Tool Marks: Fundamentals, Striations and Impressions

Tool taxonomy. Forensic tool-mark examiners classify tools by their mode of action rather than their common name, because the mode of action determines what type of mark the tool produces:

Cutting tools (knives, saws, chisels, tin snips, angle-grinder discs) have a blade edge that separates material by shear or abrasion. In cutting, the blade edge moves relative to the substrate, and the striated texture of the edge is transferred to the cut surface.

Shearing tools (bolt-cutters, wire-cutters, scissors, aviation snips) have two opposing blades that generate a compressive-then-shear action. The mark left on a cut wire or shackle has two faces: the cut surface (striated from the blade edge) and an end-deformation zone where the blades compressed and closed.

Gripping tools (pliers, pipe wrenches, vise grips) impress the serrated jaw texture into the gripped substrate. The mark reflects the jaw pattern geometry as a class characteristic and the jaw surface texture as a potential individual characteristic.

Prying tools (crowbars, screwdrivers, chisels, tyre irons) are used to lever material by applying force perpendicular or oblique to the working surface. When the tool's edge is pressed and then slid, it leaves a striated mark. When pressed without sliding, it leaves an impressed mark.

Drilling and punching tools (drill bits, punches, hole saws) produce circular or annular impressions, with the cutting flutes or punch face transferring their texture to the hole edge or punched-out disc.

Mechanics of mark formation. A tool mark is produced by one of two force geometries. In striated-mark formation, the tool slides against the substrate under force. The cutting edge, file face, or blade carries microscopic surface irregularities (asperities) from the manufacturing process and from previous use wear, and as the tool slides, each asperity either cuts a groove into the substrate or is deflected by the substrate's surface features. The result is a series of parallel grooves and ridges (striations) whose spatial pattern directly reflects the tool's surface texture at the point of contact.

In impressed-mark formation, the tool is pressed against the substrate without relative sliding. The compressed surface receives a three-dimensional negative impression of the tool face. A hammer face pressed into soft wood, a bolt-cutter blade pressed into a lead-seal, or a screwdriver tip pressed without slipping into a soft-metal screw head all produce impressed marks. The impression records the macroscopic geometry (class characteristic) and the microscopic surface texture (potential individual characteristic) of the contacting tool surface.

Striated marks carry the highest individual-characteristic content of any tool-mark category, because the striation pattern reflects the surface topography of the tool's cutting edge at the moment of marking. That topography develops from three sources: the manufacturing process, the finishing process, and use-induced wear.

Manufacturing striations. During production, the cutting edge of a tool is formed by a sequence of material-removal operations: casting or forging shapes the blank, grinding refines the edge geometry, and honing or filing puts the final cutting bevel. Each stage leaves microscopic marks on the surface. Grinding with a grinding wheel creates ridges and grooves parallel to the grinding direction; polishing blunts the peaks but does not eliminate the underlying striation pattern. Two nominally identical tools from the same production run will have similar macroscopic geometry (class characteristics) but different microscopic surface textures (potential individual characteristics) because the specific sequence of abrasive grits, grinding wheel positions, and tool orientations will differ in detail.

Use-induced modifications. As a tool is used, the edge acquires additional surface changes: small chips where the edge contacts a harder substrate, new wear grooves from specific contact events, and smoothing from repeated sliding. These changes progressively differentiate the tool's surface from other examples of the same make and model, and they also mean that the tool's striation-producing character changes over time. A mark made by a knife before and after a significant chipping event will not match each other, which is why the evidentiary reference is always to the specific condition of the tool at the time of the questioned mark.

Identifying striated mark features. Under a comparison microscope, striated marks appear as parallel lines in the substrate, running in the direction of tool travel. The class characteristics tell the examiner what kind of tool was used: the width of the marked channel, the depth of the striations, the angle of the edge, and the overall mark geometry. The individual characteristics are the specific widths, spacings, and shapes of individual striations that match (or do not match) the test marks made with the suspect tool. The AFTE Theory of Identification (1992, 2011 revision) provides the basis for this comparison, discussed in Section 4.

Impressed marks present a different analytical question from striated marks. The comparison is not between striation patterns but between the three-dimensional surface of the tool face and the three-dimensional impression left in the substrate. The answer depends on the level of detail captured in the impression and on the surface properties of the contacting tool face.

Substrate hardness and mark fidelity. A hammer face pressed into soft lead captures a very faithful impression: the microstructural features of the hammer face (machining marks, pit defects, surface scratches) are reproduced with high fidelity in the soft metal. The same hammer pressed into hardwood captures class characteristics (the overall face geometry) but the wood's grain structure partially obscures finer individual features. A hammer pressed into concrete captures only gross geometry. The substrate's ability to record fine surface features is described by its hardness and plasticity; the optimal substrate for individual-level comparison is one that is soft enough to capture fine features but rigid enough to hold them without flowing or elastic recovery.

Bolt-cutter marks on cut shackles. The bolt-cutter is the commonest tool in padlock-cutting and chain-cutting cases. The two blades close in a scissors action, each blade leaving a striated impressed mark on its half of the cut face. The cut surface shows a shear zone and a fracture zone; the striated shear zone on each half face is the region of greatest individual-characteristic content. Laboratory test cuts on reference material (similar-hardness shackles) with the suspect tool allow direct comparison. The UK Metropolitan Police forensic tool-mark unit, the FBI Materials Analysis Unit, and the RCMP Physical Evidence Section all recognise bolt-cutter-shackle comparison as a standard examination type.

Screwdriver tip impressions. When a screwdriver is used to pry (rather than to turn a fastener), the tip presses into a doorjamb, a window frame, or a safe face, leaving a compressed impression. The impression records the width, thickness, and tip geometry as class characteristics, and the surface texture of the blade face as potential individual characteristics. A flat-blade screwdriver pressed into painted softwood will leave a clean impression that captures even minor nicks in the blade edge. Phillips or Torx tips may leave patterned impressions that combine class geometry with individual surface marks in the lobes.

The Association of Firearm and Tool Mark Examiners (AFTE) provides the principal professional framework for tool-mark and firearm-mark comparison in North America. Its guidelines are referenced internationally through collaborations with ENFSI in Europe and by the UK Forensic Science Service successors and the Royal Canadian Mounted Police (RCMP).

Class characteristics are general features determined by the design and manufacturing specifications of a tool type. They allow the examiner to identify the type of tool (flat-blade screwdriver vs Phillips, bolt-cutter vs wire-cutter) and its approximate size range, but they do not distinguish between individual tools of the same type. Class characteristics include blade width, blade thickness, bite geometry, and overall profile.

Subclass characteristics are features produced during manufacture that are shared by a subset of tools made from the same batch, using the same tooling, or from the same production run. For example, a grinding wheel used to form blade bevels may produce a specific pattern on multiple tools before the wheel is replaced; those tools share a subclass characteristic. Subclass characteristics are a critical concept in exclusion logic: if two questioned marks share subclass features but the suspect tool was made in a different production run than the mark-making tool, a match conclusion based only on subclass features could be incorrect.

Individual characteristics are features random in occurrence, arising from random manufacturing processes or from use-induced modifications. They are unique (in theory) to a specific tool and are not shared with other tools. Individual characteristics are the basis for an AFTE identification opinion: "It is the conclusion of this examiner that the marked area exhibits sufficient agreement of individual characteristics to conclude that the marks were produced by the same tool."

The AFTE Theory of Identification (1992, revised 2011) defines the evidentiary conclusions available to the examiner as: identification (sufficient agreement), elimination (sufficient disagreement), and inconclusive (insufficient quality or quantity of marks for a conclusion). The theory does not provide a numerical threshold for "sufficient" agreement, a point central to the PCAST 2016 critique.

The 2016 Report of the President's Council of Advisors on Science and Technology (PCAST), titled "Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods," reviewed tool-mark identification as one of six feature-comparison disciplines. The report's finding on tool marks was more critical than its finding on latent fingerprints: PCAST concluded that tool-mark identification had only one study (the 2014 Baldwin and colleagues Ames Laboratory study) that attempted a properly designed black-box validation, and that this single study had methodological limitations that prevented drawing firm conclusions about the false-positive rate in real casework.

PCAST identified two foundational questions that any forensic feature-comparison discipline must answer: first, is the method foundationally valid (has it been shown through properly-designed studies to be capable of accurately matching questioned to known specimens), and second, has it been demonstrated to be applied at a reliable error rate in practice? For tool marks, PCAST found the scientific evidence for foundational validity was limited and the evidence for a demonstrated error rate essentially absent.

The response from the forensic community was divided. AFTE issued a formal statement disagreeing with PCAST's framing and pointing to the accumulated casework record as evidence of validity. The US Department of Justice declined to adopt PCAST's recommendation to require error-rate evidence before admitting testimony. Several state and federal courts, including the Superior Court of the District of Columbia in United States v. Tibbs (2019), cited PCAST in limiting or qualifying tool-mark testimony, though outright exclusion remained rare.

In 2024, the Organization of Scientific Area Committees (OSAC) for Forensic Science at NIST published updated guidelines for tool-mark comparison, incorporating requirements for inter-examiner agreement studies and range-of-error statements in reports. The Canadian Association of Forensic Sciences has adopted similar guidance for RCMP laboratories. The UK Forensic Science Regulator's Codes of Practice (version 7, 2023) require that any pattern-matching discipline demonstrate validation evidence consistent with the Codes' scientific validity requirements before evidence from that discipline is submitted to a UK court.

India's DFSS laboratories and the CFSL networks have not yet formally addressed the PCAST framing in published SOP documentation, though the BSA 2023 § 39 opinion-of-expert provision and the BNSS 2023 § 176 forensic-examination requirement both imply that expert opinion must be grounded in a validated scientific method.

Tool-mark identification evidence has been admitted in US federal and state courts since the 1930s, initially under the Frye general-acceptance standard and later under Daubert. Post-PCAST, the admissibility record became more varied. In United States v. Llera Plaza (3rd Circuit, 2002), the court reviewed latent fingerprint evidence under Daubert but ultimately admitted it; subsequent cases used similar reasoning for tool marks. United States v. Tibbs (D.C., 2019) is the most frequently cited post-PCAST federal decision, where the court admitted limited tool-mark testimony but restricted the expert to "more likely than not from the same source" language rather than "identification to the exclusion of all other tools."

In UK Crown Court practice, tool-mark evidence has been admitted routinely under the Criminal Justice Act 2003 expert-witness provisions. The FSR Codes require method validation and uncertainty quantification, which creates an implicit pressure toward the kind of black-box validation PCAST called for, but there has been no landmark UK appellate decision equivalent to Tibbs.

Canadian courts have admitted tool-mark evidence under the Daubert-equivalent R v. Mohan (1994) framework, supplemented by the independence and impartiality threshold from White Burgess Langille Inman v. Abbott and Haliburton (2015). Applied to tool-mark testimony, the combined framework requires the examiner to demonstrate methodology grounded in AFTE Theory of Identification, to articulate class-subclass-individual reasoning explicitly, and to satisfy the duty-of-court independence threshold before opinion evidence is admitted.

In India, tool-mark evidence is submitted through CFSL or state FSL reports and admitted under BSA 2023 § 39. Cross-examination under BSA § 142 and the Supreme Court guidance in Daxaben Harjivanbhai Patel v. State of Gujarat (2022) on expert-opinion evidence both suggest that expert testimony claiming a definitive tool identification without a stated uncertainty range may face challenge by an informed advocate.