EPA-600/3-76-004
March 1976
HERBICIDE TOXICITY IN MANGROVES
by
Howard J. Teas
University of Miami
Coral Gables, Florida 33124
Grant No. R 801178
Project Officer
Gerald E. Walsh
Environmental Research Laboratory, Gulf Breeze
Gulf Breeze, Florida 32561
U. S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY, GULF BREEZE
GULF BREEZE, FLORIDA 32561
-------
DISCLAIMER
This report has been reviewed by the Office of Research
and Development, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the
U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
ABSTRACT
The amine salts of 2,4-D and picloram were applied to
the Florida species of mangroves: red mangrove (Rhizophora
mangle), white mangrove (Laguncularia racemosa) and black
mangrove (Avicennia germinans).Treatments were to soil
or water, by aerial spray and to single leaves as droplets.
The effects on radiochloride uptake and on localization of
radiocarbon-labelled picloram after leaf application were
studied in red mangrove.
"Lethal doses" for young seedlings were 2.7 kg/ha for
white mangrove, 13 kg/ha for red and 13 kg/ha for black; for
mature plants they were 2.7, 13 and>53 kg/ha respectively.
"Tolerance doses" for young seedlings were 0.01, 5.3 and
5.3 kg/ha; for mature plants they were 0.5, 5.3 and 53 kg/ha.
"No effect doses" for seedlings were <0.01 kg/ha for all
species; for mature plants they were <0.1, 0.5 and 2.7 kg/ha.
Spray applications of 6.3 - 12.2 kg/ha of commercial
mixture to the canopy of a mixed-species forest caused par-
tial defoliation within three weeks. Within 16 months it
killed all of the white, 78 - 100% of the mature red, but
none of the mature black mangroves.
Radiocarbon-labelled picloram concentrated in dormant
buds of red mangrove and it is concluded that the tree is
killed by the mixture because of its effects on them.
The research reported here was funded through EPA Grant
No. R 801178.
13.1
-------
CONTENTS
Section Page
I Conclusions 1
11 Recommendat ion 2
III Introduction 3
IV Materials and Methods 5
V Results 11
VI Discussion 23
VII References 29
VIII Abbreviations 32
IX Glossary of Terms 33
v
-------
TABLES
No. Page
1 Herbicide dosages tested by soil application
(water over soil) 6
2 "Lethal", "tolerance" and "no-effect" concentrations of
herbicide applied to the soil (water over soil) 11
3 Herbicide dosages received at canopy and ground levels 12
4 Defoliation of tree mangrove species after herbicide
spraying (visual ratings) 13
5 Cumulative leaf fall in experimental plots after spraying.. 14
6 Killing of red mangrove after herbicide spraying 15
7 Single leaf applications: results after six weeks 17
8 Translocation of ^C-labelled picloram from leaf
application in red mangroves 18
9 Radioactive chloride uptake by red mangrove seedlings
treated with herbicide 19
10 Chlorophyll content of leaves of red mangrove treated
with 2,4-D and picloram 20
11 Translocation of sucrose in red mangroves treated with
2,4-D and picloram 21
12 Transpiration by red mangrove treated with 2,4-D and
picloram 22
13 Ethylene evolved from leaves of red mangrove treated with
2,4-D and picloram 22
VI
-------
FIGURES
No.
1 Forms of stem tips found in red mangrove
after herbicide spraying 16
VII
-------
ACKNOWLEDGEMENTS
This research was supported by the Environmental
Protection Agency under research grant R-801178. Studies
in South Vietnam were supported by the National Academy
of Sciences Committee on the Effects of Herbicides in
Vietnam. The author is grateful to the Dow Chemical
Company for Agent White herbicide and Carbon-14 labeled
picloram, to the Environmental Protection Agency for
facilities at the Perrine Laboratory, to James Vensell
of the Deltona Corporation for use of the mangrove forest
land, to Jeff Chell for work with labeled picloram; and
to Dr. A. K. Burditt, Wallace E. Manis and William F.
Reeder of the U.S. Department of Agriculture for the use
of facilities at the Sub-tropical Horticulture Research
Station.
Vlll
-------
SECTION I
CONCLUSIONS
1. The Florida black mangrove is most resistant to her-
bicide, the red mangrove intermediate and the white
mangrove most sensitive to Agent White by soil (water)
or aerial spray application.
2. Species of the genus of the Florida black mangrove,
Avicennia, are generally resistant to herbicide;
species of the genus of the red mangrove, Rhizophora,
are generally sensitive.
3. The Herbicide sensitivity of red mangrove is not due
to root membrane alteration or effects on transpiration
or translocation.
4. Sensitivity of red mangrove derives from the trans-
location of herbicide to and killing of all the tree's
dormant buds.
5. Spray drift hazard to mangroves from amine salt forms
of Agent White herbicide should be minimal.
6. Because of soil binding of herbicide, agricultural
runoff from areas of Agent White use are unlikely to
pose a hazard to mangroves
-------
SECTION II
RECOMMENDATION
Although soil uptake and tidal washing minimize the
likelihood of effects of herbicide on mangroves, it is
recommended that the use of Agent White be restricted in
areas where runoff might flow directly into a mangrove
estuary or where spray drift might reach mangroves.
-------
SECTION III
INTRODUCTION
Mangroves are trees or bushes that grow between the
high water penetration of spring tides and mean low tide.
They occur along tropical and subtropical shores where the
wave energy is low and are found in the estuarine portions
of rivers. The importance of coastal areas and estuaries
has been recognized increasingly in recent years, for example
in volumes edited by Thomas (1956), Lauf (1967), and Ketchum
(1972). It has been estimated that as much as 98.5 percent
of all commercial fish and shellfish species caught in the
Gulf of Mexico off Florida spend part of their lives in
estuarine environments (Glooschenko, 1968). Mangrove es-
tuaries play an important role in the life cycles of shrimp
in southern Florida (Idyll et al., 1968). Odum and Heald
(1972) have made detailed studies of the role played by
mangrove detritus in the food chains of estuaries in Ever-
glades National Park in Florida. Mangrove estuaries clearly
have great significance in food production of coastal areas.
Mangroves have been found to be very sensitive to auxin-
type defoliants. Tschirley (1969) and Orians and Pfeiffer
(1970) surveyed military defoliation sites in South Vietnam
and reported that although only a modest kill of upland
forest trees generally followed a single spraying, the great
majority of mangroves were killed by the same treatment.
This sensitivity of mangroves to herbicides combined
with the importance of mangrove estuaries in food webs
pointed out the need to evaluate the hazard to mangroves
from herbicide spray drift and herbicide-contaminated agri-
cultural runoff water. Portions of the investigations
reported here have been presented at an international sym-
posium on mangroves and will be published in the proceedings
(Teas and Kelly, 1975).
The defoliants that had proven so toxic to mangroves
in South Vietnam were: Agent Orange (2,4,5-trichlorophenoxy-
acetic acid + 2,4-dichlorophenoxyacetic acid as n-butyl
esters) and Agent White (2,4-dichlorophenoxyacetic acid -t
picloram, both as triisopropanolamine salts). All three of
the active ingredients are absorbed by roots and tops and
are translocated and concentrated in actively growing parts
of plants (Weed Science Society of America, 1970). These
three compounds function as defoliants and herbicides ac-
cording to dosage, growing condition of the plants, and
-------
other factors. All three active compounds meet the stereo-
chemical requirements for auxins (Eisinger and Morre, 1971).
Auxin herbicides are often selective with respect to
tree species as shown, for example, by a study of picloram
toxicity to red maple and white ash; red maple was killed
by concentrations of picloram that only slightly injured
white ash (Mitchell and Stephenson, 1973).
The United States military rate of application of Agent
White in Vietnam was 3 gal/a which corresponds to 8.54 kg/ha
of active ingredients (6.72 kg 2r4-D and 1.82 kg picloram)
or to 7.62 Ibs/a of active ingredients (6.0 Ibs 2,4-D and
1.62 Ibs picloram) (Lang, 1974).
In the present study the red mangrove, Rhizophora
mangle; the black mangrove, Avicennia germinans;and the
white mangrove, Laguncularia racemosa, were tested to deter-
mine "lethal", "tolerance11 and "no effect" doses of Agent
White. This military defoliant is identical to a commercial
formulation of 2,4-D and picloram (Tordon R101) that is
widely used as a brush killer. Because of the reported ex-
treme toxicity of dioxins that are found in some samples of
Agent Orange, no tests were conducted with this material.
Soil applications of Agent White were made to the three
Florida mangrove species at three stages: 2 month old seed-
lings; 14 month old seedlings; and "mature" (estimated 4-10
year old) trees. Herbicide was applied to the water over
soil in potted mangroves to simulate the effect of dissolved
herbicide in runoff water.
Aerial spray applications were made to a mature mangrove
forest in southern Florida to predict the effect of herbi-
cide drift and in order to follow the effects of herbidides
on mangroves, data that were not obtained in Vietnam (Lang,
1974). In addition, laboratory experiments were conducted
with Agent White, with radioactive carbon-labelled picloram,
and with radioactive chloride as part of an investigation
of the basic mechanism of auxin-type herbicide toxicity to
the red mangrove.
-------
SECTION IV
MATERIALS AND METHODS
HERBICIDE
The herbicide was Tordon 101 from the Dow Chemical
Company. Herbicide or dilutions were applied to water over
the soil with stirring. The experimental dosages used are
shown in Table 1.
Picloram with carboxyl carbon-14 label, specific acti-
vity 1 mCi/mM, was obtained from the Dow Chemical Company.
The triisopropanolamine salt of the labelled picloram was
prepared and diluted approximately 9 to 1 with non-radio-
active herbicide before application so that the 2,4-D and
picloram constituents were present in almost the normal ratio.
SOIL APPLICATIONS
Mangrove plants were collected in southern Florida and
planted in. plastic dishpans, buckets or wading pools con-
taining estuarine mud. Water levels were maintained about
5-10 cm above the soil. The water salinity was adjusted
weekly to the range of one third to two thirds that of sea-
water during the first nine months of the experiments and
less often thereafter. The plants were maintained at the
Miami U. S. Plant Introduction Station under natural partial
shade.
All three Florida mangrove species produce seed or pro-
paguJ.es in the fall. "Two month" old seedlings were from
the fall crop of 1973. They were planted directly from the
field into plastic pans or buckets and treated with herbi-
cide in January 1974.
Five to ten plants were used per pan and pans repli-
cated 3 times for the white and black mangroves and 6 times
for the red mangroves. "Fourteen month" old plants were
from the fall crop of 1972. They were individually planted
in 12-cm diameter plastic pots which were placed in plastic
buckets to maintain the water level above the soil and were
treated in January or February 1974. Treatments consisted
of three plants per plastic bucket and three replicates for
the red and white mangroves; two plants per bucket and three
replicates for the black mangroves. Salinity was adjusted
as noted above.
-------
•
^N
hJ
M
O
CO
p*
W
>
O
H
EH
&
%,_ X
K
O
M
<
o
M
J
Ck
<<
J
M
O
CO
^4
CO
Q
H
£-i
CO
w
EH
CO
o
CO
g
M
o
I-H
03
fv*
W
ffi
*
iH
w
H-I
CQ
fn
0
o
CS O CO
> -H ^!
S-t S-< ^
Q) Q)^-'
CO Pi
JD
0
o o} in in in in cji o^ Oi
Tfinin mmin mmm
co
CM
m
CO
CO
CM
CM
CO
10 1-1
-p
S3 CO
0 •
M 0 CM
bo
0 CS CM
> co m
•rl O •
-p Q
O r-t
•
O
(M
O
O
I-l
o
0
0
O 0
J^ Q^
bo >,
£3 EH
OS
X X
XXX XXX XX
XXX XXX XX
XXX XXX XX
XXX XXX XXX
XXX XXX XX
XX X X X X
XX XXX
KS rS rS rN KS rS
0 ^0 ,M 0 4i
O O SH -H -d
g S 3 -P 0 CO O
-P CO -P 0 r-<
CM •* cS 0 cS bO 1
rH ^^ gcS^
-------
"Mature" plants were transplanted from the field into
individual 65-cm diameter plastic childrens' wading pools.
These plants had the mature form and were typically 1.5-2
m nigh. The red mangroves had 5 or more prop roots; the
black mangroves had well developed pneumatophores; the
white mangroves had basal trunk diameters of 2-5 cm. Sali-
nity was adjusted as noted above. The ages of these mature
trees were estimated to be 4-10 years. Space availability
limited the numbers of these large plants. Four red man-
grove plants were used per treatment, three plants per treat-
ment in the case of black and white mangroves. Plants were
examined weekly for six weeks and then less often for 40 to
59 weeks. Plants of all age classes were scored for leaf
characteristics, that is, leaf size, leaf curl, browning,
drooping, abscission and chlorosis; for growth rate and
number of leaves and branches; for tumorous growths; for
stem tip viability, and for overall plant viability.
The lowest herbicide concentration that killed the
majority of plants within the 40-59 week observation period
was considered to be the "lethal" dose. The maximum dosage
from which 80% or more of the plants recovered, i.e. were
still alive after 40-59 weeks, although not necessarily
completely normal, was considered the "tolerance" dose.
The highest dosage that gave rise to no detectable symptoms
was considered the "no effect" dose.
SPRAY APPLICATIONS
Experimental spray plots were located in Collier County
near Marco Island, Florida, on land of the Deltona Corpora-
tion. The plots were in a forest that is a well developed
mixed mature stand of red, black and white mangroves about
8-13 m tall. Nine plots were laid out, each 20 by 40 m,
0.08 ha (0.2 acres).
Plots were treated with Tordon 101. Spraying was car-
ried out in the calm weather of early morning by helicopter
at a height of about 3 m above the forest canopy. Spray
was applied from a 9-m wide boom in two passes per plot.
Three plots were treated on 15 December 1972 and three ad-
ditional plots were sprayed on 19 January 1973. Actual
field herbicide dosage was determined from samples collected
on Whatman No. 3 MM filter paper backed with aluminum foil.
Collection papers were positioned above the canopy and others
were located 0.5 m above the ground to assess the under-
canopy dosage. Sample workup and quantitative estimations
-------
of picloram and 2,4-D by gas chromatography were carried out
as the Perrine Primate Research Laboratory of the Environ-
mental Protection Agency by procedures developed at the Gulf
Breeze Environmental Research Laboratory, U. S. Environ-
mental Protection Agency. Because of an accident, many of
the filter paper collectors were lost; however, dosages had
been determined for three herbicide sprayed plots.
Effects of herbicides were assayed by degree of defol-
iation and tree survival. Visual estimates of defoliation
and leaf litter pan collections were made. Determination of
whether the trees were ""living" or "dead" was based on
machete cuts. When a cut showed that the back and wood were
normally light colored and moist, the tree was classified
as "living"; if the bark and wood were brown and dry, it
was classified as "dead". Observations were carried out
over a period of approximately 16 months following spraying.
SINGLE LEAF APPLICATIONS
Single-leaf applications of herbicide were made to in-
dividual leaves of a few red, black and white mature plants
in plastic pools. Viability, appearance of leaves, branches,
prop roots and pneumatophores were recorded.
Radiocarbon-labelled herbicide was applied to 2 year
old red mangroves brought to the laboratory. Leaves nearest
the apical bud were treated. Labelled herbicide was applied
in the form of 20 or more micro drops per leaf at the rate
of 2.63 kg/ha. Experiments were terminated at 15 hrs after
application of radioactive herbicide. Plants were harvested
and the leaves, apical buds, sections of stem, and dormant
buds were dissected, and finely cut. The mineed samples
were extracted with four successive 5-ml aliquots of acetone,
the acetone evaporated and the radioactivity counted using
a Tracerlab liquid scintillation counter with standard
scintillation fluor in toluene.
RADIOCHLORIDE UPTAKE BY RED MANGROVE
Red mangrove seedlings at the 2-leaf stage were main-
tained at room temperature in the laboratory in individual
100 ml beakers of one-third strength seawater (10-12 ppt)
containing radioactive chloride as 36ci. The roots were
shaded by aluminum foil covers and opaque paper. The plants
were lighted by a bank of flourescent lights programmed for
-------
12-hr days. Two plants were treated with Agent White via
the roots at 2.8 kg/ha, two at 28 kg/ha, and two served as
controls. They were observed for six weeks, after which
the leaves of each plant were thinly sliced and minced with
a razor blade. They were then extracted with water, with
intermittant swirling, for 12 hrs. Aliquots of the super-
natant fluid were dried at 60 C in scintillation vials.
The scintillation fluor contained Beckman BBS-3 solubilizer.
and the samples were counted on a Tracerlab Corumatic counter,
CHLOROPHYLL CONTENT OF LEAVES
Leaves for chlorophyll determination were sampled by
use of a sharp paper punch. Fresh weights of leaf samples
were taken and the leaf discs ground in a porcelain mortar
in 80% (v/v) acetone-water. The chlorophyll solution of
acetone was decanted and leaf remains reextracted three
times. The combined extracts were made to 15 ml volume in
a glass-stoppered graduated centrifuge tube. Tubes were
centrifuged and supernatant withdrawn for chlorophyll esti-
mation. The chlorophyll solutions were diluted with 80%
acetone where necessary to obtain appropriate optical den-
sities. Estimation of chlorophyll content was carried out
by use of a Spectronic 20 spectrophotometer from the general
method of MacKinney (1941) and the equation developed by
Arnon (1949):
D652 x 1000 v V
C =
34.5 " 1000 x W
Where
C = Total chlorophyll, mg/g fresh weight
D652 = Optical density at 652 nm
V = Volume of extract
W = Weight of leaf sample, g
TRANSLOCATION OF SUCROSE
Translocation of photosynthate was measured by the use
of uniformly labeled 14C sucrose (1.135 mCi/mg) applied to
punch holes in red mangrove leaves. Approximately 1 uCi was
applied per leaf and samples were tested 18 hrs later. Leaf
or stem material was thinly sliced using a single edged
stainless steel razor blade, minced and extracted in water
-------
for 12 hrs with intermittent swirling. Sucrose from the
application site was rinsed from each leaf with five aliquots
of water. Aliquots of leaf or stem extract or of leaf rin-
sate were dried at 60 C in scintillation vials. The scin-
tillation fluor contained Beckman BBS-3 solubilizer. Samples
were counted in a Packard Tri-Carb scintillation counter.
TRANSPIRATION
Possible effects of herbicide treatment on transpiration
were checked by a simple method. The plants were placed with
their roots in glass jars. The tops extended through holes
in the caps and the roots were in the dark. The jars had
volume calibration marks on their sides and the amount of
water transpired by the leaves was estimated by volume
changes of the water in the jars.
ETHYLENE EVOLUTION FROM LEAVES
Ethylene evolution by herbicide-treated leaves was
measured by chromatography of gas samples from chambers
containing leaves. Four to six leaves were used per sample
and leaf quarters or eighths incubated for 18 hrs. Samples
of the atmosphere in which leaves had been incubated were
chromatographed using an alumina column at ambient tempera-
ture. Ethylene measurement was by means of a Gow Mac hydrogen
flame ionization unit and electrometer. An Esterline Angus
recorder was used. Ethylene elution time had been checked
against several other low M.W. gases with several types of
columns. Sample values were calculated by comparison with a
series of ethylene standards.
10
-------
SECTION V
RESULTS
SOIL APPLICATIONS
The best estimates of lethal, tolerance and no effects
doses for soil application are shown in Table 2. Because
the experimental herbicide dosages were not numerous enough
to accurately determine the lethal, tolerance and no effect
levels in each case, the true lethality doses would be lower
than those listed; the true doses for tolerance would be
higher than those listed, and the true doses for no effect
would be higher. The ranges of herbicide concentrations
tested generally proved to be more suitable for assessing
lethal doses than for tolerance and no effect levels.
It can be seen in Table 2 that the black mangrove is
generally more resistant to soil applications of Agent White
herbicide than is red mangrove, and both are generally more
resistant than white mangrove. Mature plants are generally
more resistant than are younger plants.
TABLE 2.
"LETHAL", "TOLERANCE" and "NO-EFFECT"
CONCENTRATIONS OF HERBICIDE APPLIED
TO THE SOIL (WATER OVER SOIL).
Plant Age at
Treatment
2 months
14 months
Mature
Mangrove
Type
White
Red
Black
White
Red
Black
White
Red
Black
Dosage* (kg/ha)
Lethal
2.7
13
13
2.7
2.7
27
2.7
13
>53
Tolerance
0.01
5.3
5.5
0.5
0.5
2.4
0.5
5.3
53
No-Effect
<0.01
<0.01
<0.01
0.01
0.01
0.10
^0.11
0.5
2.7
*Active ingredients: 78.6% 2,4-D and 21.4% picloram
11
-------
In only one case, in a mature red mangrove, was swelling
of a stem tip noted after soil application of herbicide.
SPRAY APPLICATIONS
The herbicide dosages received by canopy top and at
ground level are shown in Table 3, The visual estimates of
defoliation are shown in Table 4.
TABLE 3.
HERBICIDE DOSAGES RECEIVED AT CANOPY AND
GROUND LEVELS (TEAS AND KELLY, 1975).
Treatment
Dose of Herbicide*
Canopy Top"
Ground Level
2,4-D Picloram 2,4-D Picloram
(kg/ha) (kg/ha) (kg/ha) (kg/ha)
Control
Herbicide
treatment A
Herbicide
treatment B
Herbicide
treatment C
0
4.95
9.59
5.11
0
1.35
2.61
1.39
0
1,10
1.10
0.67
0
0.30
0.30
0.18
*Active ingredients
Leaves of herbicide-sprayed white mangroves began to
fall in 2-3 weeks and the trees were completely defoliated
by 5 weeks. Red mangrove leaves began to fall between 2
and 3 weeks after herbicide spraying and defoliation was
approximately 85-90% by 16 weeks. Black mangrove leaf
fall began in 3-4 weeks and reached an estimated 35-50%
defoliation at the end of one year. By 16 months (66-71
weeks) the black mangroves were substantially refoliated.
The cumulative leaf falls for the three experimental
spray plots and control, measured by the use of litter pans,
are shown in Table 5. In Florida, the annual natural leaf
fall for red mangrove approximates the standing crop of
leaves, which for a forest of this type is in the range of
12
-------
o
M
!x
^J
PS
to
w
Q
M
O
1
w
a
PS
w
EH
CO
H
M
O
W
OH
M
w
o
PS
CJ
^
w
OS*'
CO
fa O
0^
M
t7" Qj
O ^1
M PS
EH
< J
HH 42
•4 P
O CO
w >
Q >^x
°
^H
H
•J
03
^3
H
OS
j^
p.
W
v
-p
=H
<5
0]
v;
CU
0)
^
[v.
}
CD
CD
^
m
i
OJ
•*
CD
r-l
CN
r-l
CD
m
•^
co
CM
000 O
0
r-l
000 O
o
r-l
o o o o
o
l-l
o o o o
o
r-l
o o o o
o
r-l
O O 0 O
0
o o o o
o
r-l
o o o o
/
r^
o o o o
0)
> M
O 0)
?-< -ri
bfi O
fl 0
rt a,
SCO
0) ,M CD
-P O -P
•rl "O OS -H
X3 Q) I-H x!
Ss PS CQ &
-p
fl
0)
a
0]
Q)
J_(
EH
<3
a? -p
PO f-4
rH -r( CD
0 OS
S-l -IH -P
•P ^2 «J
fi 5^0)
O 0> M
O ffi EH
C75
CT>
(35
o
o
CJ5
o
05
o
CJ5
m
o
m
o
I
l~1
0
^
0
PS
o
r-l
0
m
m
CM
m
CN
in
CN
m
CM
m
CN
o
^J
p^
o
rM
o
OS
r-l
PQ
O CD O
O CJi r— 1
r-l
o m m
O Oi CO
r— 1
o m m
O CO CN
r-l
000
O CD rH
r-l
O O O
O rH r-l
rH
O O O
O r-l r-l
1
P^
o o m
m i-t
o m o
000
CD ^
•P O
•rl 'O OS
J2 CU i-H
> PS CQ
CQ
^
O C5> O
O O5 r-l
r-f
o cyj o
o o in
r-l
o o m
O C5 CM
r-l
O 0 0
O C5 CN
O O 0
o en I-H
r-l
O O O
O in r-l
i
p*i
o m o
O CN rH
rH
moo
l^_ H--J — 1
C^» r^ nH
0 0 O
cu .yj
•P 0
•rl T3 BJ
& PS CQ
o
J
x^N
in
CJi
r-l
l-H
Q)
|s^
•s
rj
a
CQ
OS
o
H
a
O
^
SH
a
-P
OS
Q
•X-
13
-------
475-650 gm/m2. The leaf fall data show that this primarily
red mangrove forest suffered very extensive defoliation.
The visual estimates of defoliation at 16 weeks were: 100%
for white, 85-90% for red, and 25% for black.
TABLE 5. CUMULATIVE LEAF FALL IN EXPERIMENTAL PLOTS
AFTER SPRAYING. (TEAS AND KELLY, 1975).
Canopy Herbicide
Dose*
Herbicide
Treatment
Control
A
B
C
2,4-D
(kg/ha)
0
4.95
9.59
5.11
Picloram
(kg/ha)
0
1.35
2.61
1.39
Leaf Fall
(gm/m2
dry wt)
164
497
348
561
*Active ingredients
The white mangroves began to die in 24 to 29 weeks.
After white mangrove trees had been classified as dead by
the machete test, several showed basal sprouting of mal-
formed leaves and shoots, but these died within a few weeks.
All white mangrove trees in the herbicide treated plots
were killed.
Killing of red mangroves was first noted at 29-32 weeks
after herbicide spraying (Table 6). By 49-54 weeks, 25-50%
of randomly sampled trees were dead. With the killing of
many canopy trees and the refoliation of some undercanopy
red mangroves, it became apparent that larger trees, red,
white and black, had protected undercanopy red mangroves
from herbicide. The fraction of trees that had been killed
at 66-71 weeks was determined for only tall, top-of-canopy
trees that received the full herbicide dose. Herbicide
plots A and C showed complete kill at that time; plot B
had 11 living trees of 50 tested.
14
-------
TABLE 6. KILLING OF RED MANGROVE AFTER HERBICIDE
SPRAYING (TEAS AND KELLY, 1975).
Number of Weeks After Herbicide Spray
(Trees Dead of Total Tested)
Herbicide
Treatment
Control
A
B
C
24
0/20
0/20
—
__
29
—
--
4/20
1/20
32
0/30
4/30
—
—
37
—
—
1/30
0/30
49-54
0/20
10/20
9/20
5/20
66-71*
0/50
50/50
39/50
50/50
*0nly the tall trees that reached the top of the canopy
were scored.
Based on evidence from canopy and undercanopy trees, red
mangroves appear to have been killed by a single spray ap-
plication of Agent White at 6.3 kg/ha (2.2 gal/a), and to
survive a treatment of 1.4 kg/ha (0.5 gal/a). These doses
correspond to 4.95 kg/ha of 2,4-D and 1.35 kg/ha picloram
in the first case and to 1.1 kg/ha 2,4-D and 0.3 kg/ha pic-
loram in the second.
Several relatively small (less than about 5 cm diameter)
black mangroves appeared to have been killed by the spray
but no larger trees died. Some of the apparently dead small
black mangrove trees in herbicide plots showed basal re-
growth at 66-71 weeks.
Mature black mangrove (Avicennia) in Florida is clearly
much more resistant to Agent White than red mangrove (Rhizo-
phora). It is interesting that in the Saigon River delta,
where military maps indicated that spraying had been with
Agent White, Avicennia trees were often seen as the only
survivors in areas that had formerly been predominantly
Rhizophora forests (Teas, 1972).
Some of the stem growth patterns of red mangrove that
developed after herbicide spraying showed apparent auxin
effects. Five common patterns of red mangrove tip and bud
swelling that were observed are shown in Fig. 1. The lo-
cations of stems showing these patterns generally suggest
increasing dosage effects from I through V. Full canopy
trees that received high dosages of herbicide and died early
15
-------
LEAF
APICAL OR TERMINAL BUD
NORMAL TIP
LEAF SCAR
I TERMINAL BUD DEAD,
DEFOLIATION, REGROWTH
OF LEAVES
II TERMINAL DEAD, REGROWTH
OF LEAVES FROM DORMANT BUDS
III TERMINAL DEAD, REGROWTH
FROM BUDS, SWELLING
AT BUD SITES
IV SWELLING AT BUD SITES,
NO LEAF REGROWTH
V STEM DEAD, NO SWELLING
NO REGROWTH
Figure 1. Forms of stem tips found in red mangrove after
herbicide spraying (from Teas and Kelly, 1975)
16
-------
had pattern V stems. The main canopy portion of some red
mangrove trees within the spray plots showed pattern V stem
tips, whereas the branches outside the spray plot had nor-
mal stems and leaves. Smaller trees, or branches of large
trees, that were in positions where they would have been
protected from the highest herbicide doses by the leaves
and branches above them, often showed stems of patterns
I-IV.
SINGLE LEAF APPLICATIONS
A small experiment was carried out in which herbicide
was applied to individual leaves of mature plants in plastic
pools. The effects after 6 weeks are shown in Table 7. In
the white mangroves, all doses killed the leaves of the
small branch, the small branch itself and one or more near-
by branches. In the red mangroves, the two higher doses
killed the leaves of the branch, the branch itself and one
or more nearby branches. In the black mangroves the two
higher doses killed the leaves of the branch, but not the
branch or other nearby branches.
TABLE 7.
SINGLE LEAF APPLICATIONS:
6 WEEKS.
RESULTS AFTER
Active Ingredients
in Herbicide
Mangrove
Type
White
Red
Black
2,4-D
(kg/ ha)
11.2
5.6
1.12
11.2
5.6
1.12
11.2
5.6
1.12
Picloram
(kg/ha)
3.1
1.5
0.31
3.1
1.5
0.31
3.1
1.5
0.31
Leaves
D**
D
D
D
D
rolled
leaves
D
D
rolled
leaves
Effects
Branch
D
D
D
D
D
ne**
ne
ne
ne
Nearby
Lateral
Branches
D
D
D
D
D
ne
ne
ne
ne
*Active ingredients applied per ha of leaf area. The area
of each leaf was estimated from a series of leaf outlines
17
-------
in which areas had been determined. The herbicide was
applied with a microburet according to the leaf area.
**ne = no effect
D = dead
In two cases with 7.1 and 14.2 kg/ha doses to red man-
groves, single leaf applications of herbicide resulted in
swollen stem tips similar to patterns III and IV in Fig. 1.
The active ingredient concentrations in these cases were
5.58 kg/ha 2,4-D and 1.52 kg/ha picloram in the first case
and twice these doses in the second case.
The evidence for translocation of herbicide by adjacent
branch kill of red mangrove, together with the very limited
number of developable dormant buds, suggested that radio-
carbon-labelled herbicide might prove useful in following
local concentration points. Two experiments were carried
out using carbon-14 labelled picloram. The principal dis-
tribution of label in leaves, apical bud, stem and dormant
buds is shown in Table 8. Label was concentrated in the
stem and in the dormant buds. The dormant buds in these
plants were probably viable.
TABLE 8.
TRANSLOCATION OF 14C-LABELLED PICLORAM FROM
LEAF APPLICATION IN RED MANGROVES.
Tissue
Leaf
Leaf
Apical bud
Stem
Stem
Dormant bud
Dormant bud
Dormant bud
Expt. 1
cpm/mg
6.6
5.8
7.8
14
74
102
Radioactivity
Expt. 2
cpm/mg
4.2
6.8
1.7
18
9.9
14
50
94
18
-------
The labelled herbicide that could be rinsed from the
leaves was measured after 15 and 24 hr. In both cases,
approximately 75% of the label was recovered in the rinse,
indicating that absorption occurred within the first few
hours.
RADIOCHLORIDE UPTAKE
The possibility that the herbicides might kill red man-
groves by altering root membranes so that they were no
longer effective in excluding salt had been suggested by
Walsh (1974). The uptakes of radiochloride by red mangrove
seedlings at 0, 2.8 and 28 kg/ha Agent White are shown in
Table 9. The active ingredient concentrations were 2.24
kg/ha of 2,4-D and 0.61 kg/ha of picloram in the lower dose
and 10 times those concentrations in the higher dose.
Leaves on herbicide treated plants showed signs of stress
and appeared to be dying at 3-5 weeks. Leaves were har-
vested at 6 weeks. The results show that red mangroves
take up some chloride, but that the levels of herbicide
applied in these experiments, do not result in increased
chloride uptake.
TABLE 9. RADIOACTIVE CHLORIDE UPTAKE BY RED MAN-
GROVE SEEDLINGS TREATED WITH HERBICIDE.
Herbicide
dosage Radioactivity
kg/ha cpm/leaf pair
0 9,430
0 14,550
2.8 200
2.8 135
28 0
28 58
19
-------
CHLOROPHYLL CONTENT OF LEAVES
Agent White herbicide was applied to the water in which
approximately 2.5 year old red mangrove plants were growing
in mud in plastic buckets. Leaves were sampled at 15, 20
and 28 days by the use of a hand paper punch. Separate
samples from young and old leaves were collected on each
date. By the last sampling date some of the leaves of
plants in the highest doses were wilted. Leaves of plants
that showed only epinasty were sampled for chlorophyll at
20 and 28 days, and only green, non-wilted leaves were
used.
It is concluded that chlorophyll synthesis or degrada-
tion were not sensitive assays of herbicidal effect. Leaves
abscized green, yellow or brown, making if difficult to
rationalize a basic mechanism of action that involved chloro-
phyll. Results are given in Table 10.
TABLE 10.
CHLOROPHYLL CONTENT OF LEAVES OF RED
MANGROVE TREATED WITH 2,4-D AND PICLORAM.
Active ingredients
in herbicide (kg/ha)
2,4-D
0
11.2
picloram
0
3.1
22.4
6.1
Type
leaf
young
old
young
old
young
old
Total
chlorophyll
(average)
(mg/gm, fr. wt)
1.561
1.993
1.878
1.949
2.110
1.924
Notes
young leaves
showed some
effect at 20
and 28 days
young leaves
smaller and
curled at 20
and 28 days
20
-------
TRANSLOCATION OF SUCROSE
Translocation of C-14 sucrose from leaves of red man-
grove was studied. The plants were approximately 2.5 yrs
old and grew in mud in plastic buckets. The plants were
treated v/ith Agent White approximately 25 days previously.
Leaves used were green and turgid but had petioles that
showed epinastic bending. At the time of the translocation
experiment, some plants in the plastic buckets had withered
or abscised leaves. The counts, made 18 hrs after applica-
tion and shown in Table 11, represent sucrose translocation
from the leaf blades. It is concluded that inhibition of
translocation was not a basic herbicide toxicity mechanism.
TABLE 11. TRANSLOCATION OF SUCROSE IN RED MANGROVES
TREATED WITH 2,4-D AND PICLORAM.
Active Ingredients Total Radioactivity
in Herbicide (kg/ha) in Plant Sample (cmp) Notes
2,4-D picloram petiole stem
0 0 482 107
0 0 233 217
11.2 3.1 210 310
22.4 6.1 85 161 Untreated
leaves be-
gan to
wither and
die ca. 1
week after
experiment
TRANSPIRATION
Red mangrove plants, approximately 2.5.years old, were
matched in jars of 50% seawater. After treatment with Agent
White they were observed during the time that the leaves
began to show herbicidal effects and to wilt and/or fall.
The data (Table 12) fail to show any obvious effects of
herbicide except that transpiration rates decreased as
leaves withered and/or abscised.
21
-------
TABLE 12.
TRANSPIRATION BY RED MANGROVE TREATED
WITH 2,4-D AND PICLORAM.
Active Ingredients Transpiration Rate
in Herbicide (kg/ha) (average), ml/day Notes
2,4-D
0
0.22
2.24
11.2
22.4
picloram
0
0.06
0.61
3.1
6.1
3.24
4.32
3.60
3.12
1.80
-
-
lost some leaves
lost some leaves
ETHYLENE EVOLUTION FROM LEAVES
Ethylene evolution from Agent White herbicide-treated
and control red mangrove leaves was measured. Leaves were
taken 16 days after water treatment of 2.5 yrs old plants
grown in mud in plastic buckets. Petioles and terminal
stems exhibited epinasty at that time. That was also about
the time when leaves began to wilt, but the leaves that were
selected had not done so. Results are given in Table 13.
No effect of herbicides on ethylene production was found.
TABLE 13.
ETHYLENE EVOLVED FROM LEAVES OF RED MAN-
GROVE TREATED WITH 2,4-D AND PICLORAM.
Active Ingredients
in Herbicide (kg/ha)
2,4-D picloram
0 0
0 0
11.2 3.1
22.4 6.1
22.4 6.1
Ethylene Evolved
(^1/gm fresh wt/hr)
4.2
2.2
4.5
5.4
3.3
22
-------
SECTION VI
DISCUSSION
HERBICIDE TREATMENTS
The active herbicidal ingredients in Agent White are
water soluble and are readily taken up (absorbed) by a
variety of soils (Bovey et al., 1974; Lutz et al., 1973;
Altom and Stritzke, 1973). Thus, it is likely that the
herbicide applied to water over soil in the plastic con-
tainers in this study was effectively a soil application.
The persistences of 2,4-D and picloram in soil differ
considerably. In one study using three soils it was found
that the half-life of 2,4-D was 4 days, whereas that of
picloram was more than 100 days (Altom and Stritzke, 1973).
In other investigations it was found that soil temperature
and moisture, but not soil type or organic content, were
important factors in picloram disappearance from soil
(Merkle et al., 1973). Both active ingredients in Agent
White have been reported to be degraded in soils (Norris,
1970).
The mangroves in the present experiments were growing
under somewhat different conditions than typical plants in
their natural habitat. However, it should be noted that
mangroves may continue to grow in the absence of tidal flow.
Stoddart et al. (1973) reported on inland mangroves in
Barbuda and there are examples in the Miami area where the
three Florida species grow on dry land (Teas, 1974). The
plants used in the present experiments, although not sub-
jected to tides, appeared to be normal. Many of the larger
ones developed prop roots or pneumatophores after trans-
planting from the field into plastic pans and some of them
bloomed and set fruit that produced viable seed or propagules,
Tidal flushing would be expected to play a role in herb-
icide elution from soils. Blackman et al. (1974) carried
out studies of picloram persistence from soil applications
of Agent White in two tidally-washed mangrove forest areas
in South Vietnam, one of which was clear-cut and the other
relatively undisturbed. They found in the clear-cut plot
that a marked loss of picloram from surface soil (top 13 cm)
occured during the first day and in both cases there was a
90-99% reduction within about thirty days. They concluded
that this loss was partly due to penetration of the pic-
loram into the soil and partly due to tidal elution of
23
-------
herbicide,, It is therefore possible and likely that the
lethal, tolerance and no-effects concentrations for soil
application which are reported in the present study (Table 2),
in which no herbicide attenuation by deeper penetration into
the soil or by tidal elution were possible, are conservative.
Thus, under natural conditions of tidal washing, it is
likely that higher doses of herbicide by soil (water) ap-
plication would be required to produce the same effects as
found here using plants growing in shallow undrained con-
tainers.
TOXICITY FROM SOIL APPLICATIONS
The present experiments show that the black mangrove is
most resistant and the white mangrove least resistant to
soil applications of herbicide. The killing of mature red
mangroves by soil application required herbicide dosages
somewhat higher than the lethal dosage by canopy spraying
(approximately 13 kg/ha for lethal effect by soil appli-
cation to plants in plastic pans versus 6.3 kg/ha for spray
application to plants in the forest). The active ingre-
dients for soil applications were 10.2 kg/ha of 2,4-D and
2.8 kg/ha of picloram. The active ingredients for spray
application were 4.95 kg/ha of 2,4-D and 1.35 kg/ha of
picloram. Thus spray application is not likely to have
killed red mangroves by absorption of wash-through and
drift-through herbicide taken up from the soil. Soil ap-
plication resulted in only one case of a swollen stem tip,
a herbicide response from spray or single leaf application.
This suggests that translocation of Agent White to the sen-
sitive dormant buds of the plant is more effective from
leaf rather than from root application.
Blackman et. al (1974) found that Rhizophora apiculata
propagules were killed by the military dose levels of Agent
White (8.5 kg/ha) applied to the soil surface before planting.
Walsh et al. (1973), who treated Rhizophora mangle propagules
and young seedling plants in plastic pans with a 2,4-D 4-
picloram herbicide (of different formulation than Agent
White), obtained a complete kill from 11.2 kg/ha, but no
kill from 1.12 kg/ha. The active ingredient concentrations
were 4.4 kg/ha 2,4-D + 1.6 kg/ha picloram in the first case
and 0.44 kg/ha 2,4-D and 0.16 kg/ha picloram in the second
case. In the present investigation, 13 kg/ha of Agent White
killed young Rhizophora mangle seedlings. Blackman et al.
(1974) reported that 3-4 weeks after field applications of
Agent White at 8.5 kg/ha to soil which was subject to tidal
inundation the herbicide no longer inhibited the growth of
24
-------
Rhizophora propagules. This dose contained 6.72 kg/ha 2,4-D
and 1.82 kg/ha picloram as the active ingredients.
Calculations from the present experimental values for
the most sensitive stage of the most sensitive of the three
mangrove species, i.e. young seedlings of the white mangrove,
indicate that they are affected by, but recover from, doses
of less than 10 ppb picloram (For this comparison only the
picloram content of the Agent White is considered and a
water depth of 5 cm is assumed.). It can be noted for com-
parison that Flater et al. (1974) reported the early growth
of tomato (Lycopersicon esculentum), alfalfa (Medicago sativa),
alsike clover (Trifqlium hybridumT, and potato (Solanum
tuberosuro) were affected by picloram at 100 ppb, but not by
1 ppb. The white mangrove seedling is thus a sensitive
species and growth stage.
The red mangrove experiments with radioactive chloride
showed that herbicide toxicity in this species does .not
result from loss of chloride-exclusion ability of roots:
very little chloride was taken up by the herbicide-treated
plants.
SPRAY AND LEAF APPLICATIONS
The black mangrove is clearly the most resistant of the
Florida mangroves to spray applications of Agent White and
the white mangrove most sensitive.
Leaf fall from the red mangrove after herbicide spraying
did not follow the pattern of auxin-induced gradual yellowing
(senescence) and abscission that has been reported (Hallaway
and Osborne, 1969). The leaves that fell were a mixture of
green leaves, yellow leaves and brown leaves, which indicated
that mechanisms other than auxin-induced ethylene production
and ethylene-induced abscission were involved.
Red mangroves have poor regenerative powers compared to
black and white mangrove. This is evidenced by cases where
Florida mangrove species have been topped, as in the clearing
of survey lines, canalside trimming, or roadside mowing along
causeways. The observed poor regenerating ability has as
its basis the gradual loss of viability of older dormant
lateral buds in the red mangrove (Gill and Tomlinson, 1971).
It is well established that auxin-type herbicides trans-
locate and concentrate in growing points and buds (Weed
Science Society of America, 1970; Hamill et al., 1972).
25
-------
The finding in the red mangrove of many under-canopy trees,
each with numerous swollen dormant buds following spray
application, together with the concentration of carbon-14
labelled picloram in dormant buds, indicates that killing
results from herbicide translocation to and concentration
in the viable dormant buds. Here, the auxin-herbicide
either kills the buds outright or stimulates such tissue
proliferation that the dormant (suppressed) buds are over-
grown and cannot give rise to shoots and leaves. Gill and
Tomlinson (1971) reported that overgrowth by bark (without
stem swelling) is the normal fate of these buds within a
few years, which appears to be the anatomical basis for the
poor coppicing ability of the red mangrove. When the buds
of the red mangrove near the terminal leaf whorls have been
killed or inactivated, the tree has no buds that can give
rise to new leaves or shoots. This is undoubtedly the
reason for the observation that some mature red mangrove
trees, which had been completely defoliated and showed no
leaf regrowth, were still alive by the machete test after
several months. This survival of completely or partially
defoliated mature red mangrove trees for a period of months
indicates that the functioning of root membranes, trans-
piration and probably translocation are operational and
therefore that inhibition of these functions is not the
basis of lethality.
The pattern of mangrove kills in Vietnam provides
evidence on the nature of herbicidal effects. The tree-
kills along spray plane paths in the Saigon River delta
often were straight and sharp, even to individual trees.
They did not follow the irregular patterns that would be
expected if killing were the result of herbicide uptake
from the water (Teas, 1972). The South Vietnam mangrove
herbicide spray kill patterns thus support a direct herbi-
cidal action on tree canopies.
The Florida spray experiments also provide evidence
for the direct action of herbicide on tree canopies. The
spraying of the Marco Island Florida herbicide plots was
carried out from the minimum height practical above the
canopy and provided fairly precise spray patterns along
the edges of flight lines. Observations at 49 and 54 weeks
showed several cases where mature red mangrove trees near
the edges of plots were defoliated except for portions of
well-leafed branches that extended outside the plots. Thus,
defoliation is specific not only to the tree, but to the
individual branch. The existence of such trees further
argues against spray killing of red mangroves by root ab-
sorption of herbicide.
26
-------
ASSESSMENT OF HAZARD TO MANGROVES FROM AGRICULTURAL USE OF
AGENT WHITE
Hazard to mangroves from herbicidal treatment of agri-
cultural lands might occur from spray drift or from runoff
water that reached estuaries.
Experience from aerial applications in Vietnam provides
some evidence on spray drift. Spray plane missions were
typically flown at 45 m above the canopy (Lang, 1974) where-
as normal application under civilian conditions is usually
from a much lower altitude. As noted above, the patterns
of mangrove killing indicated that spray drift in Vietnam
was often minimal. It can be concluded that with appropriate
precautions only mangrove areas immediately adjacent to
herbicidal spraying would be expected to show damage from
the low volatility amine salt form of 2,4-D and picloram.
Herbicide concentration in agricultural runoff water
that reached mangroves in estuaries would be determined by
several factors, including: herbicide dosage applied,
area treated, soil binding and elution, rainfall, and tidal
and other dilution.
Recommended application rates for Tordon 101 (Agent
White) vary from 1.4-11.3 kg/ha (Dow Chemical Company).
Thus the initial herbicide dosages may differ by almost a
factor of 10 according to the species of plant being con-
trolled.
A number of workers have reported values for herbicide
concentrations in runoff from herbicide treated land. Davis
and Ingebo (1973) found concentrations of picloram as high
as 350-370 ppb in chaparral runoff water. Bovey et al. (1974)
reported that picloram concentrations were high (400-800 ppb)
if heavy rainfall occurred immediately after treatment but
low (< 5 ppb) if no major storms occurred until one month or
more after treatment. Norris (1969) treated small fractions
of watersheds and found that in some cases little or no 2,4-D
or picloram could be detected after even heavy rainstorms.
In addition, he noted that early light rains favored infil-
tration and soil binding of herbicide. Bovey et al. (1974)
reported no damage to cotton (Gossypium hirsutum) (a plant
that is very sensitive to picloram) in fields adjacent to
and below several herbicide treated watersheds, from either
spray drift or from runoff water.
Herbicide eluted from small treated areas by rainfall
apparently can be absorbed by untreated soil that it flows
over. Rainfall amount and timing are important in determining
27
-------
herbicide concentration in runoff water, since light rain
can bring about deeper soil penetration and therefore slower
subsequent elution. Apparently heavy rainfall is more ef-
fective in eluting herbicide from soil.
If agricultural runoff water were to flow over extensive
areas of untreated soil before reaching the mangroves, its
concentration of herbicide would be reduced. If, in addition
to herbicide loss by soil binding, the runoff water were
diluted with other waters before reaching the estuary, very
low concentrations of herbicide would be expected. Still
further reductions by soil binding within the estuary and
by tidal dilution would be expected.
If the herbicide applied to mangroves in plastic pans,
pools and buckets is calculated as concentration on a volume
basis, the tolerance doses of herbicide in water are: white
mangrove seedlings 20 ppb, for red seedlings 10,600 ppb,
and for black seedlings 10,600 ppb; for mature mangroves
the respective values are: 1,000, 10,000, and 106,000 ppb.
No effect doses for young seedlings are: 20 ppb for
all three species; for mature mangroves they are: 200,
1,000 and 5,400 ppb.
Data cited above show that picloram concentrations in
immediate plot runoff can vary from <5 ppb to 800 ppb as a
function of soil binding and rainfall elution. If high
dosages of Agent White were applied to large areas of
agricultural lands near mangroves, and there was heavy
rain soon after application, it is possible that no effect
or tolerance doses for one or more species of mangroves
would be exceeded. However, because of the attenuating
factors of soil absorption and dilution, and the typical
separation of mangroves and agricultural land, it is not
likely that such values would be found in estuaries. In
order to assure that there is no hazard to mangroves, special
attention should be given to herbicide applications near
estuaries.
28
-------
SECTION VII
REFERENCES
Altom, J.D. and J.F. Stritzke. 1973. Degradation of di-
camba, picloram, and four phenoxy herbicides in soils.
Weed Sci. 21:556-560.
Arnon, D.I. 1949. Copper enzymes in isolated chloroplasts.
Polyphenoloxidase in Beta vulgaris. Plant Physiol.
24:1-15.
Blackman, G.E., J.D. Fryer, A. Lang and M. Newton. 1974.
Effects of herbicides in South Vietnam. Part B. Per-
sistence and disappearance of herbicides in tropical
soils. National Academy of Sciences, National Research
Council, Washington, D.C. 59 pp.
Bovey, R.W., E. Burnett, C. Richardson, M.G. Merkle, J.R.
Baur and W.G. Knisel. 1974. Occurrence of 2, 4, 5-T
and picloram in surface runoff water in the blacklands
of Texas. J. Environ. Qual. 3:61-64.
Davis, E.A. and P.A. Ingebo. 1973. Picloram movement from
a chaparral watershed. Water Res. 9:1304-1313.
Eisinger, W.R. and D.J. Morre. 1971. Growth-regulating pro-
perties of picloram, 4-amino-3, 5, 6-trichloropicolinic
acid. Canad. J. Bot. 49:889-897.
Flater, R.L., W. Yarish and H. Vaartnou. 1974. Effects of
picloram on germination and development of six crop
species. Canad. PI. Sci. 54:219-221.
Gill, A.M0 and P.B. Tomlinson. 1971. Studies on the growth
of red mangrove (Rhizophora mangle L.) 3. Phenology of
the shoot. Biotropica 3:109-124.
Glooschenko, W. 1968. Statement before U.S. Senate Hearing
on Thermal Pollution, Part 2, Miami, Florida, pp. 751-760.
Hallaway, M. and D.J. Osborne. 1969. Ethylene: a factor
in defoliation induced by auxins. Science 163:1067-1068.
Hamill, A.S., L0W. Smith and C.M. Switzer. 1972. Influence
of phenoxy herbicides on picloram uptake and phytotoxicity,
Weed Sci. 20:226-229.
29
-------
Idyll, C.P., D.C. Tabb, and B. Yokel. 1968. The value of
estuaries to shrimp. Proc. Marsh and Estuary Management
Sympos., Louisiana State Univ. pp. 83-90.
Ketchum, B.H. 1972. The water's edge: critical problems
of the coastal zone.
pp. 1-393.
MIT Press, Cambridge, Massachusettes,
Lang, A. 1974. The effect of herbicides in South Vietnam.
Part A. Summary and conclusion: National Academy of
Sciences, Washington, D.C. 327 pp.
Lauf, G.H. (Ed.). 1967. Estuaries. AAAS Publication No.
83, Washington, D.C. 757 pp.
Lutz, J.F., G.E. Byers, and T.J. Sheets. 1973. The per-
sistence and movement of picloram and 2, 4, 5-T in soils,
J. Environ. Qual. 2:485-488.
MacKinney, G.
solutions.
1941. Absorption of light by chlorophyll
J. Biol. Chem. 140:315-322.
Merkle, R.W., C.R. Youngson, R.T. Hedlund, C.A.I. Goring,
J.W. Hamaker and W.W. Addington. 1973. Measurement
and prediction of picloram disappearance rates from
soil. Weed Sci. 21:549-555.
Mitchell, J.F. and G.R. Stephenson. 1973. The selective
action of picloram in red maple and white ash. Weed
Res. 13:169-178.
Norris, L.A. 1969. Herbicide runoff from forest lands
spriyed in the summer. Research progress report,
West. Soc. Weed Sci. 24-26.
Norris, L.A. 1970. Degradation of herbicides in the
forest floor. pp. 397-411. In Tree Growth and Forest
Soils. C.T. Youngberg and C.B. Davey (Eds.). Oregon
Univ. Press.
Odum, W.E. and E.J. Heald. 1972. Trophic analyses of an
estuarine mangrove community. Bull. Mar. Sci. 22:671-738.
Orians, G.H. and E.W. Pfeiffer. 1970. Ecological effects
of the war in Vietnam. Science 168:544-554.
Stoddart, D.R., G.W. Bryan and P.E. Gibbs. 1973. Inland
mangroves and water chemistry, Barbuda, West Indies. J.
Nat. Hist. 7:33-46.
30
-------
Teas, H.J. 1972. Personal observation and photographs in
South Vietnam.
Teas, H0J0 1974. Mangroves of Biscayne Bay„ Publ. by Bade
Co., Florida. 107 pp.
Teas, H.J. and J. Kelly. 1975. Effects of herbicides on
mangroves of South Vietnam and Florida. Int. Sympos. on
Biol. and Management of Mangroves. In press.
Thomas, W.L. 1956. Man's role in changing the face of the
earth. Univ. Chicago Press, Chicago. 1193 ppp.
Tschirley, F.H. 1969. Defoliation in Vietnam. Science
163:779-786.
Walsh, G.E0 1974. Mangroves: a review. pp. 51-174. In
R. Reimhold and W. Queen Eds.). Ecology of Halophytes.
Academic Press, N.Y. 605 pp.
Walsh, G.E., R. Barrett, G.H. Cook, and T.A. Hollister.
1973. Effects of herbicides on seedlings of the red
mangrove, Rhizophora mangle L. BioScience 23:361-364.
Weed Science Society of America. 1970. Herbicide Handbook.
Humphrey Press, Inc., Geneva, N.Y. 368 pp.
31
-------
SECTION VIII
ABBREVIATIONS
a — acre
cm — centimeters
gal— gallon
ha — hectare (2.47 acres)
kg — kilogram (2.2 pounds)
Ibs— pounds
ppb— parts per billion
ppm— parts per million
ppt— parts per thousand (seawater is approximately
35 ppt salinity)
32
-------
SECTION IX
GLOSSARY OF TERMS
Agent White — a formulation of 2,4-D and picloram, as
triisopropanolamine salts, used as a de-
foliant in Vietnam
coppicing ability — capability of a plant to regrow after
having been trimmed as a hedge
defoliant — a substance or formulation that causes plant
leaves to fall
herbicide — a substance or formulation that kills plants
lethal dose — minimum dosage for killing majority of plants
military application dose (for Agent White) — 8.5 kg/ha
(3 gal/a)
no effects dose — maximum dosage at which no lethality,
leaf symptoms, bud killing or other
abnormal development were noted
tolerance dose — maximum dosage from which 80% of plants
recover, or do not die
33
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-76-004
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Herbicide Toxicity in Mangroves
5. REPORT DATE
March 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Howard J. Teas
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Miami
Coral Gables, Florida 33124
10. PROGRAM ELEMENT NO.
1EA077
11. CONTRACT/GRANT NO.
R 801178
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
13. TYPE OF REPORT AND PERIOD COVERED
Final (10/72 - 10/74)
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT fhe amine salts of 2,4-D and picloram were applied to the Florida
species of mangroves: red mangrove (Rhizophora mangle), white mangrove (Laguncularla
racemosa) and black mangrove (Avicennia germinans). Treatments were to soil or water,
by aerial spray and to single leaves as droplets. The effects on radiochloride uptake
and on localization of radiocarbon-labelled picloram after leaf application were studi<
in red mangrove. "Lethal doses" for young seedlings were 2.7 kg/ha for white mangrove,
13 kg/ha for red and 13 kg/ha for black; for mature plants they were 2.7, 13 and > 53
kg/ha respectively. "Tolerance doses" for young seedlings were 0.01, 5.3 and 5.3
kg/ha; for mature plants they were 0.5, 5.3 and 53 kg/ha. "No effect doses" for
seedlings were < 0.01 kg/ha for all species; for mature plants they were < 0.1, 0.5
and 2.7 kg/ha. Spray applications of 6.3 - 12.2 kg/ha of commercial mixture to the
canopy of a mixed-species forest caused partial defoliation within three weeks.
Within 16 months it killed all of the white, 78 - 100% of the mature red, but none of
the mature black mangroves. Radiocarbon-labelled picloram concentrated in dormant
buds of red mangrove and it is concluded that the tree is killed by the mixture
because of its effects on them.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Defoliants, Herbicides, Spraying,
Chlorides, Transpiration, Ethylene,
Chlorophylls
Florida, Mangroves,
Agent White,
Translocation
6C
6F
13. DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
42
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
34
. S. GOVERNMENT PRINTING OFFICE: 1976-657-695/5382 Region No. 5-1 I
-------
>
z
rn
O
C
>
r
O
•o
\ l
IS 3
_£* C
^ *
-<
m
2
•D
r
O
<
rn
3)
PENAL
<
-n
0
35
•o
3)
<
>
H
m
C
0)
m
ifi
0)
o
o
0
~n
Tl
O
>
r
CD
C
0)
z
m
0)
0)
o
D
o
5"
3
0)
^*
O
3-
o'
j^.
Ol
rvj
O)
03
*i
— i
1
3
o'
Q)
-ti
O
i
03
>->>
o'
CO
I-+
Q)
— 1»
— 1>
O
-*»
-*i
o'
(D
0^
33
esearch ,
a>
a.
O
CD
<
(D_
X3
3
0)
^
^-^
C
05
m
z
<
33
O
S
m
z
>
i-
•o
33
O
m
0
5
z
>
Q
m
z
o
n
0)
T3
(D
O
W
-n
CD o
o c
o i
7T T
l
O
M
(A
P)
I-+
(D
C
(0
m
Z
<
x 3
o o
z (n
2 H
m >
Z g
_l m
>r i- z
CO Tl O
ft 3) -n
01 O ra
-I m
m 0)
3 !
i 3
O
m
Z
O
------- |