Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Data
Requirement:
:
EPA
DP
Barcode
D288775
EPA
MRID
458677­
10
EPA
Guideline
70­
1(
Special
Study)

Test
material:
Purity:
not
reported
Common
name:
Atrazine
Chemical
name:
IUPAC
CAS
name
6­
chloro­
N­
ethyl­
N'­(
1­
methylethyl)­
1,3,5­
triazine­
2,4­
diamine
CAS
No.
1912­
24­
9
synonyms
EPA
PC
Code:
80803
Primary
Reviewer:
Thomas
M.
Steeger,
Ph.
D.,
Senior
Biologist
Date:
April
12,
2003
Environmental
Fate
and
Effects
Division,
ERB
4,
U.
S.
Environmental
Protection
Agency
Secondary
Reviewer(
s):
Joseph
E.
Tietge,
M.
S.,
Research
Aquatic
Biologist
Date:
Mid­
Continent
Ecology
Division,
National
Health
and
Environmental
Effects
Research
Laboratory
(
Duluth),
U.
S.
Environmental
Protection
Agency
Stephanie
Irene,
Ph.
D.,
Senior
Advisor
Date:
Environmental
Fate
and
Effects
Division,
ERB
3,
U.
S.
Environmental
Protection
Agency
Mary
J.
Frankenberry,
Statistician
Date:
Environmental
Fate
and
Effects
Division,
ERB
3,
U.
S.
Environmental
Protection
Agency
EPA
PC
Code
080803
Date
Evaluation
Completed:
05/
31/
2003
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
2
of
31
CITATION:
Smith,
E.
E.,
L.
DuPreez,
and
K.
Solomon.
2003.
Gonadal
and
laryngeal
responses
to
field
exposure
of
Xenopus
laevis
to
atrazine
in
areas
of
corn
production
in
South
Africa.
The
Institute
of
Environmental
and
Human
Health,
Texas
Tech
University,
Lubbock,
TX
and
School
of
Environmental
Sciences
and
Development,
Potchefstroom
University
for
CHE,
Potchefstroom,
South
Africa.
Sponsor:
Syngenta
Crop
Protection,
Inc.,
Laboratory
Study
ID:
ECORISK
Number
SA­
01C.

EXECUTIVE
SUMMARY:

This
report
is
part
of
a
multiphase
study
examining
the
effects
of
atrazine
on
African
clawed
frogs
(
Xenopus
laevis)
in
their
native
habitat
(
South
Africa).
This
phase
of
the
study
summarizes
the
body
weights,
lengths
and
condition
index
(
weight/
length)
of
both
male
and
female
collected
from
reference
site
(
no
corn
grown
and
no
atrazine/
triazine
use)
and
experimental
site
(
corn
grown
and
atrazine/
triazines
used)
ponds.
The
study
also
examined
several
aspects
of
gonadal
growth
(
testis
weight,
gonadosomatic
index)
and
development
(
incidence
of
deformities,
cellular
morphology)
in
frog
testes
collected
from
the
two
sites
to
determine
whether
atrazine
exposure
had
affected
these
measurement
endpoints.

While
males
were
smaller
than
females
at
both
study
sites,
there
was
no
difference
in
weight
or
length
of
males
or
females
collected
at
reference
or
experimental
sites.
Condition
indices
for
males
and
females
did
not
differ
between
reference
and
experimental
collection
sites.
However,
the
mean
testes
weight
from
reference
ponds
(
0.037
±
0.021
mg)
was
statistically
different
from
mean
weight
of
testes
(
0.046
±
0.018
mg)
from
frogs
collected
at
experimental
sites.
GSI
was
significantly
different
for
males
collected
from
reference
(
0.13
±
0.04)
compared
to
males
collected
from
experimental
sites
(
0.17
±
0.04).

There
were
no
statistical
differences
in
the
cellular
morphology
of
testes
collected
at
reference
sites
from
those
collected
at
experimental
site.
The
incidence
of
external
abnormalities
(
missing
limbs)
was
1%
across
all
animals
collected
(
reference
and
experimental)
and
was
attributed
to
the
predatory
effects
of
turtles.
The
incidence
of
testicular
oocytes
was
3%
in
males
collected
from
reference
ponds
and
2%
in
experimental
ponds.
The
study
claims
that
laryngeal
somatic
index
for
males
and
females
from
reference
and
experimental
sites
were
not
significantly
different
(
p
>
0.05);
however,
no
raw
data
were
provided
to
verify
these
observations.

Although
not
discussed
at
length
in
this
phase
of
the
study,
the
interpretation
of
these
data
is
complicated
by
the
fact
that
atrazine,
its
degradates
and
other
triazines
were
detected
at
significant
levels
in
reference
ponds.
Additionally,
at
some
sites,
frogs
were
collected
over
a
protracted
period
of
time
(
6
months),
rendering
GSI
a
questionable
metric.
This
study
provides
evidence
that
under
the
conditions
studied,
the
incidence
of
testicular
oocytes
in
Xenopus
laevis
is
roughly
2
to
3%.
Because
of
the
presence
of
atrazine
in
both
sampling
areas,
a
regression­
based
analysis
would
be
more
appropriate
than
hypothesis­
testing
based
analysis.
Unfortunately,
though,
the
study
is
not
designed
for
regression
analysis.
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
3
of
31
I.
MATERIALS
AND
METHODS
GUIDELINE
FOLLOWED:
Nonguideline
Study
COMPLIANCE:
Not
conducted
under
full
GLP;
however,
most
practices
as
defined
by
40
CFR
Part
160,
August
19,
1989
were
established
for
this
study,
including
but
not
limited
to:

°
Written,
authorized
protocol
°
Written,
authorized
Standard
Operating
Procedures
for
all
key
procedures.
°
Organization
and
personnel
were
sufficient
in
terms
of
number,
education,
training
and
experience.
°
Facilities
were
of
suitable
size
and
construction
°
Equipment
used
was
of
appropriate
design
and
adequate
capacity.
°
Test
material
identity,
strength,
purity
and
composition
were
characterized
°
Independent
QA
inspections
of
raw
data
and
final
report
were
conducted.
°
Interim
Report
was
written
°
Raw
data,
documentation,
records,
protocols,
and
final
phase
report
will
be
archived.

A.
MATERIALS:

1.
Test
Material
Atrazine
Description:
Not
reported
Lot
No./
Batch
No.
:
Not
reported
Purity:
Not
reported
Stability
of
compound
under
test
conditions:
Not
reported
Storage
conditions
of
test
chemicals:
Not
reported
2.
Test
organism:

Species:
African
clawed
frog
(
Xenopus
laevis)
.
Age
at
test
initiation:
:
Adults
Weight
at
study
initiation
(
mean
and
range):
Not
reported
Length
at
study
initiation
(
mean
and
range):
Not
reported
Source:
Adult
X.
laevis
were
field­
collected
in
two
areas
(
non­
corn
growing
areas
and
corn
growing
areas)
in
the
vicinity
of
Potchefstrooom,
South
Africa,
using
traps
baited
with
liver
and
meat
scraps.
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
4
of
31
B.
STUDY
DESIGN:

Objective:
1.
To
determine
if
exposure
to
triazines
used
in
corn
production
affects
the
incidence
or
extent
of
morphological
changes
in
the
gonads
or
intersex
gonads
in
the
male
and
female
metamorphs
and
adults
of
X.
laevis.
2.
To
determine
whether
there
are
changes
in
the
morphology
of
the
male
larynx
in
metamorphs
and
adults
of
X.
laevis
in
response
to
exposures
to
triazines
used
in
corn
production.
3.
To
determine
if
there
are
changes
in
plasma
steroid
concentrations
of
adult
X.
laevis
in
response
to
exposures
to
triazines
used
in
corn
production.
4.
To
determine
if
other
frogs
present
at
the
sites
show
any
morphological
or
other
changes
in
the
gonads.
5.
To
assess
feasibility
of
determining
if
X.
laevis
in
the
exposed
sites
have
been
subjected
to
selection
pressure.

1.
Experimental
Conditions
Sampling
was
conducted
in
April
and
May
approximately
6
months
after
atrazine
was
applied
(
October
/
November);
water
samples
to
characterize
exposure
were
collected
one
month
prior
to
frog
sampling.
Environmental
parameters
that
were
measured
at
each
site
included
habitat
description
(
vegetative
cover
and
water
depth).
Water
samples
were
analyzed
for
atrazine,
related
triazines,
other
pesticides
and
metals.

A
total
of
eight
sites
were
sampled
in
two
adjacent
regions
(
Viljoenskroon
corn
growing
region
=
E;
non­
corn
growing
region
=
C)
in
the
vicinity
of
Potchefstroom,
South
Africa.

Compounds
of
interest
included
atrazine,
its
metabolites
desethylated
atrazine
(
DEA),
desisopropyl
atrazine
(
DIA),
diaminochlorotriazine
(
DACT),
plus
other
herbicides
(
terbutylazine,
simazine
and
acetochlor).

Frogs
were
sampled
after
the
rainy
season
in
April
and
May;
however,
at
atrazine
sites
E1
and
E8,
frogs
were
collected
from
April
to
June
and
from
April
to
September,
respectively,
due
to
low
sample
sizes.
The
low
sample
sizes
at
sites
E1and
E8
were
attributed
to
the
introduction
of
sharptooth
catfish
(
Clarius
gariepinus)
into
the
ponds
due
to
high
runoff.
Sampling
at
all
other
sites
occurred
once
or
twice
within
a
"
few
days."
Frogs
were
collected
in
10
baited
traps
and
harvested
after
two
days
in
"
clean
water".
Afterwards,
animals
were
housed
individually
in
2­
L
plastic
containers
for
48
hours
to
recover
from
"
capture
stress".

Total
larynx
weight
was
used
to
determine
the
response
of
the
larynx
in
animals
from
experimental
and
control
study
sites.
Adult
males
from
each
site
were
selected
randomly
for
a
total
of
170
animals.

Testes
were
processed
for
routine
paraffin
embedding,
serially
sectioned
at
5

m,
and
stained
with
hematoxylin
and
eosin.
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
5
of
31
Three
slides
were
selected:
slide
A
was
selected
from
the
rostral
terminus
of
the
testes
on
the
basis
of
the
first
observation
of
a
seminiferous
tubule;
slide
B
was
selected
from
the
middle;
and
slide
C
was
selected
from
a
similar
arbitrarily
assigned
caudal
region
of
the
testis
with
a
complete
seminiferous
tubule.
Three
fields
were
photographed
from
each
of
the
three
slides
yielding
nine
photomicrographs
of
each
testis.
The
photomicrographs
were
overlain
with
a
grid
and
the
cell/
tissue
type
under
each
crossbar
was
identified
and
counted
by
three
different
readers
to
determine
the
representative
fractional
volume
of
spermatogonia,
spermatocyte,
sperm,
connective
tissue,
blood
vessel
"
other"
morphological
features.

For
oocytes,
every
200th
micron
section
was
evaluated.

II.
RESULTS
and
DISCUSSION:
[
All
results
discussed
in
this
section
and
the
next
are
those
reported
by
the
study
authors.
Although
supplemental
data
are
typically
used
in
a
qualitative
manner
only,
EFED
verified
spreadsheet
data
and
ran
basic
statistical
analyses
on
the
major
study
parameters.
See
attached
appendix.
If
results
differed
in
any
substantive
way,
the
difference
was
reported
in
the
text
below.]

The
researchers
were
unable
to
collect
Stage
66
metamorphs
except
in
reference
site
C3;
therefore,
the
study
focused
on
adults.
According
to
the
report,
the
targeted
number
of
fish,
i.
e.,
20
males
and
20
females
per
site,
could
not
be
achieved
(
Table
1)
at
sites
E1
and
E8
due
to
the
introduction
of
sharptooth
catfish;
however,
no
explanation
is
provided
as
to
why
the
targeted
number
could
not
be
obtained
from
reference
site
C3.

Table
2
presents
the
mean
body
weight
and
snout­
vent
lengths
(
±
standard
deviation)
of
frogs
collected
in
reference
and
experimental
sites.
There
was
no
statistical
difference
in
the
average
SVL
of
females
collected
from
reference
sites
(
71.8
±
4.45
mm)
compared
with
females
collected
from
experimental
sites
(
73.9
±
10.1
mm).
Additionally,
there
was
no
difference
in
mean
weight
of
females
from
reference
sites
(
36.5
±
4.32
g)
and
experimental
sites
(
42.8
±
15.7
g).

There
was
no
statistical
difference
in
the
average
SVL
of
males
collected
from
reference
sites
(
62.4
±
8.45
mm)
compared
with
males
collected
from
experimental
sites
(
62.6
±
2.87
mm).
There
was
also
no
statistical
difference
in
mean
weight
of
males
collected
at
reference
(
25.8
±
9.93
g)
and
experimental
sites
(
27.0
±
4.14
g).

While
the
report
states
that
there
was
no
difference
in
weight
of
the
left
testis
from
reference
(
18.6
mg)
and
experimental
sites
(
23.3
mg)
nor
between
the
weight
of
the
right
testis
from
reference
(
15.8
mg)
and
experimental
(
22.1
mg)
site
collected
animals,
there
were
differences
in
the
total
testicular
weight
(
combined
right
and
left
testis)
between
reference
and
experimental
collection
sites.
The
mean
testes
weight
from
control
(
0.037
±
0.021
mg)
was
statistically
different
(
ANOVA,
p
<
0.0449)
than
mean
weight
of
testes
(
0.046
±
0.018
mg)
from
frogs
collected
at
experimental
sites.

Table
2.
Mean
snout­
vent
length
and
body
weight
of
male
and
female
adult
frogs
collected
at
experimental
(
E1
­
E8)
and
reference
(
R1
­
R3)
sites.

Pond
Snout­
Vent
Length
(
mm)
Body
Weight
(
g)

Males
Females
Males
Females
E1
59.20
±
9.49
64.48
±
14.3
23.39
±
14.4
29.15
±
19.2
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
6
of
31
E3
66.63
±
6.99
74.39
±
6.94
32.57
±
8.99
39.38
±
8.47
E4
63.83
±
3.35
77.54
±
10.9
29.90
±
4.17
48.82
±
18.5
E6
62.66
±
9.58
88.50
±
9.53
26.11
±
12.4
66.96
±
22.7
E8
60.72
±
7.28
64.47
±
13.8
23.09
±
8.79
29.80
±
18.9
R1
58.00
±
6.64
69.55
±
8.02
19.44
±
6.10
32.83
±
11.1
R3
56.99
±
10.9
68.91
±
14.5
20.75
±
13.3
35.35
±
20.2
R8
72.11
±
7.19
76.95
±
7.04
37.26
±
12.0
41.25
±
10.4
Average
condition
indices
(
body
weight
÷
snout­
vent
length)
for
females
collected
from
reference
sites
(
0.48
±
0.15
g/
mm)
were
not
significantly
different
(
p>
0.0562)
from
females
collected
at
experimental
sites
(
0.54
±
0.20
g/
mm).
Similarly,
male
CI
between
reference
(
0.39
±
0.14
g/
mm)
and
experimental
sites
(
0.42
±
0.12
g/
mm)
were
not
statistically
different
(
p>
0.237).

The
report
states
that
mean
gonadalsomatic
indices
(
GSI
=
total
weight
of
gonad
÷
body
weight)
were
lower
for
males
collected
in
reference
sites
than
in
experimental
sites
but
were
not
statistically
different.
EFED's
analysis
(
see
attached
SAS
output),
though,
indicates
that
GSI
were
significantly
different
(
ANOVA,
p
<
0.0001)
for
males
collected
from
reference
(
0.13
±
0.04)
sites
compared
to
males
collected
from
experimental
sites
(
0.17
±
0.04).

Morphological
abnormalities
(
deformed
limbs)
were
observed
in
3
out
of
212
(
0.01%)
of
the
animals
and
were
attributed
to
attacks
by
the
Cape
terrapin
(
Pelomedusa
subrufa).

Based
on
gross
morphology,
only
2
out
of
101
males
examined
had
discontinuous
gonads,
and
4
had
testes
of
unequal
size.

The
extent
of
morphological
changes
in
adult
frog
testes
was
determined
by
evaluating
56
testes
from
reference
sites
and
43
from
experimental
sites
(
Table
3).
There
were
no
statistical
differences
in
the
cell
types
comprising
testes
collected
from
reference
and
experimental
sites.

Testicular
oocytes
were
observed
in
3%
of
the
animals
collected
from
reference
sites
and
2%
of
the
frogs
collected
from
experimental
sites.

The
study
claims
that
laryngeal
somatic
index
for
males
and
females
from
reference
and
experimental
sites
were
not
significantly
different
(
p
>
0.05);
however,
no
raw
data
were
provided
to
verify
these
observations.
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
7
of
31
Table
3.
Mean
cell
type
identified
in
testes
of
male
frogs
collected
from
reference
(
C1,
C3
and
C6)
and
experimental
(
E1,
E3,
E4,
E6
and
E8)
sites
along
with
associated
t­
test
p­
values.

Site
Spermatogoni
a
Spermatocyte
s
Sperm
Connective
Tissue
and
Other
Cells
Blood
Vessels
C1
3.0
25.6
55.9
14.9
0.5
C3
4.5
34.3
46.3
14.6
0.3
C6
2.8
22.6
60.3
13.9
0.3
E1
3.4
28.8
53.4
13.7
0.8
E3
3.6
24.9
58.1
13.2
0.2
E4
3.0
24.5
59.8
12.6
0.2
E6
3.2
20.8
62.4
13.2
0.3
E8
2.9
30.1
52.2
14.5
0.2
Mean
of
C1
­
C6
3.5
27.5
54.2
14.5
0.4
Mean
of
E1
­
E8
3.2
25.8
57.2
13.5
0.3
t­
test
p
values
0.6214
0.6382
0.4782
0.0754
0.7305
C.
REPORTED
STATISTICS:
Statistical
analyses
of
frog
numbers
at
each
site
were
conducted
using
standard
methods
such
as
Chi­
square,
Kruskal­
Wallis
and
Student's
t­
test.
The
initial
analyses
of
histopathology
were
carried
out
using
standard
ANOVA
procedures
followed
by
Tukey's
multi­
comparison
test
using
Minitab
®
statistical
software.

D.
VERIFICATION
OF
STATISTICAL
RESULTS:
Statistical
analyses
conducted
using
SAS
(
Statistical
Analysis
System,
Release
8.01,
Cary,
North
Carolina).
See
attached
printout.

E.
STUDY
DEFICIENCIES:

Atrazine
and/
or
triazine
degradates
were
present
in
control
sites.
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
8
of
31
While
the
low
sample
size
at
sites
E1
and
E8
may
have
been
explained
by
the
sudden
introduction
of
predatory
catfish,
it
doesn't
explain
why
the
remaining
corn­
growing
sites
(
E3,
E4
and
E6)
also
yielded
fewer
frogs
than
non­
corn
growing
sites.

Because
frogs
were
sampled
over
disproportionate
period
of
time,
e.
g.,
6
months
for
site
E8,
animals
were
at
different
stages
of
development.

Frog
traps
were
baited
with
liver/
meat
scraps
that
may
have
contained
hormones.

Following
capture,
frogs
were
housed
individually
in
"
clean
water"
for
48
hours.
The
volume
and
characteristics
of
this
water
were
not
described
in
the
study.

F.
REVIEWER'S
COMMENTS:

Atrazine
is
typically
applied
in
October
/
November
when
corn
is
planted
in
South
Africa.
Frog
sampling
was
conducted
after
the
rainy
season
in
April
to
May,
and
water
samples
to
characterize
exposure
were
collected
within
a
time
span
of
four
weeks
before
frogs
were
sampled.
At
exposure
site
E8,
frogs
were
sampled
over
a
six­
month
time
frame;
the
authors
attributed
the
fewer
number
of
frogs
to
a
high
runoff
event
that
washed
catfish
into
the
pond
and
the
catfish
preyed
on
the
frogs.
Given
the
aerial
photos
and
the
fact
that
the
majority
of
the
ponds
relied
on
rain
(
runoff)
and
were
not
dependent
on
a
stream,
it
is
unclear
where
the
catfish
came
from.
Because
Clarius
is
an
aquaculture
species,
it
is
possible
that
aquaculture
ponds
were
in
the
vicinity
of
E1
and
E8.

Frogs
were
attracted
into
baited
traps
containing
liver
and/
or
scrap
meat
that
may
have
contained
hormones.
Apparently
trapped
animals
were
held
in
cages
for
two
days
which
would
likely
have
provided
sufficient
time
to
consume
all
of
the
bait.
Also,
depending
on
catch
rates,
the
loading
within
the
cages
could
have
stressed
the
frogs.
Although
it
doesn't
state
the
desired
sample
size
in
the
report's
methodology,
apparently
(
based
on
the
raw
data),
20
male
and
20
female
frogs
were
to
be
sampled
at
each
site.
Table
1
shows
that
while
the
desired
sample
size
was
generally
available
for
frogs
collected
in
non­
corn
growing
areas,
sample
sizes
were
minimally
50%
lower
in
corn­
growing
areas.
In
site
E1,
sampling
for
female
frogs
extended
into
late
June
and
at
site
E8
sampling
extended
into
mid­
August
for
females
and
into
mid­
September
for
males.
Extended
sampling
periods
at
these
two
sites
may
have
resulted
in
collecting
animals
that
were
at
considerably
different
states
of
development.
The
gonadosomatic
index
and
ovarian
aromatase
activity
in
females
collected
from
sites
E1
and
E8
expressed
the
highest
variability
relative
to
any
of
the
other
treatment
sites.

Table
1.
Summary
of
total
number
of
Xenopus
laevis
collected
from
April
through
May
2002
at
non­
corn
growing
(
C)
and
corn
growing
(
E)
sites
in
South
Africa.

Site
Males
Females
C1
20
(
Apr
21
­
Apr
23)
20
(
Apr
21
­
May
1)

C3
17
(
Apr
4
­
Apr
11)
20
(
Apr
4
­
Apr
11)
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
9
of
31
C6
20
(
May
13
­
May
15)
20
(
May
13
­
May
14)

E1
8
(
April
29
­
May
7)
8
(
April
29
­
June
29)

E3
10
(
Apr
9
­
Apr
17)
10
(
Apr
9
­
Apr
18)

E4
10
(
Apr
13
­
Apr
15)
10
(
Apr
14
­
Apr
16)

E6
10
(
May
5
­
May
9)
10
(
May
5
­
May
6)

E8
6
(
Apr
29
­
Sep
17)
15
(
Apr
29
­
Aug
17)

G.
CONCLUSIONS:

This
report
is
part
of
a
multiphase
study
examining
the
effects
of
atrazine
on
African
clawed
frogs
(
Xenopus
laevis)
in
their
native
habitat
(
South
Africa).
This
phase
of
the
study
summarizes
the
body
weights,
lengths
and
condition
index
(
weight/
length)
of
both
male
and
female
collected
from
reference
site
(
no
corn
grown
and
no
atrazine/
triazine
use)
and
from
experimental
site
(
corn
grown
and
atrazine/
triazines
used)
ponds.
The
study
also
examined
several
aspects
of
gonadal
growth
(
testis
weight,
gonadosomatic
index)
and
development
(
incidence
of
deformities,
cellular
morphology)
in
frog
testes
collected
from
the
two
sites
to
determine
whether
atrazine
exposure
had
affected
these
measurement
endpoints.

While
males
were
smaller
than
females
at
both
study
sites,
there
was
no
difference
in
weight
or
length
of
males
or
females
collected
at
reference
or
experimental
sites.
Condition
indices
for
males
and
females
did
not
differ
between
reference
and
experimental
collection
sites.
However,
the
mean
testes
weight
from
reference
ponds
(
0.037
±
0.021
mg)
were
statistically
different
than
mean
weight
of
testes
(
0.046
±
0.018
mg)
from
frogs
collected
at
experimental
sites.
GSI
was
significantly
different
for
males
collected
from
reference
(
0.13
±
0.04)
compared
to
males
collected
from
experimental
sites
(
0.17
±
0.04).

There
were
no
statistical
differences
in
the
cellular
morphology
of
testes
collected
at
reference
sites
and
those
collected
at
experimental
site.
The
incidence
of
external
abnormalities
(
missing
limbs)
was
1%
across
all
animals
collected
(
reference
and
experimental)
and
was
attributed
to
the
predatory
effects
of
turtles.
The
incidence
of
testicular
oocytes
was
3%
in
males
collected
from
reference
ponds
and
2%
in
experimental
ponds.
The
study
claims
that
laryngeal
somatic
index
for
males
and
females
from
reference
and
experimental
sites
were
not
significantly
different
(
p
>
0.05);
however,
no
raw
data
were
provided
with
which
to
verify
these
observations.

Although
not
discussed
at
length
in
this
phase
of
the
study,
the
interpretation
of
these
data
is
complicated
by
the
fact
that
atrazine,
its
degradates
and
other
triazines
were
detected
at
significant
levels
in
reference
ponds.
Additionally,
at
some
sites,
frogs
were
collected
over
a
protracted
period
of
time
(
6
months)
rendering
GSI
a
questionable
metric.
This
study
provides
evidence
that
under
the
conditions
studied,
the
incidence
of
testicular
oocytes
in
Xenopus
laevis
in
South
Africa
is
roughly
2
to
3%.
Because
of
the
presence
of
atrazine
in
both
sampling
areas,
a
regression­
based
analysis
would
be
more
appropriate
than
hypothesis­
testing
based
analysis.
This
study,
though,
was
not
designed
for
regression
analysis.
Can
you
include
a
summary
conclusion
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
10
of
31
statement
here?
Anything
useful
in
this
study?
After
the
study
is
completed,
a
more
definitive
conclusion
will
be
reached.

MEAN
MALE
SVL
FOR
REFERENCE
(
R)
AND
EXPERIMENTAL
(
E)
COLLECTION
SITES
109
Obs
POND
_
TYPE_
_
FREQ_
MEAN
STD
1
E1
0
8
59.2000
9.4870
2
E3
0
10
66.6300
6.9863
3
E4
0
10
63.8300
3.3549
4
E6
0
10
62.6600
9.5755
5
E8
0
6
60.7167
7.2841
6
R1
0
20
58.0000
6.6373
7
R3
0
17
56.9882
10.9170
8
R6
0
20
72.1100
7.1908
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
11
of
31
ANOVA
FOR
MALE
SLV
BETWEEN
SITES
110
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
101
Dependent
Variable:
LENGTH
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
2.016502
2.016502
0.02
0.8822
Error
99
9041.101320
91.324256
Corrected
Total
100
9043.117822
R­
Square
Coeff
Var
Root
MSE
LENGTH
Mean
0.000223
15.22363
9.556373
62.77327
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
2.01650201
2.01650201
0.02
0.8822
Levene's
Test
for
Homogeneity
of
LENGTH
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
77939.9
77939.9
6.28
0.0139
Error
99
1229054
12414.7
Bartlett's
Test
for
Homogeneity
of
LENGTH
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
5.1814
0.0228
Dunnett's
t
Tests
for
LENGTH
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
99
Error
Mean
Square
91.32426
Critical
Value
of
Dunnett's
t
1.98422
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
SITE
Between
Simultaneous
95%
Comparison
Means
Confidence
Limits
E
­
C
0.2850
­
3.5203
4.0902
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
12
of
31
NONPARAMETRIC
COMPARISON
OF
MALE
LENGTH
BETWEEN
SITES
114
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
LENGTH
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
2872.50
2907.0
145.999621
50.394737
E
44
2278.50
2244.0
145.999621
51.784091
Average
scores
were
used
for
ties.

Wilcoxon
Two­
Sample
Test
Statistic
2278.5000
Normal
Approximation
Z
0.2329
One­
Sided
Pr
>
Z
0.4079
Two­
Sided
Pr
>
|
Z|
0.8159
t
Approximation
One­
Sided
Pr
>
Z
0.4082
Two­
Sided
Pr
>
|
Z|
0.8163
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
0.0558
DF
1
Pr
>
Chi­
Square
0.8132
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
LENGTH
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
29.0
28.217822
2.503874
0.508772
E
44
21.0
21.782178
2.503874
0.477273
Average
scores
were
used
for
ties.

Median
Two­
Sample
Test
Statistic
21.0000
Z
­
0.3124
One­
Sided
Pr
<
Z
0.3774
Two­
Sided
Pr
>
|
Z|
0.7547
Median
One­
Way
Analysis
Chi­
Square
0.0976
DF
1
Pr
>
Chi­
Square
0.7547
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
13
of
31
MEAN
FEMALE
SVL
FOR
REFERENCE
(
R)
AND
EXPERIMENTAL
(
E)
COLLECTION
SITES
116
Obs
POND
_
TYPE_
_
FREQ_
MEAN
STD
1
E1
0
6
64.4833
14.3272
2
E3
0
10
74.3900
6.9437
3
E4
0
10
77.5400
10.9218
4
E6
0
10
88.5000
9.5274
5
E8
0
15
64.4733
13.8195
6
R1
0
20
89.5450
91.3941
7
R3
0
20
68.9100
14.4936
8
R6
0
20
76.9500
7.0433
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
14
of
31
ANOVA
FOR
FEMALE
SLV
BETWEEN
SITES
117
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
111
Dependent
Variable:
LENGTH
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
628.8675
628.8675
0.38
0.5363
Error
109
178119.4467
1634.1234
Corrected
Total
110
178748.3142
R­
Square
Coeff
Var
Root
MSE
LENGTH
Mean
0.003518
52.99887
40.42429
76.27387
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
628.8675382
628.8675382
0.38
0.5363
Levene's
Test
for
Homogeneity
of
LENGTH
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
1.864E8
1.864E8
0.83
0.3656
Error
109
2.461E10
2.2581E8
Bartlett's
Test
for
Homogeneity
of
LENGTH
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
70.8565
<.
0001
Dunnett's
t
Tests
for
LENGTH
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
109
Error
Mean
Square
1634.123
Critical
Value
of
Dunnett's
t
1.98197
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
Simultaneous
SITE
Between
95%
Confidence
Comparison
Means
Limits
E
­
C
­
4.776
­
20.036
10.483
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
15
of
31
NONPARAMETRIC
COMPARISON
OF
FEMALE
LENGTH
ACROSS
SITES
121
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
LENGTH
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
60
3256.0
3360.0
168.985548
54.266667
E
51
2960.0
2856.0
168.985548
58.039216
Average
scores
were
used
for
ties.

Wilcoxon
Two­
Sample
Test
Statistic
2960.0000
Normal
Approximation
Z
0.6125
One­
Sided
Pr
>
Z
0.2701
Two­
Sided
Pr
>
|
Z|
0.5402
t
Approximation
One­
Sided
Pr
>
Z
0.2707
Two­
Sided
Pr
>
|
Z|
0.5415
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
0.3788
DF
1
Pr
>
Chi­
Square
0.5383
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
LENGTH
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
60
27.0
29.729730
2.637040
0.450000
E
51
28.0
25.270270
2.637040
0.549020
Average
scores
were
used
for
ties.

Median
Two­
Sample
Test
Statistic
28.0000
Z
1.0351
One­
Sided
Pr
>
Z
0.1503
Two­
Sided
Pr
>
|
Z|
0.3006
Median
One­
Way
Analysis
Chi­
Square
1.0715
DF
1
Pr
>
Chi­
Square
0.3006
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
16
of
31
AVERAGE
LENGTH
OF
MALE
AND
FEMALE
FROGS
AT
REFERENCE
AND
EXPERIMENTAL
SITES
123
Obs
SITE
SEX
_
TYPE_
_
FREQ_
LENGTH
SD_
L
1
C
F
0
3
78.4683
10.4010
2
C
M
0
3
62.3661
8.4536
3
E
F
0
5
73.8773
10.0521
4
E
M
0
5
62.6073
2.8664
MEAN
MALE
WEIGHT
FOR
REFERENCE
(
R)
AND
EXPERIMENTAL
(
E)
COLLECTION
SITES
124
Obs
POND
_
TYPE_
_
FREQ_
MEAN
STD
1
E1
0
8
23.3888
14.4006
2
E3
0
10
32.5740
8.9938
3
E4
0
10
29.9000
4.1673
4
E6
0
10
26.1050
12.3935
5
E8
0
6
23.0917
8.7925
6
R1
0
20
19.4445
6.1022
7
R3
0
17
20.7541
13.2996
8
R6
0
20
37.2590
12.0340
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
17
of
31
ANOVA
FOR
MALE
WEIGHT
BETWEEN
SITES
125
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
101
Dependent
Variable:
WEIGHT
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
52.00466
52.00466
0.35
0.5572
Error
99
14842.45191
149.92376
Corrected
Total
100
14894.45657
R­
Square
Coeff
Var
Root
MSE
WEIGHT
Mean
0.003492
45.83106
12.24434
26.71624
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
52.00466491
52.00466491
0.35
0.5572
Levene's
Test
for
Homogeneity
of
WEIGHT
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
126859
126859
2.23
0.1387
Error
99
5637162
56941.0
Bartlett's
Test
for
Homogeneity
of
WEIGHT
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
3.0051
0.0830
Dunnett's
t
Tests
for
WEIGHT
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
99
Error
Mean
Square
149.9238
Critical
Value
of
Dunnett's
t
1.98422
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
Simultaneous
SITE
Between
95%
Confidence
Comparison
Means
Limits
E
­
C
1.447
­
3.428
6.323
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
18
of
31
NONPARAMETRIC
COMPARISON
OF
MALE
WEIGHT
BETWEEN
SITES
129
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
WEIGHT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
2742.0
2907.0
146.005999
48.105263
E
44
2409.0
2244.0
146.005999
54.750000
Average
scores
were
used
for
ties.

Wilcoxon
Two­
Sample
Test
Statistic
2409.0000
Normal
Approximation
Z
1.1267
One­
Sided
Pr
>
Z
0.1299
Two­
Sided
Pr
>
|
Z|
0.2599
t
Approximation
One­
Sided
Pr
>
Z
0.1313
Two­
Sided
Pr
>
|
Z|
0.2626
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
1.2771
DF
1
Pr
>
Chi­
Square
0.2584
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
WEIGHT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
26.0
28.217822
2.503874
0.456140
E
44
24.0
21.782178
2.503874
0.545455
Average
scores
were
used
for
ties.

Median
Two­
Sample
Test
Statistic
24.0000
Z
0.8858
One­
Sided
Pr
>
Z
0.1879
Two­
Sided
Pr
>
|
Z|
0.3757
Median
One­
Way
Analysis
Chi­
Square
0.7846
DF
1
Pr
>
Chi­
Square
0.3757
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
19
of
31
MEAN
FEMALE
WEIGHT
FOR
REFERENCE
(
R)
AND
EXPERIMENTAL
(
E)
COLLECTION
SITES
131
Obs
POND
_
TYPE_
_
FREQ_
MEAN
STD
1
E1
0
6
29.1450
19.2315
2
E3
0
10
39.3810
8.4697
3
E4
0
10
48.8210
18.5235
4
E6
0
10
66.9560
22.6873
5
E8
0
15
29.8047
18.8819
6
R1
0
20
32.8315
11.1449
7
R3
0
20
35.3455
20.2334
8
R6
0
20
41.2495
10.4261
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
20
of
31
ANOVA
FOR
FEMALE
WEIGHT
BETWEEN
SITES
132
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
111
Dependent
Variable:
WEIGHT
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
1040.14605
1040.14605
2.97
0.0874
Error
109
38122.22029
349.74514
Corrected
Total
110
39162.36634
R­
Square
Coeff
Var
Root
MSE
WEIGHT
Mean
0.026560
47.58918
18.70147
39.29775
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
1040.146048
1040.146048
2.97
0.0874
Levene's
Test
for
Homogeneity
of
WEIGHT
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
2126150
2126150
5.83
0.0174
Error
109
39770440
364866
Bartlett's
Test
for
Homogeneity
of
WEIGHT
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
9.1873
0.0024
Dunnett's
t
Tests
for
WEIGHT
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
109
Error
Mean
Square
349.7451
Critical
Value
of
Dunnett's
t
1.98197
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
Simultaneous
SITE
Between
95%
Confidence
Comparison
Means
Limits
E
­
C
6.143
­
0.917
13.202
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
21
of
31
NONPARAMETRIC
COMPARISON
OF
FEMALE
WEIGHT
ACROSS
SITES
136
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
WEIGHT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
60
3132.50
3360.0
168.996671
52.208333
E
51
3083.50
2856.0
168.996671
60.460784
Average
scores
were
used
for
ties.

Wilcoxon
Two­
Sample
Test
Statistic
3083.5000
Normal
Approximation
Z
1.3432
One­
Sided
Pr
>
Z
0.0896
Two­
Sided
Pr
>
|
Z|
0.1792
t
Approximation
One­
Sided
Pr
>
Z
0.0910
Two­
Sided
Pr
>
|
Z|
0.1820
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
1.8122
DF
1
Pr
>
Chi­
Square
0.1782
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
WEIGHT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
60
27.0
29.729730
2.637040
0.450000
E
51
28.0
25.270270
2.637040
0.549020
Average
scores
were
used
for
ties.

Median
Two­
Sample
Test
Statistic
28.0000
Z
1.0351
One­
Sided
Pr
>
Z
0.1503
Two­
Sided
Pr
>
|
Z|
0.3006
Median
One­
Way
Analysis
Chi­
Square
1.0715
DF
1
Pr
>
Chi­
Square
0.3006
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
22
of
31
AVERAGE
WEIGHT
OF
MALE
AND
FEMALE
FROGS
AT
REFERENCE
AND
EXPERIMENTAL
SITES
138
Obs
SITE
SEX
_
TYPE_
_
FREQ_
WEIGHT
SD_
L
1
C
F
0
3
36.4755
4.3213
2
C
M
0
3
25.8192
9.9288
3
E
F
0
5
42.8215
15.7066
4
E
M
0
5
27.0119
4.1411
MEAN
TOTAL
WEIGHT
OF
TESTES
(
LEFT
AND
RIGHT
COMBINED)
BY
POND
139
Obs
POND
_
TYPE_
_
FREQ_
WEIGHT
SD_
W
1
E1
0
8
0.036925
0.014242
2
E3
0
10
0.053570
0.017665
3
E4
0
10
0.057670
0.009618
4
E6
0
10
0.033560
0.018074
5
E8
0
6
0.044200
0.019885
6
R1
0
29
0.022082
0.008848
7
R3
0
17
0.030953
0.023958
8
R6
0
30
0.051621
0.013727
MEAN
TOTAL
WEIGHT
OF
TESTES
(
LEFT
AND
RIGHT
COMBINED)
BY
SITE
140
Obs
SITE
_
TYPE_
_
FREQ_
WEIGHT
SD_
W
1
C
0
76
0.037232
0.021104
2
E
0
44
0.045650
0.018130
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
23
of
31
ANOVA
FOR
TOTAL
TESTES
WEIGHT
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITES
141
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
120
NOTE:
Due
to
missing
values,
only
91
observations
can
be
used
in
this
analysis.

Dependent
Variable:
T_
WT
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
0.00161041
0.00161041
4.14
0.0449
Error
89
0.03462027
0.00038899
Corrected
Total
90
0.03623068
R­
Square
Coeff
Var
Root
MSE
T_
WT
Mean
0.044449
47.75261
0.019723
0.041302
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
0.00161041
0.00161041
4.14
0.0449
Levene's
Test
for
Homogeneity
of
T_
WT
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
2.989E­
7
2.989E­
7
1.81
0.1823
Error
89
0.000015
1.654E­
7
Bartlett's
Test
for
Homogeneity
of
T_
WT
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
1.0070
0.3156
Dunnett's
t
Tests
for
T_
WT
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
89
Error
Mean
Square
0.000389
Critical
Value
of
Dunnett's
t
1.98699
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
SITE
Between
Simultaneous
95%
Comparison
Means
Confidence
Limits
E
­
C
0.008418
0.000197
0.016639
***
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
24
of
31
NONPARAMETRIC
COMPARISON
OF
TOTAL
TESTES
WEIGHT
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITES
145
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
T_
WT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
47
1900.0
2162.0
125.912307
40.425532
E
44
2286.0
2024.0
125.912307
51.954545
Average
scores
were
used
for
ties.

Wilcoxon
Two­
Sample
Test
Statistic
2286.0000
Normal
Approximation
Z
2.0768
One­
Sided
Pr
>
Z
0.0189
Two­
Sided
Pr
>
|
Z|
0.0378
t
Approximation
One­
Sided
Pr
>
Z
0.0203
Two­
Sided
Pr
>
|
Z|
0.0407
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
4.3298
DF
1
Pr
>
Chi­
Square
0.0375
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
T_
WT
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
47
19.0
23.241758
2.396612
0.404255
E
44
26.0
21.758242
2.396612
0.590909
Average
scores
were
used
for
ties.

Median
Two­
Sample
Test
Statistic
26.0000
Z
1.7699
One­
Sided
Pr
>
Z
0.0384
Two­
Sided
Pr
>
|
Z|
0.0767
Median
One­
Way
Analysis
Chi­
Square
3.1325
DF
1
Pr
>
Chi­
Square
0.0767
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
25
of
31
MEAN
FEMALE
CONDITION
INDEX
BY
CITE
147
Obs
SITE
_
TYPE_
_
FREQ_
CI
SD_
CI
1
C
0
61
0.47898
0.15083
2
E
0
51
0.54278
0.19533
ANOVA
FOR
FEMALE
CONDITION
INDEX
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITE
148
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
112
NOTE:
Due
to
missing
values,
only
110
observations
can
be
used
in
this
analysis.

Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
0.11133162
0.11133162
3.73
0.0562
Error
108
3.22722541
0.02988172
Corrected
Total
109
3.33855703
R­
Square
Coeff
Var
Root
MSE
CI
Mean
0.033347
33.99081
0.172863
0.508559
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
0.11133162
0.11133162
3.73
0.0562
Levene's
Test
for
Homogeneity
of
CI
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
0.00619
0.00619
3.05
0.0835
Error
108
0.2191
0.00203
Bartlett's
Test
for
Homogeneity
of
CI
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
3.5615
0.0591
Dunnett's
t
Tests
for
CI
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
108
Error
Mean
Square
0.029882
Critical
Value
of
Dunnett's
t
1.98218
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
SITE
Between
Simultaneous
95%
Comparison
Means
Confidence
Limits
E
­
C
0.06380
­
0.00172
0.12931
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
26
of
31
NONPARAMETRIC
COMPARISON
OF
FEMALE
CONDITION
FACTOR
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SIT
152
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
CI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
59
2965.0
3274.50
166.833000
50.254237
E
51
3140.0
2830.50
166.833000
61.568627
Wilcoxon
Two­
Sample
Test
Statistic
3140.0000
Normal
Approximation
Z
1.8522
One­
Sided
Pr
>
Z
0.0320
Two­
Sided
Pr
>
|
Z|
0.0640
t
Approximation
One­
Sided
Pr
>
Z
0.0334
Two­
Sided
Pr
>
|
Z|
0.0667
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
3.4416
DF
1
Pr
>
Chi­
Square
0.0636
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
CI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
59
24.0
29.50
2.627047
0.406780
E
51
31.0
25.50
2.627047
0.607843
Median
Two­
Sample
Test
Statistic
31.0000
Z
2.0936
One­
Sided
Pr
>
Z
0.0181
Two­
Sided
Pr
>
|
Z|
0.0363
Median
One­
Way
Analysis
Chi­
Square
4.3832
DF
1
Pr
>
Chi­
Square
0.0363
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
27
of
31
MEAN
MALE
CONDITION
INDEX
BY
SITE
154
Obs
SITE
_
TYPE_
_
FREQ_
CI
SD_
CI
1
C
0
57
0.39435
0.13640
2
E
0
44
0.42482
0.11530
ANOVA
FOR
MALE
CONDITION
INDEX
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITE
155
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
101
Dependent
Variable:
CI
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
0.02304412
0.02304412
1.41
0.2372
Error
99
1.61346081
0.01629758
Corrected
Total
100
1.63650493
R­
Square
Coeff
Var
Root
MSE
CI
Mean
0.014081
31.31858
0.127662
0.407624
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
0.02304412
0.02304412
1.41
0.2372
Levene's
Test
for
Homogeneity
of
CI
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
0.000694
0.000694
1.37
0.2444
Error
99
0.0501
0.000506
Bartlett's
Test
for
Homogeneity
of
CI
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
1.3333
0.2482
Dunnett's
t
Tests
for
CI
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
99
Error
Mean
Square
0.016298
Critical
Value
of
Dunnett's
t
1.98422
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.

Difference
SITE
Between
Simultaneous
95%
Comparison
Means
Confidence
Limits
E
­
C
0.03046
­
0.02037
0.08130
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
28
of
31
NONPARAMETRIC
COMPARISON
OF
MALE
CONDITION
INDEX
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITES
159
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
CI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
2680.0
2907.0
146.006849
47.017544
E
44
2471.0
2244.0
146.006849
56.159091
Wilcoxon
Two­
Sample
Test
Statistic
2471.0000
Normal
Approximation
Z
1.5513
One­
Sided
Pr
>
Z
0.0604
Two­
Sided
Pr
>
|
Z|
0.1208
t
Approximation
One­
Sided
Pr
>
Z
0.0620
Two­
Sided
Pr
>
|
Z|
0.1240
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
2.4172
DF
1
Pr
>
Chi­
Square
0.1200
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
CI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
57
23.0
28.217822
2.503874
0.403509
E
44
27.0
21.782178
2.503874
0.613636
Median
Two­
Sample
Test
Statistic
27.0000
Z
2.0839
One­
Sided
Pr
>
Z
0.0186
Two­
Sided
Pr
>
|
Z|
0.0372
Median
One­
Way
Analysis
Chi­
Square
4.3426
DF
1
Pr
>
Chi­
Square
0.0372
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
29
of
31
MEAN
MALE
GONADOSOMATIC
INDEX
(
GSI)
BY
SITE
161
Obs
SITE
_
TYPE_
_
FREQ_
GSI
SD_
GSI
1
C
0
76
0.13027
0.037421
2
E
0
44
0.16847
0.042300
ANOVA
FOR
MALE
GONADOSOMATIC
(
GSI)
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SITE
162
The
ANOVA
Procedure
Class
Level
Information
Class
Levels
Values
SITE
2
C
E
Number
of
observations
120
NOTE:
Due
to
missing
values,
only
91
observations
can
be
used
in
this
analysis.

Dependent
Variable:
GSI
Sum
of
Source
DF
Squares
Mean
Square
F
Value
Pr
>
F
Model
1
0.03315187
0.03315187
20.87
<.
0001
Error
89
0.14135759
0.00158829
Corrected
Total
90
0.17450946
R­
Square
Coeff
Var
Root
MSE
GSI
Mean
0.189972
26.79357
0.039853
0.148742
Source
DF
Anova
SS
Mean
Square
F
Value
Pr
>
F
SITE
1
0.03315187
0.03315187
20.87
<.
0001
Levene's
Test
for
Homogeneity
of
GSI
Variance
ANOVA
of
Squared
Deviations
from
Group
Means
Sum
of
Mean
Source
DF
Squares
Square
F
Value
Pr
>
F
SITE
1
3.249E­
6
3.249E­
6
0.72
0.3994
Error
89
0.000403
4.531E­
6
Bartlett's
Test
for
Homogeneity
of
GSI
Variance
Source
DF
Chi­
Square
Pr
>
ChiSq
SITE
1
0.6603
0.4164
Dunnett's
t
Tests
for
GSI
NOTE:
This
test
controls
the
Type
I
experimentwise
error
for
comparisons
of
all
treatments
against
a
control.

Alpha
0.05
Error
Degrees
of
Freedom
89
Error
Mean
Square
0.001588
Critical
Value
of
Dunnett's
t
1.98699
Comparisons
significant
at
the
0.05
level
are
indicated
by
***.
Difference
SITE
Between
Simultaneous
95%
Comparison
Means
Confidence
Limits
E
­
C
0.038194
0.021583
0.054806
***
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
30
of
31
NONPARAMETRIC
COMPARISON
OF
MALE
GONADOSOMATIC
INDEX
BETWEEN
REFERENCE
AND
EXPERIMENTAL
SIT
166
The
NPAR1WAY
Procedure
Wilcoxon
Scores
(
Rank
Sums)
for
Variable
GSI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
47
1645.0
2162.0
125.915315
35.000
E
44
2541.0
2024.0
125.915315
57.750
Wilcoxon
Two­
Sample
Test
Statistic
2541.0000
Normal
Approximation
Z
4.1020
One­
Sided
Pr
>
Z
<.
0001
Two­
Sided
Pr
>
|
Z|
<.
0001
t
Approximation
One­
Sided
Pr
>
Z
<.
0001
Two­
Sided
Pr
>
|
Z|
<.
0001
Z
includes
a
continuity
correction
of
0.5.

Kruskal­
Wallis
Test
Chi­
Square
16.8587
DF
1
Pr
>
Chi­
Square
<.
0001
Median
Scores
(
Number
of
Points
Above
Median)
for
Variable
GSI
Classified
by
Variable
SITE
Sum
of
Expected
Std
Dev
Mean
SITE
N
Scores
Under
H0
Under
H0
Score
ââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââââ
C
47
14.0
23.241758
2.396612
0.297872
E
44
31.0
21.758242
2.396612
0.704545
Median
Two­
Sample
Test
Statistic
31.0000
Z
3.8562
One­
Sided
Pr
>
Z
<.
0001
Two­
Sided
Pr
>
|
Z|
0.0001
Median
One­
Way
Analysis
Chi­
Square
14.8701
DF
1
Pr
>
Chi­
Square
0.0001
Data
Evaluation
Report
on
Gonadal
and
Laryngeal
Responses
to
Field
Exposure
of
Xenopus
laevis
to
Atrazine
in
Areas
of
Corn
Production
in
South
Africa
EPA
MRID
Number
458677­
10
Page
31
of
31
MEAN
CELL
TYPES
FOR
EXPERIMENTAL
SITES
168
Obs
POND
_
TYPE_
_
FREQ_
total
Gonia
Sperma
Sperm
other
BV_
avg
1
E1
0
8
33.725
3.425
28.7625
53.35
13.675
0.75
2
E3
0
10
33.340
3.640
24.8800
58.06
13.220
0.22
3
E4
0
10
32.500
2.970
24.4900
59.82
12.580
0.16
4
E6
0
10
33.960
3.240
20.7900
62.41
13.240
0.31
5
E8
0
5
33.740
2.880
30.1400
52.24
14.540
0.16
MEAN
CELL
TYPES
FOR
REFERENCE
SITES
169
Obs
POND
_
TYPE_
_
FREQ_
total
Gonia
Sperma
Sperm
other
BV_
avg
1
C1
0
19
34.9158
3.02105
25.5737
55.9316
14.9474
0.50526
2
C3
0
17
34.3941
4.53529
34.2882
46.2647
14.6353
0.28824
3
C6
0
20
34.3800
2.82000
22.6050
60.3200
13.9050
0.33500
AVERAGE
CELL
TYPES
BY
SITE
170
Obs
SITE
_
TYPE_
_
FREQ_
total
Gonia
Sperma
Sperm
other
BV_
avg
1
C
0
3
34.5633
3.45878
27.4890
54.1721
14.4959
0.37617
2
E
0
5
33.4530
3.23100
25.8125
57.1760
13.4510
0.32000
