Analysis of Burned Human Remains - Mercyhurst College ...

U.S. Department of Justice 

Office of Justice Programs 

National Institute of Justice 

Solicitation SL# 000802: Research and Development in Forensic Anthropology and Forensic Odontology. 

Grants.gov Funding No. 2008-NIJ-1793 

Project Title: Recovery and Interpretation of Burned Human Remains 

Principal Investigators: 

Steven A. Symes, Ph.D, D.A.B.F.A. 

Dennis C. Dirkmaat, Ph.D, D.A.B.F.A. 

Stephen D. Ousley, Ph.D. 

Department of Applied Forensic Sciences, 

Mercyhurst Archaeological Institute, 

Mercyhurst College 

Erie, PA

Index 

Index ………………………………………………………………………………………………….………. i 

Abstract ……………………………………………………………………………………………………… iii 

Project Narrative: 

Purpose of the Project .……………………………...………………………………………………….… 1 

Current State of the Problem - Review of Relevant Literature …………………………………….... 3 

Fire Modification to Bone ……………………………………………………………………….........…. 3 

Fire Alteration of Skeletal Trauma Evidence ………………………………………………………….. 4 

Fatal Fire Scene Recovery ……………………………………………………………………………… 5 

Project Goals and Objectives ……………………..……………………………………………………... 7 

Research Design and Methods ……………………………………………………………………….…. 8 

Research Component 1. Recovery of Burned Human Remains ………………………………….…. 9 

Archival Research …………………………………………………………………………...……. 9 

Processing Mock Fatal Fire Scenes ………………………………………………………….…. 11 

Processing Actual Fatal Fire Scenes ……………………………………………………….…… 15 

Research Component 2. Analysis and Interpretation of Heat Altered Bone: Normal 

Burn Patterns ………………………………………………………..……………………………………. 17 

Research Component 3. Heat Alterations in Traumatized Bone ………….…………………………. 21 

Implications for Policy and Practice …………………..………………………………………………… 24 

Management Plan and Organization ………………………………………………………………...….. 25 

Timeline ……………………………………………………………………………………………………… 25 

Dissemination Strategy …………………………………………………………………………………… 25 

Appendices 

Appendix 1: References Cited ……………………………………………………………………………. 26 

Appendix 2: List of Previous and Current NIJ Awards .………………………………………………. 30 

Appendix 3: Figures – Case Examples………………………………………………………………….. 31 

Appendix 4: Data Collection Form Employed for the Archival Pilot Study …………………….…. 35 

Symes, Dirkmaat and Ousley – Project Narrative - Solicitation: No 2008–NIJ–1793 - SL# 000802 - page i

Appendices (Cont.) 

Appendix 5. Example of Mock Scene (Results and Procedures) ………………………..………. 37 

Appendix 6. Example of Real Scene (Results and Procedures) ………………………..………. 39 

Appendix 7. Example of Mapped Burn Patterns in the Human Body …………………………… 41 

Appendix 8. Timeline …………………………………………………………………………………..… 43 

Appendix 9. List of Key Personnel ……………………………………………………………………. 44 

Appendix 10. Resumes of Key Personnel ……………………………………………………………. 46 

Steven A. Symes ……………………………………………………………………………………….. 48 

Dennis C. Dirkmaat …………………………………………………………………………………….. 63 

Stephen D. Ousley …………………………………………………………………………………….. 97 

Gregory O. Olson ………………………………………………………………………………………. 108 

Appendix 11. Letters of Support ………………………………………………………………………. 117 

Symes, Dirkmaat and Ousley – Project Narrative - Solicitation: No 2008–NIJ–1793 - SL# 000802 - page ii

Abstract 

Victim remains at fatal fire scenes are typically difficult to detect, recover and handle. All of the burned 

material at the scene, including biological tissue, is often modified to a similar appearance, and bones, in 

particular, become discolored, brittle, and highly fragmented. As a consequence, these remains are often 

missed, disturbed, altered, or even destroyed during scene processing with the existing protocols. 

The added postmortem fracturing, fragmentation and bone loss resulting from these recovery techniques 

hinder the already difficult task of autopsy and laboratory analysis of burned human remains. This is especially 

problematic for bone trauma analysis, as its most immediate goal is distinguishing perimortem (forensically 

significant) trauma, from postmortem (not forensically significant) alteration. The substantial addition of trauma 

features created by fire and then recovery can result in a daunting analytical task. 

Lack of on-scene recordation of relevant information related to body positioning and contextual relationships 

of remains as well as other physical evidence at the scene, further complicate trauma analysis, biological profile 

estimation, and event reconstruction. For the trauma analyst, it is arguably difficult to detect and characterize 

atypical, potentially forensically significant trauma, if the extent of exposure of individual portions of the body to 

fire is unknown. In addition, very little and often contradictory information regarding what is considered “normal” 

fire alterations of the human body has been presented. This information lacuna notably includes specific burn 

sequences of soft tissue and patterns of hard tissue modification (e.g., fracturing, color changes, distortion and 

shrinkage). The same problem affects estimates as simple and relevant as whether a missing element was ever 

present at the scene, missed during recovery, or totally consumed by the fire. 

The proposed project addresses these problems by linking rigorous scene recovery and documentation 

methodologies with subsequent laboratory analyses (in particular, bone trauma analysis) of heat altered human 

remains from fatal fire scenes. This will be accomplished by: 1) developing and testing effective fatal fire scene 

recovery protocols and guidelines, to maximize the location, documentation and recovery of biological tissues 

(including bone), while minimizing postmortem bone alteration and damage due to collection and transport 

Symes, Dirkmaat and Ousley – Project Narrative - Solicitation: No 2008–NIJ–1793 - SL# 000802 - page iii

methods, 2) precisely documenting “normal” soft tissue burn sequences and resulting bone modification (bone 

color changes and fracturing patterns) in fully fleshed human bodies, burned under controlled (crematorium) 

conditions, and 3) analyzing the macro- and microscopic effects of fire and heat on previously well-described 

diagnostic characteristics of skeletal trauma (primarily, fractures) and tool marks in fresh bone. 

Symes, Dirkmaat and Ousley – Project Narrative - Solicitation: No 2008–NIJ–1793 - SL# 000802 - page iv

Purpose of the Project 

Fire scenes that include human victims provide some of the most difficult investigative challenges for fire 

responders, investigators, forensic experts, and agents of law enforcement. In addition to determining the 

cause/origin of fire, investigators must reconstruct circumstances of death of the victim, utilizing evidence related 

to cause and manner of death. In normal (non-burned) cases, this evidence is derived from body location and 

positioning at the scene, and identification of perimortem trauma. 

Both of these assessments are significantly more difficult if the scene and the victim are subjected to fire. It 

is not surprising that fire is such a common method for attempting to conceal evidence of criminal activity inflicted 

on human victims. Fire may be used by the perpetrator to: 1) totally destroy the body, 2) destroy features 

allowing for victim identification (facial features, fingerprints), or 3) destroy evidence related to the circumstances 

surrounding the death. Consumption of soft tissues by fire significantly hampers or totally impedes analysis by 

other specialists (such as forensic pathologists) and, therefore, analyzing burned human remains is a common 

task ascribed to forensic anthropologists. 

The forensic anthropological analysis of any set of human remains, whether burned or not, first requires an 

understanding of the context of the remains and the identification of specific factors that have led to disturbance 

of the elements from their original anatomical position, loss of bone, and modification of individual elements . 

Contextual scene information is best understood when documented and recovered via exacting 

archaeological methods, which are absent in current fatal fire scene recovery protocols. As a result, forensically 

significant contextual data pertaining to body location and positioning within the burned structure are not carefully 

noted (particularly via precise mapping). 

In addition, burned human remains and other evidence may go undetected or worse, damaged by recovery 

efforts. Fatal fire scenes are often much more complex not only because the body and individual elements are 

dramatically modified by fire, but because the entire surrounding contextual environment is also modified in the 

same way, resulting in a homogeneous coloration and intermingling of all materials. This modification makes it 

Symes, Dirkmaat and Ousley – Project Narrative - Solicitation: No 2008–NIJ–1793 - SL# 000802 - page 1

difficult to distinguish the body and burned skeletal elements from the surrounding substrate debris, increasing 

the chance of missing some of the burned and fragmented remains during a cursory examination of the scene 

(Figure 1). If not detected, these important skeletal remains may even be trampled upon at the scene (Figure 2). 

Additionally, given the frail and brittle nature of burned bone, transport methods that include placing the body in 

flexible body bags, make them susceptible to further fragmentation and postmortem damage before they even 

get to the forensic pathologist’s examination table. The significance of these problems for fatal fire case 

investigation, and the importance of recovering all osteological materials present at the scene, can be better 

understood after considering the case discussed in Figure 3. 

Once and if the surviving detected bones arrive to the laboratory, one key inference to be made for their 

forensic anthropological analysis is whether a missing skeletal element was actually present at the scene before 

the fire, or consumed by combustion. A consideration of recovery methods employed is vital (as discussed 

above), but a reasonable assessment of the probability of the missing element surviving a typical burn episode, 

is also needed. This determination requires an in-depth knowledge of typical burning sequences and 

modification patterns of each area of the body, including normal fire-induced postural changes, the role played 

by soft tissue, and specific color changes in bones related to temperature and exposure time factors. This critical 

information regarding how human bodies burn in fires, however, is currently not available in the literature. 

Fire alteration of human bone is seemingly more detrimental to the analysis of skeletal trauma in the 

laboratory. The typical analysis and interpretation of perimortem skeletal trauma in non-burned bone relies 

primarily on the understanding of timing of bone modification in the case of fracture patterns, or the type of 

implement used to inflict tool marks or cut marks on the bones. The introduction of additional bone fractures 

resulting from fire modification (without even considering recovery and transport factors), further complicates the 

analysis and requires an understanding of how bodies are typically modified by fire. 

With respect to tool marks on bones, much information regarding general tool class characteristics of the 

inflicting tool or implement can be obtained through macroscopic, but especially microscopic analysis of the 


inflicted mark in unburned bone (see Relevant Literature section).Before interpreting the absence of a diagnostic 

feature in a fracture or cut mark as evidence that a particular tool was not used as the inflicting weapon, it is 

imperative to assess whether these diagnostic traits could have been altered or erased as a result of the fire. A 

review of the current anthropological literature indicates that we do not have a clear answer to that query. 

In consideration of the issues discussed above, it becomes obvious that in order to perform an appropriate 

forensic anthropological analysis of burned human remains it is necessary to: 1) ensure that all surviving skeletal 

elements and information related to body location and positioning, as well as other relevant taphonomic 

influences, are recovered at the scene, and that further alterations of the evidence from scene to laboratory are 

minimized, 2) recognize the extent and relative sequence patterns in which different body areas are typically 

consumed by the fire, taking into account tissue thickness and the temperatures typically reached depending on 

body positioning, and 3) be cognizant of the effect of fire on the diagnostic traits employed to infer fracture and 

tool class characteristics in fresh unburned bone, as well as their limitations and necessary modifications when 

applied to burned bone. In this consideration, the scene, fire, and osteological analysis are intimately 

interrelated. The purpose of this proposal is to improve all three of these essential and interrelated components 

of the forensic anthropological analysis of burned bone. 

Current State of the Problem - Review of Relevant Literature 

A review of the extant anthropological literature reveals a striking scarcity of research regarding fatal fire 

scene processing, normal burn patterns of the body as a whole, and unambiguous effects of fire on skeletal 

trauma evidence. Instead, most of the research on burned bone has focused on heat alteration of individual 

bones (often sans flesh). 

Fire Modification to Bone: In spite of the long history of research on this topic, our current knowledge on the 

subject of burned bone suffers from two key problems: 1) a lack of consistency in terminology and study sample 

design and outcomes, and 2) a low applicability of the results to real forensic contexts. 


To a great extent, the lack of agreement of the results of previous research on burned bone, and even the 

inability to scientifically compare research results and conclusions, is a product of inconsistencies in terminology, 

experimental methods and the type and variety of skeletal materials employed. These problems were already 

noted by Mayne Correia (1997), and reiterated in Schmidt and Symes (2008), and references therein. 

Apart from the difficulty of replicating real forensic conditions in laboratory simulations, the focus on fracture 

patterns on a single burned bone, most often without flesh, probably derives from the archaeological roots of 

burned bone research. Early studies in the field largely relied on post-facto examination of archaeological 

cremations, or experimental studies focused on distinguishing fleshed and de-fleshed burning patterns. For 

instance, Krogman (1943) proposed studying the patterning of surface patina as a way to infer whether the bone 

was fresh when burned. This was later confirmed by Baby (1954) and Binford (1963), who added that certain 

cremated dry bone characteristics (e.g., warping) were not present in fresh bone. However, Buikstra and Swegle 

(1989) questioned these conclusions, finding warping also in fresh bone. These studies have a low degree of 

application to forensic settings, but illustrate the focus and tradition of the field. 

Fire Alteration of Skeletal Trauma Evidence: As a consequence of burned bone research primarily focusing 

on archaeological materials, the effect of heat alterations on forensically significant traumatized bone remains 

has rarely been researched until recently. Hermann and Bennett (1999) used an animal model to study the 

persistence of recognizable trauma after exposure to fire, but do not discuss specific diagnostic traits. Pope and 

Smith (2004) studied the persistence of some trauma characteristics in fleshed human heads after controlled 

cremation, and reported results that confirmed previous findings of the first author of this proposal (Symes et al. 

1996, 1999a, 1999b). Emanovsky et al. (2002), Devlin et al. (2006), Schmidt and Symes (2005) and Symes et 

al. (1996, 1999a, 1999b, 2005a, 2005b, 2008) all show that careful examination of the heat altered skeletal 

remains can differentiate perimortem trauma (sharp force and blunt force trauma, respectively) from postmortem 

thermal destruction. Still, there is an almost complete lack of experimental or observational studies based on 


fleshed remains, and all of the extant research is typically limited to isolated body areas and tissues (e.g., 

Christensen 2002, Pope and Smith 2004). 

Recent research by the first investigator (Schmidt and Symes 2005, Symes and Dirkmaat 2005, Symes et 

al. 1996, 2008) has served to identify three major process “signatures” recognizable in normal burned bone 

destruction. The recognition and analysis of these signatures mandates examination of the body at a number of 

different levels, from full body to individual bone fragments. Specifically, factors related to: 1) body 

positioning/tissue thickness, 2) bone color change, and 3) fracture biomechanics, are considered. 

Body positioning refers to the characteristic body and limb postures (often termed pugilistic pose) induced 

by the heating and shrinking of muscle fibers. Symes et al. (2008) demonstrates how departures from the 

normal expected patterns of body positioning allow for the detection and interpretation of forensically significant 

perimortem trauma (Figure 4). Of relevance to this proposal, these studies emphasize the importance of 

precisely noting victim position and orientation at the fire scene (context) and understanding normal burn 

patterns in any attempt to detect and analyze bone trauma. The importance of understanding body positioning 

and its relationship to fire characteristics has also been recently noted by DeHaan (2008). 

Fatal Fire Scene Recovery: In a recent book on fire investigation, Joe DeHaan, a former fire investigator 

and well known authority on the subject, has noted the relevance of incorporating better recovery routines and 

the employment of professionals devoted to the detection, analysis and preservation of all types of evidence, in 

fire scene investigation protocols (DeHaan 2007:4-5). 

As with conventional scenes, contextual evidence provides a key element not only for understanding trauma 

patterns, but also for identifying potential indicators of “foul play” or evaluating witness accounts. Still, despite the 

importance and added difficulty of the task, exact procedures for the location, recovery and documentation of 

human remains at fatal fire scenes are virtually absent from the fire investigation and forensic anthropology 

literature. The protocol for victim recovery typically consists of contacting medical or mortuary professionals to 

recover and remove the body from the scene (Churchward 2004, Lentini 2006). Olson (2006) has also noted 


that indeed a common practice is circumventing, rather than modifying, forensic and fire scene investigation 

protocols when a victim is present, in order to speed victim removal to quickly return the remains to the family. 

Mayne Correia and Beattie (2002) provide a critical review of fatal fire recovery techniques, stressing the 

need for improvement. One of the principal investigators of this proposal (DCD), provides a review of the scene 

recovery literature, and proposes an approach based on high resolution archaeological recovery methodologies 

(Lovis 1992, Dirkmaat and Adovasio 1997, Morse et al. 1976, Sigler-Eisenberg 1985), suitably modified to fit the 

characteristics of fatal fire scenes (Dirkmaat 2002). The two key elements identified by Dirkmaat as essential to 

the recovery process of the fatal fire scene are: 1) the initial recognition and identification of potentially significant 

physical evidence, and 2) the comprehensive documentation of contextual and associational relationships of the 

body relative to the environmental setting and other physical evidence, potentially related to the death (i.e., fire- 

altered debris). From this point of view, the core of fatal fire scene investigation also includes comprehensively 

locating, recovering and documenting the human remains and associated artifacts, rather than simply 

determining the cause of the fire. This information cannot be gleaned through casual documentation and low 

resolution recovery methods (random searches, minimal photography and rapid collection of the remains). It is 

only procured through exacting forensic archaeological recovery methods, the benefits of which are well 

documented in all manner of outdoor forensic scenes (Dirkmaat and Adovasio 1997, and references therein). 

In summary, a review of the literature reveals that the analysis and interpretation of burned human remains 

from fatal fire scenes is currently hampered by: 1) the lack of standard recovery protocols specific to this type of 

scene, 2) a poor understanding of general patterns of fire destruction of the human body, and 3) the lack of 

standards (including terminology), and guidelines for the laboratory analysis of burned human remains, 

particularly regarding the detection and analysis of perimortem trauma. The proposed research will address 

these issues by developing and testing new fatal fire scene recovery protocols, analyzing “normal” sequences of 

fire destruction to human remains, and studying the effects of fire alteration on perimortem sharp force trauma. 


Project Goals and Objectives 

The ultimate goal of the proposed research project is to address the problems outlined above through the 

construction and validation of basic standards and guidelines of utility to forensic professionals and fire 

investigators involved in the interpretation of burned human remains from fatal fire scenes. This goal will be 

implemented through a set of interrelated experimental and observational studies designed to enhance onsite 

location, evaluation and preservation of burnt human remains; forensic laboratory analysis of this type of 

evidence (especially skeletal trauma); and ultimately, fatal incident reconstruction. From an operational point of 

view, this translates into the following specific Research Objectives: 

Research Objective 1: Produce comprehensive guidelines and a simple, user friendly data collection 

database for the recovery of human remains from fatal fire scenes Enhanced fatal fire scene recovery protocols 

and guideline will be aimed at enhancing the success rates of the location, recovery and preservation of human 

skeletal elements and tissue at the scene, while maximizing the compatibility of forensic anthropology protocols 

with those of standard fire investigation. This will result in timely scene processing and efficient onsite 

cooperation between different types of investigators. Better in situ recovery methods will also benefit subsequent 

laboratory analyses by providing better preserved bone elements and relevant information on contextual factors 

such as body location and orientation (essential for biological profile, taphonomic and trauma analyses), thus 

resulting in better past event reconstruction. 

Research Objective 2: Describe basic, meaningful patterns of fire alteration to the human body, depending 

on temperature exposure of the corresponding anatomical area, with their relative frequencies. This will produce 

a Daubert-compliant baseline for the recognition and interpretation of forensically significant perimortem trauma 

to the body or other forms of intentional body manipulation or modification prior or during the fire episode. 

Research Objective 3: Assess and validate the applicability of conventional (non-burned) protocols for the 

analysis of sharp trauma to burned bone. This can be accomplished by assessing rates of preservation and 


patterns of alteration of quantitative and qualitative tool class mark characteristics on bone, both under laboratory 

and near-real conditions, and with animal and human models. 

The main methodological assumption underlying these operational objectives is that scene recovery and 

documentation, and trauma analysis of burned human remains cannot be uncoupled. As with other types of 

evidence, it can be said that trauma analysis begins at the scene. Effective, reliable, court-defendable 

interpretations of burned bone evidence (especially related to trauma analysis) require a comprehensive 

recovery of all human remains and associated physical evidence, including contextual and spatial data, as well 

as proper handling of the evidence, avoiding additional postmortem alteration. 

Research Design and Methods 

The main rationale behind the proposed methodology is that the analysis of perimortem trauma in burned 

bone requires first identifying bone alteration resulting from the recovery process, as well as identifying the 

normal effects of heat exposure on the human body in fatal fire circumstances. The research objectives 

described above will be pursued through three primary Research Components, which integrate previous and 

current independent research by two of the principal investigators (SAS and DCD). Combined, they have over 

50 years of experience in the recovery, forensic analysis and court presentation of human remains from different 

contexts, from surface-scattered and buried remains to mass disasters and, of course, fatal fire scenes. 

These three research components are aimed at identifying, assessing and, when appropriate, controlling: 1) 

additional trauma and bone loss inflicted during scene recovery, 2) the patterns of bone trauma resulting from 

exposure to fire, and 3) the heat alterations to be normally expected on inflicted perimortem trauma, particularly 

diagnostic characteristics normally employed to assess forensic significance and inflicting tool characteristics in 

fresh, unburned bone. The three research components include Recovery of Burned Human Remains, Analysis 

and Interpretation of Burn Patterns; and Heat Alterations in Traumatized Bone. These research components are 

intimately linked, and are designed to provide comparative data and complementary materials to one another. 


Research Component 1. Recovery of Burned Human Remains 

Research Component 1 is aimed at two main objectives: 1) the development and testing of new and 

efficient fatal fire scene recovery protocols, and 2) providing comparative osteological materials for Research 

Component 3, and comparative data regarding contextual information, such as temperature distribution around 

the body, and burn patterns of human remains in real fires, for Research Component 2 of this study. 

Current fatal fire scene recovery protocols must improve success rates with respect to the location, recovery 

and preservation of evidentiary items, as well as the documentation of relevant contextual information. 

Concurrently, they must avoid delaying or adding unnecessary burdens to current fire investigation protocols. 

New protocols must also be as economical as possible in terms of time, personnel and budgetary costs. 

Research Component 1 is sub-divided into three interrelated areas of study: Archival Research, Processing 

Mock Fatal Fire Scenes and Processing Actual Fatal Fire Scenes. 

Archival Research 

Purpose: Since a critical objective of the research is to streamline and maximize fatal fire data recovery, thus 

improving existing protocols, it is extremely important to review and analyze existing documentation of fatal fire 

scenes. This will serve to identify: 1) key variables regarding fire and victim investigation, 2) current deficiencies 

in variable recordation, and 3) inconsistencies in variable coding affecting data management, sharing and 

analysis. Through comparison with real fatal fire conditions, these data will also serve to assess the validity and 

representativeness of the experimental conditions selected for the simulation studies discussed below. 

Data Sources: The Office of the Fire Marshal, Province of Ontario, Ministry of Community Safety and 

Correctional Services, has granted access to all the documentation pertaining to their investigation of fire 

scenes, both current and historical. In addition, similar data access agreements have been discussed with fire 

investigators in Pennsylvania (e.g., City of Erie, PA, Fire Chief’s Office, PA State Police Fire Marshall’s Office). 

Research Design: The methodology for this study is based on a pilot study conducted during the last year 

by one of the key consultants for the project, Gregory O. Olson (GOO). Based primarily on current standard fire 


investigation protocols in use in the US and Canada (Churchward et al. 2004), a set of 30 qualitative and 

quantitative variables describing the most relevant parameters of fire scenes was created (Appendix 4).These 

variables were then recorded from the investigation and court files of 60 recent fatal fire scenes. No data 

regarding victim identity, geographical location of the structure, or any other detail that could allow for the 

identification of a specific fire event were or are to be recorded. Both variable presence/absence in the record, 

and coding of the information were taken into account. 

The present project will continue this study, with a conservative estimate of 150 new cases added to the 

already recorded ones. In order to allow for reliable data collection and sharing between more observers, 

minimize recordation errors, and expedite the recordation process, the current improved paper version of the 

recordation form, will be translated into a list menu-based electronic database requiring minimal typing. This 

simple database will be available for PC and PDA versions. 

Data collection will focus on obtaining two balanced subsamples of cases from the years 2000-01 and 2005- 

06, in order to assess recent developments in data recording methods or recovery protocols. Variable 

recordation (information recorded/non recorded) will be compared between the two subsamples through a χ 2 

test for independence. Additionally, a non-metric multidimensional scaling analysis (Young and Hamer 1994) 

will serve to assess the presence of relevant fatal fire scene groupings based on the variables under study, as 

well as the relevance and contribution of each of the variables to this classification (including processing year). 

Expected outputs: The pilot study already indicates that current data recording is highly inconsistent, with a 

pronounced diversity of both the variables taken into consideration and the way in which they are coded and 

expressed. Significantly, important contextual data regarding body position and taphonomic factors highly 

relevant for forensic anthropological analysis are rarely recorded. The electronic database-forms developed to 

record the historic data, can simplify scene data collection by providing a user-friendly electronic check list of key 

variables that are to be recorded. The primary significance of this electronic checklist is that important and 


elevant scene data will be collected consistently from scene to scene, enhancing scene comparison. This data 

is useful to both forensic anthropologists and fire investigators, and thus reduces redundancies in data collection 

efforts and allows for easy data sharing between the two professional groups. 

Information on number, agency and training of investigators and first responders, response and scene 

processing times, physical characteristics of the burned structure, or fire classification will provide a baseline for 

comparison with the scenes processed through the newly proposed protocols, in order to assess their efficiency 

in terms of processing time and personnel, and protocol applicability in real situations. 

Information on number and biological characteristics of the victims, their position and location within the 

scene, and graphic and autopsy documentation of the burned remains, will serve as a baseline for comparison 

with the patterns observed in the controlled cremations of human remains (Research Component 2). These 

factors will also be employed to assess the degree of realism of the experimental conditions set for mock scenes 

and laboratory experiments. 

Processing Mock Fatal Fire Scenes 

Purpose: Comparative (mock) fatal fire scene exercises will serve to: 1) test and refine the new forensic 

archaeological protocols in controlled near-actual conditions, and then compare them with the current protocols, 

2) provide detailed temperature readings from different areas of the fire scene near the body (for comparison 

with the experimental conditions in Research Component 2), and 3) provide comparative osteological materials 

for Research Component 3. 

Data Sources: The Office of the Fire Marshal, Province of Ontario, Ministry of Community Safety and 

Correctional Services, has already granted permission to conduct this research during fire response and 

investigation training exercises at their facilities. A conservative estimate of 4 to 6 mock scenes will be conducted 

in Ontario under the project. Cooperative support with the Ontario Fire College, and the Ontario Police College, 

has also been obtained for these exercises. The City of Erie, Pennsylvania, Fire Chief’s Office, and 

Pennsylvania State Police Fire Marshall’s Office have already expressed support and interest in conducting 


similar exercises. Similar agreements will be sought with other fire investigation agencies throughout 

Pennsylvania, Ohio, New York, and West Virginia in order to conduct another 4-6 mock fire scene recoveries. 

Research Design: The basic methodology for this study is based primarily on previous research and case 

experience by one of the principal investigators (DCD). The proposed recovery protocols are a modification of 

conventional forensic archaeological techniques, aimed at non-burned human remains (Dirkmaat 2002, 

Dirkmaat and Adovasio 1997, and references therein). These include: 1) detailed mapping and excavation of the 

human remains, using both grid system, electronic total station and GPS data, 2) careful written, photographic 

and videographic scene documentation of evidence and recovery process, and 3) evidence collection and 

treatment (Dirkmaat, 2002). Data collection is not limited to evidentiary items, but also includes information 

related to the contextual and physical characteristics of the scene which may influence taphonomic factors 

associated with the human remains. In the proposed extension of these protocols, as applied in this project, 

collected data also include fire-specific parameters from standard fire investigation protocols (Churchward et al. 

2004), which are of potential utility to forensic anthropological analysis. 

Apart from the comprehensive documentation of physical and contextual data, an important outcome of 

these protocols is simplification of procedures and reduction of time required for scene processing. This is 

accomplished through the application of technological enhancements readily available to law enforcement (e.g., 

the total station), as well as through the concurrent implementation of appropriate search, location, 

documentation and recovery steps. This has been shown to actually reduce recovery time and personnel, while 

dramatically improving recovery rates and documentation, even in large, complex mass fatality scenes, with 

highly altered and fragmented remains (Dirkmaat et al. 2001, 2005). 

As with the archival studies above, the methodology for this study was developed and tested through a pilot 

study comprising three mock (comparative) fire scenes, taking advantage of regular training exercises by the 

Office of the Fire Marshal, Province of Ontario, Canada (see above Data Sources). The proposed research will 

utilize this methodology with some modifications. In these exercises, a number of pig (Sus scrofa) limbs and 


carcasses (serving as an animal proxy to human remains), and other mock evidentiary items, will be emplaced 

at different locations inside a real structure (a donated building as noted above) by the assistant to primary 

consultant Fire Marshall Olson (GOO) who is not involved in this process. In addition, the pig limbs will be pre- 

processed with a variety of carefully documented fresh-bone trauma (fractures, cut marks and a variety of saw 

cuts, described below).This role will be filled by Mark Duval (MD), a crime scene investigator from the York 

Regional Police, New Market, Ontario, who has forensic archaeological training and experience. The scene and 

the distribution of evidence will then be carefully documented via written notes, photography, and maps by the 

assistant. An important question in forensic anthropological analysis of burned human remains is whether 

missing bone elements were originally not present at the scene, or, instead, went undetected. Consequently, 

metallic plates are used to tag the animal remains, and their exact location recorded, to allow for later location 

and identification even if the tissues were completely degraded by the fire, or disturbed by recovery efforts. 

A number of thermocouples (up to 20) will be placed in different key areas of the house, as well as close to, 

above, underneath and inside the pig carcasses. The structure is then set on fire and allowed to burn for a 

specified amount of time or until total collapse. Temperature readings from the thermocouples will be recorded 

with a 2635T Fluke recording thermometer in 10 second intervals during the combustion process (Figure 5). 

After the fire is extinguished, the structure will be first processed following conventional fire investigation 

protocols (Churchward et al. 2004), by fire responders with fire-fighting background (fire investigators and 

students currently involved in fire fighting training), but with no anthropological or archaeological training (see 

Appendix 5 for a more detailed explanation of the protocol and results of one of the pilot mock scenes). 

Once the first recovery team has completed scene processing, the scene will be re-processed by a team of 

10-15 Mercyhurst graduate students (Anthropology Masters program in Forensic and Biological Anthropology), 

following forensic archaeological protocols, under the supervision of the primary consultant (GOO). Besides 

detailed maps, temperature readings, and the information recorded in the mock scenes will make use of the 

recordation forms developed from the archival studies, and include more precise information regarding total and 


individual processing times and training. This will allow for comparison of protocol efficiency in terms of 

processing efforts, and learning curves for the new protocols. A Student’s t analysis for repeated measurements 

will be employed to compare total times between methods. If necessary, a correction for small sample sizes will 

be applied (Vallejo and Livacic-Rojas 2005). Factors such as the correlation between number of responders and 

processing times will also be controlled through regression models. 

As noted briefly above, the mock scenes in this proposal will also include additional pre- and post- 

processing of the pig carcasses. Pre-processing will consist of the infliction of cut marks, saw cuts, and fresh 

fractures on the pig limbs. The protocols are identical to those described in Symes et al. (2005, 2006). The heat- 

altered bones will be examined post-burning in order to analyze the effect of fire on the evidentiary value of 

perimortem sharp trauma (see Research Component 3 below for further details). Post-processing will include 

the estimation of taphonomic parameters, such as minimum number of individuals (MNI), number of identified 

specimens (NISP), or number of fragments (Lyman 2001), which will serve to compare bone fragmentation and 

alteration between recovery protocols. 

After each mock scene has been processed following these revised protocols, each fire investigation 

agency involved will complete a questionnaire regarding the utility, effectiveness, efficiency and user-friendliness 

of the new protocols. In this way, professional fire fighters will serve as independent testers and provide 

feedback on the protocols, particularly regarding the difficulty of their implementation in terms of learning curve, 

and their applicability to real situations. 

Expected outputs: Preliminary results derived from the mock scenes processed in 2007 by primary 

consultant (GOO) indicate that processing fire scenes through conventional protocols resulted in missing 

evidence (later recovered through the archaeological protocols), some trampled and damaged evidence, 

substandard documentation, and no significant decrease in recovery times. It was also noted that the 

participants in the archaeological recovery showed a steep increase in their efficiency and effectiveness after 

just one training session and mock scene, suggesting that the new protocols are not too costly in terms of 


training requirements. 

Mock scene exercises will provide enhanced and more realistic recovery protocols and electronic 

recordation forms and guidelines. These exercises will also serve to evaluate the advantages or disadvantages 

of the new guidelines over current standard protocols in terms of efficiency, effectiveness and costs, prior to their 

implementation and testing in real situations. Testing of the protocols will be conducted by independent trained 

professionals in realistic, non-compromised conditions. 

Continuous temperature readings at different scene and body locations, combined with the pattern of 

alteration observed in the animal model, will provide an opportunity to compare data and propose corrections to 

the burning patterns observed in the controlled cremations (Research Component 2), which will be conducted 

under ideal conditions and with an homogeneous distribution of temperatures around the body. Finally, the 

mock scenes will provide osteological materials for Research Component 3. 

Processing Actual Fatal Fire Scenes 

Purpose: Comprehensive analysis of victim recovery methods used at actual fatal fire scenes will serve to: 

1) test the applicability, advantages and weaknesses of the new protocols in real situations, 2) provide a 

comparison with past cases processed with current protocols (archival study part of this proposal), in terms of 

information gained, processing times, effort and costs, and 3) provide a realistic baseline for assessing the 

validity of the results and observations obtained at the mock scenes and controlled cremations. 

Data Sources: The data for this part of the study will be collected during the processing of real fatal fire 

scenes investigated by the Office of the Fire Marshal, Province of Ontario, Canada, and the forensic recovery 

team of the Applied Forensic Sciences Department (AFSD) of Mercyhurst College, Erie, PA. 

Research design: One of the primary consultants (GOO), a fire investigator, will process actual fatal fire 

scenes as part of his job with the Ontario Office of the Fire Marshal. He will apply forensic archaeological 

recovery protocols to each new scene he investigates, and will record the same data described above for the 

mock scenes. The principal investigators (DCD primarily) will utilize similar recovery methodologies for all fatal 


fire cases processed by AFSD. A conservative estimate of the number of scenes to be processed in this way 

during the duration of the project is between 10 and 15 cases. The proposed improvements developed during 

the project will be gradually incorporated into the real-scene protocols, as they are tested and refined in the 

project phases above. This will serve to test and fine-tune them in real world situations. 

Although comprehensive protocol testing in real situations may seem risky in a forensic setting, the validity 

of this approach has been already proven by the customary application of forensic archaeological protocols in 

the fatal fire scenes processed by AFSD, Mercyhurst College to date. Further, the exact methodology proposed 

for this project was applied to six (6) fatal fire scenes investigated by the Office of the Fire Marshal, Province of 

Ontario, Canada in the past year during a pilot study carried out by GOO. In these scenes GOO acted as the 

first responder and primary fire investigator and applied the same methodologies and data recording techniques 

described above for the mock scene pilot study, although without the assistance of archaeology students. 

Olson’s pilot study showed that the research protocols described above, even in their gestational stage of 

development, can be realistically applied in real fatal fire scenes, without adding an unnecessary burden to the 

investigation or delaying scene processing (see Appendix 6 and Figure 1 for a more detailed description of the 

applied protocols and outputs). In virtually all cases, data pertaining to: 1) body location and positioning, 2) 

contextual information pertaining to relative location of evidentiary items, and/or 3) the comprehensive in situ 

identification and preservation of small skeletal elements and evidentiary items; was critical to the reconstruction 

of the events surrounding fire and death. 

The information obtained from real scenes originally processed using archaeological protocols will be 

statistically compared with the archive records using the same methodology described above to compare 

archive subsamples. Processing times between past and archaeologically processed scenes will be compared 

through a Student’s t test, with small sample size corrections if appropriate. 

In the cases in which AFSD is also in charge of the anthropological analysis of the body, the observed burn 

patterns will be recorded and mapped in the same way as the remains cremated under controlled circumstances 


(Research Components 2 and 3), providing direct comparison between real and experimental patterns. 

Expected Outputs: The real scenes will provide the final product in the recovery protocols, with fully 

operational methodologies and data-recording databases and protocols. Additionally, they will provide realistic 

estimates of the relative efficiency, effectiveness and costs of the new protocols, as compared with the ones 

reflected in the archival material. Finally, they will provide comparative information regarding the consistency and 

representativeness of the normal burn patterns observed in the controlled cremations. 

Research Component 2. Analysis and Interpretation of Heat Altered Bone: Normal Burn Patterns 

Purpose: One of the key objectives of the project is improving our ability to distinguish heat-inflicted trauma 

from perimortem trauma in bone. This requires recognizing normal burn patterns expected in burned bodies. 

There is a lack of understanding of how bodies burn naturally, and interpretations of heat-induced bone trauma 

are often based on misconceptions such as ‘the limbs were completely consumed by the fire’ or ‘skulls explode 

in a fire.’ This component of the research intends to: 1) study and document normal burn patterns in fully fleshed 

human remains cremated under controlled conditions (in a crematorium), and 2) assess the validity of these 

burn patterns when applied to real situations. These goals will be accomplished through comparison of normal 

burn patterns with the patterns observed in forensic cases, incorporating scene information regarding body 

position and temperature distributions around the body, obtained from real and mock scenes. 

Data Sources: Data for this component will be gathered from three main sources: 1) existing documentation, 

including the extensive written documentation of dozens of cases charted for burn patterns over the past 23 

years by SAS, and archival information collected from Ontario, Canada fatal fire scenes, 2) graphic 

documentation, mapping and analysis of the burn process exhibited by fully fleshed human bodies, incinerated 

under controlled temperature conditions in modern crematoriums, and 3) the analysis of current forensic cases 

processed at AFSD, Mercyhurst College, the New York Office of the Chief Medical Examiner, New York, Erie 

County Medical Examiner’s Office, Buffalo, NY, and the Erie County Coroner’s Office, Erie, PA . 

Research design: This component of the research is based on previous research developed by one of the 


principal investigators (SAS), who documented all fatal fire victims (and individual bones) brought to the 

Regional Forensic Center, Memphis, Tennessee, since 1992 (Symes et al. 1996, 1999a-b). See Appendix 7 for 

an example of mapped burn patterns based on preliminary data. 

During the burn pattern documentation and analysis phase of the project, whole fleshed bodies will be 

incinerated at a modern crematorium under homogeneous controlled temperatures, similar to the flash-over 

temperatures and atmospheres reached in a burning house or car. Different designs have been explored for this 

purpose, however, the most efficient, realistic and economic design consists of video-recording the cremation 

process at regular times, from the exterior of the oven, without changing the temperature or removing the 

remains from the oven. Following this protocol, and considering the number of cremations processed by the 

crematoria consulted, an estimated sample size of 50 cremations is proposed for the duration of the project. 

The study requires high resolution color images to record color changes and tissue alteration. The presence 

of moving flames and dynamic heat artifacts between the lens and the body, does not allow for the recording of 

these elements through conventional photography or video-recording equipment. Additionally, the high 

temperatures registered around the oven viewing window are beyond the functional temperatures of 

conventional video cameras. Therefore, the proposed methodology is based on the use of a ruggedized, 

temperature-resistant, high-speed, high-resolution color video camera (Fastec Ranger HR) to record the burning 

process in sessions of five second continuous filming bursts (500-1000 frames per second), recorded at different 

time intervals. Based on preliminary research for protocol development, heat-induced alterations progress 

rapidly at conventional crematorium temperatures. It is estimated that most of the relevant changes will occur 

within the first 3 to 10 minutes of the process. According to this estimate, data collection at 1 or 1.5 minute 

intervals would be desirable. Starting recording times for each individual will be randomized from 1 to 60 

seconds in order to obtain pattern readings over a continuous time range across the sample. 

Temperature readings will be recorded during the same time intervals, and stored with the corresponding 

video file. Medical and biological data of each individual will also be collected, but no data allowing individual 


identification will be recorded or included either in the study or in published results. 

The resulting images will then be treated and analyzed, mapping the progression of body positioning, color 

and tissue changes for each anatomical area. Cremation patterns at consecutive times will then be 

superimposed, to reveal the progression of the heat-alteration in each bone area, and refined using kernel filters 

based on pixel counts (Fortin and Dale 2005). The average timing and confidence intervals of each key 

modification (event), defined by the presence of a single, unambiguous criterion, and coded as binary variables 

(i.e., taking either the values 1 or 0), for the range of temperatures applied, will be estimated by fitting the binary 

data to a cumulative normal distribution (essentially, probit analysis), using a curve fitting routine commonly 

employed to estimate the timing of physiological events (Fuiman et al. 1998). The dependence of character 

timing on temperature will be then assessed through Spearman’s rank order correlation between average event 

timing and temperature at event. 

The uniformity and consistency of the sequence of alteration across body parts (i.e., which elements are 

altered before or after each other one), will be assessed through parsimony analysis and consensus trees 

(Kitching et al. 1998). With this purpose, each of the binary key alterations described above will be recorded for 

each data point (e.g. a matrix {i1, i2, i3, i4} = {0, 0, 1, 0} would indicate that the individual had undergone at that 

time heat alteration 3, while alterations 1, 2, and 4 were still not present). All the trait combinations present in the 

sample will then be entered into a parsimony analysis, in which each trait combination will be considered a 

taxon, and the presence/absence of an alteration as a character state. The tree extraction method will depend 

on the number of observed states (as it relies heavily on computation times), but if there are less than 12 

alteration combinations, a branch-and-bound algorithm, with Dollo optimization, is proposed (Kitching et al. 

1998). The strict and majority consensus trees will then be calculated, representing the most common heat 

alteration sequence patterns. The number of trees supporting the obtained classification will serve as an 

indication of the regularity of the observed “normal” pattern across the sample. The Bremer support (decay 

index, Kitching et al. 1998) calculated over the consensus tree will give an estimate of the typicality of pairs of 


character states (for example, of simultaneous absence of alteration x and presence of alteration y). 

The introduction of key alteration patterns from real cases, analyzed in the same manner as described 

above, will serve to assess the validity of the heat alteration sequences obtained in the controlled cremations. 

Fiedl or case patterns inconsistent with the crematory ones are expected to result in an increase of tree length, 

and in low decay indices for that combination of heat alterations. In a more intuitive way, in real cases the 

presence of atypical patterns, or of combinations of heat alteration pairs would indicate the possibility of 

perimortem trauma inflicted before cremation. 

Heat alteration patterns in real cases will depend on the amount and duration of heat exposure of the 

different body areas, while in the crematory experiments temperature will be, in general, homogenously 

distributed around the body. Information on body placement and positioning at real cases, when available, will 

be compared with the temperature readings recorded at different body regions (over, inside and underneath the 

body) in the mock scenes from Research Component 1. This will serve to further validate and refine the 

application of the experimental patterns to real situations with anisotropic temperature distributions. If atypical 

patterns of heat alteration, as defined above, cannot be explained by differences in temperature exposure 

related to body placement and positioning, the presence of trauma inflicted before fire initiation becomes a more 

likely possibility. This would serve to identify suspicious anatomical areas requiring closer examination. 

Heat-altered skeletal remains from real cases will be documented with respect to posturing, bone color 

change, as well as fracture patterns through close-up photographic and microscopic examination. Color and 

physical alteration of each bone will be charted, following a mapping protocol similar to the one employed in the 

crematory experiments. In order to allow for better inter-observer comparison, as well as to better describe and 

quantify color change that will permit statistical analysis, bone surface color in each area will be recorded in the 

CIELAB 1976 (L*a*b*) three-dimensional color space., following a methodology similar to that described in 

Devlin and Herrmann (2008) and Beary and Cabo (2008). A hand-held X-Rite CA22 tethered spectrophotometer 

will be used for this purpose. 


Expected Outputs: The charted fire alteration patterns and anatomical burning sequences derived from 

these analytical methods, will provide a baseline for detection of trauma patterns not readily attributable to heat 

alteration. These patterns and sequences will be described at both the full body and isolated bone level, and will 

include general corrections for body positioning and temperature distribution at the scene. 

Research Component 3. Heat Alterations in Traumatized Bone 

Purpose: Once normal, expected heat-induced bone trauma and damage due to recovery (including bone 

loss or destruction at the scene) have been identified and controlled for, the remaining question is whether the 

same diagnostic traits used to detect, assess and identify forensically significant perimortem trauma in non- 

burned bone (including the identification of inflicting tool class characteristics), are still effective after fire 

alteration. Ultimately, the question can be posed as whether the absence of certain class characteristics in a 

trauma feature means that the tool typically producing those characteristics can be ruled out as the inflicting 

weapon or, on the contrary, exposure to fire is actually expected to significantly alter or even destroy them. This 

component of the research addresses this question for sharp trauma, through the analysis and documentation of 

saw marks in human bone and an animal model (Sus scrofa limbs), before and after burning in near-real and 

laboratory conditions. Additionally, the study will evaluate the validity of the use of animal models for this kind of 

research, through the comparison of results obtained from animal and human limbs. 

Data Sources: Data for this component will be collected from three sources: 1) Data Source A, a sample of 

human bones previously subjected to trauma under experimental conditions in a research project currently 

conducted by the first investigator (SAS) with funding by the NIJ, aimed at the study of saw mark analysis in 

fresh bone and its accuracy in Daubert-type analyses (Symes et al. 2005a, 2006), 2) Data Source B, a sample 

of animal bones that will have been generated during the mock scene exercises in Research Component 1 and 

subjected to the exact same treatment as applied to the human sample, and 3) Data Source C, a second animal 

sample that will have been generated by applying the same trauma-induced treatment as the other two samples, 

and then burned under controlled conditions in the laboratory. 


Research design: This component of the research takes advantage of an existing sample of human 

remains, derived from an existing NIJ-funded experimental project (described above), and a comparative sample 

generated in Research Component 1 of this proposal. The simultaneous availability of all these resources, 

coupled with the information obtained from previous studies and the other components of this proposal, 

represents a unique opportunity to conduct a relevant study at minimal cost. 

This study relies on an opportunistic experimental design, intended to take advantage of these existing 

resources through the generation of Data Source C, which will serve to link the two existing samples through two 

pair-wise sample comparisons via a simple but powerful statistical design. Both the Data Source A sample of 

human remains and Data Source C sample of animal bones will receive the exact same treatment, serving to 

compare the effects of fire on trauma class characteristics in human and animal remains. 

In the second comparative study, Data Source B and C samples, both animal (pig) remains, will be burned 

under different conditions, after receiving the same pre-burning treatment (i.e., controlled vs. field burning). The 

results will then be compared. If the first comparison (human to animal) reveals no differences in class- 

characteristic preservation, the results of the field experiments can be easily extrapolated to human remains. It is 

also possible that adequate corrections for extrapolation of experiments based on animal models to human 

remains can be realistically assessed and estimated. 

The pre-burning treatment for all samples will consist of the infliction of a number of cut marks with different 

tools, following the protocols described in Symes et al. (2005a and 2006). Saw marks, cut marks and fractures in 

the fresh bones will be examined macro- and microscopically utilizing a Leica MZ16A microscope with In-Focus 

capability. All of the diagnostic characteristics (knife vs. saw, teeth per inch, etc) will be carefully noted. 

Documentation will include microphotography and the creation of casts of each tool mark. Saw marks are 

specifically utilized here so that both qualitative and quantitative comparisons can be tested on the burned bone. 


The bones will then be cremated in a Fisher, programmable, forced draft furnace at a temperature of 500ºC. 

An initial sample of 5 to 10 bone fragments will first be exposed before cremation to a temperature of 60° C, 

monitoring weight loss and coloration change throughout the process, until the weight curve becomes 

asymptotic. This will serve to determine dry weight. The difference between dry and post-cremation weights will 

provide an estimate of the average organic matter consumed during the cremation process. This set of bones 

will also be monitored at regular intervals during the 500ºC cremation process, to determine times of color 

change and until complete calcination. The results of this initial study will serve to determine the time parameters 

for the cremation of the rest of the sample. 

The three samples of bones will be re-examined after cremation and analyzed in two independent tests: 

Random (jackknifed) subsamples will be examined in blind trials by initial observers. The number of correct 

classifications before and after burning will be compared through paired χ 2 tests for independence, , and a 

multivariate U– Mann-Whitney unplanned test for count data (Sokal and Rohlf, 2003). This will serve to assess 

the degree of qualitative alteration of the bones and trauma characteristics in terms of their diagnostic value. 

Concurrently, pre-burning casts and post-burning cutmarks and fractures will be compared quantitatively, 

using a Leica FS C Comparison Microscope in terms of feature length (shrinkage and destruction), and the 

presence/absence of discrete quantifiable traits (e.g., number of teeth impressions in saw marks), employing 

ANOVA models for repeated measurements, This will assess and quantify fire mark alteration and destruction 

rates in an objective manner. 

Expected outputs: This component of the research will provide a baseline for the forensic analysis of sharp 

trauma in burned bone, particularly in terms of the interpretation of the absence of diagnostic traits relative to 

general tool class characteristic, as well as the reliability of quantitative estimates such as teeth per inch, or 

blade dimensions. Additionally, the comparison of results from identical experiments in human and animal 

remains will serve to assess the validity for future research of observations based on animal models. 


Implications for Policy and Practice: The integration of comprehensive field documentation and recovery 

of significantly-modified human remains from fatal fire scenes, with equally comprehensive, empirically-based 

documentation of patterns of thermal modification of bone, results in the production of much more scientifically- 

defendable (Daubert compliant) taphonomic interpretations of events surrounding the death of fire victims. 

The recovery aspect of this research will document the applicability, effectiveness and efficiency of forensic 

archaeological methods to document the context and association of human and physical evidence at fatal fire 

scenes. The implementation of these methods is expected to optimize evidence detection, recovery, 

preservation and trauma analysis. The final product of the research will be a series of guidelines (protocols) for 

recovery and documentation techniques realistically applicable to most fatal fire scenes. This will benefit the 

analyses of all professionals involved, from anthropologists, to arson, and crime scene and fire investigators. 

The enhanced contextual data and evidence integrity (precise body position, detection and recovery of all bone 

elements, relationship of other physical evidence to the body) will benefit the morgue/laboratory analysis of 

burned human remains, and the final determination of events surrounding death. 

The documentation of normal burn patterns of each bone of the human body drawn from crematory and 

fatal fire victim examinations will provide a better understanding of thermal modifications to bone (fracture, 

warping, discoloration patterns), enhancing our ability to distinguish perimortem trauma from postmortem burn 

trauma. Additionally, a more standardized and precise terminology for the description of this type of trauma will 

result in more robust, interpretable diagnoses, providing a baseline for research evaluation and replication, as 

well as for presentation in court. Finally, the comprehensive analysis of heat alteration patterns on well- 

documented sharp force trauma evidence, in both controlled and mock scene fire exposure contexts, further 

enhances the reliability of trauma analyses in the fatal fire victim. The melding of these three areas of research 

will result in more Daubert-compliant reconstructions of events surrounding death, with particular relevance to 

determinations of cause and manner of death. 


Management Plan and Organization: Dr. Steven Symes, Dr. Dennis Dirkmaat and Dr. Stephen Ousley 

will serve as the Principal Investigators on this project. Dr. Symes will oversee the normal burn patterns and 

traumatized bone components of the project. Christopher Rainwater, MS, New York City Medical Examiner 

Investigator will serve as the primary consultant to Dr. Symes, reviewing cases and providing important 

supplementary data from his case load from New York City. Dr. Dirkmaat will oversee the scene processing 

component. He will primarily be assisted by Gregory O. Olson a Mercyhurst College graduate student and Fire 

Marshall in Ontario, Canada. Dr. Ousley will be responsible for the sampling and experimental designs, and 

statistical analyses. One full-time graduate research fellow will be utilized in each of the three important 

components of the research, serving to overseeing operations and conduct primary research (see Budget 

Justification for details). 

Timeline: Each phase of research can be conducted concurrently and is staffed accordingly. In a 

conservative estimate, data gathering for Research Components 1 and 2 can be completed within 18 months 

after the awarding of the grant. Research Component 3 requires analysis of elements generated in the two other 

research components, and, therefore, will be carried out in part once these are completed. Thus, as correctly 

suggested by one of the concept paper reviewers, an extended timeline of two (2) years is conservatively 

estimated for completion of the project, including report production (see Appendix 8 for further timeline details). 

Dissemination Strategy: The results of this research will be disseminated through presentations at 

professional meetings and publications in indexed peer-reviewed journals and edited volumes. Parts of this 

research may be available for presentation at the Annual Meeting of the American Academy of Forensic 

Sciences in 2010 and 2011. Presentations will also be made in 2011 at other national meetings. Mercyhurst 

College’s Department of Applied Forensic Sciences has been regularly conducting training short courses for law 

enforcement professionals, forensic specialists, policy-makers and students from 18 countries and 6 continents 

for the last 17 years, thus providing an optimal opportunity for the dissemination and discussion of the results. 


Appendix 1: References Cited 

Baby, R. (1954): Hopewell Cremation Practices. Ohio Histor. Soc. Pap. Archaeol. 1: 1-7. 

Beary M.O. and L.L. Cabo (2008): Estimation of Bone Exposure Duration Through the Use of 

Spectrophotometric Analysis of Surface Bleaching and its Applications in Forensic Taphonomy. To be presented 

at the 60th Anniversary Scientific Meeting of the American Association of Forensic Sciences, Washington DC, 

February 18-23. 

Binford, L. (1963): An Analysis of Cremations from Three Michigan Sites. Wiscon. Archaeol. 44(2): 98-110. 

Buikstra, J. and M. Swegle (1989): Bone Modification Due to Burning: Experimental Evidence. In R. 

Bonnichsen and M.H. Sorg (eds.): Bone Modification. ME, Center for the Study of the First Americans, University 

of Maine, Orono, pp. 247-258 

Churchward, D. (2004). Guide for Fire and Explosion Investigations. National Fire Prevention Association. 

Christensen A.M. (2002): Experiments in the Combustibility of the Human Body. J. For. Sci. 47(3): 466–470. 

DeHaan, J.D. (2007): Kirk’s Fire Investigation. Pearson Prentice Hall, Upper Saddle River. 

DeHaan, J.D. (2008): Fire and Bodies. In C.W. Schmidt and S.A. Symes (eds): The Analysis of Burned 

Human Remains. Academic Press, London, pp. 1-13. 

Devlin J.B. and N.P. Herrmann (2008): Bone Color as an Interpretive Tool of the Depositional History of 

Archaeological Cremains. In Schmidt C.W. and S.A. Symes: The Analysis of Burned Human Remains. 

Academic Press, London, pp. 109-128. 

Devlin, J.B, A.M. Kroman, S.A. Symes, and N.P. Herrmann (2006): Heat Intensity vs. Exposure Duration 

Part I: Macroscopic Influence on Burned Bone. Paper and proceedings presented to the 59th annual American 

Academy of Forensic Sciences 12:310-311. 

Dirkmaat, D. (2002). Recovery and Interpretation of the Fatal Fire Victim: The Role of Forensic 

Anthropology. In W. H. Haglund and M. Sorg (eds): Advances in Forensic Taphonomy: Method, Theory, and 

Archaeological Perspectives. CRC Press, Boca Raton, pp. 451-472. 


Dirkmaat, D. and J. Adovasio (1997): The Role of Archaeology in the Recovery and Interpretation of Human 

Remains. In W. H. Haglund and M. Sorg (eds): Forensic Taphonomy: The Postmortem Fate of Human 

Remains. CRC Press, Boca Raton, pp. 39-64. 

Dirkmaat, D.C., J. Hefner and M. Hochrein (2001): Forensic Processing of the Terrestrial Mass Fatality 

Scene: Testing New Search Documentation and Recovery Methodologies. Paper presented at the Annual 

Meeting of the American Academy of Forensic Sciences, Seattle, WA. 

Dirkmaat D.C., L. Cabo, J.M. Adovasio, and V. Rozas (2005): Mass Graves, Human Rights and 

Commingled Remains: Considering the Benefits of Forensic Archaeology. Paper presented at the 57th Annual 

Meeting of the American Academy of Forensic Sciences. February 21-26, 2005, New Orleans, LA. (Available 

online at http://mai.mercyhurst.edu) 

Emanovsky, P., J.T. Hefner and D.C. Dirkmaat (2002): Can Sharp Force Trauma To Bone Be Recognized 

After Fire Modification? An Experiment Using Odocoileus virginianus (White-Tailed Deer) Ribs. Paper presented 

at the Annual Meeting of the American Academy of Forensic Sciences. Atlanta, GA. 

Fortin M.J. and M. Dale (2005): Spatial Analysis. Cambridge University Press, Cambridge. 

Fuiman L.A., K.R. Poling and D.M. Higgs (1998): Quantifying Developmental Progress for Comparative 

Studies of Larval Fishes. Copeia 1998(3): 602-611. 

Hermann, N. and J. Bennett (1999): The Differentiation of Traumatic and Heat-related Fractures in Burned 

Bone. J. For. Sci. 44(3): 461-469. 

Oxford. 

Kitching I.J., P.L. Forey, C.J. Humphries and D.M. Williams (1998). Cladistics. Oxford University Press, 

Lentini, J.J. (2006): Scientific Protocols for Fire Investigation. CRC Press, Boca Raton. 

Lovis, W. (1992): Forensic Archaeology as Mortuary Anthropology.Soc. Sci. Med. 34(2): 113-117. 

Lyman, R.L. (2001): Vertebrate Taphonomy. Cambridge University Press, Cambridge. 


Mayne Correia, P. (1997): Fire Modification of Bone: A Review of the Literature. In W. Haglund and M. Sorg 

(eds): Forensic Taphonomy: The Postmortem Fate of Human Remains. CRC Press, Boca Raton, pp. 275-93. 

Mayne Correia, P.M. and O. Beattie (2002): A critical look at methods for recovering, evaluating, and 

interpreting cremated human remains. In W. H. Haglund and M. Sorg (eds): Advances in Forensic Taphonomy: 

Method, Theory, and Archaeological Perspectives. CRC Press, Boca Raton, pp. 435–450. 

Morse, D., D. Crusoe and H.G. Smith (1976): Forensic Archaeology.J. For. Sci. 21(2): 323-332. 

Olson, G. (2006): Science vs. Sympathy. Paper presented at the Northeast Forensic Anthropology 

Association. Erie, PA. 

Pope, E. and O. Smith (2004): Identification of Traumatic Injury in Burned Cranial Bone: An Experimental 

Approach. J. For. Sci. 49(3): 431-440. 

Schmidt, C. and S. Symes (2005) Beyond Recognition: The Analysis of Burned Human Remains, presented 

at the Annual Meeting of the American Association of Physical Anthropology, Milwaukee, Wisc. 

London. 

Schmidt, C.W. and S.A. Symes (eds) (2008): The Analysis of Burned Human Remains. Academic Press, 

Sigler-Eisenberg, B. (1985): Forensic research: Expanding the Concept of Applied Archaeology. Am. 

Antiquity 50: 650-655. 

Sokal R.R. and F.J. Rohlf (2003): Biometry: The Principles and Practice of Statistics in Biological Research. 

W. H. Freeman and Co., New York. 

Symes, S.A., O.C. Smith, H.E. Berryman, C.E. Peters, L.A. Rockhold, S.J. Haun, J.T. Francisco, and T.P. 

Sutton (1996): Bones: Bullets, Burns, Bludgeons, Blunderers, and Why. Workshop presented to the 48th annual 

meeting of the American Academy For. Sci., Nashville, TN. 

Symes, S.A., O.C. Smith, and H.E. Berryman (1999a): Burned Bone: Looking a Little Closer. Paper 

presented to the 13th Annual Mountain, Swamp and Beach Meeting of Practicing Forensic Anthropologists, 

Columbia, SC. 


Symes, S.A., O.C. Smith, H.E. Berryman, and E. Pope (1999b): Patterned Thermal Destruction of Human 

Remains. Conference Sponsored by the Smithsonian Institution and Central Identification Laboratory, Hawaii. 

Symes, S., A. Kroman, et al. (2003a): Bone Trauma: A text of fracture biomechanics. Grant awarded by the 

University of Tennessee Law Enforcement Innovation Center, National Forensic Academy. Award No. LEIC 

R01-1007-087. 

Symes, S., A. Kroman, et al. (2003b): Dismemberment and Mutilation Saw Marks in Bone: Analysis and 

Interpretation. Grant awarded by the University of Tennessee Law Enforcement Innovation Center, National 

Forensic Academy. Award No. R01-1007-088. 

Symes, S., and D. Dirkmaat (2005). Perimortem Bone Fracture Distinguished from Postmortem Fire 

Trauma: A Case with Mixed Signals. 57th Annual Meeting of American Academy of Forensic Sciences. New 

Orleans, LA. 

Symes, S. and A. Kroman, et al. (2005a): Knife and Saw Toolmark Analysis in Bone: Research Designed 

for the Examination of Criminal Mutilation and Dismemberment. National Institute of Justice, US Department of 

Justice, Award No.2005-IJ-CX-K016. 

Symes, S.A., A.M. Kroman, C.W. Rainwater, and A.L. Piper (2005b): Bone Biomechanical Considerations in 

Perimortem vs. Postmortem Thermal Bone Fractures: Fracture Analyses on Victims of Suspicious Fire Scenes. 

Presented at the 74th Annual Meeting of American Association of Physical Anthropologists. Milwaukee, WI. 

Symes, S.A., C.W. Rainwater, et al. (2006): Knife and Saw Tool Mark Analysis in Bone: A Manual Designed 

for the Examination of Criminal Mutilation and Dismemberment- Phase II. National Institute of Justice, US 

Department of Justice, Award No. 2005-IJCK016 Continuance. 

Symes, S.A., C.W. Rainwater, E.N. Chapman, D.R. Gipson, and A.L. Piper (2008): Patterned Thermal 

Destruction of Human Remains in a Forensic Setting. In C.W. Schmidt and S.A. Symes (eds): The Analysis of 

Burned Human Remains. Academic Press, London, pp. 15-54. 


Vallejo G. and P. Livacic-Rojas (2005): Comparison of Two Procedures for Analyzing Small Sets of 

Repeated Measures Data. Multivar. Behavior. Res. 40(2): 179-205 

Young F.W. and R.M. Hamer (1994): Theory and Applications of Multidimensional Scaling. Eribaum 

Associates. Hillsdale, NJ. 

Appendix 2: List of Previous and Current NIJ Awards 

Knife and Saw Toolmark Analysis in Bone: Research Designed for the Examination of Criminal Mutilation 

and Dismemberment. Award No.2005-IJ-CX-K016. Principal Investigator: S.A. Symes. 

Knife and Saw Tool Mark Analysis in Bone: A Manual Designed for the Examination of Criminal Mutilation 

and Dismemberment- Phase II. Award No. 2005-IJCK016 Continuance. Principal Investigator: S.A. Symes. 


Appendix 3: Figures – Case Examples 

Figure 1. Frontal view of a burned vehicle with the severely burned remains of three victims: a child and two 

commingled adults. Note that while the large concentrations of soft tissue can be relatively easy to spot (blue 

arrows) burnt bone remains are difficult to distinguish from the surrounding debris. The problem is more relevant 

in the case of very fragmented and small elements (e.g., carpal bones). Can you spot the 3 rd victim? (the pelvic 

remains in the red square belong to the one of the adult victims, blues arrow). 

In this case, processed by one of the project consultants (GOO) in 2007, the individualization and 

identification of the commingled victims depended heavily on onsite identification and precise location of each 

individual bone element. The child was identified on the basis of a set of deciduous teeth found during debris 

screening. 


Figure 2. A set of burned bones trampled, altered 

and displaced during conventional recovery (note 

the boot marks around them), discovered during 

careful archaeological removal of the surrounding 

debris. Note the homogeneous coloration of all 

the materials (this is a color photograph). 

Figure 3. The photograph and map illustrate the 

position of two sets of human remains at a case 

scene processed by one of the principal investigators (DCD), applying forensic archaeological techniques. 

(a) Note the difficulty in distinguishing the human remains from their surroundings in the photography (and at 

the scene) and the advantages for report and court presentation offered by precise and detailed archaeological 

maps, created onsite from accurate spatial data (b). 

The position of the remains (two 3 year olds) served to support the account given by their mother, who 

claimed that she had been unable to find the boys in the house when the fire started. Her cold and impersonal 

manner during the account (later attributed to mental shock) raised the suspicions of law enforcement. 

Based on archaeological and osteological data, it was possible to determine that the children were located 

within an unlocked closet, the oldest (by 10 months) farthest from the closet door, in unforced anatomical 

positions. No indications of trauma or restraint were found, and background interviews diminished suspicion. 

Concurrently, the fire investigators found no evidence of accelerants and traced the most possible origin of 

the fire to the children, who likely had been playing with matches. After examining all the information from the 

forensic and fire investigations, no charges were filed against the mother. 

Archaeological recovery also served to individualize and identify all the skeletal elements of each victim. 


Figure 4. In this case, investigated by one of the 

principal investigators (SAS), the victim’s right hand did 

not show the pugilistic posture, even when the arm and 

forearm muscles had contracted, flexing the arm (a, 

left). In addition, the distal portions of the fingers were 

more severely burnt than the knuckles and dorsal area 

of the hand (b and c). This is an atypical pattern, as the 

fingertips and palm are sheltered from heat by the 

dorsal tissues during pugilistic posture. This served to 

infer that the hand had actually not adopted the 

pugilistic posture, rather than having been straightened 

during recovery or autopsy (note left hand in (a), right). 

This directed attention toward the right wrist area, 

where no trauma had been detected in the initial 

autopsy examination. Closer examination revealed a 

suspicious fracture above the right wrist (d). 

Microscopic examination of this fracture indicated that 

it extended uniformly across both heat-altered and nonaltered 

areas (demarcation of these areas was based 

on coloration and translucency changes). It also 

exhibited the biomechanical properties of a bending, 

green bone fracture. This provided an unequivocal 

diagnosis as a perimortem fracture that preceded the 

fire event. 

A more detailed description and discussion of this case example can be found in Symes et al. (2008) 


Figure 5. Temperature readings obtained in one of the mock fire scenes described in the proposal (pp. 11- 

14). Note the low temperatures registered under the body, and the acute differences observed between the 

north and south ends of the body, matching the orientation observed in the surrounding environment. 

In the new exercises, additional thermocouples will allow for more precise mapping of temperatures in and 

around the body. 


Appendix 4: Data Collection Form Employed for the Archival Pilot Study 

*Note: the form has been condensed from the original one, removing filling spaces and check boxes, for 

formatting purposes. The new data collection form and database include variable codes, that will be displayed as 

scrolling list menus, reducing the need for written notation. 

Statistics Data Collection Form 

(1) Structure: 

Type of structure: 

Square footage: 

Roof style and material: 

Date of construction: 

Material used in structure construction: 

Amount of destruction: 

Number of windows/doors: 

(2) Excavators: 

Number of excavators involved in recovery: 

Background education and training of excavators: 

Start/Finish times for each excavator: 

Number of hours on scene for each excavator: 

(3) Fire Service: 

Fire Service response time: 

Distance from the Fire Service to fire scene: 

Was the Fire Service a fulltime station or volunteer? 

Classification of the fire: 

(3) Fire Service (cont) 


Were there charges laid in this fire: 

If so, number of individuals charged and charges laid: 

(4) Human Remains And Context 

Number of deceased persons located within the structure: 

Condition of deceased persons: (calcined etc.) 

Location of deceased within the structure: 

Inventory of deceased persons: (osteological inventory) (outstanding remains or all accounted for) 

Was there a strata correlation made by excavators? 

Where were the deceased persons located within the debris strata? 

What was the approximate distance of the deceased persons from the nearest means of egress? 

Were there associated artifacts located near or in association to the deceased persons, if so what 

artifacts were located? (I.e. Fire extinguishers, accelerants, sources of ignition etc.) 

What form or forms of measurement were utilized? (Tape measure, total station etc.) 

Cause of death for deceased persons: 

Extent of thermal injuries and placement on deceased persons: 

Carbon monoxide levels for deceased persons: 

(5) Other Agencies Involved: (i.e. TSSA, ESA etc.) 

(6) Smoke Alarms: 

Present: 

Number present: 

Type and placement: 

Were they fully operational, if not, why not? 

(7) Human behavior of fire victims if able to determine: 


Appendix 5. Example of Mock Scene (Results and Procedures) 

COMPARATIVE STRUCTURE FIRE #3 

OBJECTIVE: Determining what evidence/remains students with both education and 

experience in the fields of archaeology and Osteology can locate with the 

use of a proper grid-style search in a fire ravaged structure as compared to 

a search conducted by persons with a fire fighting background. 

DATE: 23-25 November 2007 

DESCRIPTION: Single-story wood frame farmhouse – 100 to 110 years old. 

Pitched roof with asphalt shingles. 

ARTIFACTS: (1) 2 pig cadavers – 60 to 70 lbs. 

(2) 20 pig limbs of various sizes (with stainless steel tags) 

(3) 2 rifles 

(4) 1 rifle bolt 

(5) 3 shotgun shells (red) 

(6) 6 handgun shells 

(7) 12 2’ x 2’ carpet sections (with stainless steel tags) 

SUMMARY: This wood frame farm house was “prepared” on 23 November 2007 with 

the above noted artifacts placed indiscriminately throughout the main floor 

and corresponding rooms. The artifacts were photographed in place while 

the structure was “wired” with thermocouples. 

Personnel from the Centre of Forensic Sciences, Toronto, placed 12 2’ x 

2’ pieces of carpet containing samples of alcohol. Each carpet sample was 

tagged with a stainless steel tag. 

On 24 November 2007, the fire control officer with the fire department 

poured a quantity of gasoline inside the main structure and caused it to be 

set on fire by the external application of a common road flare. The initial 

reaction of fire to gasoline caused a predictable “push” inside the structure 

with the thermocouples registering an immediate 600° Celsius increase 

within the interior temperatures. 

The structure grew from the incipient stage into a fully working structure 

fire within fifteen minutes. The building was allowed to totally burn to the 

ground before some measure of suppression was applied. The total time of 

the burn was slightly in excess of one hour. 

On 25 November 2007, a crew of volunteer fire fighters commenced a 

pedestrian search of the structure remnants with a pre-search briefing 


RESULTS: 

which involved a homicide scenario. All or any artifacts/remains were 

then flagged and the search was concluded. 

Following this initial primary search by the fire crew, 13 fourth year 

archaeology students from an area university commenced setting up an 

archaeological style grid over the fire debris by the utilization of tape 

measures, chaining pins and colored string. The artifacts located initially 

by the fire crews were measured, mapped and removed. Once this was 

done, the students, who also have an osteological background, commenced 

the scene excavation, GRID by GRID and the debris run through a ¼ “ 

screen. 

All artifacts/remains located by the students were also measured, mapped 

and removed from the areas where they were located. 

Once the secondary search was completed, the artifacts/remains located by 

both search teams were compared and documented. 

FIRST CREW SECOND CREW 

7 pig limbs/tags 1 - rifle bolt 

1 - Shotgun shell 

2 - Rifle barrels 

12 - Pig limbs 

7 - Steel tags/carpet samples 

3 - Handgun casings 

2 - Pig cadavers 

1 - Shotgun shell 

NOTE: the handgun casings and shotgun shell were found during the screening process, along 

with two pig limbs the remainder were found in situ. The carpet sections were totally destroyed 

in the fire leaving the stainless steel tags. 

THERMOCOUPLE READINGS: 

During the course of this structure fire, there were six thermocouples placed throughout 

the structure at various heights, four of which were associated to the pig cadavers both on top 

and underneath. The data logger unit, situated a short distance from the fire recorded the 

internal structure temperatures in ten second intervals. With the utilization of an accelerant, the 

temperature immediately climbed from 4° Celsius to 577° Celsius in 10 seconds before 

dropping to a predictable level to just below 200° Celsius in just a few moments. 

The internal structure temperatures continued at this level for approximately eighteen 

minutes where there were visible upward spikes, although slight, they continued to rise while 

the fuel load within the structure reached ignition temperatures and became fully involved. 

Interestingly enough, the thermocouples placed under the pig cadavers remained at a 

constant low temperature throughout the fire until approximately the forty minute mark, where 

the internal temperatures reached in excess of 1100° Celsius where the thermocouples became 

compromised with heat and fire impingement. 


Appendix 6. Example of Real Scene (Results and Procedures) 

DATE: 20 October 2007 

WORKING FIRE #6 

SYNOPSIS: 

On Saturday 20 October 2007 at approximately 0230 hours, a lone male in a 2007 

Hyundai four door motor vehicle was observed on closed-circuit television driving around to the 

rear of an industrial area. The vehicle sat for almost an hour before the male is observed exiting 

the vehicle and attending the back portion/trunk area. Within several moments, a bright flash is 

seen at the rear of the vehicle and died almost immediately. 

The male is then observed getting back into the driver’s side of the vehicle when 

approximately 20 minutes later another flash is seen, this time inside the vehicle. The vehicle 

immediately burst into flames and became completely engulfed by fire. A short time later, the 

fire department attended and applied suppression. Following an initial assessment of the fire, the 

fire personnel located the body of the deceased within the vehicle. 

SCENE: 

The scene took place at the rear of a large industrial complex, hidden from the front 

roadway. Located there was a mid-size 2007 four door Hyundai motor vehicle with the interior 

totally consumed by fire. 

The interior of the vehicle is as follows: 

Front: two captain-type seats 

Rear: one full length bench seat 

Deceased: 45 year old male 

Body Condition: 

The deceased was almost completely calcined in the extremities with the remaining tissue 

confined to the torso area. At the time of post mortem, the deceased weighed approximately 70 

lbs. 

SCENE PROCESSING: 

A decision was made to process the scene by utilizing archaeological techniques. The 

vehicle was bisected into two equal parts and further broken down into four grid sections: 

a) Grid #1 – driver’s side compartment to centre dividing line. 

b) Grid #2 – passenger’s side compartment to centre dividing line. 

c) Grid #3 – left rear passenger’s compartment to centre dividing line. 

d) Grid #4 – right rear passenger’s compartment to centre dividing line. 

The excavators consisted of a forensic anthropologist, Ph.D. with extensive forensic 

anthropological experience and originally trained in the discipline of archeology and a forensic 


anthropologist Masc. with a background and experience in archaeology and forensic 

anthropology. 

The excavators commenced the scene recovery by excavating grids in a numerical 

fashion starting with Grid #1, the driver’s seat and compartment. The debris was de-layered 

using a 5” masons trowel and put through a ¼” screen. A sample for fire debris was collected 

from each grid and placed into separate glass Mason jars for subsequent analysis at the Centre of 

Forensic Sciences. 

The male deceased was located partially in the driver’s seat and compartment with the 

right arm splayed over the passenger’s seat and the head and a portion of the upper torso, located 

behind the right front passenger’s seat. The extremities were badly calcined, some to the point of 

ash. 

The deceased was removed from the vehicle by placing the calcined bone into aluminum 

foil and boxed for transport. 

A careful excavation of the remnants of the right front passenger’s seat found the remains 

of the right carpals and metacarpals along with the metal head and inner springs of a disposable 

butane lighter. Also located in this area, were the wire rims to a pairs of men’s prescription 

eyeglasses. 

Information received from the police indicated that this individual had attempted suicide 

on two occasions prior to this date also that he was right handed and had left a suicide note at his 

residence. 

The Centre of Forensic Sciences analyzed the debris samples along with the slight 

clothing remnants collected on the floor of the diver’s compartment. The resulting analysis 

indicated the presence of a medium petroleum distillate (gasoline) present. 

RECOVERED EVIDENCE/REMAINS: 

The following is a list of recovered evidence: 

a) One metal head to disposable butane lighter. 

b) One inner spring from disposable butane lighter. 

c) Metal frame for a men’s pair of prescription eyeglasses. 

COMMENTS: 

The Centre of Forensic Sciences was able to locate gasoline in the debris and clothing 

pieces located on the floor of the driver’s compartment. There was a history of suicide attempts 

for the individual located in the vehicle and a suicide note had been left at the residence. The 

deceased was right handed and after careful excavation of the area of the right hand, the 

remnants of disposable butane lighter were located. 


Appendix 7. Example of Mapped Burn Patterns in the Human Body 

*Taken from Symes et al. (2008). See this reference for further details and discussion. 



Appendix 8. Timeline 


Appendix 9. List of Key Personnel 

Steven A. Symes, Ph.D., is an assistant professor in the Anthropology and Applied Forensic Sciences 

Departments of the Mercyhurst Archaeological Institute at Mercyhurst College in Erie, Pennsylvania and a 

Diplomate of the American Board of Forensic Anthropology. Dr. Symes’ research interests involve human 

skeletal biology with an emphasis on forensic toolmark and fracture pattern interpretation in bone. While a 

leading expert in the interpretation of sharp-force trauma, areas of research include burned bone patterns, and 

the recognition of child abuse with special attention to acute and healing fractures of ribs. Dr. Symes is an 

international lecturer on the topic of bone trauma and has been qualified as an expert for both the prosecution 

and defense in the fields of forensic anthropology, trauma analysis, and toolmark analysis. In the recent past, 

Dr. Symes has presented research on burned bone trauma at professional meetings in places as widespread as 

Guatemala, Hawaii, Milwaukee, and New Orleans. 

Dennis C. Dirkmaat, Ph.D., is a full professor of Anthropology and Director of the Applied Forensic 

Sciences Department at Mercyhurst College, and a Diplomate of the American Board of Forensic Anthropology. 

He has conducted over 300 forensic anthropology cases, including nearly 50 field recoveries involving the 

processing of evidence from human death scenes. He has published articles on the role of archaeology in 

forensic investigation in general, and fatal fire scenes and mass fatalities, in particular. In addition, he has 

presented over 60 lectures and papers discussing forensic investigation and anthropology at numerous regional 

and national meetings. 

Stephen Ousley, Ph.D., is an assistant professor in the Department of Applied Forensic Sciences at 

Mercyhurst College. He earned his Ph.D. Degree at the University of Tennessee, Knoxville. For nine years he 

was the Director of the Repatriation Osteology Laboratory at the National Museum of Natural History, 

Smithsonian Institution. He is best known for his work in quantitative methods in forensic anthropology, 

especially the computer program Fordisc, coauthored with Richard Jantz of the University of Tennessee, 


Knoxville. His other research interests include geometric morphometrics, skeletal biology, human variation, and 

quantitative genetics. 

Gregory O. Olson, BA, is a fire investigator with the Office of the Fire Marshal, Central Region, Midhurst, 

Ontario (Canada). Mr. Olson possesses a unique curriculum, combining his current experience as fire 

investigator with more than 30 years of previous experience as a police officer, 11 of them as Officer-in-Charge 

of the Archaeological-Forensic Recovery Team of the York Regional Police in Ontario. He has also ample 

experience in protocol training and education, having presented numerous talks, both on crime and fire scene 

processing, in different colleges, universities and law enforcement training centers. 

See the Budget Narrative and Management Plan and Organization (p. 25) for details on other project 

participants.

Analysis of Burned Human Remains - Mercyhurst College ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?