Forensic palynology and the search for geolocation: Factors for analysis and the Baby Doe case

Abstract

First used over 50 years ago, forensic palynology is an important tool for law enforcement agencies. In most countries that use forensic palynology, microscopic pollen grains and spores are traditionally used in criminal investigations to link suspects to crime scenes or items. While still underutilized in many parts of the world, forensic palynology is increasingly being used to determine the region of origin, or geolocation, for persons and items of interest. Drawing upon the experience of the authors using trace pollen and spores to geolocate forensic samples, the types, methods, and variables of this type of analysis are discussed and demonstrated using the Baby Doe case from Massachusetts, USA as a case study. This is not an exhaustive list and every forensic sample is unique so the methods and experience presented here are intended to be a guide for future forensic and anti-terrorism investigations as forensic palynology becomes more commonplace in law enforcement agencies around the world.

Full text

Forensic
palynology
and
the
search
for
geolocation:
Factors
for
analysis
and
the
Baby
Doe
case
Andrew
R.
Laurence
a,
*,
Vaughn
M.
Bryant
b
a
U.S.
Customs
and
Border
Protection,
Laboratories
and
Scientific
Services
Directorate,
Chicago,
IL
60607,
USA
b
Palynology
Laboratory,
Department
of
Anthropology,
Texas
A&M
University,
College
Station,
TX
77843,
USA
A
R
T
I
C
L
E
I
N
F
O
Article
history:
Received
26
April
2019
Received
in
revised
form
24
June
2019
Accepted
27
June
2019
Available
online
27
July
2019
Keywords:
Forensic
science
Pollen
Spores
Geolocation
Sampling
Interpretation
A
B
S
T
R
A
C
T
First
used
over
50
years
ago,
forensic
palynology
is
an
important
tool
for
law
enforcement
agencies.
In
most
countries
that
use
forensic
palynology,
microscopic
pollen
grains
and
spores
are
traditionally
used
in
criminal
investigations
to
link
suspects
to
crime
scenes
or
items.
While
still
underutilized
in
many
parts
of
the
world,
forensic
palynology
is
increasingly
being
used
to
determine
the
region
of
origin,
or
geolocation,
for
persons
and
items
of
interest.
Drawing
upon
the
experience
of
the
authors
using
trace
pollen
and
spores
to
geolocate
forensic
samples,
the
types,
methods,
and
variables
of
this
type
of
analysis
are
discussed
and
demonstrated
using
the
Baby
Doe
case
from
Massachusetts,
USA
as
a
case
study.
This
is
not
an
exhaustive
list
and
every
forensic
sample
is
unique
so
the
methods
and
experience
presented
here
are
intended
to
be
a
guide
for
future
forensic
and
anti-terrorism
investigations
as
forensic
palynology
becomes
more
commonplace
in
law
enforcement
agencies
around
the
world.
©
2019
Published
by
Elsevier
B.V.
For
the
last
50
years,
pollen
and
fungal
spore
studies
have
gained
an
increasingly
important
role
as
trace
evidence
used
in
forensic
applications
[1–4].
Forensic
palynology
has
traditionally
focused
on
the
use
of
pollen
and
spore
evidence
to
associate
suspects
or
items
with
a
specific
crime
scene
[5].
Other
applications
have
centered
on
using
trapped
pollen
and
spores
on
or
in
an
item
as
a
way
to
determine
the
probable
geolocation
of
that
item’s
origin
[6].
In
this
regard,
during
the
past
few
decades,
forensic
palynology
has
been
most
often
used
in
situations
pertaining
to
homicides,
acts
of
terrorism,
war
crimes
and
genocide,
to
detect
the
forgery
of
documents,
help
solve
crimes
related
to
theft,
kidnapping,
rape,
arson,
and
identifying
the
manufacturing
and
distribution
sources
of
counterfeit
medications
sold
worldwide
[7].
Other
applications
have
focused,
to
a
limited
degree,
on
the
source
and
distribution
of
illegal
drugs
such
as
marijuana,
heroin,
cocaine,
and
methamphetamine
as
well
as
investigated
cases
of
assault,
hit
and
run,
poaching,
and
identity
theft
[2,5,8–10,11 ].
Currently,
the
application
of
pollen
and
spore
evidence
in
forensics
is
highly
underutilized
in
some
regions
of
the
developed
world,
especially
in
the
United
States
[6,12].
The
use
of
forensic
palynology
in
criminal
cases
is
well
documented
but
there
are
few
published
examples
or
discussions
that
detail
cases
involving
the
resolution
of
geolocation
using
pollen
and
spores
[13].
A
primary
reason
for
this
lack
of
published
information
is
the
sensitive
nature
of
many
cases
or
the
varied
situations
where
those
techniques
have
been
used.
For
this
contribution,
we
will
detail
some
of
the
factors
involved
in
analyzing
forensic
pollen
samples
for
geolocation.
For
this,
we
will
draw
upon
our
experiences
and
use
the
Baby
Doe
case
as
an
example.
1.
Sample
collection
Ideally,
forensic
pollen
samples
should
be
collected
in
the
field
by
a
forensic
palynologist
or
someone
trained
under
his/her
direction.
Many
different
methods
for
collecting
forensic
pollen
samples
have
been
published
elsewhere
(e.g.
Refs.
[14,11 ])
so
they
will
not
be
discussed
here.
When
possible,
sample
collection
strategies
should
be
developed
and
agreed
upon
before
inves-
tigators
arrive
at
a
crime
scene
in
order
to
avoid
contamination.
Every
case
is
different,
so
sampling
strategies
are
usually
developed
on
a
case-by-case
basis.
2.
Forensic
samples:
pollen
sampling
and
processing
We
have
analyzed
a
wide
range
of
sample
types
including
many
different
types
of
drugs
(such
as
marijuana,
cocaine,
heroin,
crystal
methamphetamine,
etc.),
explosives,
clothing,
and
human
hair.
Because
every
sample
is
different
and
even
similar
sample
types
can
have
unique
challenges,
forensic
palynologists
need
to
adapt
their
in-lab
sampling
and
processing
methods
based
on
the
*
Corresponding
author.
E-mail
address:
andrew.r.laurence@cbp.dhs.gov
(A.R.
Laurence).
http://dx.doi.org/10.1016/j.forsciint.2019.109903
0379-0738/©
2019
Published
by
Elsevier
B.V.
Forensic
Science
International
302
(2019)
109903
Contents
lists
available
at
ScienceDirect
Forensic
Science
International
journal
homepage:
www.elsevier.com/locat
e/f
orsciint

individual
characteristics
of
the
sample.
Due
to
the
need
to
adapt
to
the
characteristics
of
each
sample,
we
usually
modify
the
methods
of
Brown
[15]
Erdtman
[16]
and
Faegri
et
al.
[17]
to
process
sample
types.
Whenever
possible,
the
same
methods
should
be
used
for
similar
sample
types
to
maintain
consistency
and
to
adhere
to
standard
protocols.
All
methods
used
during
sampling
and
processing
are
documented
in
the
final
report.
Furthermore,
if
multiple
types
of
analyses
are
to
be
performed
in
conjunction
with
pollen
analysis,
(such
as
DNA
or
fiber
analysis),
subsamples
need
to
be
taken
so
each
analysis
can
be
undertaken
with
minimal
risk
of
contamination
or
destruction
of
other
evidence.
3.
Analysis
Most
microscopic
examinations
of
forensic
pollen
samples
are
performed
using
a
light
microscope
rather
than
scanning
electron
microscopy
[17,18].
Normally
for
samples,
a
total
of
at
least
200–
300
palynomorphs
(pollen
and
spores)
should
be
counted
[19].
Higher
palynomorph
counts
can
be
used
in
exceptional
circum-
stances,
but
higher
counts
increase
both
the
time,
expense
of
the
analysis,
and
often
do
not
change
the
inferred
geolocation.
They
can,
however,
increase
the
number
of
rare
taxa
seen
in
a
sample,
and
in
special
cases
must
be
conducted
to
conclusively
prove
a
specific
geolocation.
Because
time
is
often
a
factor
in
our
casework,
we
typically
count
200–300
palynomorphs
and
then
scan
additional
slides
for
new
pollen
and
spore
types,
which
might
help
to
identify
a
site’s
location.
Pollen
and
spore
identifications
are
often
a
problem,
especially
when
forensic
samples
are
from
areas
of
the
world
where
there
is
little
published
information
on
the
palynomorphs
characteristic
of
those
regions.
A
number
of
published
pollen
atlases,
modern
pollen
and
spore
reference
collections
can
aid
in
identification.
In
addition,
there
are
web
sites
that
provide
images
of
some
pollen
and
spore
taxa,
and
published
articles
in
professional
journals
with
informa-
tion
about
pollen
and
spore
types
in
some
regions
of
the
world
are
often
helpful.
Nevertheless,
one
of
the
greatest
concerns
in
forensic
palynology
is
the
accuracyof
identification.
Inaccurate
palynomorph
identifications
can
result
in
misleading
interpretations;
therefore,
it
is
imperative
that
the
individuals
conducting
the
collection,
processing,
and
analysis
of
forensic
samples
be
thoroughly
trained
and
highly
competent
to
conduct
these
types
of
studies.
Even
with
the
aid
of
digitized
reference
collections,
the
most
accurate
way
to
identify
palynomorphs
is
through
direct
comparisons
with
slides
of
modern
pollen
and
spores
prepared
as
reference
material.
4.
Interpretations
Analysts
examining
forensic
pollen
samples
need
to
understand
aspects
related
to
pollen
production,
dispersal,
long
distance
transport,
pollen
degradation,
recycling
of
dispersed
pollen
and
spores,
and
the
potential
for
different
items
retaining
different
amounts
of
pollen
and
spore
spectra.
For
example,
we
know
that
different
types
of
fabrics
potentially
retain
different
amounts
and
types
of
pollen
and
spores
[14,20,21].
Every
forensic
pollen
sample
is
unique.
The
retained
pollen
spectrum
of
a
sample
results
from
a
number
of
variables
some
of
which
may
not
precisely
reflect
the
pollen
spectrum
at
the
original
location.
Therefore,
the
following
discussion
is
by
no
means
exhaustive,
but
it
does
point
out
some
of
the
more
common
variables
that
forensic
analysts
should
consider
when
analyzing
samples
for
geolocation
purposes.
5.
Types
of
geolocation
There
are
essentially
two
main
types
of
forensic
samples
used
for
geolocation
purposes.
The
first
type
includes
samples
where
an
agency
or
an
individual
believes
the
sample
originated
from
a
specific
location.
The
goal
of
the
pollen
analyst
examining
those
samples
is
to
search
for
confirming
pollen
types,
or
the
lack
of
them,
that
might
confirm
or
reject
the
suspected
location.
The
second
type
of
sample
is
one
that
has
no
known
point
of
origin,
but
it
is
hoped
that
a
location
can
be
established.
The
first
type
of
samples
pertains
to
very
specific
situations
involving
victims,
suspects,
crime
scenes,
and
items
suspected
of
being
associated
with
one
of
the
previous
situations
[22–25].
For
this
first
type
of
investigation,
the
successful
use
of
pollen
and
spores
will
depend
on
the
palynologist’s
knowledge,
experience,
and
familiarity
with
the
crime
scene
and/or
other
places
of
interest
related
directly
to
answering
specific
questions
about
a
victim,
suspect,
or
aspect
related
to
those
situations.
The
palynologist
must
understand
the
plant
ecology
and
plant
communities
in
and
around
a
crime
scene,
and
be
aware
of
other
ecological,
soil,
and
climatic
features,
including
topographic
features
of
the
landscape
in
and
around
the
general
region.
For
the
second
type
of
samples,
there
are
various
steps
that
one
can
follow.
Generally,
the
first
step
is
to
scan
one
or
more
microscope
slides
made
from
the
processed
sample
in
search
of
“marker
pollen
and
spore
types.”
Marker
types
are
pollen
grains
and
spores
that
come
from
plants
that
either
have
specific
ecological
needs
or
are
the
most
prominent
in
one
geographical
region.
Marker
types
would
include
pollen
taxa
such
as
Abies
(Fir),
which
would
suggest
a
location
in
a
cold
or
northern
area
of
the
Northern
Hemisphere.
On
the
other
hand,
finding
Phoenix
(date
palm)
pollen
would
suggest
a
tropical
and
perhaps
semi-arid
region.
When
a
series
of
marker
pollen
types
are
identified
in
a
sample
and
they
all
come
from
the
same
general
type
of
environment,
then
that
becomes
the
first
major
clue
needed
to
narrow
the
origin
of
a
sample.
Although
the
key
marker
types
may
suggest
a
specific
“type”
of
environment,
it
often
takes
a
group
of
marker
types
to
narrow
the
search
to
a
specific
continent
or
even
a
specific
region
within
a
continent.
Next,
the
amounts
of
each
pollen
type
in
a
sample
might
further
narrow
the
search
area.
For
example,
high
percentages
of
Artemisia
(sagebrush),
grass,
Sarcobatus
(greasewood),
sedges,
and
low
or
trace
amounts
of
Pinus
(pine),
Picea
(spruce),
Pseudotsuga
(Douglas
fir)
and
fir
would
suggest
a
probable
area
in
the
steppe
regions
of
North
America
[26].
6.
Variables
while
searching
for
geolocation
Examining
forensic
pollen
samples
(e.g.
clothing,
hair,
vehicle
air
filters,
sediment,
electronics,
etc.)
for
geolocation
purposes
differs
from
many
of
the
other
types
of
pollen
analyses
(e.g.
based
on
swamp
or
lake
deposits)
because
it
must
account
for
anthropogenic
variables
and
samples
and
are
often
not
a
complete
pollen
record,
as
might
be
found
typically
in
a
core
from
a
bog
or
lake.
One
of
the
most
common
encountered
problems
when
searching
for
geolocation
is
the
presence
of
pollen
from
ornamental
plants.
In
some
situations,
that
information
becomes
invaluable
because
the
pollen
from
a
specific
type
of
ornamental
might
come
from
a
plant
at
a
specific
crime
scene
[3,27–29].
At
other
times,
ornamental
pollen
can
add
confusion
when
attempt-
ing
to
determine
the
geolocation
of
a
sample
from
its
pollen
spectrum.
For
example,
palm
trees,
pine
trees,
Eucalyptus
(gum)
trees,
Callistemon
(bottlebrush),
and
many
other
plants
are
now
grown
as
ornamentals
in
areas
outside
their
normal
ecological
range.
Because
the
pollen
from
many
of
those
plant
types
are
often
considered
marker
pollen
in
samples,
their
presence
in
some
samples
could
provide
misleading
geolocation
interpretations
The
other
major
problem
in
trying
to
establish
the
geolocation
of
a
sample
relates
to
the
practice
of
plant
introduction
and
landscape
modification.
At
one
time,
it
would
have
been
2
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903

reasonable
to
assume
that
Casuarina
(she
oak)
pollen
indicated
either
Australia
or
a
very
limited
area
in
Southeast
Asia,
but
today,
that
plant
has
become
a
pandemic
ornamental
in
the
warm
and
arid
regions
of
the
world.
Likewise,
there
were
no
pine
trees
or
pine
pollen
in
prehistoric
or
even
historic
Australian
and
New
Zealand
sediments
until
Europeans
introduced
that
plant
[30,31].
Humans
have
been
modifying
the
landscape
for
over
200,000
years
[32],
which
has
altered
and
transformed
vegetation
communities
to
the
point
that
sometimes
the
pollen
profiles
of
the
area
no
longer
reflect
the
expected
pollen
rain
of
the
natural
landscape
[33].
This
becomes
a
special
problem
when
searching
for
the
geolocation
of
pollen
samples
from
urban
areas.
Often,
pollen
samples
taken
in
an
urban
area
contain
a
pollen
spectrum
influenced
by
pollen
from
trees
and
other
vegetation
lining
the
streets,
including
introduced
ornamentals
in
parks
and
in
the
yards
of
local
residents,
and
sometimes
even
remnant
traces
of
the
original
vegetation.
Green
Lake
Park
in
Seattle,
Washington
(USA),
for
example,
has
had
a
continuous
introduction
of
many
non-native
and
exotic
trees
and
other
vegetation
since
it
was
established
in
1903
[34].
Thus,
the
current
pollen
profiles
for
some
areas
of
Seattle
do
not
reflect
the
Pacific
Northwest
but
instead
could
be
mistaken
for
locations
in
California
or
even
the
Southeastern
United
States.
Apart
from
the
pollen
and
spores
observed
in
forensic
samples,
there
are
often
other
clues
that
can
help
determine
a
specific
geolocation.
Since
most
forensic
samples
are
modern
samples,
the
condition
of
the
pollen
grains
themselves
can
offer
clues
about
recycling,
possible
microbial
degradation,
and
differential
preser-
vation
of
fragile
types.
The
presence
of
certain
types
of
debris
in
a
sample,
such
as
soot
from
diesel-powered
vehicles
or
coal
dust
from
industrial
activities,
might
help
to
include
or
exclude
potential
regions
of
origin.
Finally,
the
type
of
item
from
which
a
sample
was
collected
may
provide
information
as
to
the
types
of
pollen
grains
that
should
be
expected.
For
example,
a
vehicle
seized
in
a
rural
coastal
area
should
contain
pollen
evidence
of
that
coastal
area.
However,
it
might
contain
pollen
and
spore
evidence
suggesting
the
vehicle
had
been
driven
in
high-elevation
forests
or
in
nearby
urban
regions
rather
than
being
used
exclusively
along
local
coastal
areas.
Similarly,
forensic
pollen
sampling
of
rare
paintings
that
were
painted
in
areas
of
Europe,
or
antique
furniture
purported
to
come
from
France
may
contain
marker
pollen
types
confirming
those
points
of
origin
or
might
instead
provide
questions
about
their
authenticity
[14].
A
final
point
that
must
be
considered
is
the
potential
biases
that
may
be
inherent
in
the
type
of
forensic
sample
that
is
collected.
Samples
exposed
for
long
periods
will
often
contain
a
much
richer
suite
of
pollen
taxa
than
a
sample
that
was
only
briefly
exposed.
Furthermore,
samples
from
certain
seasons
of
the
year
will
often
reflect
a
bias
skewed
toward
local
plants
that
were
pollinating
at
that
time.
Pollen
grains
can
also
be
transported
long
distances
by
wind,
animals,
or
human
activities,
so
the
presence
of
a
few
non-local
pollen
grains
does
not
necessarily
suggest
that
a
sample
travelled
to
another
region.
Finally,
the
surface
areas
from
which
samples
are
collected
might
retain
certain
types
of
pollen
more
effectively
than
other
types
[21,29].
7.
Case
study:
Baby
Doe
The
Baby
Doe
case
from
Massachusetts
provides
an
example
of
how
forensic
palynology
is
used
effectively
for
geolocation.
June
25,
2015,
a
jogger
and
her
dog
discovered
a
sealed,
black
trash
bag
that
had
floated
to
the
shore
of
Deer
Island
in
Boston
Harbor
(Fig.
1).
The
trash
bag
contained
the
remains
of
a
murdered
female
child
that
was
partly
decomposed.
Wave
action
had
distorted
the
child’s
facial
features
and
fingerprints
making
an
initial,
positive
identification
unlikely.
The
infant,
estimated
to
have
been
2–4
years
of
age,
was
wearing
a
pair
of
black
polka-dot
pants
and
was
wrapped
in
a
Zebra
pattern
blanket,
but
no
other
items
were
present.
The
clothing
did
not
provide
immediate
clues
because
the
Fig.
1.
Location
where
body
was
found.
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903
3

Polka-dot
pants
were
sold
at
Wal-Mart
stores
throughout
the
Northeastern
U.S.
and
the
Zebra-Stripe
blanket
was
a
popular
item
sold
in
K-Mart
stores
in
the
same
region.
It
was
a
composite
facial
reconstruction
for
Baby
Doe
by
a
forensic
specialist
that
was
of
the
most
value
and
the
picture
was
put
on
posters,
large
billboards
and
on
the
Boston
and
Massachusetts
State
Police
web
site
(Fig.
2).
The
U.S.
Coast
Guard
was
asked
to
determine
potential
places
along
the
eastern
coast
of
the
United
States
where
the
plastic
bag
might
have
originated.
Based
on
ocean
currents,
the
estimated
time
since
the
trash
bag
was
thrown
into
the
ocean,
and
wind
and
wave
action,
the
Coast
Guard
estimated
that
the
potential
origin
could
have
been
hundreds
of
miles
north
or
even
south
of
the
Boston
Harbor
area
and
Deer
Island
where
the
body
was
discovered.
The
Boston
Police
Facebook
Page
describing
the
Baby
Doe
case
and
their
request
for
information
about
the
baby
received
over
50
million
hits,
and
hundreds
of
tips
poured
in
from
all
over
the
eastern
region
near
Boston
as
well
as
other
countries.
Unfortu-
nately,
none
of
the
suggested
tips
proved
successful
in
determining
either
where
the
baby
had
been
killed,
or
where
it
had
been
discarded
into
the
ocean.
John
Walsh,
co-founder
of
the
National
Center
for
Missing
and
Exploited
Children
that
was
working
with
the
Boston
and
Massachusetts
State
Police,
suggested
asking
help
from
federal
agencies
in
hopes
of
solving
the
murder
of
Baby
Doe.
One
of
the
suggested
forensic
tests
focused
on
the
potential
information
that
might
be
gained
from
a
pollen
study
of
the
garments
and
blanket
associated
with
the
dead
baby.
Those
agencies
sent
the
girl’s
clothing,
blanket,
and
a
sample
of
her
hair
to
the
U.S.
Customs
and
Border
Protection
Laboratories
and
Scientific
Services
Directorate
lab
in
Houston,
Texas,
hoping
that
maybe
pollen
evidence
might
provide
new
clues.
When
the
items
arrived,
one
of
the
authors
(Laurence),
who
was
working
at
the
Houston
lab
at
the
time,
processed
and
analyzed
the
samples.
8.
Methods
A
forensic
vacuum
with
a
cellulose
filter
cartridge
was
used
to
remove
pollen
and
other
particulates
from
the
blankets
and
pants
(Fig.
3).
Separate
filter
cartridges
were
used
to
process
each
sample.
Once
vacuumed,
the
filter
element
from
each
cylinder
was
carefully
removed
and
placed
into
a
separate,
sterile
15
ml
centrifuge
tube
filled
with
95%
ethanol
(EtOH).
Ethanol
is
our
preferred
sample
liquid
because
its
lower
specific
gravity
ensures
that
all
palynomorphs
will
sink
to
the
bottom
of
a
centrifuge
tube
during
centrifugation
[18].
The
samples
were
centrifuged
and
the
supernatant
decanted.
Each
sample
was
then
filled
with
glacial
acetic
acid,
centrifuged,
and
decanted.
Once
these
steps
were
completed,
the
residue
was
treated
with
acetolysis
to
remove
unwanted
cellulose
and
lipids
in
the
pollen
grains,
which
could
obscure
the
identity
of
the
recovered
pollen.
The
standard
acetolysis
procedure
is
to
use
a
mixture
of
9:1
acetic
anhydride
to
sulfuric
acid.
The
acetolysis
mixture
with
the
samples
was
then
heated
in
an
aluminum-heating
block
for
10
min
at
a
temperature
of
85

C.
After
acetolysis,
the
samples
were
rinsed
and
washed
in
EtOH
several
times.
Once
that
was
completed,
the
residue
was
washed
again
in
distilled
water
and
then
heated
in
48%
hydrofluoric
acid
(HF)
at
85

C
for
30
min
to
remove
silicates.
After
HF,
each
sample
was
rinsed
several
times
with
distilled
water,
and
then
rinsed
in
36.5%
hydrochloric
acid
(HCl)
to
remove
any
fluorosilicates.
The
samples
were
again
rinsed
in
distilled
water
and
a
final
rinse
of
EtOH.
Some
large
organic
debris
remained
so
the
samples
were
carefully
screened
using
continuous
sprays
of
EtOH
to
remove
debris
larger
than
150
mm.
The
final
step
was
to
stain
the
residue
with
organic
stain,
Safranin-O,
rinse
a
final
time
in
EtOH,
and
then
store
it
in
a
2
ml
plastic
centrifuge
tube
before
adding
three
drops
of
glycerin.
Each
sample
tube
contained
a
top
with
an
O-ring
to
ensure
secure
closure
and
to
prevent
possible
airborne
palyno-
morph
contamination.
The
hair
sample
was
placed
in
a
beaker
containing
a
50:50
mixture
of
EtOH
and
water,
then
agitated
until
it
appeared
to
be
clean
(Fig.
4).
Next,
the
recovered
liquid
was
transferred
to
centrifuge
tubes
and
concentrated
by
centrifugation.
Once
concentrated,
the
sample
was
processed
in
the
same
manner
as
the
vacuum
filters.
9.
Results
There
was
enough
pollen
in
all
samples,
except
for
the
hair,
to
complete
a
standard
pollen
count
of
200–300
pollen
grains
(Table
1).
Additional
slides
from
each
sample
were
scanned
for
additional
pollen
taxa
which
might
provide
clues
for
geolocation.
The
pollen
spectra
recovered
from
all
clothing
samples
were
very
similar
and
suggested
a
suburban
Northeastern
U.S.
environment.
The
clothing
samples
were
dominated
by
background
pollen
types
Fig.
2.
Composite
image
of
Baby
Doe
produced
by
NCMEC.
Fig.
3.
Forensic
vacuum
used
to
remove
pollen
from
Baby
Doe’s
pants.
4
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903

including
oak
(Quercus)
and
pine
pollen
grains
along
with
minor
amounts
of
composite
(Asteraceae)
and
grass
pollen
(Poaceae).
No
pollen
from
known
cultivated
or
agricultural
plants
was
found
during
the
initial
counts
or
during
the
additional
scanning
of
slides.
Other
pollen
taxa
recovered
from
the
samples
were
types
native
to
the
Northeastern
U.S.,
such
as
maple
(Acer),
walnut
(Juglans),
hickory
(Carya),
birch
(Betula),
basswood
(Tilia),
ash
(Fraxinus),
spruce,
hemlock
(Tsuga),
elm
(Ulmus),
alder
(Alnus),
and
beech
(Fagus).
Although
pollen
from
taxa
common
in
boreal
forests
was
found,
such
as
spruce
and
hemlock,
their
frequency
was
minimal
(Table
1).
9.1.
Analysis
Our
initial
impression
of
the
pollen
analyses
suggested
a
typical
type
of
pollen
rain
(all
the
pollen
deposited
at
a
location)
found
in
North
American
temperate,
deciduous-pine
forests
[35]
that
are
common
along
the
eastern
seaboard
of
the
United
States
from
Georgia
to
Maine
(Fig.
5).
The
lack
of
pollen
from
any
agricultural
or
commonly
cultivated
plants
suggested
an
urban
rather
than
a
rural
location.
The
samples
contained
high
quantities
of
soot,
which
is
often
common
in
forensic
samples
from
urban
regions.
The
lack
of
many
pollen
grains
from
taxa
such
as
such
as
weedy
plant
members
of
the
composites
(Asteraceae)
and
amaranths
(Amar-
anthaceae),
suggested
an
urban
area
that
did
not
contain
disturbed
areas
or
vacant
lots
full
of
weeds
and
debris.
The
combination
of
pollen
data
from
our
samples
suggested
that
the
child
had
lived
in
a
developed
urban
or
suburban
area.
Usually,
homes
and
yards
in
more
affluent
areas
of
a
city,
have
few
weeds
growing
in
vacant
lots
and
unkempt
vegetation
along
roadsides,
but
still
have
enough
vehicle
traffic
to
contribute
high
amounts
of
soot
into
the
atmosphere
that
can
become
part
of
the
pollen
rain
or
embedded
in
the
fabric
of
people’s
clothing.
Had
the
child
been
from
a
forested
area,
we
would
have
expected
higher
ratios
of
arboreal
pollen
that
were
found.
Although
most
of
the
recovered
pollen
suggested
a
northeast-
ern
U.S.
urban
or
suburban
area
south
of
the
conifer
forest
zone,
the
Fig.
4.
Baby
Doe’s
hair
soaking
in
50:50
distilled
water
and
EtOH
solution.
Table
1
Palynomorphs
recovered
from
Baby
Doe
samples.
Pollen
taxa
%
%
%
%
Sample
name
Baby
blanket
Zebra-print
blanket
Pants
Hair
Acer
(maple)
3
1.2%
1
0.4%
3
1.2%
0
0.0%
Alnus
(alder)
0
0.0%
3
1.2%
0
0.0%
0
0.0%
Artemisia
(sagebrush)
0
0.0%
1
0.4%
1
0.4%
0
0.0%
ASTERACEAE
(ragweed-type)
1
0.4%
1
0.4%
2
0.8%
1
2.7%
ASTERACEAE
(dandelion-type)
1
0.4%
3
1.2%
2
0.8%
0
0.0%
Betula
(birch)
17
6.7%
14
5.6%
20
7.8%
2
5.4%
Cannabis/Humulus
0
0.0%
2
0.8%
0
0.0%
2
5.4%
CARYOPHYLLACEAE
(carnation
family)
2
0.8%
0
0.0%
1
0.4%
0
0.0%
Carpinus
(hornbeam)
0
0.0%
2
0.8%
0
0.0%
0
0.0%
Carya
(hickory)
3
1.2%
1
0.4%
2
0.8%
1
2.7%
Cedrus
(cedar)
0
0.0%
1
0.4%
2
0.8%
0
0.0%
CHENO-AMS
(goosefoot)
1
0.4%
1
0.4%
2
0.8%
1
2.7%
Diporate
1
0.4%
1
0.4%
3
1.2%
0
0.0%
ERICACEAE
(heath
family)
1
0.4%
1
0.4%
0
0.0%
0
0.0%
EUPHORBIACEAE
(spurge
family)
0
0.0%
1
0.4%
1
0.4%
0
0.0%
FABACEAE
(bean
family)
0
0.0%
0
0.0%
1
0.4%
0
0.0%
Fagus
(beech)
0
0.0%
2
0.8%
3
1.2%
1
2.7%
Fern
spores
2
0.8%
1
0.4%
1
0.4%
0
0.0%
Fraxinus
(ash)
3
1.2%
8
3.2%
5
2.0%
2
5.4%
Juglans
(walnut)
1
0.4%
1
0.4%
1
0.4%
0
0.0%
Ligustrum
(privet)
2
0.8%
0
0.0%
0
0.0%
0
0.0%
Liquidambar
(sweetgum)
0
0.0%
0
0.0%
1
0.4%
0
0.0%
Lycopodium
(clubmoss)
2
0.8%
0
0.0%
1
0.4%
0
0.0%
Picea
(spruce)
1
0.4%
1
0.4%
1
0.4%
1
2.7%
Pinus
(pine)
38
14.9%
44
17.6%
47
18.4%
7
18.9%
Pinus
body/bladder
6
2.4%
10
4.0%
12
4.7%
2
5.4%
Plantago
(plantain)
1
0.4%
13
5.2%
1
0.4%
1
2.7%
POACEAE
(grass
family)
43
16.9%
8
3.2%
2
0.8%
1
2.7%
Populus
(cottonwood)
0
0.0%
3
1.2%
1
0.4%
3
8.1%
Quercus
(oak)
90
35.3%
94
37.6%
112
43.8%
8
21.6%
Rhus
(sumac)
11
4.3%
8
3.2%
10
3.9%
2
5.4%
Tilia
(basswood)
1
0.4%
0
0.0%
0
0.0%
0
0.0%
Trifolium
(clover)
0
0.0%
2
0.8%
0
0.0%
0
0.0%
Tsuga
(hemlock)
1
0.4%
1
0.4%
1
0.4%
0
0.0%
Typha
tetrad-type
(cattail)
1
0.4%
3
1.2%
0
0.0%
0
0.0%
Typha
monad-type
(cattail)
4
1.6%
2
0.8%
0
0.0%
0
0.0%
Ulmus
(elm)
15
5.9%
13
5.2%
17
6.6%
2
5.4%
Unknown
pollen
(other)
3
1.2%
3
1.2%
0
0.0%
0
0.0%
Totals
255
100 %
250
100 %
256
100 %
37
100 %
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903
5

recovery
of
two
different
species
of
cedar
(Cedrus)
pollen
was
a
significant
clue
for
determining
the
geolocation
(Fig.
6).
Current
taxonomic
information
list
only
four
species
of
true
cedar,
all
of
which
are
native
to
temperate
mountain
zones
of
North
Africa
and
the
Mediterranean
region
such
as
the
mountainous
regions
in
Lebanon
and
the
island
of
Cyprus
[36].
Historical
records
suggest
that
the
first
cedar
trees
were
imported
into
the
Southeastern
and
Western
United
States
during
the
180 0 s
for
their
economic
utility
[36].
Later,
additional
cedar
trees
were
planted
in
more
regions
of
the
U.S.
where
they
were
often
used
in
parks
and
in
urban
areas
as
ornamental
plants
[37].
A
major
problem
was
that
these
initial
species
of
cedar
trees
were
not
acclimatised
to
extremely
cold
winters
and
thus
could
not
survive
in
many
of
the
colder
areas
of
North
America.
In
the
1900s,
seeds
from
a
new
variety
of
Lebanon
cedar
(Cedrus
libani
A.
Rich.)
from
the
Taurus
Mountains
in
Turkey
were
imported
and
planted
at
Harvard
University’s
Arnold
Arboretum
in
1902.
The
success
of
that
initial
planting
produced
seeds
that
became
the
“breeding
stock”
for
cedar
trees
planted
in
parks
and
arboretums
throughout
many
regions
of
the
United
States,
including
the
Morris
Arboretum
of
The
University
of
Pennsylvania
[36].
Unfortunately,
our
available
modern
pollen
reference
slides
included
only
two
of
the
four
recognized
species
of
cedar;
therefore,
we
took
the
con-
servative
approach
and
listed
the
cedar
pollen
only
at
the
generic
level
even
though
it
was
apparent
from
the
pollen
morphology
there
were
two
distinct
species
represented.
Because
the
cedar
pollen
grains
were
in
pristine
condition
and
showed
no
damage
that
often
results
during
recycling,
we
believed
the
cedar
pollen
came
from
ornamentals
close
to
where
the
victim
had
lived,
perhaps
near
an
arboretum
or
near
a
park.
Because
the
recovered
pollen
profiles
contained
soot,
suggesting
an
urban
or
suburban
environment,
it
was
not
likely
that
the
cedar
pollen
was
the
result
of
long
distance
transport
from
a
distant
source.
Most
parks
and
personal
residences
in
the
Northeastern
region
of
the
U.S.
only
contain
the
variety
of
the
Lebanon
cedar
that
can
grow
without
extensive
habitat
modification,
but
arboretums
often
contain
several
species
of
cedar.
For
the
Northeastern
region,
there
are
only
three
arboretums
that
grow
multiple
species
of
cedars.
These
three
include
the
New
York
Botanical
Gardens
in
New
York
City,
the
Morris
Arboretum
of
The
University
of
Pennsylvania
in
Philadelphia,
and
the
Arnold
Arboretum
of
Harvard
University
in
Boston.
Of
these,
the
Arnold
and
Morris
Arboretums
have
three
different
species
of
cedar,
which
could
account
for
the
two
different
types
of
cedar
pollen
found
in
these
samples.
Based
on
our
pollen
records
it
was
less
likely
the
victim
came
from
the
areas
of
either
New
York
City
or
Philadelphia.
Instead,
it
appeared
more
likely
that
the
child
may
have
come
from
an
area
near
the
Arnold
Fig.
5.
Location
of
deciduous-pine
forest
(Appalachian
oak
section)
from
Dyer
[38].
Fig.
6.
Types
of
Cedrus
pollen
grains
recovered
from
Baby
Doe
samples.
Note
the
different
surface
features
between
A
and
B.
6
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903

Arboretum
in
Boston
since
the
body
was
found
just
outside
of
Boston.
Armed
with
the
new
information,
provided
by
the
pollen
samples,
the
Boston
Police
Department
began
a
thorough
search
for
a
missing
baby
from
the
areas
closest
to
the
Arnold
Arboretum.
Their
initial
earlier
questioning
of
residents
in
that
region
had
not
provided
any
usable
clues,
but
many
recent
immigrants
live
in
those
areas,
and
some
may
be
illegal
and
thus
did
not
want
to
talk
to
the
police.
Nevertheless,
with
a
probable
geolocation
of
the
victim,
additional
and
thorough
questioning
did
provide
needed
clues
to
arrest
the
mother
of
the
baby
and
her
boyfriend.
A
neighbour
in
one
of
the
apartment
houses
told
police
she
has
not
seen
one
of
the
residents
and
her
baby
in
weeks.
That
information
encouraged
the
police
to
interview
and
then
arrest
the
mother
of
Baby
Doe,
Rachelle
Bond,
who
pleaded
guilty
of
accessory
after
the
fact
and
with
a
plea
bargain
was
given
two
years
of
probation
plus
credit
for
time
served
in
jail.
Her
live-in
boyfriend
(Michael
McCarthy)
was
high
on
drugs
when
he
abused
Baby
Doe,
causing
her
death.
He
has
been
convicted
of
second-degree
murder
and
is
serving
a
minimum
sentence
of
20
years.
Finally,
it
should
be
noted
that
while
we
observed
a
few
marijuana-type
(Cannabis)
pollen
grains
in
two
of
the
samples,
it
is
unknown
if
they
were
related
to
McCarthy’s
drug
use.
Male
marijuana
plants
produce
copious
amounts
of
pollen
and,
apart
from
illicit
and
licit
growing
operations,
the
plants
themselves
grow
feral
throughout
North
America,
including
in
cities.
For
this
reason,
marijuana
pollen
is
a
very
common
component
of
modern
pollen
profiles
throughout
North
America,
and
such
low
amounts
are
consistent
with
modern
air
and
sediment
samples
from
the
northeastern
United
States.
10.
Conclusions
Baby
Doe
was
only
two
years
old
when
she
died
and
lived
about
4
km
southeast
of
the
Arnold
Arboretum
in
Boston.
In
this
case,
the
geolocation
information
provided
by
forensic
pollen
analyses
of
the
victim’s
clothing,
associated
blanket,
and
hair
helped
inves-
tigators
narrow
their
search
to
a
few
specific
areas
in
Boston,
and
helped
them
identify
and
arrest
suspected
individuals
responsible
for
the
victim’s
death.
The
value
of
using
forensic
palynology
as
an
important
geolocation
tool
for
generating
investigative
leads,
or
eliminating
some
suspected
locations
was
dramatically
shown
and
demon-
strated
by
the
solving
of
the
Baby
Doe
case.
However,
as
with
other
types
of
forensic
trace
evidence,
pollen
studies
cannot
always
be
applied
in
every
situation
and
to
all
types
of
investigations
[11 ].
Nevertheless,
when
investigating
various
types
of
crimes,
one
should
always
consider
the
potential
value
of
pollen
studies
and
perhaps
discuss
the
possible
use
of
palynology
with
individuals
familiar
with
that
forensic
technique.
Under
ideal
circumstances,
forensic
palynology
can
focus
and
guide
the
efforts
of
law
enforcement
personnel
to
a
particular
area,
or
eliminate
suspected
areas
thereby
saving
time
and
resources.
Adopting
forensic
palynology
has
been
slow
in
some
countries,
such
as
the
U.S.,
mainly
because
it
remains
unknown
and
because
there
are
so
few
individuals
trained
to
assist
in
the
collection
and
analysis
of
forensic
pollen
samples
(6,39]).
We
hope
that
as
this
technique
becomes
more
publicized,
the
potential
use
will
increase
and
it
may
become
more
commonplace
as
an
important
forensic
tool.
Acknowledgements
The
information
reported
here
was
conducted
as
part
of
the
authors’
duties
working
with
the
Department
of
Homeland
Security,
U.S.
Customs
and
Border
Protection.
Earlier
drafts
of
this
article
benefitted
substantially
from
comments
by
anonymous
reviewers
that
helped
us
clarify
our
discussion
and
better
explain
our
results.
We
would
especially
like
to
thank
Crawford
White
for
making
the
map
in
Fig.
1,
the
National
Centre
for
Missing
and
Exploited
Children
(NCMEC)
for
allowing
us
to
use
their
composite
image
of
Baby
Doe,
and
NCMEC
and
the
Massachusetts
State
Police
for
allowing
us
to
present
results
from
the
Baby
Doe
case.
We
also
extend
our
thanks
to
Patricia
Hawes,
Steven
Goldfarb,
Robert
Harvey,
and
Melanie
Glass
for
their
tireless
promotion
of
the
benefits
of
forensic
palynology.
References
[1]
D.L.
Hawksworth,
P.E.J.
Wiltshire,
J.A.
Webb,
Rarely
reported
fungal
spores
and
structures:
an
overlooked
source
of
probative
trace
evidence
in
criminal
investigations,
Forensic
Sci.
Int.
264
(2016)
41–46.
[2]
D.C.
Mildenhall,
Forensic
palynology
in
New
Zealand,
Rev.
Paleobot.
Palynol.
64
(1990)
227–234.
[3]
P.E.J.
Wiltshire,
Mycology
in
palaeoecology
and
forensic
science,
Fungal
Biol.
120
(2016)
1272–1290.
[4]
J.
Woodall,
The
tales
pollen
tells:
the
rise
of
forensic
palynology,
a
once
obscure
science,
Scientist
21
(2009)
68.
[5]
D.C.
Mildenhall,
P.E.
Wiltshire,
V.M.
Bryant,
Forensic
palynology:
why
do
it
and
how
it
works,
Forensic
Sci.
Int.
163
(2006)
163 –172 .
[6]
A.R.
Laurence,
V.M.
Bryant,
Forensic
palynology,
in:
J.S.
Albanese
(Ed.),
The
Encyclopedia
of
Criminology
and
Criminal
Justice,
Wiley-Blackwell
Publishing
Company,
Chichester,
2013,
pp.
1741 –1754.
[7]
D.C.
Mildenhall,
The
use
of
spores
and
pollen,
Civ.
Crim.
Investig.
4
(2008)
35–52.
[8]
G.
Erdtman,
Handbook
of
Palynology,
Hafner
Publishing,
New
York
(NY),
1969.
[9]
M.
Frei-Sulzer,
Nine
years
of
palinological
studies
on
the
shroud,
Shroud
Spectr.
Int.
3
(1982)
3–7.
[10]
L.A.
Milne,
V.M.
Bryant,
D.C.
Mildenhall,
Forensic
palynology,
in:
H.M.
Coyle
(Ed.),
Forensic
Botany:
Principles
and
Applications
to
Criminal
Casework,
CRC
Press,
Boca
Raton
(FL),
2005,
pp.
217–252.
[11]
P.E.J.
Wiltshire,
Protocols
for
forensic
palynology,
Palynology
39
(2015)
4–24.
[12]
V.M.
Bryant,
J.G.
Jones,
Forensic
palynology:
current
status
of
rarely
used
techniques
in
the
United
States
of
America,
Forensic
Sci.
Int.
163
(2006)
183 –197.
[13]
V.M.
Bryant,
M.K.
Bryant,
Forensic
palynology,
in:
K.S.
Moran,
C.L.
Gold
(Eds.),
Forensic
Archaeology:
Multidisciplinary
Perspectives,
Springer
Nature
Switzerland
AG,
Cham,
Switzerland,
2019.
[14]
V.M.
Bryant,
D.C.
Mildenhall,
Forensic
palynology:
a
new
way
to
catch
crooks,
in:
V.M.
Bryant,
J.W.
Wrenn
(Eds.),
New
Developments
in
Palynomorph
Samples,
Extraction,
and
Analysis,
33,
American
Association
of
Stratigraphic
Palynologists
Contributions
Series,
Dallas,
1998,
pp.
145–15 5.
[15]
C.A.
Brown,
Palynology
Techniques,
American
Association
of
Stratigraphic
Palynologists
Foundation,
Dallas
(TX),
2008.
[16]
G.
Erdtman,
The
acetolysis
method:
a
revised
description,
Svensk
Bot.
Tidskr.
54
(1960)
561.
[17]
K.
Faegri,
P.E.
Kaland,
K.
Krzywinski
(Eds.),
Textbook
of
Pollen
Analysis,
4th
edn.,
Blackburn
Press,
Caldwell,
1989.
[18]
J.G.
Jones,
V.M.
Bryant,
A
comparison
of
pollen
counts:
light
versus
scanning
electron
microscopy,
Grana
46
(2007)
20–33.
[19]
F.A.
Barkley,
The
statistical
theory
of
pollen
analysis,
Ecology
15
(1934)
283–289.
[20]
J.B.
Riding,
B.G.
Rawlins,
K.H.
Coley,
Changes
in
soil
pollen
assemblages
on
footwear
worn
at
different
sites,
Palynology
31
(2007)
135–151.
[21]
M.
Zavada,
S.
Mcgraw,
M.
Miller,
The
role
of
clothing
fabrics
as
passive
pollen
collectors
in
the
northeastern
United
States,
Grana
46
(2007)
285–291.
[22]
D.C.
Mildenhall,
Hypericum
pollen
determines
the
presence
of
burglars
at
the
scene
of
a
crime:
an
example
of
forensic
palynology,
Forensic
Sci.
Int.
163
(2006)
231–235.
[23]
P.E.J.
Wiltshire,
D.L.
Hawkswork,
K.J.
Edwards,
Light
microscopy
can
reveal
the
consumption
of
a
mixture
of
psychotropic
plant
and
fungal
material
in
suspicious
deaths,
J.
Forensic
Legal
Med.
34
(2015)
73–80.
[24]
P.E.J.
Wiltshire,
D.L.
Hawksorth,
K.J.
Edwards,
A
rapid
and
efficient
method
for
evaluation
of
suspect
testimony:
palynological
scanning,
J.
Forensic
Sci.
60
(2015)
1441 –1450.
[25]
P.E.J.
Wiltshire,
D.L.
Hawksworth,
J.A.
Webb,
K.J.
Edwards,
Two
sources
and
two
kinds
of
trace
evidence:
enhancing
the
links
between
clothing,
footwear
and
crime
scene,
Forensic
Sci.
Int.
254
(2015)
231–242.
[26]
R.N.
Mack,
V.M.B.
Bryant,
Modern
pollen
spectra
from
the
Columbia
Basin,
Washington,
Northwest
Sci.
48
(1974)
183 –194.
[27]
V.M.
Bryant,
The
bottlebrush
&
hibiscus
get
their
man,
Odyssey
1
(2004)
38–41.
[28]
D.C.
Mildenhall,
A
grain
of
evidence,
N.
Z.
Sci.
Mon.
9
(1998)
9.
[29]
L.A.
Milne,
A
Grain
of
Truth:
How
Pollen
Brought
a
Murderer
to
Justice,
Reed
New
Holland,
Sydney,
2005.
[30]
P.B.
Lavery,
D.J.
Mead,
Pinus
radiata:
a
narrow
endemic
from
North
America
takes
on
the
world,
in:
D.M.
Richardson
(Ed.),
Ecology
and
Biogeography
of
Pinus,
Cambridge
University
Press,
Cambridge,
1998,
pp.
432–449.
[31]
D.C.
Le
Maitre,
Pines
in
cultivation:
a
global
view,
in:
D.M.
Richardson
(Ed.),
Ecology
and
Biogeography
of
Pinus,
Cambridge
University
Press,
Cambridge,
1998,
pp.
407–431.
[32]
D.
Western,
Human-modified
ecosystems
and
future
evolution,
PNAS
98
(10)
(2001)
5458–5465.
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903
7

[33]
K.J.
Edwards,
R.M.
Fyfe,
S.T.
Jackson,
The
first
100
years
of
pollen
analysis,
Nat.
Plants
3
(2017)
170 01,
doi:http://dx.doi.org/10.1038/nplants.2017.1.
[34]
C.
Thrush,
Native
Seattle:
Histories
from
the
Crossing-Over
Place,
University
of
Washington
Press,
Seattle
(WA),
2007.
[35]
J.W.
Williams,
B.
Shuman,
P.J.
Bartlein,
J.
Whitmore,
K.
Gajewski,
M.
Sawada,
T.
Minckley,
S.
Shafer,
A.E.
Viau,
I.I.I.T.
Webb,
P.
Anderson,
L.
Brubaker,
C.
Whitlock,
O.K.
Davis,
An
Atlas
of
Pollen-Vegetation-Climate
Relationships
for
the
United
States
and
Canada,
American
Association
of
Stratigraphic
Palynologists
Foundation,
Dallas,
2006.
[36]
A.S.
Aiello,
M.S.
Dosmann,
The
quest
for
the
hardy
Cedar-of-Lebanon,
Arnoldia
65
(2007)
26–35.
[37]
E.H.
Wilson,
Aristocrats
of
the
Garden,
The
Stratford
Company,
Boston
(MA),
1926.
[38]
J.
Dyer,
Revisiting
the
deciduous
forests
of
eastern
North
America,
BioScience
56
(2006)
341–352.
[39]
V.M.
Bryant,
J.G.
Jones,
D.C.
Mildenhall,
Forensic
palynology
in
the
United
States
of
America,
Palynology
14
(1990)
193–208.
8
A.R.
Laurence,
V.M.
Bryant
/
Forensic
Science
International
302
(2019)
109903