Cremation Weights in East Tennessee

Abstract

In spite of increasing number of cremations in the U.S., little is known about weights of cremated remains. This research was undertaken in order to add to the limited literature on cremains weights and to explore variation. Weights of cremated remains were obtained from the East Tennessee Crematorium. The sample consists of 151 males and 155 females. Age, sex, and race were obtained for each individual. Males are about 1000 g heavier than females. Both sexes lose weight with age, but females lose weight at about twice the rate of males. East Tennessee cremation weights were compared with those from Florida reported by Warren and Maples, and those from Southern California reported by Sonek. East Tennessee results were also compared with an earlier study on ash weight of anatomical human skeletons carried out by Trotter and Hixon. East Tennessee cremations weigh about 500 g more than the samples from Florida and California, and about the same as the earlier anatomical samples. We hypothesize that variation reflects variation in body weight and activity. This variation must be taken into account when cremation weights are at issue.

Full text

JForensicSci,Sept. 2004, Vol. 49, No. 5
Paper ID JFS2004002
Available online at: www.astm.org
William M. Bass,1Ph.D. and Richard L. Jantz,1Ph.D.
Cremation Weights in East Tennessee∗
ABSTRACT: In spite of increasing number of cremations in the U.S., little is known about weights of cremated remains. This research was
undertaken in order to add to the limited literature on cremains weights and to explore variation. Weights of cremated remains were obtained from
the East Tennessee Crematorium. The sample consists of 151 males and 155 females. Age, sex, and race were obtained for each individual. Males
are about 1000 g heavier than females. Both sexes lose weight with age, but females lose weight at about twice the rate of males. East Tennessee
cremation weights were compared with those from Florida reported by Warren and Maples, and those from Southern California reported by Sonek.
East Tennessee results were also compared with an earlier study on ash weight of anatomical human skeletons carried out by Trotter and Hixon.
East Tennessee cremations weigh about 500 g more than the samples from Florida and California, and about the same as the earlier anatomical
samples. We hypothesize that variation reflects variation in body weight and activity. This variation must be taken into account when cremation
weights are at issue.
KEYWORDS: forensic science, cremation, forensic anthropology, bone mass, sex dimorphism
Literature reporting weights of adult cremated remains is surpris-
ing sparse, and much of it consists only of rough approximations.
For example, Iserson (1) reports a range of 3–9 lb, Quigley (2) a
range of 6–12 lb, Carlson (3) a range of 3–7 lb, and Maples and
Browning (4) a range of 2.2 to 8.8 lb. The only study in the primary
literature we were able to locate reporting data from a specified
sample in the U.S. is Warren and Maples (5). A few other data
sources were located, and these will be discussed later.
The present study was stimulated by the involvement of one
of us (WMB) in litigation surrounding cremated remains returned
to families by the Tri-State crematorium in Noble, GA. Questions
were raised during a deposition August 19, 2002 and a court hearing
August 22, 2002, concerning cremation weights and whether the
weights reported by Warren and Maples (5) could be extrapolated
to other situations. This paper will present data and analysis of
a sample obtained from the East Tennessee region and compare it
with the available data in order to address the question of variability
in cremains weight.
Materials and Methods
The weights of all the cremations were obtained by WMB be-
tween December 6, 2002 and July 23, 2003. All weights were taken
on an OHAUS digital scale, Model CS5000 which was purchased
new for this project. The scale measures in 1 g increments from 0
to 5000 g. The accuracy of the scales was checked monthly using
certified weights from the Denver Instrument Company of 200 g
(Serial No. 98-J36238-24) and 1000 g (Serial No. 98-J20326-5) at
200, 1000, and 1200 g.
All weights were recorded at the East Tennessee Cremation Com-
pany (ETCC) in Maryville, TN. The ETCC has two natural gas fired
Industrial Engineering and Equipment Company (IEE) cremation
1Department of Anthropology, University of Tennessee, Knoxville, TN.
∗Presented at the annual Mountain, Swamp and Beach regional forensic
meeting, Chattanooga, TN, Aug. 30, 2003, and at the 56th annual meeting of
the American Academy of Forensic Sciences, Dallas, TX. Feb. 19, 2004.
Received 10 Jan. 2004; and in revised form 20 Mar. 2004; accepted 28 Mar.
2004; published 3 Aug. 2004.
furnaces in a small neat building at the edge of an industrial park
near the Knoxville airport. ETCC is a clean and well managed
facility that does approximately 1000 cremations per year.
All cremations are performed individually. Bodies arrive in a
variety of clothing attire, from nude to completely dressed. Indi-
viduals who have died in a hospital often have on only a hospital
gown, and occasionally the life support systems of plastic tubes
and needles are still attached. Cremation is performed by placing
the deceased in a casket or other container. The bodies arrive in a
variety of containers from body bags through fancy and expensive
wooden coffins. It is difficult to place body bags into the cremation
chamber, the funeral industry’s name for the furnace, so that almost
all bodies are at least in a cardboard carton especially made for
the cremation process. The cardboard container usually contains a
pressboard or plywood bottom because the cardboard tends to bend
or buckle under the weight of a body. If the body is leaking fluids,
the cardboard bottom tends to break.
In most cases, the furnace temperature stays somewhere between
1600 and 1800 degrees Fahrenheit. The length of the cremation
process varies depending on the size of the body and the bone
structure. A dehydrated elderly person burns slower than a body
with more fat. The usual cremation takes 2 to 3 h, comparable to
the time required presented by Bohnert et al. (6). Following the
cremation process, there is a cool down period of a few hours.
After the remains are cooled, they are removed from the cre-
mation chamber. The burned bones inside clearly outline a human
skeleton. The long bones are fractured but in most cases are intact
and the rib cage still sketches the framework of the chest. The skull
is fractured but may still be somewhat intact. It breaks into small
pieces when the long handle metal broom is inserted to rake the
cremains out of the cremation chamber.
The bone fragments and ashes are removed and placed on a
worktable beneath a vented exhaust fan. Hinges, screws, staples
and small nails that were part of the cardboard carton or wooden
box are often mixed with the cremains. Often still burning pieces of
plywood need to be removed along with charred pieces of burned
plywood that will darken the final cremains if they are not removed.
A heavy magnet is drawn through the cremains to remove all mag-
netic items. Many individuals today have orthopedic devices such
Copyright C
2004 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. 1
Copyright by ASTM Int'l (all rights reserved); Mon Jul 30 13:59:20 EDT 2018
Downloaded/printed by
TENNESSEE UNIVERSITY (TENNESSEE UNIVERSITY) pursuant to License Agreement. No further reproductions authorized.

2JOURNAL OF FORENSIC SCIENCES
as hip or knee replacements and various metal items and screws or
bolts for the multitude of fracture repair devices. All of these metal
items that could be detected, either visually or magnetically, are
removed before the cremains are pulverized.
The final stage in the cremation process is the pulverization of the
remaining pieces of bone. This was originally done by a hammer or
board but the modern standard is a commercial pulverizer. ETCC
has an IEE processor that resembles a soup kettle grafted on top
of a garbage disposal. Essentially, it is an industrial size kitchen
blender. The cremains, minus all the metal that could be detected,
are put in the kettle and after the heavy top is secured, a switch is
flipped and the fragments are reduced to a grainy powder in 60 s.
The processed cremains are then poured into a plastic bag po-
sitioned inside a rectangular plastic box measuring 90 ×155 ×
210 mL and a metal tag containing the name, date and sometimes
the social security number of the deceased is placed into the bag
and the bag is sealed tightly with a plastic cable tie. The box is
then snapped shut. It is this box of cremains that was weighed. The
plastic box, the plastic bag, cable tie and the name tag weighed
310 g. When the total weight is obtained, the 310 g are subtracted
so that the weight reported here is the true weight of the cremation.
During the weighing process, all cremains were weighed twice to
eliminate any errors in recording.
Information obtained from each of the cremated individuals con-
sists of age, sex, and race. It was not possible to obtain pre-cremation
weight, but unusually heavy individuals were noted. Since few non-
whites were present in the sample, they were omitted and the analy-
sis limited to whites. Also omitted were amputees. The final sample
TAB LE 1—Descriptive statistics for age and weight of East
Tennessee sample.
Males Females
Age Weight Age Weight
n151 151 155 155
Mean 62.834 3379.768 70.680 2350.170
Median 65.0 3375.0 77.0 2260.00
Std 17.173 634.975 16.78 536.43
Minimum 18 1865.0 19 1050.0
Maximum 99 5379.0 98 4000.0
TAB L E 2 —Regression equations of cremation weight on age.
Males Females
Statistic Estimate s.e. Estimate s.e.
Regression coefficient −8.19∗2.95 −16.5507∗2.21
Constant 3894.17∗∗ 192.37 3520.036∗∗ 160.55
Correlation 0.2214∗... .5179∗...
∗P<0.01.
∗∗ P<0.001.
TAB L E 3 —Classification of sex by age group using cremation weight.
Means by Age Group Sex Classification by Age Group
Males Females Males Females
Age Group NMean NMean No. % No. %
17–39 15 3528.43 7 2821.43 11/15 73.3 6/7 85.7
40–69 80 3497.46 51 2667.47 60/80 75.0 39/51 76.5
70+57 3167.42 97 2149.33 47/57 82.5 86/97 88.7
consists of 151 males and 155 females. Standard summary statis-
tics were computed, and the relationship between age and cremains
weight was investigated using linear regression.
Results
Table 1 presents the summary statistics for the present sample.
The male sample is slightly younger than female, and heavier by
over 1000 g. Figure 1 presents the relationship between age and
weight, with fitted linear regression lines, and Table 2 gives the
regression statistics and tests of significance. Both sexes exhibit
statistically significant declines in weight with age. Females lose
weight at almost twice the rate of males, 16.55 g vs. 8.19 g per
year. The slopes of the two regression lines also differ significantly
(F=5.14, P<0.025, df =1,302), allowing rejection of the null
hypothesis that the two sexes lose weight at the same rate.
Table 1 shows that the sexes differ by approximately 1000 g and
Fig. 1 shows the sex distributions are to some extent distinct, es-
pecially in the older ages. Sex differences of this magnitude raise
the question of whether sexes could be discriminated from their
cremains weight. Because the sexes lose weight as they age, age
must be controlled in some fashion. The different rate of change
of the two sexes complicates this, because covariance adjustment
to a common age is not possible. This question may be examined
in a preliminary way by dividing the sample into broad age cat-
egories. Table 3 shows means and correct classification rates for
three broad age categories. Sample sizes for the youngest category
are small, so these results inspire little confidence. In the second
category about 3
/4are correctly classified, and in the oldest it rises to
over 85%, suggesting that for those over 70 cremains weight alone
can estimate sex as reliably as many morphologically complete
FIG. 1—Plot of cremation weights (grams) on age for East Tennessee
males and females.
Copyright by ASTM Int'l (all rights reserved); Mon Jul 30 13:59:20 EDT 2018
Downloaded/printed by
TENNESSEE UNIVERSITY (TENNESSEE UNIVERSITY) pursuant to License Agreement. No further reproductions authorized.

BASS AND JANTZ rCREMATION WEIGHTS 3
TABLE 4—Comparison of East Tennessee and other samples.
Males Females
Group NMean S.D. NMean S.D.
E. Tenn. 151 3379.77 634.98 155 2350.17 536.43
Florida 50 2898.70 499.20 40 1829.38 406.53
California 76 2801.38 589.47 63 1874.87 528.82
Anatomical 30 3410 ... 30 2297 ...
skeletal elements. These results also demonstrate that age must be
considered in evaluating cremains weights. For example, cremains
weighing 3000 g could be consistent with a 30-year-old female or
a 75-year-old male.
Discussion
The results reported above raise several issues regarding weights
of cremated remains. Most important is the issue of variation among
samples. The only published study on a U.S. sample of which we
are aware is Warren and Maples (5). In addition to the formally
published results, cremains weights from the San Diego, California
area were distributed by the late Alexander Sonek in handout form
in connection with his 1992 paper (7). The pooled sex mean of
Sonek’s data was included in Murad’s (8) summary of forensic
implications of cremation, but we have calculated the sex spe-
cific means and standard deviations. Criteria for inclusion of the
Florida and California samples were the same as ours, limited to
individuals 18 and over and excluding amputees and non-whites.
Finally, Trotter and Hixon (9) presented ash weights of skeletons
from anatomical specimens collected during the first half of the
20th century. These are not cremations in the usual sense, because
soft tissue was removed and the skeletons reduced to a dry fat free
state prior to burning. Ash weights should be comparable to cre-
mation weights, because in both cases one is left with the inorganic
component.
Summary statistics for these samples are presented in Table 4.
Both Warren and Maples (5) and Sonek (7) are significantly lighter
than our sample, (t=4.88 and 5.73, for Warren and Maples, and
t=6.63 and 5.95 for Sonek, males and females respectively, p<
0.001 in all cases). Trotter and Hixon presented ash weights for
both Blacks and Whites, but only their White data are presented
in Table 4. Unlike the Florida and California cremation weights,
Trotter and Hixon’s ash weights are very similar to the cremation
weights presented in the present study. They do not present standard
deviations, so a test of significance is not possible, but a difference
of 30 g and 50 g for males and females respectively, is not likely to
be significant.
It is not immediately clear what accounts for the variation among
samples presented in Table 4. One possibility concerns age dif-
ferences among the samples. The mean age of both the Florida
and California samples is slightly older than the our Tennessee
sample: 66.34 and 74.06 years for Florida males and females respec-
tively, and 64.11 and 75.65 for California males and females respec-
tively. The maximum age difference is about 5 years (Tennessee vs.
California females). Using the slope of the regression line in Table 2,
assuming it applies to other groups, only accounts for about 80 g
of the difference. We also controlled age by breaking the samples
into 10 year age cohorts. The Florida and California age cohorts are
lighter than their Tennessee counterparts. The anatomical sample
age means are 63.5 and 63.8 for males and females respectively.
Males are about the same as the present sample, and females are
about 6 years younger. If the anatomical sample females were ad-
justed to our mean age, they would be about 100 g lighter, (using the
regression slope in Table 2) still substantially heavier than Florida
or California. Age differences therefore do not account for differ-
ences in cremains weight.
Another possibility, which we put forth in the spirit of a sugges-
tion for future research, concerns regional variation in body mass.
A relationship between bone mass and body weight has been estab-
lished, to the point where femur cortical thickness can be used to
estimate weight (10). The Center for Disease control statistics give
an obesity rate for 1999 of 23% for Tennessee, 18.2% for Florida
and 19.5% for California. The greater obesity rate in Tennessee
would predict greater bone mass, which would be reflected in cre-
mains weight.
The ash weight study presents a different situation. Since these
skeletons were collected during the first half of the 20th century,
giving them birth years from the mid 19th century to the early
20th century, they predate the obesity epidemic in America. How-
ever, it was also a time when Americans were considerably more
active than they currently are. Activity may also maintain bone
mass.
The loss of cremains weight with age seen in our data reflects the
well-known loss of bone associated with aging, and the more rapid
decline of cremains weight in females reflects the well-documented
acceleration of bone loss with menopause. Trotter and Hixon report
loss of ash weight at the rate of 15.6 g/year, not much different from
our figure of 16.6 g/year in females. They detected no difference
between races or sexes, possibly because sample sizes were smaller
than ours.
What emerges from this analysis is that cremains weight varies,
perhaps regionally, but insufficient evidence exists to specify any
particular pattern of variation. The causes of variation remain to be
identified, although factors that affect bone mass, such as activity
and body weight, must play a role. It should be clear that variation in
cremains weight should be treated as a problem in human variation
to be examined and understood, and that forensic anthropologists
and others concerned with cremains weight should be aware of this
variation.
Acknowledgments
We thank Helen Taylor, Manager of the East Tennessee Crema-
tion Company and a licensed Funeral Director for permission to
conduct this study.
References
1. Iserson K. Death to dust. Tucson, AZ: Galen Press, 1994.
2. Quigley C. The corpse: a history. Jefferson, NC: McFarland and Co.,
1996.
3. Carlson L. Caring for the dead: your final act of love. Hinesburg, VT:
Upper Access, Inc., 1998.
4. Maples WR, Browning M. Dead men do tell tales: The strange and
fascinating cases of a forensic anthropologist. New York: Doubleday,
1994.
5. Warren MW, Maples WR. The anthropometry of contemporary commer-
cial cremation. J Forensic Sci 1997;42:417–23. [PubMed]
6. Bohnert M, Rost T, Pollak S. The degree of destruction of human bodies
in relation to the duration of the fire. Forensic Sci Int 1998;95:11–21. [PubMed]
7. Sonek A. The weight(s) of cremated remains. Paper presented at the 44th
annual meeting of the American Academy of Forensic Sciences.1992.
Feb. 21. New Orleans, LA.
Copyright by ASTM Int'l (all rights reserved); Mon Jul 30 13:59:20 EDT 2018
Downloaded/printed by
TENNESSEE UNIVERSITY (TENNESSEE UNIVERSITY) pursuant to License Agreement. No further reproductions authorized.

4JOURNAL OF FORENSIC SCIENCES
8. Murad TA. The growing popularity of cremation versus inhumation:
some forensic implications. In: Reichs KJ, editor. Forensic osteology:
Advances in the identification of human remains. 2nd ed. Springfield,
IL: Charles C. Thomas, 1998:86–105.
9. Trotter M, Hixon BB. Sequential changes in weight, density, and percent-
age of ash weight of human skeletons from an early fetal period through
old age. Anat Rec 1973;179:1–18.
10. Ruff CB, Scott WW, Liu AYC. Articular and diaphyseal remodeling of
the proximal femur with changes in body-mass in adults. Am J Phys
Anthropol 1991;86:397–413. [PubMed]
Additional information and reprint requests:
William M. Bass, Ph.D.
Department of Anthropology
250 South Stadium Hall
University of Tennessee
Knoxville, TN 37996-0720
Copyright by ASTM Int'l (all rights reserved); Mon Jul 30 13:59:20 EDT 2018
Downloaded/printed by
TENNESSEE UNIVERSITY (TENNESSEE UNIVERSITY) pursuant to License Agreement. No further reproductions authorized.