THE EFFECT OF DENTAL RESTORATIONS ON TWO DENTAL AGING METHODS UTILIZING TOOTH
ROOT TRANSLUCENCY
By
Kristin E. Perrine

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE
Forensic Science
2012

ABSTRACT
THE EFFECT OF DENTAL RESTORATIONS ON TWO DENTAL AGING METHODS UTILIZING TOOTH
ROOT TRANSLUCENCY
By
Kristin E. Perrine
.
The purpose of this study is to investigate the effects of dental restorations on the
accuracy of the Lamendin (1992) and Bang & Ramm (1970) dental aging methods, which rely on
measurement of tooth root translucency. It has been established that root canal treatment can
have a significant effect on the development of tooth root translucency (Thomas et al., 1994),
but there has been no documentation of the effects of restorations. The sample for this study
consists of 50 premolar teeth with no dental restorations and 50 premolar teeth with
restorations from the William M. Bass Donated Skeletal Collection. Measurement of
periodontosis, root height, and root translucency were taken in millimeters with sliding calipers
and age at death was recorded from the collection database. Age at the time of death was
estimated using the Lamendin and Bang & Ramm methods. The difference between the
estimated age and known age was calculated for each tooth. Error was compared between
teeth with no restorations and teeth with restorations using a student’s T-test. No significant
difference (p < 0.05) was found between the errors of the teeth with restorations and the teeth
without restorations for either method. It is concluded that the presence of dental restorations
does not significantly impact the accuracy of the Lamendin method or the Bang & Ramm
method, and that teeth with restorations may be used to estimate age using dental aging
methods that rely on tooth root translucency.

ACKNOWLEDGEMENTS

I would like to thank my thesis committee, Dr. Norm Sauer, Dr. Todd Fenton, and Dr.
Steve Dow for their help throughout this process. Thanks are also due to Ryan Tubbs and
Wendy and Jered Cornelison for their support and feedback. Special thanks to the University of
Tennessee Forensic Anthropology Center for arranging access to the William M. Bass Donated
Skeletal Collection.
I would like to thank my parents, who have stood by me and provided endless
encouragement and support. Last, but certainly not least, thanks to my husband Gary, for his
unconditional love and unending encouragement (and peanut M&M’s).

iii

TABLE OF CONTENTS

LIST OF TABLES

v

LIST OF FIGURES

vi

CHAPTER 1
INTRODUCTION
Biological Profile
Purpose of this Study

1
1
2

CHAPTER 2
BACKGROUND
Dental Physiology
Age Estimation
Age Estimation in Subadults
Age Estimation in Adults
Research Objectives

4
4
7
7
9
19

CHAPTER 3
MATERIALS AND METHODS

20

CHAPTER 4
RESULTS
Lamendin Method
Bang & Ramm Method
Other Results

25
25
28
31

CHAPTER 5
DISCUSSION
Lamendin Method
Bang & Ramm Method
Utility of the Two Methods

32
33
37
38

CHAPTER 6
CONCLUSION
Areas for Further Research

41
42

APPENDIX

44

BIBLIOGRAPHY

48

iv

LIST OF TABLES

Table 1: Teeth included in this study by type.

21

Table 2: Comparison of measured values for root height, periodontosis height, and
25
root translucency height between teeth with restorations and teeth without restorations.
Table 3: Constants provided by Bang & Ramm (1970) for use with their formulae for
calculating age at death.

28

Table 4: Mean error for age estimations by the Lamendin method by decade.

34

Table 5: Data.

45

v

LIST OF FIGURES

Figure 1: Tooth macroscopic and microscopic anatomy.

4

Figure 2: Frequency of teeth in this study by decade of age at death.

22

Figure 3: Root height and root translucency; periodontosis height.

23

Figure 4: Overestimation and underestimation of age at death by decade using the
Lamendin method.

27

Figure 5: Actual age vs. estimated age using the Lamendin method.

27

Figure 6: Overstimation and underestimation of age at death by decade using the
Bang & Ramm method.

30

Figure 7: Actual age vs. estimation age using the Bang & Ramm method.

30

Figure 8: Brown, nontranslucent tooth roots.

31

Figure 9: Actual ages and estimated ages for the Lamendin method by decade.

34

Figure 10: Mean error by decade for the Lamendin and Bang & Ramm methods for
this study compared to the mean errors published by Lamendin et al. (1992) and
Prince & Ubelaker (2002).

35

Figure 11: Actual ages and estimated ages for the Bang & Ramm method by decade.

38

vi

CHAPTER 1: INTRODUCTION
Biological profile
In forensic anthropology, when human remains are discovered, one of the primary goals
is to identify the individual. Positive identification of an individual relies on unique
characteristics, including DNA, fingerprints, dentition, and bony landmarks used in x-ray
comparison. But before a positive identification can be made, the biological profile provides a
starting point that guides the search for identification. Without a biological profile, one would
not know which records to compare with the unknown individual.
The biological profile consists of the sex, height, ancestry, and age at death of an
individual, as well as any additional information such as anatomical variation, signs of
pathology, or trauma. Physical anthropologists develop and use a biological profile of an
individual in a number of settings. In archaeology, the biological profile reveals valuable
demographic information about the population being studied. In forensic anthropology, the
biological profile is a necessary step in the identification of human remains.
The biological profile is generated by physical anthropologists using a variety of
methods. Macroscopic features, such as the morphology of the pelvis and skull, robusticity of
the bones, and measurements of bony features such as the head of the femur are helpful in
determining an individual’s sex (Ubelaker, 1999). Ancestry is also based on morphological
features, mainly of the skull, but also including other bony features, such as the curvature and
torsion of the femur (Ubelaker, 1999; Bass, 1995). Height can be estimated by measuring all of
the bones that contribute to height and adding a set amount to account for soft tissue (Raxter
et al., 2006), or by measuring a single or multiple long bones and performing calculations

1

(Trotter, 1970). Age at death is based on features such as bone and dental growth and
development, and degenerative changes.
Purpose of this Study
In adults, one way of determining age at death is to make observations and
measurements of the dentition. There are many different methods that can be used to
evaluate the dentition (Bang & Ramm, 1970; Burns & Maples, 1976; Gustafson, 1950; Johanson,
1971; Kvall & Solheim, 1994; Lamendin et al., 1992; Nalbandian et al., 1960; Maples, 1978;
Miles, 1963; Solheim, 1989). Some of these methods (Bang & Ramm, 1970; Lamendin et al.,
1992) rely on the development of tooth root translucency with increasing age.
The purpose of this study is to analyze the effects of dental restorations on the accuracy
of the Lamendin method (Lamendin et al., 1992) and the Bang and Ramm method (1970), both
of which utilize tooth root translucency in intact teeth to estimate age at death. Tooth root
translucency begins at the tip of the root and proceeds toward the crown with advancing age.
It is believed to be caused by calcification within dentinal tubules (Bang & Ramm, 1970). This
changes the refractive index within the dentinal tubule so that it is similar to that of the
material surrounding the dentinal tubules, making the area appear translucent. Although the
effects of root canal treatment on tooth root translucency have been studied (Thomas et al.,
1994), there has been no published work documenting the effects of restorations. Several
methods of age estimation using intact teeth are currently in use, many of which include tooth
root translucency as a criterion, most notably the methods developed by Lamendin et al. (1992)
and Bang and Ramm (1970). Therefore, it is important to have a clear understanding of the

2

impact that restorations might have on the development of tooth root translucency and age
estimation.

3

CHAPTER 2: BACKGROUND
Dental Physiology
Teeth consist of a crown and a root, with a central pulp chamber which contains
connective tissue, blood vessels, and nerves. The pulp chamber is surrounded by a layer of
dentin, a mineralized connective tissue formed by specialized cells called odontoblasts, which
makes up the entirety of the tooth root and the majority of the crown. The denting of the
crown is covered with enamel, which is more highly mineralized than dentin and is secreted by
ameloblasts. The dentin of the root is covered by a thin mineralized layer called cementum.
The line where the enamel of the crown meets the cementum covering the dentin of the root is
called the cementoenamel junction (Fitzgerald & Rose, 2000) (Figure 1).
Predentin
Enamel
Odontoblast
cell bodies
in pulp
chamber

Dentin
with
dentinal
tubules

Cementoenamel
junction

Cementum

Pulp
chamber
Figure 1: Tooth macroscopic and microscopic anatomy.

4

In the forming tooth, a single layer of odontoblasts lines the outer layer of the pulp
chamber. As the odontoblasts lay down predentin from the periphery toward the pulp, the
cells themselves are pushed toward the pulp chamber. As a result, the size of the pulp chamber
decreases as the dentin layer increases. Although the odontoblasts are pushed into the pulp
chamber by the predentin they are depositing, a thin process remains, extending to the
periphery of the tooth through a channel in the dentin called a dentinal tubule (Mjor, 1984)
(Figure 1).
Mineralization of predentin by the deposit of hydroxyapatite extends from the
periphery toward the pulp. Two types of dentin are seen in the developing tooth. Peritubular
dentin is a zone of highly mineralized dentin that surrounds the odontoblast process within the
dentinal tubule. Intertubular dentin is found between dentinal tubules, and is less highly
mineralized than peritubular dentin. Collagen fibers are found in both zones of dentin (Mjor,
1984).
Secondary dentin is dentin that is deposited after the tooth is formed. Regular
secondary dentin forms over time over the entire pulpal surface of the dentin, gradually
reducing the size of the pulpal cavity. Another type of secondary dentin is sclerotic, or
intratubular, dentin, which is deposited within the dentinal tubule. Intratubular dentin is
harder and more brittle than regular dentin, and its deposition may result in partial or complete
obliteration of dentinal tubules (Mjor, 1984).
The origin and pattern of deposition of this sclerotic dentin is not entirely understood.
Although earlier researchers thought that sclerotic dentin formed as a result of continuous
appositional growth of the peritubular dentin (Nalbandian et al., 1959), more recent data

5

suggest that the occlusion of dentinal tubules is a separate process unrelated to the peritubular
dentin (Vasiliadis et al., 1983b; Takuma & Eda, 1966). Vasiliadis et al. (1983b) studied the
structure of dentin in sectioned teeth with light microscopy, microradiography, and scanning
electron microscopy. It was found that, although peritubular dentin and intratubular dentin
have similar characteristics, such as radiodensity, a clear annular space can be seen between
the two, suggesting that continued growth of peritubular dentin is not the cause of dentinal
tubule occlusion.
It has been observed that intratubular dentin is deposited in a predictable pattern
starting soon after the tooth erupts (Vasiliadis et al., 1983a). Vasiliadis et al. (1983b) found that
dentin deposition begins at the periphery and proceeds toward the pulp cavity. This process
begins at the root tip and proceeds coronally over time. It was also observed that occlusion of
dentinal tubules starts on the mesial and distal aspects of the tooth root before proceeding to
the buccal and lingual aspects, forming a distinctive butterfly appearance when the root is
sectioned transversely.
Intratubular dentin appears to be the result of deposition of hydroxyapatite or another
calcium phosphate mineral within the tubule (Balooch et al., 2001). The mechanical properties
of intratubular dentin are similar to those of peritubular and intertubular dentin, with
intratubular dentin being slightly less dense than peritubular dentin but slightly more dense
than intertubular dentin (Balooch et al., 2001). Along with the restoration of dentinal tubules in
regions of transparent dentin, Balooch et al. (2001) also observed chemical changes in the
surrounding intertubular dentin. These chemical changes appear to occur later than the
restoration of dentinal tubules, because a 250-300 um transition zone between normal and

6

transparent dentin was noted in which the tubules are filled, but the chemical composition of
the intertubular dentin is normal.
Age Estimation
Estimation of age at death is an integral part of the biological profile. It is easy to
distinguish a subadult’s remains from those of an adult. Subadults exhibit rapid growth and
development following a predictable pattern, which makes it relatively easy to determine a
subadult’s age at death (Scheuer et al., 2000). Such estimations can usually be narrowed down
to a span of a few years or less. Adults, however, do not continue to grow, so age at death
estimation is based upon degenerative changes which occur slowly and often with great
variability (Ubelaker, 1999). Adult age estimations may have a range of a decade or more.
Age Estimation in Subadults
Age estimation in subadults is most often based on the growth of bones and the
eruption of dentition. The formation and eruption sequence of the deciduous and permanent
dentition is strongly controlled by genetics and is considered to be a reliable indicator of age
(Ubelaker, 1999). Development and fusion of epiphyseal plates of long bones can also be used
to estimate age at death in subadults. However, there are differences in the timing of these
events between the sexes, which necessitate a wider age range if the sex of the individual is not
known (Scheuer & Black, 2000).
The use of age dentition to estimate age was first documented in 1836, with the ruling
that if the first permanent molar had not erupted, a child could be said to be younger than 7
years old. The primary application of dental aging at this time was to determine the age of
children as relevant to child labor laws (Miles, 1963). Since then, much progress has been made

7

in the estimation of age in children, both from dentition and the skeleton. Methods for
estimating the age in children focus either on skeletal development or dental development.
Length of long bones can be useful in determining age at death, as the long bones
increase in length with age (Hoffman, 1979; Fazekas & Kosa, 1978; Scheuer et al., 1980).
However, factors such as nutritional status and disease can cause variation in the rate of bone
growth. Determination of age by long bone length is most accurate for fetal remains (Stewart,
1979). Epiphyseal union is another skeletal technique that can be used to determine age at
death for children. The union of primary and secondary centers of ossification occurs regularly
as a child ages, as the centers of ossification are first completely separate, then fuse together
with a line demarcating their prior separation, and finally the line is obliterated (McKern &
Stewart, 1957; Stewart, 1979). This method is most useful for individuals between the ages of
10 and 25 years, as this is when most of the epiphyses are in the process of fusing (Buikstra &
Ubelaker, 1994).
Dental development can be more useful for estimation of age in children, because the
dentition is less likely than bones to be affected by nutritional deficits and disease status
(Fanning, 1961). Throughout childhood, the deciduous dentition first forms within the jaws,
then erupts. As the permanent dentition forms within the jaws and begins eruption, the roots
of the deciduous teeth resorb. All of these processes (formation of the crown and root of
deciduous and permanent dentition, eruption of deciduous and permanent dentition, and
resorption of deciduous dentition roots) occur in a regular and predictable order and can be
used to estimate at age death in children (Moorees et al., 1963a,b; Schour & Massler, 1941;

8

Ubelaker, 1999). However, estimation of age in adults, after dental and skeletal development
has reached completion, remains problematic.
Dental Age Estimation in Adults
Age estimation in adults is an integral part of any biological profile, which may be
generated as part of archaeological or forensic investigations. Reliable methods for age
estimation using the adult skeleton have been developed, but dental aging methods must be
used in cases where the postcranial remains have been destroyed through trauma or burial
conditions (Maples, 1978). Skeletal methods of age estimation in adults include pubic
symphysis analysis (Suchey & Katz, 1998), auricular surface analysis (Buckberry & Chamberlain,
2002; Lovejoy et al., 1985), changes in the sternal end of the fourth rib (Iscan et al., 1985a;
Iscan et al., 1985b), and cranial suture closure (Meindl & Lovejoy, 1985; Nawrocki, 1998). The
best results for age estimation are obtained when a variety of methods are applied, along with
the overall impression of an experienced forensic anthropologist (Baccino et al., 1999).
The first scientific method of adult dental age estimation was published by Gustafson in
1950 (Solheim & Sundnes, 1980). Gustafson’s (1950) method relied on six observable changes
in the teeth. These changes had previously been described in the literature, but no objective
method for estimating age from them had been developed. The six traits considered by
Gustafson (1950) are attrition, periodontosis, secondary dentin within the pulp cavity,
cementum apposition at the tooth root, root resorption, and root transparency. Examination
of teeth using Gustafson’s (1950) method requires the preparation of ground sections followed
by microscopic examination. Each trait is scored on a scale of 0 to 3, which introduces a level of
subjectivity into the method. The scores for each trait are totaled and the estimated age is

9

based on a regression formula. Gustafson developed his regression formula from the
examination of 41 teeth and tested its accuracy using 22 teeth. Although Gustafson claimed an
error of only ± 3.63 years, later researchers were not able to replicate this low error rate (Burns
& Maples, 1975).
The next advance in the development of a reliable method for age estimation using
dental changes was made by Miles (1963), who attempted age estimation by considering only
tooth root translucency. Rather than subjectively score the degree of translucency as
Gustafson (1950) did, Miles (1963) took measurements of tooth root translucency on sectioned
teeth, making his method more objective. Miles found slightly less favorable results with
measuring root translucency than when he applied the Gustafson method (33% of estimates
within 3 years of actual age, versus 38% with the Gustafson method). However, this study
confirmed the correlation between root translucency and age (Miles, 1963).
Bang and Ramm (1970) also studied tooth root translucency in isolation. Bang and
Ramm (1970) developed formulae to estimate age using either intact or sectioned teeth from a
sample of 978 tooth roots from 926 teeth from 265 individuals. The teeth were collected from
201 patients after extraction, and from 64 individuals at autopsy. Bang and Ramm’s (1970)
formulae allow the estimation of age from intact tooth roots using only the measurements of
tooth root translucency and total root length, both of which were measured on the buccal side
of the tooth for both intact and sectioned roots. Tooth root length was measured from the
cementoenamel junction to the apex of the root. Root translucency was measured from the
apex to the junction with normal dentin. If this junction occurred in a relatively horizontal line
across the buccal surface of the tooth, one measurement was taken. If this junction did not

10

appear to be a relatively horizontal line, a minimum and maximum measurement were taken
and then averaged.
Regression equations were calculated for each of the 32 teeth individually to avoid bias
from using multiple teeth from the same individual in the calculations. Separate regression
equations were developed from the measurements taken on intact and sectioned teeth.
Separate formulae are also provided for teeth with root translucency 9.0 mm or less and teeth
with root translucency greater than 9.0 mm, because the equation that was most accurate for
younger individuals with less root translucency predicted a decrease in tooth root translucency
after age 70. The addition of a second equation for teeth with greater amounts of translucency
fixes that error. The authors found the highest correlation coefficients were associated with
the molars, but had a relatively small number of molars and upper premolars in their study.
Therefore, the authors recommend using incisors and canines for the best results. The authors
did not find any significant difference in the coefficients for sectioned and intact teeth, and
therefore conclude that teeth do not need to be sectioned to obtain the best results (Bang &
Ramm, 1970).
Bang and Ramm (1970) tested their technique using 168 teeth from 24 individuals, and
found it to generally be reliable (58% of individuals fell within the 35% confidence interval of +/4.2 to 4.7 years, and 79% fell within the 65% confidence interval of +/- 9.2 to 10.5 years).
However, the testing sample was very small, and even in this sample it was noted that the
formulae had a tendency to underestimate age in older individuals (especially those over the
age of 75).

11

Although the Bang and Ramm (1970) method is fairly reliable and does not require
sectioning of teeth or complicated measurements, it is not necessarily a user-friendly method
because there are multiple regression equations available, and the user must refer to several
tables to find the correct equation and coefficients to use.
Although tooth root translucency had been studied several times in the 25 years
following Gustafson’s publication (1950), no further research had been published on the
remainder of Gustafson’s variables. Maples (1978) revisited Gustafson’s six traits with the goal
of increasing the precision of the method while reducing the number of variables to be
measured. Using multiple regression, Maples (1978) examined the correlation of each trait
with age on a sample of 355 sectioned teeth. It was found that root translucency is the trait
most closely correlated with age, followed by secondary dentin, attrition, periodontosis, and
cementum. Root resorption was not found to be a useful trait in age estimation. Maples
(1978) used multiple regression to determine the combination of traits that would give the best
estimate of age and concluded that the use of root translucency combined with secondary
dentin deposition is sufficient in most cases.
Tooth root translucency was emerging as a reliable indicator of age at death in adults.
Solheim (1989) further investigated root translucency and the possible influences of sex and
periodontal destruction using a sample of 1000 single-rooted teeth. They also investigated the
difference between measuring the translucent zone on intact and sectioned teeth. In this
study, measurement of the translucent area on intact teeth was found to have a stronger
correlation with age than measurement on sectioned teeth. One significant finding is that
periodontal destruction as a reason for extraction was not found to be correlated with the size

12

of the translucent area, suggesting that periodontal pathology does not have an effect on tooth
root translucency (Solheim, 1989).
Despite the research published since Gustafson (1950), no new method of dental aging
in adults had arisen that was widely accepted to the exclusion of other methods (Solheim &
Sundnes, 1980), until a new method for dental aging was proposed by Lamendin et al. (1992).
This method combined tooth root translucency and periodontosis in a user friendly, simple
method. Lamendin et al. (1992) criticized Gustafson’s (1950) work based on small sample size,
subjective scoring, and the destructive examination. Lamendin et al. (1992) focused on two
traits that had been used by Gustafson: periodontosis and root transparency. Both of these
traits can be objectively measured without sectioning the tooth. Periodontosis is the recession
and degeneration of the gingival attachment on the tooth neck. It is seen as a “smooth and
yellowish area” that is “darker…but clearer than the rest of the root” (Lamendin et al., 1992). It
is not uncommon to see deposits of tartar at this level as well. Periodontosis was measured
from the cementoenamel junction to the line of periodontosis. Tooth root translucency was
measured from the apex to the maximum height of translucency. Root height was measured as
the distance from the apex to the cementoenamel junction. All of these features were
measured on the buccal side of the tooth to the nearest millimeter (Lamendin et al., 1992).
Lamendin et al. (1992) developed a regression equation based on a study of 306 teeth
from 208 individuals. The equation cannot be used for individuals under age 25, because the
constant in the regression equation yields a minimum age of 25. The mean error between the
actual age and estimated age was calculated for each tooth. This calculation revealed that the
method is most accurate for individuals aged 40-79 years and less accurate for younger and

13

older individuals. No difference in accuracy was seen between the sexes or for different tooth
types (Lamendin et al., 1992).
The equation was tested using 45 teeth from 24 forensic cases (Lamendin et al., 1992).
The equation was found to be most accurate for individuals over 40 years (90% fell within the
mean error range for the method), and less accurate for individuals aged 25-40 (46% fell within
the mean error range for the method). The equation was also compared to the Gustafson
(1950) method on a sample of 39 individuals, and the mean error of the Lamendin method was
found to be significantly lower than that of the Gustafson method (Lamendin et al., 1992).
The Lamendin method has gained favor for several reasons. Most importantly, it does
not require any special preparation or destruction of the tooth, whereas many other dental
aging methods require sectioning of the teeth (Gustafson, 1950; Johanson, 1971). Another
reason is that the Lamendin method requires no special equipment to gather data, such as xrays or color scales (Solheim, 1993; Kvaal & Solheim, 1994). Finally, the accuracy of the
Lamendin method is not impacted by the researcher’s inexperience, meaning that one does not
have to be an expert in dentition to obtain accurate results (Baccino et al., 1999).
Recently, some researchers have compared multiple techniques for estimating age from
the dentition to determine which methods are considered to be most useful (Soomer et al.,
2003; Meinl et al., 2008; Reppien et al., 2006). Soomer et al. (2003) examined eight different
methods. Two of the methods can be used with teeth in situ, three methods use intact teeth
that have been extracted, and the remaining methods require sectioning of the teeth. The
most user friendly methods are those that do not require any specialized equipment or
sectioning of the teeth (Soomer et al., 2003). The two methods examined by Soomer et al.

14

(2003) that fit these criteria are those by Lamendin et al. (1992) and Bang and Ramm (1970).
Meinl et al. (2008) compared three methods for dental aging. One method was histological, the
other two methods were those of Lamendin et al. (1992) and Bang and Ramm (1970). Although
histological examination of tooth cementum annulations proved to be the most accurate
method across all age groups, this method requires special equipment and expertise (Meinl et
al., 2008). Finally, Reppien et al. (2006) performed a retrospective study of dental methods
used to estimate age in adults at the Institute of Forensic Medicine in Denmark from 1984 to
2004. The three most frequently used methods were Bang and Ramm (1970), Gustafson
(1950), and clinical evaluation (an observer’s overall subjective impression of the dentition). Of
the most user-friendly methods for estimating age from the adult dentition, the methods of
Lamendin and Bang and Ramm have emerged as the most frequently used (Soomer et al., 2003;
Reppien et al., 2006).
The utility of both the Lamendin method and the Bang and Ramm method is clear.
These methods are fast, easy to use, and do not require any special equipment. Both methods
provide a relatively accurate estimation of age that is useful in both forensic and archaeological
contexts. These methods both rely on a measurement of tooth root translucency. Tooth root
translucency appears to steadily develop as an individual ages. However, it is important to
determine if any external factors affect the rate of development of tooth root translucency. If
any factors are discovered that do affect the rate of translucency, these factor would influence
the accuracy of dental aging methods that rely on tooth root translucency.
Maples (1978) concluded that sex and ancestry do not have any effect on the
development of tooth root translucency. In a sample of 355 teeth that were evaluated for

15

tooth root translucency and secondary dentin development, little difference was found in age
estimation between blacks and whites and between males and females (Maples, 1978). Several
other researchers who have compared tooth root translucency between males and females
have found no significant differences (Lamendin et al., 1992; Bang & Ramm, 1970; Drusini,
1991).
Whittaker and Bakri (1996) conducted a study using 198 teeth of Caucasian, Malay,
Chinese Malay, and Indian Malay origin. It was found that root translucency correlated most
closely with age among Caucasian teeth, and least closely among Chinese Malay teeth. The
authors speculate that these differences may be due to ancestry or cultural differences, as the
groups included in the study have different eating habits. They recommend caution in using
tooth root translucency to estimate age in cases where ancestry or cultural practices are
different from the populations which were used to construct the regression equation being
applied (Whittaker & Bakri, 1996).
The most comprehensive study of the effects of ancestry and sex on the accuracy of the
Lamendin method was performed by Prince and Ubelaker (2002). In this study, a sample of 400
teeth from 94 black females, 72 white females, 98 black males, and 95 white males was used to
test the accuracy of the Lamendin method. They obtained similar overall error rates as were
obtained in Lamendin et al.’s (1992) original study. Analysis of sex and ancestry individually
revealed that sex could have some influence on the accuracy of the method, with females
showing greater errors than males (Prince & Ubelaker, 2002). However, the females in this
study were not evenly distributed, with more older females than older males included in the
sample. Because the Lamendin method has greater errors in older individuals, the greater

16

errors in females might be due to the older age, rather than an actual sex difference. Although
ancestry alone was not found to be a significant factor in the accuracy of the Lamendin method,
the lowest errors were obtained when regression formulae were used that took both sex and
ancestry into consideration (Prince & Ubelaker, 2002).
Another complicating factor in the estimation of age from tooth root translucency is the
postmortem interval, which can affect the appearance of teeth and the ability to measure
periodontosis and tooth root translucency (Megyesi et al., 2006; Sengupta et al., 1999; Kvall &
During, 1999; Mandojana et al., 2001; Solheim, 1993). Solheim (1993) observed that teeth
from deceased individuals were darker than teeth extracted from living individuals. Kvall and
During (1999) found that the translucent zone in teeth could be observed even after 300 years
in an underwater archaeological site. However, they also noted that some teeth were badly
damaged by the postmortem environment, and note that teeth which do not appear to be
damaged may still not be as accurate in age estimation given a long postmortem interval.
Sengupta et al. (1999) found, in their study of 56 modern teeth and 61 archaeological
specimens, that a microscopically “chalky” appearance of the dentin was common among
sectioned teeth from the archaeological collection. The authors considered this to be a postdepositional change, although the origin was not clear. The amount of “chalkiness” was not
correlated with the length of time that the remains were buried. It was also not clear whether
the “chalky” dentin invaded or replaced the sclerotic dentin of the translucent area of the tooth
root, although the “chalky” dentin did appear to originate in the root canal of the tooth and
spread into the root from there. Overall, Sengupta et al. (1999) concluded that many
postmortem processes involving microorganisms, dissolution and remineralization of inorganic

17

components, and removal of organic components have the potential to affect the appearance
of tooth root translucency and therefore age estimation. Megyesi et al. (2006) studied 1,188
teeth from 220 individuals from an archaeological collection, and noted that approximately 36%
of the teeth exhibited a lack of root translucency macroscopically. The teeth were not
sectioned, so the authors speculated that this lack of translucency is correlated with the
“chalky” dentin mentioned by Sengupta et al. (1999), but this was not confirmed. No mention
was made of the macroscopic appearance of the tooth root other than the lack of translucency.
It was not stated whether the non-transparent teeth were occasional findings, or if they were
clustered in certain individuals. Megyesi et al. (2006) also noted that there was some difficulty
in identifying the periodontal line in archaeological specimens. In specimens that were in good
condition with an intact and smooth root (even with the presence of hypercementosis and
stains), the periodontal line was observed as a darkened line on the root surface. The
periodontal line was much harder to observe on teeth in poor condition with cracked and
degraded roots (Megyesi et al., 2006).
Thomas et al. (1994) studied the effects of root canal treatment on the amount of root
translucency. This study, which measured both the area and height of root translucency, used
age-matched pairs of teeth, both vital and non-vital at the time of extraction. It was found that
non-vital teeth contain a greater amount of translucent dentin than vital teeth. This would
result in an over-estimation of age when using the Lamendin or Bang and Ramm method. Bang
and Ramm (1970) also studied a small sample (16 teeth) with root canal treatment. It was
found that root transparency in these teeth was variable, with many of these teeth exhibiting
no transparency at all. These studies show that teeth with root canal treatment can be more

18

variable in development of root translucency. Given the large sample size of Thomas et al.
(1994) and the very small sample size of Bang and Ramm (1970), one can expect that teeth with
root canal treatment will usually lead to an overestimation of age.
Research Objectives
Although some studies have included teeth with restorations, no investigation was
made into the effects that these restorations might have on the accuracy of age estimation
(Bang and Ramm, 1970). The purpose of this research is to determine whether the Lamendin
and Bang and Ramm methods can be used to estimate age at death from teeth with dental
restorations as accurately as they are used in teeth without dental restorations. It is expected
that restorations in otherwise intact teeth will have no effect on the amount of tooth root
translucency or accuracy of the Lamendin and Bang and Ramm methods. Previous research has
shown that root canal treatment does affect the amount of tooth root translucency (Bang &
Ramm, 1970; Thomas, 1994). However, dental restorations are surface treatments of the
tooth, extending through the enamel and into the dentin, but not into the root canal. Tooth
root translucency related to aging begins at the apex of the tooth and proceeds toward the
crown. Dentin translucency may occur in the crown or the cervical region of the tooth, but this
is in response to trauma or pathology rather than age (Thomas et al, 1994). Dental restorations
may result in translucency in areas other than the apex of the tooth, but should not accelerate
apical tooth root translucency.

19

CHAPTER 3: MATERIALS AND METHODS
For this study, one hundred premolars were selected from the William M. Bass Donated
Skeletal Collection. This collection was selected for this research because it is well-documented
and of recent origin. Although Lamendin et al. (2002) found that the maxillary central incisors
generate the most accurate age estimate and Bang and Ramm (1970) reported that incisors and
canines were easiest to measure reliably, premolars were chosen because they are the most
likely single rooted tooth type to have restorations. Molars are not appropriate for use with
the Lamendin method, as they have multiple roots (Lamendin et al., 1992). Although Bang and
Ramm (1970) included molars in their study, they reported difficulty with getting accurate
measurements. Teeth were only selected from individuals aged 30 or older at the time of
death. The Lamendin method cannot be used for individuals less than 25, and tends to be less
accurate at younger ages (Lamedin et al, 1992).
A list of all individuals in the database aged 30 or older at time of death and with
premolars present was obtained. There was some uncertainty at the beginning of data
collection about how many premolars with restorations might be encountered, which was
further bolstered by the discovery that many premolar teeth in the collection were glued or
otherwise lodged in the alveolar processes, making it impossible to examine the tooth root.
Data collection therefore involved taking measurements on all premolars that were
encountered, rather than just one tooth per individual. Once it was established that
measurements had been taken on 50 premolars with restorations and 50 premolars without
restorations, from 100 unique individuals, data collection ceased.

20

Fewer teeth with restorations were measured than teeth without restorations, so the
first step in narrowing down the dataset was to select teeth with restorations, and then select
teeth without restorations from the remaining individuals. Within these constraints, an
arbitrary system was devised to ensure impartiality in the selection of teeth to be included in
the study. Selection priority was given first to the left mandibular first premolar, then to left
mandibular second premolar, followed by the right mandibular first and second premolars, the
left maxillary first and second premolars, and finally the right maxillary first and second
premolars until 100 teeth had been selected for inclusion in the study (Table 1).
Selection
Priority
1

Tooth (premolar)
Left mandibular first

Number Included
in Study
31

2

Left mandibular second

15

3

Right mandibular first

18

4

Right mandibular second

12

5

Left maxillary first

4

6

Left maxillary second

11

7

Right maxillary first

5

8

Right maxillary second

4

Table 1: Teeth included in this study by type.
The individuals in the study range in age from 31 to 96 years at time of death, with a
mean age of 58 years. The sample of teeth with restorations tended to be from slightly older
individuals, with a mean age of 62, versus a mean age of 54 for individuals with teeth without

21

restorations (Figure 2). This is not surprising, as the frequency of restorations increases with
age. Among the 50 teeth with restorations, 47 of the restorations were amalgam, while 3 were
composite restorations. Broken teeth were excluded from this study because the presence or

Number of Teeth

absence of a restoration could not be adequately assessed.
18
16
14
12
10
8
6
4
2
0

Restorations
No Restorations

30-39

40-49 50-59 60-69 70-79
Age at Death Decade

80+

Figure 2: Frequency of teeth in this study by decade of age at death.
Teeth from 71 males and 29 females were included in the study. The majority of the
individuals were classified as white (89). The study also included 8 individuals classified as
black, 2 classified as Hispanic, and 1 individual of unknown race. Although some studies have
found racial differences in the development of tooth root tranclucency (Whittaker & Bakri,
1996), this question is not addressed here because of the small sample size.
All measurements were taken to the hundredth of a millimeter as described by
Lamendin et al. (1992) on the labial side of the tooth with digital sliding calipers. Root
translucency was observed using a light box. Age at death was recorded from the collection
database. Tooth number was recorded at the time of measurement, as well as the presence or

22

absence of restorations. Root height was measured as the distance between the
cementoenamel junction and the apex of the root. Periodontosis was measured as the
maximum distance from the cementoenamel junction to the line of soft tissue attachment.
Lamendin et al. (1992) describe the line of soft tissue attachment as “a smooth and yellowish
area below the enamel, and darker than it, but clearer than the rest of the root.” It is noted
that tartar buildup is often seen at this line. Root translucency is defined as the distance from
the apex of the root to the maximum height of translucency (Figure 3).

Figure 3: Left: Root height (longer arrow) and root translucency (shorter arrow).
Right: Periodontosis height. For interpretation of the references to color in this and all
other figures, the reader is referred to the electronic version of this thesis.
To check for errors in measurement, the regression equation of the Lamendin method
was applied to each tooth included in the study, and any age estimate that was 20 years or
more different from the actual age at death was flagged. Twenty years was chosen because it
is greater than the mean errors provided by Lamendin et al. (1992) for each age category
included in this study. These flagged teeth were re-measured, with the new measurements
replacing the old. Five teeth were re-measured, with three periodontosis values found to be
more than 1mm different from the previously measured values. All three of these teeth were

23

measured early in the data collection process, when the researcher was not as familiar with
determining the line of periodontosis.
The primary objective of this study is to assess whether restorations affect the accuracy
of the Lamendin or Bang & Ramm methods, and therefore whether restorations affect the
development of tooth root translucency. The null hypothesis of this study is that no difference
will be found between estimated ages for teeth with restorations and teeth without
restorations. A t test will be used to compare estimated ages for teeth with restorations to
teeth without restorations. The second objective of this study is to assess the error rates of
each method, and to use this information to discuss the overall utility of each method in
forensic science.

24

CHAPTER 4: RESULTS
For each tooth, root length, periodontosis, and root translucency were measured. The
root height measurements ranged from 11.22 to 18.87, with an average of 15.07 and standard
deviation of 1.73. The periodontosis measurements ranged from 6.53 to 16.95, with an
average of 12.49 and standard deviation of 1.83. The root translucency measurements ranged
from 0.00 to 14.82, with an average of 6.52 and standard deviation of 3.05. Measured values
were very similar for teeth with restorations and teeth without restorations (Table 2).
Measurement

Average Root Height
Standard Deviation of Root
Height
Average Periodontosis Height
Standard Deviation of
Periodontosis Height
Average Root Translucency
Height
Standard Deviation of Root
Translucency Height

All
Teeth

Teeth With
Teeth
Restorations Without
Restorations
15.07
14.98
15.16
1.73
1.63
1.84
12.49
1.83

12.22
1.99

12.76
1.63

6.52

6.86

6.17

3.05

3.00

3.09

Table 2: Comparison of measured values for root height, periodontosis height, and root
translucency height between teeth with restorations and teeth without restorations.
Lamendin Method
For each individual, age at the time of death was calculated using the Lamendin method
(Lamendin et al., 1992):

25

Overall, estimated ages ranged from 41.20 to 79.82, with an average of 58.70 and standard
deviation of 8.45. For teeth with restorations, estimated ages ranged from 41.94 to 71.32, with
an average of 59.41 and standard deviation of 7.84. For teeth without restorations, estimated
ages ranged from 41.20 to 79.82, with an average of 58.00 and standard deviation of 9.05.
The error was calculated for each tooth as the difference between the estimated age
and the known age. The absolute value of the error for teeth with restorations ranged from
0.15 to 28.17, with an average of 8.79 and standard deviation of 6.83. The absolute value of
the error for teeth without restorations ranged from 0.29 to 32.15, with an average of 10.48
and standard deviation of 7.69.
The absolute value of the error was compared between teeth with restorations and
teeth without restorations using a two-tailed t test. The null hypothesis, that the mean
absolute value of the error is the same for teeth with restorations as for teeth without
restorations, cannot be rejected with any level of statistical significance, with p = 0.25.
Overall, the average error by age group was 13.5 for ages 30-39, 9.6 for ages 40-49, 6.7
for ages 50-59, 6.1 for ages 60-69, 9.6 for ages 70-79, and 21.7 for individuals 80 and older. For
the 8 individuals aged 30-39, the Lamendin method overestimated age at death of all of them.
For the 23 individuals aged 40-49, the Lamendin method overestimated age at death for 19 of
them (83%). For the 30 individuals aged 50-59, the Lamendin method overestimated age at
death for 23 of them (77%). For the 18 individuals aged 60-69, the Lamendin method
overestimated age at death for 3 of them (17%). For the 11 individuals aged 70-79, the
Lamendin method overestimated age at death for 1 of them (9%). For the 10 individuals over
age 80, the Lamendin method overestimated age at death for none of them (Figures 4 and 5).

26

Number of Teeth

25
20
15
10
5
0
30-39

40-49

50-59
60-69
70-79
Age at Death by Decade

overestimates

80+

underestimates

Figure 4: Overestimation and underestimation of age at death by decade using the Lamendin
method.
100
90

Estimated age

80
70

60
50
40
30
30

40

50

60
70
Actual age

80

90

100

Figure 5: Actual age vs. estimated age using the Lamendin method. Thin line is the regression
line. Thick line indicates actual age = estimated age.
27

Bang & Ramm Method
For each individual, age at the time of death was calculated using the appropriate
formula from Bang and Ramm (1970) (Table 3):
For teeth with transparency ≤ 9.0mm:
For teeth with transparency < 9.0mm:

Transparency less than or equal
to 9.0 mm
Upper first premolars
Upper right second premolar
Upper left second premolar
Lower right first premolar
Lower left first premolar
Lower right second premolar
Lower left second premolar

B0
23.91
23.78
25.95
24.83
29.17
29.42
18.72

B1
3.02
5.06
4.07
6.85
5.96
4.49
5.79

B2
0.203
-0.064
-0.067
-0.237
-0.173
-0.065
-0.082

Transparency
greater than 9.0mm
B0
18.42
25.33
26.92
30.83
34.97
30.68
20.87

B1
5.4
4.28
3.37
4.05
3.74
3.76
4.79

Table 3: Constants provided by Bang & Ramm (1970) for use with their formulae for calculating
age at death.

Overall, estimated ages ranged from 18.72 to 90.40, with an average of 55.78 and
standard deviation of 14.16. For teeth with restorations, estimated ages ranged from 23.91 to
77.45 with an average of 56.30 and a standard deviation of 14.19. For teeth without
restorations, estimated ages ranged from 18.72 to 90.94 with an average of 55.27 and a
standard deviation of 14.27.
The error was calculated for each tooth as the difference between the estimated age
and the known age. The absolute value of the error for teeth with restorations ranged from

28

0.09 to 28.09, with an average of 10.62 and a standard deviation of 6.70. The absolute value of
the error for teeth without restorations ranged from 0.18 to 27.28, with an average of 9.53 and
a standard deviation of 7.74.
The absolute value of the error was compared between teeth with restorations and
teeth without restorations using a two-tailed t test. The null hypothesis, that the mean
absolute value of the error is the same for teeth with restorations as for teeth without
restorations, cannot be rejected with any level of statistical significance, with p = 0.45.
Overall, the average error by age group was 6.6 for ages 30-39, 11.2 for ages 40-49, 8.8
for ages 50-59, 8.1 for ages 60-69, 10.5 for ages 70-79, and 17.3 for individuals 80 and older.
For the 8 individuals aged 30-39, the Bang & Ramm method overestimated age at death for 5 of
them (63%). For the 23 individuals aged 40-49, the Bang & Ramm method overestimated age
at death for 10 of them (43%). For the 30 individuals aged 50-59, the Bang & Ramm method
overestimated age at death for 19 of them (63%). For the 18 individuals aged 60-69, the Bang
& Ramm method overestimated age at death for 6 of them (33%). For the 11 individuals aged
70-79, the Bang & Ramm method overestimated age at death for 1 of them (9%). For the 10
individuals over age 80, the Bang & Ramm method overestimated age at death for none of
them (Figures 6 and 7).

29

Number of Teeth

20
18
16
14
12
10
8
6
4
2
0
30-39

40-49

50-59
60-69
70-79
Age at Death by Decade

overestimates

80+

underestimates

Figure 6: Overestimation and underestimation of age at death using the Bang & Ramm method.
100
90

Estimated Age

80
70

60
50
40
30
20

30

40

50

60
70
Actual Age

80

90

100

Figure 7: Actual age vs. estimated age using the Bang & Ramm method. Thin line is the
regression line. Thick line indicates actual age = estimated age.

30

Other Results
Another interesting observation was that of dark tooth roots. A small number of teeth
in the collection were observed to have strikingly dark brown roots on visual inspection. When
these teeth were observed using the light box, no translucency was detected, even when
considerable translucency was observed in other teeth from the same individual (Figure 8).
Megyesi et al. (2006), Kvaal & During (1999), and Sengupta et al. 1999) mention teeth with a
lack of translucency in historical samples, suggesting that the phenomenon may be due to
taphonomic processes. However, the observation in this study of dark teeth with no
translucency in a more recent collection, as well as the presence of such teeth adjacent to
normal teeth in a single individual, suggests that the lack of translucency could be due to a
pathological process that occurred during life, such as trauma to the tooth, disease, or root
canal treatment. Further research is needed to determine the cause of this phenomenon.

Figure 8: Brown, nontranslucent tooth roots.

31

CHAPTER 5: DISCUSSION
The purpose of this study was to determine whether the presence of restorations
affects the accuracy of the Lamendin and Bang and Ramm methods. For the Lamendin method,
comparison of the absolute values of the estimated and known ages at death by t test showed
that there was no statistically significant difference between the two (p = 0.25). For the Bang &
Ramm method, comparison of the absolute values of the estimated and known ages at death
by t test showed that there was no statistically significant difference between the two (p =
0.45). It was found that restorations have no statistically significant effect on the accuracy of
either method. Because both of these methods rely on tooth root translucency for age
estimation, it can be concluded that restorations do not have any significant effect on the rate
of development of tooth root translucency.
One difficulty that was encountered in the study was properly identifying the
periodontal line in dry specimens. Megyesi et al. (2006) discussed some difficulty in identifying
the periodontal line in archaeological specimens, but stated that the line was still readily
identifiable in teeth with roots in good condition. The teeth in this study were generally in good
condition, and relatively recent compared with Megyesi et al.’s (2006) sample. In teeth where
the periodontal line was not readily identifiable on the buccal surface of the tooth, rotating the
tooth to view the mesial and distal surfaces was helpful. The line was often easier to identify
on the mesial and distal surfaces, and, once identified there, could be followed onto the buccal
surface for measurement.
As the researcher became more skilled in observation of the periodontal line, it became
easier to identify. However, in some of the first specimens measured, the periodontal line was

32

misidentified. At the end of data collection, comparison of the estimated age at death (via the
Lamendin method) and the known age at death revealed several specimens where the
estimated age at death was more than 20 years different from the known age at death. In
these specimens, measurements were taken again, and in three of the specimens, it was found
that the periodontal line had been incorrectly identified. The correct measurement replaced
the previous measurement for data analysis.
Lamendin Method
It was found that the Lamendin method tends to overestimate the age of younger individuals
and underestimate the age of older individuals. This is consistent with Lamendin et al.’s (1992)
original findings, as well as the findings of later researchers (Prince & Ubelaker, 2002). Age was
overestimated in all of the individuals age 30 to 39, overestimated in the majority of individuals
age 40-59, and underestimated in all of the individuals over age 80. This can be seen when
looking at the dispersion of the actual ages versus the dispersion of the estimated ages.
Although the mean for both is almost identical, the standard deviation of the estimated ages
(8.5) is much lower than that of the actual ages (15.1). This means that the estimated ages
were more tightly clustered around the mean than the actual ages (Figure 9).
The average error was greater for the youngest and oldest age categories. For this
sample, the Lamendin method was found to be most accurate for individuals aged 50 to 69
years at the time of death. Even in this age range, however, one individual’s age was
overestimated by 28 years. The largest errors were found in individuals over age 80, and the
next largest errors were found in individuals aged 30 to 39. Lamendin et al. (1992) presented a

33

chart of mean error between actual and estimated age by decade. The mean error for this
study was similarly calculated by decade (Table 4).

Number of Teeth

25
20
15
10

actual
estimated

5

31-35
36-40
41-45
46-50
51-55
56-60
61-65
66-70
71-75
76-80
81-85
86-90
91-95
96-100

0

Age at Death
Figure 9: Actual ages and estimated ages for the Lamendin method by decade.
Age intervals

30-39

40-49

50-59

60-69

70-79

80+

Lamendin et
al. (1992)
mean error
Current study
mean error

15.5

9.9

7.3

6.3

11.6

18.9

13.5

9.6

6.7

6.1

9.6

21.7

Table 4: Mean error for age estimations by the Lamendin method by decade.
The error rates obtained in this study using the Lamendin method are very similar to
those originally obtained by Lamendin et al. (1992), and also similar to those obtained by Prince
& Ubelaker (2002) in their test of the Lamendin method (Figure 10). The error rates obtained in
this study using the Bang & Ramm method are also similar to those obtained using the
Lamendin method. The author began data collection with very little experience in using these

34

methods, so these similar error rates confirm the findings of Baccino et al. (1999), that
experience in the method is not necessary to apply the method correctly.

Mean Error (years)

25

Lamendin method,
current study

20

15

Bang & Ramm
method, current study

10

Lamendin et al.

5
0
30-39

40-49
50-59
60-69
70-79
Age at Death by Decade

80+

Prince & Ubelaker
using the Lamendin
method

Figure 10: Mean error by decade for the Lamendin and Bang & Ramm methods for this study
compared to the mean errors published by Lamendin et al. (1992) and Prince & Ubelaker
(2002).
In recent years, several researchers have critiqued the Lamendin method (Schmitt et al,
2010; Foti et al., 2001; Sperber, 1993). Most significantly, the large error values have been
criticized (Foti et al., 2001). Although the statistical methods of Lamendin et al. (1992) have
been questioned, a recent attempt by Schmitt et al. (2010) to develop a more accurate
technique using different statistical methods failed to match the accuracy of the Lamendin
method. Schmitt et al. (2010) concluded that, although imperfect, the Lamendin method is still
useful as a quick and easy way to estimate age at death.
Another criticism of the Lamendin method is the inclusion of periodontosis, which is less
strongly correlated with age than root translucency (Sperber, 1993). Severe periodontal

35

disease may cause premature recession of the periodontal ligament and an artificially high
measurement of periodontosis, resulting in an overestimation of age at death.
A new regression equation was calculated using the data from this study. Age at death
= -0.34P + 0.44T + 66.48. The negative coefficient on P means that, for individuals with similar T
values, those individuals with higher P values will be estimated to be younger. In other words,
in two individuals with similar amounts of root translucency, this equation will assign a younger
age to the individual with more periodontal ligament recession. The very large constant is a
result of this sample having a larger overall age than Lamendin’s sample.
The new regression equation that was based on the data gathered in this study had a
negative coefficient for periodontosis. This means that, among the individuals in this study,
younger individuals had higher values for periodontosis than older individuals, which is
opposite of what is expected. Dental aging methods which include periodontosis in their
criteria generally assume that age and periodontosis are directly correlated, rather than
inversely correlated as is shown in this study. This anomaly affirms the concerns that
periodontosis may not be a good variable to consider in dental aging methods.
One possible explanation for this anomaly is the small sample size of this study. It
appears that in this case, the negative coefficient is likely the result of a few individuals with
very high P values for their age, rather than an overall trend in the population that was
sampled. It is also possible that socioeconomic factors have influenced the composition of the
skeletal collection used to gather data. Younger individuals with lower socioeconomic status
would not have access to routine dental care, and therefore could be at greater risk for
developing periodontal disease. Skeletons in the Bass Collection can be donated by the

36

individual before death, by the family after death, or by the medical examiner. A wide range of
socioeconomic statuses is represented in the collection. It is possible that, among the
individuals included in this study, the younger individuals tended to have poorer dental care
and therefore increased periodontal disease. Another explanation is that, among the older
individuals in the study, those with severe longstanding periodontal disease are more likely to
have lost affected teeth. This would mean that, among older individuals, teeth remaining
would be from individuals with less periodontal disease (Loe et al., 1986).
Bang & Ramm Method
Bang & Ramm (1970) reported, in a test of their method using 168 teeth from 24
individuals, that 58.3% of the estimated ages fell within ± 4.7 years (35% confidence interval),
and 79.2% fell within ± 10.5 years (65% confidence interval). In the current study, the mean
error for teeth with restorations was 10.62, ranging from 0.09 to 28.09. The mean error for
teeth without restorations was 9.53, ranging from 0.18 to 27.28. For teeth with restorations,
24% of the individuals fell within ± 4.7 years, and 50% fell within ± 10.5 years. For teeth
without restorations, 34% of the individuals fell within ± 4.7 years, and 64% fell within ± 10.5
years. The errors rates obtained in this study are larger than those reported by Bang & Ramm
(1970), but are similar to those obtained in this study using the Lamendin method and reported
by Lamendin et al. (1992) and Prince & Ubelaker (2002) for the Lamendin method.
Bang and Ramm (1970) stated that their method was most reliable in individuals
younger than age 75. Over age 75, they noticed a tendency to underestimate age, as was seen
with the Lamendin method (Lamendin et al., 1992). This tendency is shown in the current
sample, in which the ages of 20 out of 21 individuals over the age of 70 were underestimated.

37

However, unlike the Lamendin method, the Bang and Ramm method shows fewer tendencies
to overestimate the ages of young people. Out of 31 individuals aged 30-49, the Bang and
Ramm method only overestimated the age of 15 individuals (48%), whereas the Lamendin

Number of Teeth

method overestimated the age of 27 individuals (87%) (Figure 11).
20
18
16
14
12
10
8
6
4
2
0

overestimates
underestimates

30-39 40-49 50-59 60-69 70-79 80+
Age at Death by Decade
Figure 11: Actual ages and estimated ages for the Bang & Ramm method by decade.
The mean error in this study for the Bang & Ramm method does not show a clear
pattern as was seen with the Lamendin method. The lowest mean error was found in the
youngest age group (30-39), and the largest mean error was found in the oldest age group (80
or older), but the mean error of the remaining age groups did not show a clear trend (11.2 for
individuals 40-49, 8.8 for individuals 50-59, 8.1 for individuals 60-69, and 10.5 for individuals 7079).
Utility of the Two Methods
Based on the absolute values of the errors found in this study, the Lamendin method
and the Bang and Ramm method are nearly equally accurate. For all 100 teeth in the sample,

38

the Lamendin method had an average absolute value error of 9.63, while the Bang and Ramm
method had an average of 10.08. Based on these figures alone, the Lamendin method and the
Bang and Ramm method should be equally effective. However, there are some differences in
the ways that the error of each method is distributed (See Figure 4 and Figure 6). Both the
Lamendin method and the Bang and Ramm method underestimate the age of older individuals.
The Lamendin method also consistently overestimates the age of younger individuals, whereas
the Bang and Ramm method has more equal levels of underestimation and overestimation in
younger individuals.
The tendency to overestimate the age of younger people and underestimate the age of
older people causes problems with using the Lamendin method for age estimation. Although
the method has very high levels of accuracy in the middle age range, when confronted with an
individual of unknown age it is impossible to know whether the individual falls into the ideal
age range to use the Lamendin method. For example, consider the following three individuals
in this study: UT52-04D had an actual age at death of 50 and estimated age at death of 51.4.
UT49-03D had an actual age at death of 86 and estimated age at death of 53.9. UT17-00D had
an actual age at death of 35 and an estimated age at death of 52.5. All three individuals had an
estimated age at death of around 50 years old. When the first individual is considered in
isolation, the Lamendin method is very accurate at estimating the age at death. Presented only
with data from the middle age ranges, a forensic anthropologist might be tempted to give a
fairly small range for age at death in the biological profile. However, when the other two
individuals are considered, the Lamendin method appears to be very inaccurate, requiring an
age span of over 50 years to encompass all of the individuals with an estimated age at death of

39

50 years. Such a large age span is generally not helpful for identification as part of the
biological profile. Therefore, although the Lamendin method appears to be useful, in practice
large age spans on either side of the estimated age are needed to be sure that that actual age
at death is included. These large age spans severely limit the utility of the Lamendin method in
forensic anthropology practice. There is a similar problem with the Bang and Ramm method,
with some individuals’ ages overestimated or underestimated by nearly 30 years.
The question of whether the Lamendin and Bang and Ramm methods are useful in
forensic anthropology is difficult to answer. Although both methods tend to require wide age
estimate ranges, many other commonly-used age estimations do as well. For example, in the
later phases, the Suchey-Brooks method of pubic symphysis aging has age ranges of 40-50
years, similar to those one might report with the Lamendin or Bang and Ramm methods
(Suchey & Katz, 1998). One advantage that pubic symphysis aging has over these dental
methods is that the age ranges for earlier phases are smaller, allowing for closer estimation of
age in younger individuals. In the dental methods, there is no real way to know whether one is
dealing with a young, middle age, or old individual simply based on the age estimate calculated
from the formula. However, with the knowledge that age at death estimates for middle-aged
individuals are most accurate, other dental characteristics such as wear and staining could be
combined with the calculated estimates from Lamendin or Bang and Ramm to give a smaller
age range. It is always important in forensic anthropology to consider all of the available
evidence when producing a biological profile, rather than to rely on a single calculation or
observation.

40

CHAPTER 6: CONCLUSION
In the field of forensic anthropology, the biological profile is one of the most important
undertakings. Although sex and ancestry are relatively simple to assess in complete adult
skeletal remains, age can be problematic. Age estimation in adults becomes even more difficult
when the remains are incomplete or damaged. Several methods for age determination in
adults are based on degenerative changes in the dentition that occur with advancing age. Two
of the most popular methods, the Lamendin method and the Bang and Ramm method, both
utilize the progression of tooth root translucency to estimate age at death.
This study has shown that the development of tooth root translucency with age as used
in the Lamendin and Bang and Ramm methods is not affected by the presence of restorations in
a tooth. T tests show that there is no significant difference for either method between the
mean errors for teeth without restorations (10.48 for the Lamendin method and 9.53 for the
Bang & Ramm method) and teeth with restorations (8.79 for the Lamendin method and 10.62
for the Bang & Ramm method). Therefore, these methods can be used to estimate age at
death of adults with restorations without altering the published accuracy for the methods.
This study confirmed the findings of Lamendin et al. (1992) that the Lamendin method
tends to uniformly overestimate the age of younger individuals and underestimate the age of
older individuals. Mean error for the Lamendin method in this study was largest in the
youngest (ages 30-39, mean error 13.5) and oldest (over 80, mean error 21.7) age groups.
Mean errors for the remainder of the age groups (40-79) were less than 10, which is
comparable to the overall mean error of 10 presented by Lamendin et al. (1992).

41

The Bang & Ramm method does not overestimate the age of younger individuals as
consistently as the Lamendin method does, however the Bang & Ramm method does
consistently underestimate the age of older individuals (over age 70), as was seen with the
Lamendin method. The mean error in this study for the Bang & Ramm method also does not
show a clear pattern. The lowest mean error was found in the youngest age group and the
largest mean error was found in the oldest age group, but the mean error of the remaining age
groups did not show a clear trend.
The utility of these two methods is clear. They are fast, easy to use, and do not require
any special equipment. The methods provide an estimation of age that can be useful in both
forensic and archaeological contexts. Unfortunately, neither method is ideal. Both methods
can have large errors in estimated ages, especially among younger and older adults. In order to
ensure that the actual age of the individual is included in the range given, large age ranges may
be needed. However, if the estimated age from one of these methods is combined with other
skeletal aging methods and observations, a smaller age range may be arrived at.
Like any age estimation method in forensic anthropology, dental aging methods relying
on tooth root translucency can assist the forensic anthropologist in arriving at an appropriate
age range, but additional observations and professional judgment are necessary to produce the
more accurate biological profile possible. It is important that forensic anthropologists are
aware of the strengths and the shortfalls of any method that they employ.
Areas for Further Research
The phenomenon of dark brown tooth roots is an interesting and possibly unreported
occurrence. Further research into the cause of this phenomenon is warranted.

42

Further improvements on adult dental aging methods are needed. An ideal method
would be as easy to use as the Lamendin method, but provide a higher degree of accuracy. The
routine overestimation of age in younger adults and underestimation of age in older adults is
problematic. A method that was able to estimate age without such tendencies would be more
useful in anthropological practice.

43

APPENDIX

44

Root Translucency Height

Lamendin Method Estimated Age

Error: Lamendin Method

Bang & Ramm Method Estimated
Age

Error: Bang & Ramm Method

Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Composite
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam

Periodontosis Height

Tooth
Rxp2
Rnp1
Lnp2
Rnp2
Lxp1
Rxp1
Lnp1
Rnp1
Lxp1
Lnp2
Rnp2
Lnp1
Rnp2
Rnp1
Lnp1
Rnp2
Rxp1
Lnp2
Rnp1
Rnp1
Lxp2
Rnp2
Lnp2
Rnp2
Rxp2
Lxp1
Lnp1

Root Height

Age at Death
47
31
32
78
82
54
69
41
56
69
52
94
81
73
78
84
54
58
96
79
63
61
42
61
47
55
73

Restoration

UT ID Number
UT12-88D
UT82-07D
UT14-93D
UT18-93D
UT21-93D
UT38-93D
UT44-93D
UT22-95D
UT11-97D
UT10-98D
UT21-98D
UT33-99D
UT23-00D
UT24-00D
UT28-00D
UT20-01D
UT26-01D
UT41-01D
UT01-02D
UT08-02D
UT09-02D
UT10-02D
UT25-02D
UT30-02D
UT01-03D
UT19-03D
UT43-03D

13.91
13.24
16.35
14.58
14.09
15.23
15.65
15.47
13.70
15.94
15.63
16.95
18.87
16.50
15.52
15.14
16.46
18.29
15.41
14.12
11.90
17.83
16.45
12.88
13.91
13.92
12.57

12.06
12.83
13.03
11.18
10.91
13.65
13.61
12.84
10.11
14.67
13.49
15.32
14.03
14.00
13.45
13.29
13.34
14.49
10.76
11.52
7.75
14.87
14.11
9.62
11.12
12.74
9.83

4.92
0.00
2.86
7.61
9.80
4.18
11.05
0.73
7.84
10.32
10.47
10.92
12.24
8.99
7.88
7.86
4.84
3.89
11.51
8.08
6.25
11.08
3.00
5.10
3.71
3.48
6.54

55.99
42.97
47.22
61.25
68.68
53.19
70.84
42.45
62.85
69.29
69.20
68.86
66.16
63.69
62.45
63.14
52.47
48.72
69.47
64.25
59.31
66.64
48.63
55.60
51.12
52.50
61.46

8.99
11.97
15.22
-16.75
-13.32
-0.81
1.84
1.45
6.85
0.29
17.20
-25.14
-14.84
-9.31
-15.55
-20.86
-1.53
-9.28
-26.53
-14.75
-3.69
5.64
6.63
-5.40
4.12
-2.50
-11.54

47.05
60.75
64.21
65.39
66.38
70.39
75.81
76.30
34.61
35.35
38.71
40.00
45.22
55.32
66.90
67.91
68.58
70.30
36.88
60.06
73.00
41.42
50.71
54.30
55.59
57.86
68.21

4.05
-12.25
-16.79
-12.61
12.38
-20.61
-18.19
7.30
2.61
-6.65
-7.29
-18.00
-7.78
4.32
-2.10
-0.09
9.58
1.30
-18.12
4.06
-9.00
-2.58
-12.29
-0.70
-13.41
11.86
-7.79

Table 5: Data.
45

Table 5 (cont’d)
UT49-03D
UT02-04D
UT17-04D
UT26-04D
UT38-04D
UT41-04D
UT48-04D
UT56-04D
UT57-04D
UT73-04D
UT07-05D
UT17-05D
UT20-05D
UT25-05D
UT34-05D
UT35-05D
UT41-05D
UT50-05D
UT59-05D
UT60-05D
UT61-05D
UT66-05D
UT91-05D
UT15-90D
UT19-90D
UT20-91D
UT26-91D
UT35-93D
UT39-93D
UT19-94D
UT23-94D
UT11-98D
UT20-99D
UT26-99D
UT28-99D
UT17-00D
UT21-00D
UT29-00D
UT07-01D

86
68
91
69
54
68
46
76
81
46
40
58
51
51
44
43
86
68
53
69
55
41
59
54
67
76
31
61
55
54
67
49
55
74
44
35
71
39
50

Rnp1
Rnp1
Lnp1
Lxp2
Lnp1
Rnp2
Lxp2
Lxp2
Lnp1
Lnp2
Rxp1
Rnp2
Rnp1
Lnp2
Lxp2
Lnp1
Rnp1
Lnp2
Lnp2
Lnp2
Lxp2
Rnp2
Lnp2
Lnp1
Lnp1
Lnp1
Rxp2
Rxp1
Lxp1
Lnp1
Lnp1
Rnp2
Lnp1
Lnp1
Lxp2
Rnp1
Lnp1
Lnp1
Rxp1

Amalgam
Amalgam
Amalgam
Composite
Amalgam
Amalgam
Amalgam
Amalgam
Composite
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
Amalgam
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None

16.68
17.41
13.71
14.73
15.48
15.83
17.02
11.81
14.33
14.93
13.36
12.72
13.08
13.86
13.93
12.86
13.77
15.00
14.62
16.54
14.46
16.10
16.48
18.00
14.99
14.30
13.91
17.24
11.22
16.08
12.11
17.54
15.29
13.42
15.58
12.95
14.85
15.84
11.49

11.94 6.13 53.85
14.30 6.46 55.90
10.07 9.47 67.76
12.21 7.10 60.69
14.42 8.19 64.52
12.56 5.69 54.91
14.30 7.48 59.11
8.53 9.22 71.32
9.58 7.52 59.60
12.09 3.64 50.35
12.18 0.00 41.94
10.75 7.87 66.73
6.53 5.88 53.40
12.88 7.02 63.53
10.65 4.46 52.74
12.77 3.32 54.25
8.16 9.20 64.26
13.01 9.82 68.64
13.62 4.92 56.43
13.07 9.61 64.16
11.23 6.88 59.49
13.77 8.25 62.45
13.75 9.96 65.93
13.91 11.34 65.90
11.86 8.64 63.98
12.92 8.63 67.14
12.49 1.35 45.77
11.72 8.62 58.77
9.47 7.41 68.46
12.53 5.06 52.77
10.63 6.75 64.74
11.77 2.72 44.12
10.79 7.38 58.50
12.88 8.35 68.94
13.40 4.08 52.01
11.06 3.58 52.51
13.29 8.12 64.60
14.75 4.92 55.34
10.48 9.91 78.17

46

-32.15
-12.10
-23.24
-8.31
10.52
-13.09
13.11
-4.68
-21.40
4.35
1.94
8.73
2.40
12.53
8.74
11.25
-21.74
0.64
3.43
-4.84
4.49
21.45
6.93
11.90
-3.02
-8.86
14.77
-2.23
13.46
-1.23
-2.26
-4.88
3.50
-5.06
8.01
17.51
-6.40
16.34
28.17

24.83
29.70
56.91
57.91
59.19
64.71
67.26
68.09
77.45
50.63
52.86
59.82
60.70
60.73
62.04
70.05
72.34
76.70
23.91
40.08
43.28
41.67
47.13
29.17
45.54
53.71
54.18
54.31
54.90
56.88
58.35
61.52
63.73
64.97
65.17
66.16
66.87
67.57
67.72

-6.17
-11.30
5.91
-28.09
-8.81
-14.29
-5.74
-17.91
-18.55
-10.37
-15.14
-18.18
-23.30
2.73
21.04
18.05
11.34
-4.30
-16.09
-13.92
-10.72
-5.33
0.13
-3.83
-1.46
-9.29
4.18
15.31
0.90
6.88
14.35
-5.48
8.73
5.97
17.17
-4.84
-7.13
13.57
-8.28

Table 5 (cont’d)
UT10-01D
UT11-01D
UT12-01D
UT27-01D
UT30-01D
UT32-01D
UT12-02D
UT34-02D
UT35-02D
UT08-03D
UT11-03D
UT13-03D
UT14-03D
UT24-03D
UT27-03D
UT28-03D
UT44-03D
UT37-03D
UT12-04D
UT25-04D
UT44-04D
UT45-04D
UT50-04D
UT52-04D
UT63-04D
UT66-04D
UT15-05D
UT16-05D
UT56-05D
UT74-05D
UT06-06D
UT19-06D
UT30-05D
UT10-95D

75
88
50
73
64
33
49
58
55
51
47
48
50
44
46
54
46
43
60
40
39
54
51
50
62
47
53
43
63
38
43
59
69
56

Lnp1
Lnp1
Lnp1
Lnp1
Rnp1
Lnp1
Lnp2
Lxp2
Rnp1
Lxp2
Lnp2
Lnp1
Lnp1
Lnp1
Lnp2
Rnp1
Rnp1
Rnp1
Rxp2
Lxp2
Rnp1
Lnp1
Lnp2
Rnp2
Rnp1
Lnp1
Lxp2
Lnp2
Lnp1
Rnp2
Lnp1
Lnp1
Lnp1
Rnc

None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None
None

14.33
14.47
15.02
15.41
17.83
15.21
15.60
15.58
14.02
12.56
17.46
16.13
12.89
16.19
14.93
15.09
11.98
12.78
13.84
15.94
15.73
15.13
14.36
15.42
18.45
16.16
14.53
17.32
18.12
18.49
17.78
13.91
13.92
16.53

12.88 8.88 67.74
12.95 9.86 70.26
11.72 5.54 55.07
11.90 14.82 79.82
15.81 4.88 52.99
14.44 0.00 42.62
13.89 8.44 64.28
12.70 5.73 55.65
11.51 5.38 56.42
11.54 7.98 68.75
15.97 4.64 53.16
14.15 7.81 61.66
10.91 4.89 56.70
14.72 5.91 57.23
13.54 0.00 41.85
11.99 5.24 54.42
10.43 0.00 41.20
10.16 2.88 49.30
12.83 1.84 47.80
12.94 9.99 66.46
13.64 2.83 48.69
12.13 8.58 63.78
11.36 4.22 52.11
12.59 4.11 51.42
14.36 8.82 59.62
12.39 3.01 47.15
12.87 4.58 54.71
15.78 7.88 61.04
12.66 4.78 49.19
14.52 4.48 49.84
16.95 9.42 64.94
11.71 7.75 64.08
11.35 9.02 67.42
14.58 7.68 60.92

47

-7.26
-17.74
5.07
6.82
-11.01
9.62
15.28
-2.35
1.42
17.75
6.16
13.66
6.70
13.23
-4.15
0.42
-4.80
6.30
-12.20
26.46
9.69
9.78
1.11
1.42
-2.38
0.15
1.71
18.04
-13.81
11.84
21.94
5.08
-1.58
4.92

67.75
68.45
68.70
70.20
71.85
77.38
90.40
18.72
41.69
43.82
59.25
61.75
57.43
41.44
43.19
47.07
54.16
60.59
67.67
24.83
42.32
42.59
46.32
52.61
54.22
54.82
66.81
41.15
46.78
48.23
65.03
71.93
30.49
32.87

0.75
-6.55
-0.30
27.20
-16.15
23.38
17.40
-27.28
-9.31
-3.18
16.25
12.75
2.43
-2.56
-9.81
-10.93
3.16
20.59
11.67
-21.17
3.32
-0.41
11.32
-11.39
0.22
-0.18
4.81
-7.85
-3.22
10.23
4.03
21.93
-0.51
-27.13

BIBLIOGRAPHY

48

BIBIOLOGRAPHY

Baccino E., Ubelaker D.H., Hayek L.C., Zerilli A. 1999. Evaluation of seven methods of
estimating age at death from mature human skeletal remains. Journal of Forensic Sciences
44(5): 931-936.
Bass, WM. 1995. Human Osteology: A laboratory and field manual. 4th ed. Columbia, MO:
Missouri Archaeological Society.
Bang G., Ramm E. 1970. Determination of age in humans from root dentin transparency. Acta
Odontologica Scandinavica 28(1):3-35.
Buckberry JL, Chamberlain AT. 2002. Age estimation from the auricular surface of the ilium: a
revised method. American Journal of Physical Anthropology 119(3):231-239.
Buikstra JE, Ubelaker DH. 1994. Standards for Data Collection from Human Skeletal Remains.
Fayetteville, AR: Arkansas Archaeological Survey Research Series 44.
Burns K.R., Maples W.R. 1976. Estimation of age from individual adult teeth. Journal of
Forensic Sciences 21(2):343-356.
Fanning EA. 1961. A longitudinal study of tooth formation and root resorption. New Zealand
Dental Journal 57:202-217.
Fazekas IG, Kosa F. 1978. Forensic Fetal Osteology. Akademiai Kaido, Budapest.
Fitzgerald CM, Rose JC. 2000. Reading between the lines: dental development and subadult
age assessment using the microstructural growth markers of teeth. In Biological Anthropology
of the Human Skeleton, edited by MA Katzenberg & SR Saunders. Wiley-Liss, Inc., 163-186.
Gustafson G. 1950. Age determinations on teeth. The Journal of the American Dental
Association 41:45-54.
Hoffman, JM. 1979. Age estimations from diaphyseal lengths: two months to twelve years.
Journal of Forensic Sciences 24:461-479.
Iscan MY, Loth SR, Wright RW. 1985a. Age estimation from the rib by phase analysis: white
females. Journal of Forensic Sciences 30(3):853-863.
Iscan MY, Loth SR, Wright RW. 1985b. Age estimation from the rib by phase analysis: white
males. Journal of Forensic Sciences 29(4):1094-1104.
Johanson G. 1971. Age determinations from human teeth. Odontologisk Revy 22S: 1-126.
49

Kvall S.I., During E.M. 1999. A dental study comparing age estimations of the human remains
from the Swedish warship Vasa. International Journal of Osteoarchaeology 9: 170-181.
Kvall S, Solheim T. 1994. A non-destructive dental method for age estimation. Journal of
Forensic Odontology 12: 6-11.
Lamendin H, Baccino E, Humbert JF, Tavernier JC, Nossintchouk RM, Zerilli A. 1992. A simple
technique for age estimation in adult corpses: the two criteria dental method. Journal of
Forensic Sciences 37(5): 1373-1379.
Loe H, Anerud A, Boysen H, Morrison E. 1986. Natural history of periodontal disease in man.
Journal of Clinical Periodontology 13(5):431-440.
Lovejoy CO, Meindl RS, Pryzbeck TR, Mensforth RP. 1985. Chronological metamorphosis of the
auricular surface of the ilium: a new method for the determination of adult skeletal age at
death. American Journal of Physical Anthropology 68:15-28.
Mandojana JM, Martin-de las Heras S, Valenzuela A, Valenzuela M, Luna JD. Differences in
morphological age-related dental changes depending on postmortem interval. Journal of
Forensic Sciences 2001 46(4):889-892.
Maples W.R. 1978. An improved technique using dental histology for estimation of adult age.
Journal of Forensic Sciences 23(4):764-770.
McKern TW, Stewart TD. 1957. Skeletal age changes in young American males. Headquarters,
Quartermaster Research and Developmental Command, Technical Report EP-45. Natick, MA.
Megyesi M.S., Ubelaker D.H., Sauer N.J. 2006. Test of the Lamendin aging method on two
historic skeletal samples. American Journal of Physical Anthropology 131: 363-367.
Meindl RS, Lovejoy CO. 1985. Ectocranial suture closure: a revised method for the
determination of skeletal age at death based on the lateral-anterior sutures. American Journal
of Physical Anthropology 68:57-66.
Meinl A., Huber C.D., Tangl S., Gruber G.M., Teschler-Nicola M., Watzek G. 2008. Comparison
of the validity of three dental methods for the estimation of age at death. Forensic Science
International 178: 96-105.
Miles AEW. 1963. Dentition in the estimation of age. Journal of Dental Research 42(1):255263.
Mjor IA. 1984. The morphology of dentin and dentinogenesis. In Dentin and Dentinogenesis,
Volume I, edited by Anders Linde. Boca Raton, Florida: CRC Press: 1-18.

50

Moorees CFA, Fanning EA, Hunt EE. 1963a. Formation and resorption of three deciduous teeth
in children. American Journal of Physical Anthropology 21:205-213.
Moorees CFA, Fanning EA, Hunt EE. 1963b. Age variation in formation stages for ten
permanent teeth. Journal of Dental Research 42:1490-1502.
Nalbandian J, Gonzales F, Sognnaes RF. 1960. Sclerotic age changes in root dentin of human
teeth as observed by optical, electron, and x-ray microscopy. Journal of Dental Research
39(3):598-607.
Nawrocki SP. 1998. Regression formulae for estimating age at death from cranial suture
closure. In Forensic Osteology: Advances in the Identification of Human Remains, edited by
Reichs. Springfield, IL, Charles C. Thomas: 276-292.
Prince D.A., Ubelaker D.H. 2002. Application of Lamendin’s adult aging technique to a diverse
skeletal sample. Journal of Forensic Sciences 47(1): 107-116.
Raxter M.H., Auerbach B.M., Ruff C.B. 2006. Revision of the Fully technique for estimating
statures. American Journal of Physical Anthropology 130(3): 374-384.
Reppien K., Sejrsen B., Lynnerup N. 2006. Evaluation of post-mortem estimated dental age
versus real age: a retrospective 21-year survey. Forensic Science International 159S: S84-S88.
Scheuer JL, Musgrave JH, Evans SP. 1980. The estimation of late fetal and perinatal age from
limb bone length by linear and logarithmic regression. Annals of Human Biology 7:257-265.
Scheuer JL, Black S, Christie A. 2000. Developmental Juvenile Osteology. San Diego: Academic
Press.
Schmitt A, Saliba-Serre B, Tremblay M, Martrille L. 2010. An evaluation of statistical methods
for the determination of age of death using dental root translucency and periodontosis. Journal
of Forensic Sciences 55(3):590-596.
Schouer, Massler. 1941. The development of the human dentition. Journal of the American
Dental Association 28:1153.
Sengupta A., Whittaker D.K., Shellis R.P. 1999. Difficulties in estimating age using root dentin
translucency in human teeth of varying antiquity. Archives of Oral Biology 44: 889-899.
Solheim T., Sundnes P.K. 1980. Dental age estimation of Norwegian adults – a comparison of
different methods. Forensic Science International 16: 7-17.
Solheim, T. 1989. Dental root translucency as an indicator of age. Scandinavian Journal of
Dental Research 97: 189-197.
51

Solheim T. 1993. A new method for dental age estimation in adults. Forensic Science
International 59: 137-147.
Soomer H., Ranta H., Lincoln M.J., Penttilä A., Leibur E. 2003. Reliability and validity of eight
dental age estimation methods for adults. Journal of Forensic Sciences 48(1): 149-152.
Sperber N. 1993. A simple technique for age estimation in adult corpses: the two criteria
dental method (letter to the editor). Journal of Forensic Sciences, 38:763-764.
Stewart TD. 1979. Essentials of Forensic Anthropology. Springfield, IL: Charles C Thomas.
Suchey JM, Katz D. 1998. Applications of pubic age determination in a forensic setting. In
Forensic Osteology: Advances in the Identification of Human Remains, edited by Reichs.
Springfield, Charles C. Thomas: 204-236.
Takuma S, Eda S. Structure and development of the peritubular matrix in dentin. Journal of
Dental Research Supplement to No 3 45:683-692.
Thomas G.J., Whittaker D.K., Emberly G. 1994. A comparative study of translucent apical
dentine in vital and non-vital human teeth. Archives of Oral Biology 39(1): 29-34.
Trotter M. 1970. Estimation of Stature from Intact Long Bones. In Personal Identification in
Mass Disasters, edited by T. D. Stewart. Washington D.C., Smithsonian Institution Press: 71-83.
Ubelaker DH. 1999. Human Skeletal Remains. 3rd ed. Washington, D.C.: Taraxacum.
Vasiliadis L, Darling AI, Levers BGH. 1983. The histology of sclerotic human root dentine.
Archives of Oral Biology 28(8):693-700.
Vasiliadis L, Darling AI, Levers BGH. 1983. The amount and distribution of sclerotic human root
dentine. Archives of Oral Biology 28(7):645-649.
Whittaker DK, Bakri MM. 1996. Racial variations in the extent of tooth root translucency in
aging individuals. Archives of Oral Biology 41(1):15-19.

52