nelson. Roger Parian, direc­tor of the Savannah Branch of the State Crime Laboratory, tes­ti­fied that he had mount­ed a hair found on Valerie’s body on a slide, exam­ined it, and deter­mined it came from the arm of a Black per­son. On the stand, he claimed he could nar­row down the source of the hair to about 120 Black peo­ple out of Chatham County’s 60,000 Black res­i­dents, with Mr. Nelson includ­ed among those 120 indi­vid­u­als. District Attorney Andrew J. Ryan III empha­sized this evi­dence in his open­ing state­ment to the jury, telling them that the hair that was found on the body and the known hair from the arm of Gary Nelson have, in (Mr. Parian’s) opin­ion, the same origin.” 

Based on this and oth­er evi­dence, the jury con­vict­ed Gary Nelson and sen­tenced him to death. Shortly there­after, an Atlanta law firm agreed to rep­re­sent Mr. Nelson on appeal pro bono. After ten years of inves­ti­ga­tion, the defense team uncov­ered cru­cial evi­dence that pros­e­cu­tors delib­er­ate­ly with­held despite prop­er motions and a direct order from the tri­al judge to com­ply with discovery. 

The hair evi­dence, which had been piv­otal to the state’s case, fell apart. Contrary to Mr. Parian’s sworn tes­ti­mo­ny, he had nev­er actu­al­ly exam­ined the hair found on Valerie Armstrong’s body. Rather, at his request, the FBI’s crime lab in Washington, D.C., exam­ined the hair and sent its report to Mr. Parian. The FBI concluded that this hair is not suit­able for sig­nif­i­cant com­par­i­son pur­pos­es.” This report had been with­held from Mr. Nelson’s coun­sel. Myron Scholberg, the for­mer Unit Chief of the Microscopic Analysis Unit of the FBI’s crime lab, pro­vid­ed an affi­davit stating that the limb hair obtained from the vic­tim could have come from any Black per­son, includ­ing, but not lim­it­ed to, oth­er sus­pects in this case of the vic­tim.” This affi­davit exposed Mr. Parian’s false tes­ti­mo­ny and the sup­posed foren­sic evi­dence link­ing Mr. Nelson to the crime scene. 

The pros­e­cu­tion’s files con­tained sub­stan­tial evi­dence that was delib­er­ate­ly with­held from the defense: state­ments from mul­ti­ple wit­ness­es cor­rob­o­rat­ing Mr. Nelson’s ali­bi that he was else­where when Valerie dis­ap­peared; con­crete evi­dence point­ing to some­one else as the per­pe­tra­tor, includ­ing a con­fes­sion; and infor­ma­tion that mul­ti­ple wit­ness­es saw Valerie after she left the house where Mr. Nelson was alleged­ly work­ing on his car. 

Despite the acknowl­edg­ment of the with­held evi­dence, in May 1990, Judge E. Byron Smith upheld Mr. Nelson’s con­vic­tion, claim­ing the with­held evi­dence would not have made a dif­fer­ence in the tri­al out­come. In November 1991, the Georgia Supreme Court unan­i­mous­ly dis­agreed with Judge Smith and held that Mr. Nelson’s con­vic­tion and death sen­tence were obtained ille­gal­ly. Then-Chatham County District Attorney Spencer Lawton, who had not pros­e­cut­ed the case, acknowledged, There is no mate­r­i­al ele­ment of the state’s case in the orig­i­nal tri­al which has not sub­se­quent­ly been deter­mined to be impeached or con­tra­dict­ed.” The dis­trict attor­ney chose not to retry Mr. Nelson, and he was released from prison on November 61991

Sources

Rene Ebersole, How the Junk Science of Hair Analysis Keeps People Behind Bars, Mother Jones, December 15, 2023; David S.H. Funes, Kaitlyn Bonilla, Mathieu Baudelet, Candice Bridge, Morphological and chem­i­cal pro­fil­ing for foren­sic hair exam­i­na­tion: A review of quan­ti­ta­tive meth­ods, Forensic Science International, 2023; Samuel D. Hodge and Amelia Holjencin, A Post-Mortem Review of Forensic Hair Analysis – A Technique Whose Current Use in Criminal Investigations is Hanging on by a Hair, Saint Louis University Law Journal, April 7, 2020; Douglas W. Deedrick, Hair, Fibers, Crime, and Evidence, Part 1: Hair, FBI Laboratory Division, July 2000

See also Killing Justice: Government Misconduct and the Death Penalty