United States
Environmental Protection
Agency
Environmental Research
Laboratory
Gulf Breeze FL 32561
EPA-600/3-79-111
November 1979
Research and Development
Atrazine Fate and
Effects in a
Salt Marsh
•^
>
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Reseaich
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
PROPERTY OF THE
OFFICE OF SUPERFUND
ATRAZINE FATE AND EFFECTS IN A SALT MARSH
by
Donald E. Davis
J.D. Weete, C.G.P. Filial, F.G. Plumley
J.T. McEnerney, J.W. Everest, B. Truelove, A.M. Diner
Department of Botany & Microbiology
Alabama Agricultural Experiment Station, Auburn University
Auburn, Alabama 36830
Grant No. R803835
Project Officer
Frank G. Wilkes
Environmental Research Laboratory
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
GULF BREEZE, FLORIDA 32561
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
Gulf Breeze, Florida, 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 endorse-
ment or recommendation for use.
-------
FOREWORD
The protection of our estuarine and coastal areas from damage caused by
toxic organic pollutants requires that regulations restricting the introduc-
tion of these compounds into the environment be formulated on a sound scien-
tific basis. Accurate information describing dose-^response relationships
for organisms and ecosystems under varying conditions is required. The EPA
Environmental Research Laboratory, Gulf Breeze, contributes to this informa-
tion through research programs aimed at determining:
•the effects of toxic organic pollutants on individual species and
communities of organisms;
•the effects of toxic organics on ecosystem processes and components;
•the significance of chemical carcinogens in the estuarine and marine
environments.
Atrazine is the herbicide most widely used in the United States. This
report describes the fate and effects of this pesticide in salt-marsh eco-
systems, both in the field and in laboratory microecosystems. Data such as
these are useful in the U.S. Environmental Protection Agency in developing
strategies that will minimize the harmful impact of toxic substances on the
environment.
Henry FY Enos
Director
Environmental Research Laboratory
Gulf Breeze, FL 32561
m
-------
ACKNOWLEDGMENTS
We thank the Environmental Protection Agecny for financial support
and the project officer, Dr. Frank G. Wilkes, for guidance and understanding
of the innumerable problems that invariably arise in an investigation of
this magnitude. We also acknowledge the gift of l^C-ring labeled atrazine
and N-dealkylated derivates of atrazine by the CIBA-GEIGY Corporation.
Some of the atrazine metabolites were identified by Dr. John L. Laseter of
the Center for Bioorganic Studies, University of New Orleans, and some of
the amino acids were identified by Dr. Paul Melius of the Auburn University
Chemistry Department. Ms. Ingrid Kircher and Mr. Craig Weatherby provided
technical assistance in the crab studies; and Drs. Ruth Patrick and Francis
R. Trainor provided technical advice and training that made the study with
the diatoms possible.
iv
-------
CONTENTS
Foreword iii
Acknowledgments iy
Abstract vii
Figures ix
Tables xi
1. Introduction 1
2. Conclusions and Recommendations 6
3. Atrazine metabolism by Spartina alterniflora:
Chloroform-soluble Metabolites 8
Objectives 8
Materials and Methods , 8
Results and Discussion 9
Summary 15
4. Atrazine Metabolism by Spartina alterniflora:
Water-soluble Metabolites 16
Objectives 16
Materials and Methods 16
Results and Discussion 19
Summary 26
5. Effect of Atrazine on Uoa pugnax 28
Objectives 28
Materials and Methods 28
Results and Discussion 31
Summary 36
6. Metabolism of Atrazine in the Spartina alterniflora-
detritus-[/ca pugnax Food Chain 37
Objectives 37
Materials and Methods 37
Results and Discussion 40
Summary 43
7. Metabolism of Atrazine by Sesarma oinereum 45
Objectives 45
Materials and Methods 45
Results and Discussion 46
Summary 51
8. Effect of Atrazine on Marine Diatoms 53
Objectives 53
Materials and Methods 53
Results 56
Discussion 56
Summary 62
-------
9. Atrazine Residues in Salt Marsh Ecosystem Components:
A Comparison of Field and Microecosystem Results. ... 63
Objectives 63
Materials and Methods 63
Results and Discussion 65
Summary 69
10. Mi croecosys terns 71
Literature Review 71
Microecosystem s Used 72
Summary 77
References 78
-------
ABSTRACT
A series of experiments were conducted to determine the effect of
atrazine on various biological components of a salt marsh ecosystem such
as exists along the Georgia coast and the associated offshore islands and
the fate of atrazine when introduced into this ecosystem. Investigative
procedures involved supplying either formulated atrazine, AAtrex , or
'^C-labeled technical atrazine to individual organisms in the laboratory,
to groups of organisms in microecosystems of various levels of complexity,
and to the salt marsh itself.
Spavtina alterniflora Loisel was grown for 2 days with its roots in
14C-labeled atrazine solution (2.6 \i M) followed by 28 days in atrazine-free
solution. S. altemiflora was moderately resistant to this herbicide.
Radioactivity in the chloroform fraction of 80% methanol extract was
ca 82, 38, and 24% after 2, 10, and 30 days, respectively. In the aqueous
fraction the values were ca 15, 57, and 60%, respectively. The chloroform
fraction contained atrazine and three N-dealkylation products. About half
of the 14 water-soluble metabolites detected contained fully N-alkylated
triazine rings while almost all of the others contained the 4-amino-6-
isopropylamino derivative. 2-Hydroxyatrazine and 2-hydroxy-4-amino-6-
isopropylamino-s-triazine were identified as water-soluble metabolites of
atrazine.
Box crabs (Sescama c-inerewn} were fed for 10 days with leaves from
S. alterniflora grown for 2 days in 14c_]abeled atrazine solution followed
by 3 days in atrazine-free solution with no significant effect on crab
survival or behavior. At the end of the 10-day feeding period, only 1.2%
and 0.5% of the total radioactivity in the crabs and their feces, respec-
tively, was atrazine compared to 24% in the S. alterniflora used as a food
source.
Leaves from S. altern-iflora plants grown in solutions containing ^C
ring-labeled atrazine was converted to detritus. During the 20-day con-
version to detritus, the chloroform fraction of the 80% methanol extract
decreased from 55 to 9%. Uca pugnax fed this detritus or detritus wetted
with l^c-labeled atrazine decreased the percent radioactivity in the
chloroform fraction. Thus, atrazine concentration decreased as leaves were
converted to detritus and again when the detritus was consumed by u. pugnax.
A 10-5 M (2.2 ppm) atrazine concentration significantly reduced photo-
synthesis rate, chlorophyll content, and cell numbers in unialgal cultures
of Nitzschia sigma Grun. and Thalassiosim fluviatilis Hustedt isolated from
a salt marsh habitat. Atrazine effects were less in the field than in
vi i
-------
microecosysterns or cultures. Diatom species diversity was not affected
by 10-5 M atrazine in microecosystems or in the field, but the number of
Cymatosire belgiea was increased in both situations.
Atrazine was sprayed at 0.0, 0.05, 0.5, and 5.0 g/m2 on plots on Sapelo
Island and on microecosystems. Residue determinations were made 3 months
later. Atrazine concentration in the 0 to 1-, 1- to 10-, and 10- to 25-cm
layers of soil from plots on Sapelo treated with 5 g/m^ averaged 1.20, 0.77,
and 0.25 ppm, respectively, and together accounted for 3% of that applied.
Atrazine concentrations in s. alterniflora from plots on Sapelo treated with
5.0 g/m2 were 21.6 and 12.8 ppm for 5. alterniflora < 0.5 m and > 0.5 m tall,
respectively, and for similarly treated microecosystems were 16.8 and 21.1
ppm, respectively. Periwinkle snails, horse mussels, and fiddler crabs from
microecosystems receiving 5.0 g/m2 contained 7.8, 3.5, and 0.31 ppm atrazine,
respectively. Less atrazine was found in animals from Sapelo. Lower rates
of atrazine application gave lower residue levels, often too low to be
detectable.
This report was submitted in fulfillment of Grant No. R803835 by the
Auburn University Agricultural Experiment Station under the sponsorship of
the U.S. Environmental Protection Agency. This report covers the period of
July 1, 1975, to March 1, 1979.
vm
-------
FIGURES
Number Page
1. Diagram of energy flow in a Georgia Spartina altemiflora salt
marsh ecosystem 2
2. Distribution of methanol extractable radioactivity in roots and
shoots of Spartina altemiflora treated for 2 days with [14r]-
atrazine and then transferred to an atrazine-free nutrient
solution for 28 days 11
3. Change to radioactivity with time; in water, chloroform, and
insoluble fractions of an 80% methanol extract of shoots of
Spartina alterniflora plants grown in [14,0-atrazine for 2 days
and then transferred to an atrazine-free nutrient solution . . 11
4. Autoradiograph of a TLC plate showing radioactive components
present in the chloroform fraction of 80% methanol extract of
Spartina alterniflora treated with [14~]-atrazine 2 days and
;nen transferred to a herbicide-free nutrient solution; also,
(A) radiochromatogram of the chloroform fraction, (B) diagram
of the autoradiograph shown on the left, and (C) R^ values
of standards of atrazine and its metabolites. . . ". 13
5. Change in radioactivity with time in atrazine extracted from
shoots and roots of Spartina alterniflora treated for 2 days
with [14p]-atrazine and subsequently transferred to an atrazine-
free nutrient solution for 28 days 14
6. Outline of the kind of procedures used in the isolation and
purification of water-soluble metabolites of atrazine of
Spartina alterniflora 18
7. Change in radioactivity with time in the aqueous, chloroform, and
insoluble fractions of an 80% methanol extract of shoots of
Spartina altemiflora plants grown in [14r]-atrazine for 2 days
ana then transferred to an atrazine-free nutrient solution.
(Repeat of Figure 3) 19
8, Radioactive fractions of the aqueous fraction of the S. alterniflora
extract passed through a cation-exchange (50W-X2 Aminex resin,
200-400 mesh) column. Samples were taken 0, 3, 18, and 28 days
after a 2-day exposure to [14p]-atrazine 21
ix
-------
Number Page
9. Separation of components in each fraction from the cation-exchange
column (AG-lx2, 200-400 mesh) 23
10. TLC of fraction 6A from the anion-exchange column 25
11. Diagrammatic representation of experimental setup used to
make detritus 38
12. Autoradiogram showing the '^C separation of radioactive components
of the chloroform extracts from (A) S. aJU-zHYii^toMi leaves,
(B) crabs, and (C) crab feces 50
13. Radioactivity in subfractions of the aqueous fractions of the
leaves of 5. alteA^lo^a. grown in ^C-labeled atrazine solution,
in crabs fed leaves from these plants, and in feces of these
crabs 50
14. Effect of 7 days of exposure to various atrazine concentrations
on the chlorophyll content, cell number, and rate of photo-
synthesis of cultures of TkaJLaAi>jj> and N-£t6du&
-6-ujma 57
15. Percentage dominance for each of the five most common diatom
species in atrazine-treated and control areas in microecosystems
and in the field 58
16. Arrangement of water reservoirs on the tide machine support
frame 74
17. Arrangement of mechanical drive and electrical timing system of
the tide machine 75
18. Schematic diagram of the electrical system of the tide machine . 75
-------
TABLES
Number Page
1. Atrazine effects investigated, where investigated, and where
reported 4
2. Effect of atrazine on the growth of Spartina alterniflora 45 days
after treatment 10
3. Radioactivity in N-dealkylated products of atrazine metabolism
isolated from shoots of Spartina alterniflora collected at
intervals over a 28-day period 15
4. Summary of results of the isolation of some water-soluble
metabolites of atrazine from s. alterniflora shoots 22
5. Chemical nature of the s-triazine ring in the individual radio-
active components from the aqueous fraction of the 80% methanol
extract of S. alterniflora grown in ^C-labeled atrazine
solution 24
6. Effects of atrazine on four classes of fiddler crabs in systems
with and without marsh soil, Aug. 6, 1977,test 33
7. Effect of atrazine on small male crabs in systems without soil,
Nov. 7, 1977, test 34
8. Effect of atrazine on small male crabs in systems without soil,
March 29, 1978, test 34
9. Effect of atrazine on small male crabs in systems without soil,
Aug. 1, 1978, test 35
10. Amounts of radioactivity in the chloroform-soluble, water soluble,
and insoluble fractions in physical components of the system. . 41
11. Amounts and percentages of total radioactivity in the various
fractions extracted from radio!abeled detritus residue fed to
fiddler crabs and in the fiddler crab themselves 42
-------
Number
12. Amounts and percentages of the total radioactivity in the various
fractions extracted from detritus residues wetted with radio-
labeled atrazine and fed to fiddler crabs, and in the fiddler
crabs themselves 42
13. The amounts and concentrations of water-soluble and chloroform-
soluble atrazine or atrazine metabolites in s. alterniflora
leaves, in detritus derived from the leaves, and in fiddler
crabs fed detritus derived from the leaves 43
14. Radioactivity in chloroform, aqueous, and insoluble fractions of
S. alterniflora, crabs, and feces 48
15. Radioactivity in atrazine, N-dealkylation products, and unidenti-
fied polar components of the chloroform fraction of extracts
from S. alterniflora leaves 49
16. Effects of atrazine on photosynthesis, cell numbers, and
chlorophyll content of diatoms in microecosysterns 58
17. Diatom species diversity (Habits/individual) and number of
species (S) for atrazine-treated and untreated microecosystems
and in the field 59
18. Average concentrations of atrazine in various soil layers in
the field and in microecosystems 10 weeks after herbicide
application 66
19. Average concentrations of atrazine in two-sized classes of
S. alterniflora (<0.5 m and > 0.5 m tall) in the field
and in microecosystems 10 weeks after herbicide application . . 67
20. Average concentrations of atrazine in periwinkle snails,
horse mussels, and fiddler crabs in the field and in
microecosystems 10 weeks after herbicide application 68
21. Average amounts of atrazine collected in water reservoirs
attached to microecosystems 69
XI1
-------
SECTION I
INTRODUCTION
Over the past few years a growing concern has developed regarding the
pollution and misuse of our coastal salt marshes. These areas are very im-
portant because they provide substantial amounts of food substances and
protection for the many organisms that use them as breeding grounds and
nurseries (Odum, 1971 a). A variety of energy flow models have been prepared
for the Spartina alterniflora marshes of Georgia which were the site of these
investigations (Odum,1961 and 1971; Teal, 1962). Figure 1 is similar to
these proposed models, but it identifies all of the species used in this
study. Spartina alterniflora is the dominant macrophyte on the 192,508 ha
of Georgia coastal salt marshes (Reimold,1977). Net annual primary produc-
tion of S. alterniflora is as high as 3800 g/m2 (Gallagher et al., 1979)
with approximately 90% of the fixed carbon entering the detritus food web
(Smalley, 1959). Teal (1962) estimated that only 7% of the S. alterniflora
productivity was consumed directly by herbivores.
The soils in the S. alterniflora marsh are high in organic matter and
may be water-saturated for extended periods of time (Cotnoir, 1974); sand,
silt, and clay constitute the bulk of the inorganic matter with the percent-
age of each varying with the position relative to tidal creeks. The pH of
the aerobic surface sediments is 7-8 while the anaerobic layers (beneath
approximately 2 mm) have pH values as low as 5. The high rate of S. alterni-
flora production is due to lack of competition and adequate nutrient supply
from the sediments and flooding tides; only nitrogen limits the growth of
S. alterniflova in the Georgia salt marshes (Gallagher, 1975). A relatively
high rate of mineralization and/or oxidation causes a fast turnover of
nutrients (Christian et al., 1978) and some nitrogen is added by nitrogen
fixation in the surface sediments (Hanson, 1977).
The second major group of primary producers in these marshes is the
surface-inhabiting algae, primarily diatoms, blue greens, and euglenoids
(Williams, 1962). The importance of the edaphic algae has been variously
estimated (Eaton and Moss, 1966; Gallagher, 1971; Round, 1971; Sullivan and
Daiber, 1975; Van Raalte et al., 1974; Williams, 1962). Although the data
are incomplete, it now appears that the algae supply a major portion of the
utilizable fixed carbon to the estuarine ecosystem (Haines, 1977). The
biomass, productivity, and species composition of these algae vary widely
between the different zones in the marsh.
Within the marsh, there exist at least four readily distinguishable
zones (Teal, 1962) which appear to be delineated by their degree of tidal
-------
PRIMARY PRODUCER
Sportina oltemiflora
PRIMARY PRODUCER
diatoms, blue green
algae, euglenoids
PRIMARY CONSUMER
bo« crabs, grasshoppers,
other insects
DETRITUS FEEDERS
box crabs, fiddler
crabs
PRIMARY CONSUMER
box crabs, fiddler crabs,
periwinkle snails
Figure 1. Diagram of energy flow in a Georgia Spartina alterniflora salt
marsh ecosystem.
exposure. Zone 1, the creek bank zone, with its muddy and/or sandy banks,
is devoid of S. alterniflora but it is here that the unshaded algae reach
their maximum productivity of 245 mg C/m2/h (Darley et al., 1976). Grazing
by the mud snail, Nassarius obsoletus, significantly reduces algal produc-
tivity in this area (Shimmel, 1979). Immediately adjacent to the creek
bank is Zone 2, the 1-3 m wide streamside marsh where S. alterniflora
reaches its maximum height (> 2m) and productivity (3800 g C/m^/yr). The
shading by S. alterniflora reduces algal productivity to approximately 75%
of that in Zone 1. The largest individuals of the fiddler crab (Uoa spp.)
population occupy Zones 1 and 2 and graze heavily on the algae and detritus
along the creek bank during low tide. Zone 3 consists of the naturally
occurring levees bordering the creeks and is occupied by intermediate-height
S. alterniflora. Due to their higher elevation, the levees are only covered
by spring tides and their exposed surface sediments are often dry and
cracked. Zone 4, the short S. alterniflora marsh, lies directly behind the
levees and consists of large flat areas which occupy 40-60% of the total
S. alterniflora marsh. TKe productivity of S. alterniflora in Zone 4 is
relatively low (1300 g C/m2/yr) (Gallagher et al., 1979). Horse mussels
(Geukensia demissa} and snails (Littorina irrorata] reach their maximum
population density in this area and are rare in the others.
-------
All field work for this project was conducted on Cabretta Island, a
small strip of land separated from Sapelo Island (31°N Latitude) by Big Hole
Creek and Cabretta Creek. The specific zones used for field studies were
approximately 50 m WSW of the bridge connecting Cabretta Island with Sapelo
Island. Algal studies were conducted in Zone 1 on Cabretta Creek and s.
alterniflora studies were in Zone 3. Specimens used in laboratory studies
were collected from areas adjacent to those used for field studies.
Many pollutants pose potential threats to these salt marshes. Among
those of particular concern are oil from oil spills, industrial and domestic
sewage, and pesticides. This particular investigation deals with the effect
of a very widely used pesticide, atrazine, on various components of the salt
marsh ecosystem. Atrazine was selected for use in this study because twice
as much atrazine is used as any other single pesticide in the United States.
The average amount used yearly is approximately 40 million kg. As much as
2 to 3% of the atrazine may be removed from the soil surface by runoff water
(Correll et al., 1978). Atrazine is one member of a very important family
of herbicides, the s-triazines. It is used at rates of 2 to 4 kg/ha for
selective control of broadleaf and some grassy weeds in corn, sugarcane, and
sorghum (WSSA Herbicide Handbook Committee, 1979). It is usually applied
by ground equipment to the surface of the soil after crop planting and prior
to weed emergence. It may sometimes be lightly incorporated into the soil
surface and has on occasions been applied from the air (WSSA Herbicide Hand-
book Committee, 1979).
Atrazine is classified as a photosynthesis inhibitor. More specifically,
it inhibits the Hill reaction at the step where a H20 molecule is split to
form 02 and electrons (Esser et al., 1975). Evidence also suggests that
atrazine may affect protein synthesis (Esser et al., 1975). Atrazine is
toxic to essentially all broad-leaved plants although some deep-rooted peren-
nial species are unaffected by surface applications because relatively little
of the applied herbicide is taken up by their roots (Gunther and Gunther,
1970). (Narrow-leaved annual species such as corn and grain sorghum are able
to rapidly convert absorbed atrazine into nontoxic metabolites [Gunther and
Gunther, 1970].)
The reported effects of atrazine on animals vary. Rats administered
large oral doses of atrazine were relatively unaffected (LD5Q 3,080 mg/kg)
whereas rainbow trout were sensitive to atrazine concentrations of a few
ppm (LCso 12.0 ppm) (Pimentell, 1971). Common invertebrates of the salt
marsh such as clams, water bugs, mayfly nymphs, common midges, mosquitoes,
biting midges, caddice fly larvae, aquatic worms, and brown shrimp have
been reported to be reduced in numbers by as much as 50% after applications
of atrazine to the soil in concentrations ranging from 0.5 to 2 ppm (Balinke
and Bilodub-Pantera, 1964; Pimentell, 1971). However, other invertebrates
such as damselfly nymphs and water beetles have doubled their numbers after
applications of 1 ppm atrazine (Pimentell, 1971). Since these studies were
conducted in the field, it is not known whether atrazine was directly in-
volved or whether the changes in numbers were due to atrazine-induced
changes in some of the food sources.
-------
Although considerable information has been gathered about the toxicity
and metabolism of atrazine by important crop and some weed species and about
its fate in well-drained agricultural soils (Gunther and Gunther, 1970),
very little is known about its toxicity to the components of the salt marsh
ecosystem or its ultimate fate in the salt marsh.
A series of experiments were performed to expand our knowledge about
the fate and effects of atrazine in a salt marsh ecosystem. The major com-
ponents of this ecosystem are shown in Table 1. As indicated, experiments
dealing with the toxicity of atrazine to S. altevniflora and the metabolism
of atrazine to chloroform-soluble and water-soluble metabolites are reported
in Sections 3 and 4 of this paper. Section 5 covers studies of the toxicity
of atrazine to the fiddler crab, Uoa pugnax ^ar\d Section 6 the metabolism of
atrazine as it passes down the S. alterniflora-detritus-Uca pugnax food chain.
Section 7 reports on studies dealing with the toxicity of atrazine to the
TABLE 1. ATRAZINE EFFECTS INVESTIGATED, WHERE INVESTIGATED, AND WHERE
REPORTED
Atrazine effects
Where
Investigated
Reported
Atrazine toxicity to:
Smooth cordgrass, Spartina alterniflova
Box crab, Sesarma cinereum
Fiddler crab, Uoa pugnax
Marine diatoms, Thallasiosira fluviatilis,
Nitzsehia sigma, Cymatosi-Ta belgiaa,
Melosira sp.3 Nawicula sp.
Atrazine metabolism by:
Smooth cordgrass, Spartina alterniflora
Box crab, Sesavma oi-nevewn
Fiddler crab, Uaa pugnax
Detritivores
Atrazine residues in:
Smooth cordgrass-detritus-fiddler crab
food chain
Smooth cordgrass; snails, Littorina
irrorata; horse mussels, Geukensia
dem-issa; Soil
Laboratory
Lab. & field
Lab. & field
Lab. & field
Sec. 3
Sec. 7
Sec. 5
Sec. 8
Laboratory
Laboratory
Laboratory
Laboratory
Laboratory
Field &
microecosysterns
Sec. 3&4
Sec. 7
Sec. 6
Sec. 6
Sec. 6
Sec. 9
Effectiveness of microecosysterns
Laboratory
Sec. 10
-------
box crab, Sesarma cinereum and metabolism of atrazine
Section 8 deals with a series of investigations of the
to marine diatoms in monocultures, in microecosystems,
includes an evaluation of the effect of this herbicide
The amount of atrazine remaining in various components
after a single application of three different rates of
field and to microecosystems in the subject of Section
marizes some of the advantages and limitations of the microecosystems used
in various aspects of these investigations.
by this animal.
toxicity of atrazine
and in the field and
on species diversity,
of the ecosystem
atrazine in the
9. Section 10 sum-
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
These studies on the fate and effects of atrazine on the salt marsh
have largely dissipated fears of adverse effects on the marsh by atrazine
in runoff water from herbicide-treated fields. Concentrations of atrazine
present in runoff waters entering estuaries seldom reach parts per billion
(ppb). Based on the results of this study, this concentration is far below
the concentration needed to adversely affect the conversion of 5. alterni-
flora leaves to detritus or to be toxic to box crabs, fiddler crabs, or
S. alterniflora. The least effect level for diatoms, the most sensitive
organisms tested, is approximately 5xlO~?M (ca 100 ppb). This is about
50-fold greater than the maximum atrazine concentration expected in runoff
water, but much of the atrazine removed from the application site is adsorbed
on soil particles. It is conceivable that when such particles are deposited
in the marsh that they might adversely affect the growth of diatoms on the
soil surface.
Because atrazine was found to be readily metabolized by S. alterniflora,
box crabs, fiddler crabs, and detritivores, there seems to be little probabil-
ity that atrazine introduced in the marsh would remain as a problem for more
than a very few months unless present in concentrations far in excess of
those used for weed control in agricultural crops. This conclusion is sub-
stantiated by the finding, that 10 weeks after atrazine was applied to the
marsh at rates over 10 times those used for weed control, less than 5% of
that applied remained in the marsh. Nearly half of the atrazine that was
applied was removed in the tidal water. There was no evidence of bioaccumu-
lation of atrazine by any species tested.
Microecosystems worked well in giving qualitative answers to questions
about atrazine toxicity or metabolism. However, the amounts of atrazine
residues remaining in various components in the microecosystems were some-
times significantly different from the residues in those same components
in the field. Apparently, the major problem of developing a microecosystem
that accurately predicts behavior in the field is the difficulty of getting
a flux of water onto and off of the microecosystem comparable to that in
the marsh. The relatively small areas involved have different flow patterns
than occur in a large marsh.
Additional work is needed to develop microecosystems capable of more
accurately predicting the amount of a pesticide remaining in various eco-
system components in the field. Research is also needed to determine
whether atrazine, in the amounts found adsorbed on soil particles in runoff
-------
SECTION 3
ATRAZINE METABOLISM BY SPARTINA ALTERNIFLORA:
CHLOROFORM SOLUBLE METABOLITES^
OBJECTIVES
Spartina alt erni flora, a marsh grass commonly known as smooth cordgrass,
is the primary autotroph in the salt marsh under investigation. The objec-
tives of the research presented in this section were: 1) determine the
tolerance of S. alterniflora to atrazine; 2) describe the metabolism of this
herbicide to chloroform-soluble forms, and 3) identify the metabolites of
atrazine produced by s. alterniflora that may be ingested by consumers of
this ecosystem component.
MATERIALS AND METHODS
Ring-labeled p4c]_atrazine (24.9 yCi/mg) and the following atrazine
metabolites were obtained from the Agricultural Division of Ciba Geigy Cor-
poration, Greensboro, N.C.: 2-chloro-4-amino-6-ethylamino-s-triazine,
2-chloro-4-amino-6-isopropylamino-s-triazine, and 2-chloro-4,6,-diamino-s-
triazine.
Plants were collected from the marsh at Sapelo Island, Ga. They were
maintained at Auburn University in 2-liter (2-L) beakers containing Hoagland's
solution placed in a growth chamber having a 14-h photoperiod with 60% RH,
a temperature of 28 C, and 30 klux of light provided by a mixture of incan-
descent and fluorescent lamps. The photoperiods were followed by 10-h dark
periods at 60% RH and 24 C.
In a preliminary experiment to determine the tolerance of s.
flora to atrazine, plants were divided into five groups of 16 plants each.
The plants were weighed and the number of leaves per clump of plants and
plant heights recorded. Plants were then transferred to 1-L plastic beakers
containing 900 ml of Hoagland's solution with 0, 5 x 10~8, 5 x 10-7,
5 x 10-6, or 5 x 10-5 M atrazine. Each beaker contained four plants, and
each treatment was replicated four times.
In time course experiments, a uniform lot of vigorously growing plants
was divided into 10 groups of four plants each. Each group of plants was
Most of the material in this section is reprinted from the Journal of
Agricultural and Food Chemistry 25:852-855. Copyright 1977, American
Chemical Society. Used by permission.
-------
placed in 300 ml of Hoagland's solution containing 2.0 yCi of [14C>atrazine
(2.6 yM). After 2 days, the plants were removed from the [14c]-atrazine
solutions, the roots were rinsed, and all but two groups were placed in
atrazine-free Hoagland's solution. The two groups not transferred to the
atrazine-free solution and two additional groups collected at 3, 8, 18, and
28 days after transferring them to the atrazine-free nutrient solution were
extracted as described below.
The extraction methods used in this study were essentially those des-
cribed by Shimabukuro et al. (1973). Roots and shoots were separated and
extracted with 10 ml of 80% methanol for each gram of tissue. The extracts
were concentrated by flash evaporation at 37 C, diluted with water, and then
washed with chloroform. Each phase was brought to volume, and the amount of
radioactivity in each was determined by liquid scintillation spectrometry
(Beckman LS-200B). Radioactivity in the insoluble plant residue was estimated
grinding a portion of this material to a fine powder, suspending it in
Aquasol (Beckman) liquid scintillation cocktail containing Cab-0-Sil
(Beckman), and counting as before.
Radio!abeled components of the chloroform fractions were separated by
thin-layer chromatography (TLC),using glass plates coated with a 250-ym layer
of silica gel HF-254 and activated for 1 h at 110 C. Plates were developed
initially in benzene-acetic acid-water (60:40:3,v/v/v), and radioactivity
was located using a Berthold TLC-Scanner. Silica gel from radioactive areas
was removed from the plates, washed with methanol, and removed from the
solvent by centrifugation. Atrazine and its metabolites in these extracts
were again spotted on TLC plates and developed in chloroform-ethanol (90:10,
v/v) and identified by comparing their R^ values with those of authentic
standards. Standards of atrazine and its N-dealkylation products were
visualized on the TLC plates with ultraviolet light (254 nm). Radioactivity
in each component was determined by liquid scintillation spectrometry after
removing the compound from silica gel as before.
RESULTS AND DISCUSSION
Tolerance of_S. altevniflora to Atrazine
For S. alterniflora to be a successful component of the model ecosystem
for long-term studies, it must be at least moderately resistant to atrazine.
To determine the effects of atrazine on this species, plants were grown for
45 days in the different atrazine solutions and fresh and dry weights of
roots and shoots, plant heights, and number of leaves were determined. There
were no obvious symptoms of atrazine toxicity such as chlorosis, necrosis,
or wilting. However, with the exception of plant height, significant de-
creases in all growth parameters measured were obtained with atrazine con-
centrations of 5 x 10-5 and 5 x 10~6 M when compared with the control
(Table 2). There was an average of 37% reduction of the growth parameters
measured at the most concentrated atrazine solution used (5 x 10-5 M).
Root and shoot fresh weights and root dry weights were significantly less
than the control at all atrazine concentrations down to 5 x 10-7 M. Only
root dry weight was significantly less than the control at 5 x 10~8 M concen-
tration.
-------
TABLE 2. EFFECT OF ATRAZINE ON GROWTH OF SPAHTINA ALTERNIFLORA 45 DAYS
AFTER TREATMENT*
Atrazine
concn
M
Control
5 x 10-5
5 x 10-6
5 x 10-7
5 x 10-8
Fresh wt.
of roots
g
19.7 a
7.5 a
7.9 b
9.4 b
13.7 ab
Fresh wt.
of shoots
9
27.3 a
7.9 c
13.8 be
16.6 b
24.1 a
No. of
leaves
15.6 a
5.1 c
9.9 be
11.2 ab
15.0 a
Total
height
cm
68.3 a
45.9 b
61.1 a
63.1 a
69.7 a
Dry wt.
of roots
g
3.3 a
0.7 c
1.1 be
1.4 be
2.0 b
Dry wt.
of shoots
g
7.2 a
2.6 e
4.6 be
5.5 ab
7.9 ab
*
Each value is the average of four replications with four plants in
each replication. Values in a column followed by the same letters are not
significantly different at the 5% level using Duncan's multiple range test.
These data are similar to those obtained for resistant species. Corn
is considered a resistant plant, but its degradation system can be over-
loaded. Couch and Davis (1966) reported a 50% decrease in fresh and dry
weights when corn was grown in a 5 x 1Q-5 M atrazine solution, and ca
5 x 10"5 atrazine decreased photosynthesis about 60%. A correlation between
net C02 exchange (NCE) and atrazine resistance was reported for certain
grasses (Jensen et al., 1977). Grasses with NCE recovery rates exceeding
1.2 mg of C02 dm"2 h"l were considered tolerant to 1.0 kg/ha preemergence
and 1.25 kg/ha postemergence atrazine applications. When representatives
of the subfamilies Festucoideae, Eragrostoideae, and Panicoideae were
screened; only some members of the latter subfamily had NCE recovery rates
exceeding 1.2 mg dm"2 h~l. Although the NCE recovery rates for s. alterni-
flora, a member of the Festucoideae, are not known; our data suggest that
this species is resistant to atrazine.
Uptake and Translocation of [14c]-Atrazine
[14C]-Atrazine was readily adsorbed and translocated by S. alterniflora.
After 2 days of continuous exposure to the radiolabeled herbicide, approxi-
mately 90% of absorbed atrazine was present in the shoots. Methanol extract-
able radioactivity in the roots and shoots remained relatively constant
throughout the 28-day period in the atrazine-free nutrient solution (Figure
2). This is consistent with the concept that absorption and translocation
limitations are not the primary factors that determine susceptibility to
atrazine (Davis et al., 1959).
Extracts of s. alterniflora roots and shoots were separated into
chloroform, aqueous, and insoluble fractions, and the radioactivity in each
was determined. Radioactivity in the insoluble fraction represented atrazine
10
-------
80
> 60
O O RECOVERY
D D SHOOTS
A A ROOTS
WATER-SOLUBLE FRACTION
CHLOROFORM-SOLUBLE FRACTION
INSOLUBLE FRACTION
TIME (DAYS)
TIME (DAYS)
Figure 2.
Figure 3.
Figure 2. Distribution of methanol extractable radioactivity in roots and
shoots of Spartina alterniflora treated for 2 days with [14C]-atrazine and
then transferred to an atrazine-free nutrient solution for 28 days.
Figure 3. Change in radioactivity with time in the water, chloroform, and
insoluble fractions of an 80% methanol extract of shoots of Spartina alterni-
flora plants grown in P4C]-atrazine for 2 days and then transferred to an
atrazine-free nutrient solution.
or its metabolites in the plant residue that remained after exhaustive
washing with 80% methanol. The chloroform fraction, which contained atra-
zine, had approximately 80% of the radioactivity after the initial 2-day
exposure to the radiolabeled herbicide (Figure 3). Radioactivity in this
fraction declined rapidly during the first 5 to 6 days after transferring
the plants to an atrazine-free solution. This was followed by a slower
decrease. There was a corresponding increase in radioactivity of the
aqueous fraction, which is consistent with the expected precursor-product
relation between components of the chloroform and aqueous fractions. Similar
results were reported for the distribution of radioactivity between the
chloroform and aqueous fractions of extracts of sorghum treated with [14c]
atrazine (Lamoureux et al., 1973). It is well established for atrazine-
resistant corn and sorghum that the chloroform fraction contains atrazine,
its N-dealkylated products, and some conjugated metabolites, whereas the
aqueous fraction contains 2-hydroxyatrazine and most of the conjugated
metabolites (Shimabukuro, 1967b; Shimabukuro et al., 1970; Shimabukuro et
al., 1971; Lamoureux et al., 1972). Conversion of atrazine to water-soluble
metabolites by s. alterniflova is slower than in sorghum. After 2 days of
continuous exposure to p4c]-atrazine, radioactivity in the aquequs fraction
from sorghum was near maximum (Lamoureux et al., 1973), whereas in our
11
-------
studies, the increase in radioactivity in the same fraction from the
smooth cordgrass did not level off until 10 days after exposure to the
herbicide. Radioactivity in the 80% methanol-insoluble plant residue in-
creased slowly with time, reaching approximately 20% of the total radio-
activity after 30 days.
It appears that in S. alterniflova the root plays a relatively minor
role in the metabolism of atrazine, since about 90% of the radioactivity
was present in the shoots when the plants were transferred to the atrazine-
free solution. However, the changes in radioactivity with time in the
three fractions from the roots and shoots were compared, and a similar re-
lationship between the fractions was found, suggesting that at least some
of the same reactions occur in both tissues (data not given).
Chloroform-Soluble Metabolites of Atrazine
The basis of atrazine resistance in higher plants is due primarily to
conversion of the herbicide to nontoxic metabolites (Shimabukuro et al.,
1970). The primary types of reactions in the atrazine degradation are well
known and include 2-hydroxylation, N-dealkylation, and conjugation. The
three possible N-dealkylation products of atrazine are 2-chloro-4-amino-6-
isopropylamino-s-triazine, 2-chloro-4-amino-6-ethylamino-s-triazine, and
2-chloro-4,6-diamino-s-triazine. Along with unchanged atrazine, each of
these N-dealkylation products was detected in the chloroform fraction of
extracts of S. alternifloTa by TLC and cochromatography with authentic
standards (Figure 4). N-dealkylation appears to be the principal reaction
in the degradation of atrazine by soil fungi (Kaufman and Kearney, 1970)
and pea plants (Shimabukuro et al., 1966; Shimabukuro, 1967a) and seems to
be a universal reaction in higher plants, animals, and microorganisms
(Shimabukuro et al., 1970). The two monodealkylated products of atrazine
degradation seem to be most common in higher plants, but the diamino products
has been identified in sorghum (Shimabukuro et al., 1973). In smooth cord-
grass, atrazine and its N-dealkylation products were accompanied by a polar
metabolite(s). The identity of the polar metabolite(s) in the chloroform
fraction was not determined, but they may be similar to those reported in
the chloroform fraction from sorghum which are intermediates in the con-
jugation pathway of atrazine metabolism (Shimabukuro et al., 1973).
Change in Atrazine and Its Metabolites with Time
Davis et al. (1959) showed that there was a correlation between the
amount of atrazine in exposed plants and susceptibility. After 2 days of
continuous exposure to ['4c]_atrazine, 77.9 and 57.1% of the total radio-
activity in the cordgrass roots and shoots, respectively, was present as
atrazine (Figure 5). Atrazine in each tissue declined rapidly as shown by
the decrease in radioactivity between 2 and 10 days after the initial
exposure. After 10 days from the initial exposure to [l^C] atrazine, the
rate of decline of absorbed radioactive atrazine was similar in the roots
and shoots.
As noted above, S. alt-ernifloTa can be classified as resistant to
atrazine. Although differences in absorption and translocation are possible
12
-------
ATRAZINE
ISOPROPYL-
AMINO-
ETHYLAMINO-
OIAMINO-
12 16 20
TIME (MINI
O CD ©0 ©
034
061 072
Figure 4. Left, an autoradiograph of a TLC plate showing radioactive
components present in the chloroform fraction of an 80% methanol extract
of Spartina alterniflora treated with P4C]- atrazine for 2 days and then
transferred to a herbicide-free nutrient solution. Right, (A) radio-
chromatogram of the chloroform fraction, (B) diagram of TLC autoradiograph
shown on the left, and (C) Rf values of standards of atrazine and its
metabolites. (1) Polar metabolites, (2) 2-chloro-4,6-diamino-s-triazine,
(3) 2-chloro-4-amino-6-ethylamino-s-triazine, (4) 2-chloro-4-amino-6-
isopropylamino-s-triazine, and (5) 2-chloro-4-ethvlamino-6-isopropylamino-
s-triazine (atrazine). The TLC plate was developed two times in benzene-
acetic acid (50:4, v/v).
factors chat determine the degree of resistance or susceptibility of a
plant to atrazine, it is well-established that resistance and selectivity
are due primarily to the plant's ability to degrade the herbicide to non-
toxic substances (Shimabukuro, 1967a; Lamoureux et al., 1970; Robinson and
Greene, 1977; Lamoureux et al., 1973). Eight days after transferring the
plants to an atrazine-free solution, radioactivity in the chloroform
fraction remained nearly constant for 20 days at about 34.5 to 37.3% of the
total. The chloroform fraction contains primarily atrazine and its
N-dealkylation products which at least partially contribute to detoxification
of the herbicide (Shimabukuro, 1967a; Shimabukuro, 1967b). 2-Chloro-4-amino-
6-isopropylamino-s-triazine is the primary product of N-dealkylation in
S. alterniflora. Radioactivity in this metabolite extracted from shoot
tissue ranged from 5.8 to 20.4% of the total, while the corresponding
mono-N-dealkylated product ranged between 2.5 and 12.5% over a 30-day
period (Table 3). 2-Chloro-4-amino-6-isopropylamino-s-triazine was present
13
-------
80-
£ 60-
§40
20
10
15
TIME (DAYS)
20
25
30
Figure 5. Change in radioactivity with time in atrazine extracted from
shoots and roots of Spartina altemiflora treated for 2 days with P4C]-
atrazine and subsequently transferred to an atrazine-free nutrient solution
for 28 days.
at the highest levels between 5 and 10 days after the initial herbicide
treatment while the corresponding mono-N-dealkylated product was detected
at the highest levels at 20 days. Although the level of each mono-N-
dealkylated atrazine product decreased between 20 and 30 days after the
initial exposure, it seems that 2-chloro-4-amino-6-isopropylamino-s-triazine
is the favored substrate for the second N-dealkylated or possibly the con-
jugation reaction. The diamino atrazine metabolite resulting from the
second dealkylation reaction represented a relatively minor component of
the chloroform fraction but did tend to increase with time through the
30-day experiment (Table 3).
14
-------
TABLE 3. RADIOACTIVITY IN N-DEALKYLATED PRODUCTS OF ATRAZINE METABOLISM
ISOLATED FROM SHOOTS OF SPAETINA ALTERNIFLORA COLLECTED AT
INTERVALS OVER A 3Q-DAY PERIOD
Days after
treatment
2
5
10
20
30
I*
14.4
20.4
18.2
9.4
5.8
Radioactivity, % of total
lit
2.5
4.2
4.2
12.1
5.1
III?
0.8
2.2
2.8
3.0
8.7
* I, 2-chloro-4-amino-6-isopropylamino-s-triazine.
t II, 2-chloro-4-amino-6-ethylamino-s-triazine.
$ III, 2-chloro-4,6-diamino-s-triazine.
SUMMARY
The tolerance to atrazine and the translocation and metabolism of this
herbicide by the marsh grass Spart-ina alterniflora were studied for 28 days
in an atrazine-free solution after an initial 2-day root exposure to the
radiolabeled compound. No visual symptoms of atrazine toxicity were ob-
served at the concentrations tested and s. altevniflora was considered
at least moderately resistant to this herbicide. Atrazine was readily
absorbed by the roots and translocated to the shoots; after 2 days exposure
to [14c]-atrazine, 90% of the radioactivity was present in shoots. Atrazine
was readily metabolized to chloroform, aqueous, and subsequently to insoluble
substances. The chloroform fraction, which contained atrazine, showed an
initial rapid decrease and then a steady decrease from 84.8 to 23.9% of
radioactivity in the total extract. Along with atrazine, three N-dealkyla-
tion products were identified by thin-layer chromatography: 2-chloro-4-
amino-6-ethylamino-s-triazine, 2-chloro-4-amino-6-isopropylamino-s-triazine,
and 2-chloro-4,6-diamino-s-triazine.
15
-------
SECTION 4
ATRAZINE METABOLISM BY SPARTINA ALTERNIFLORA:
WATER-SOLUBLE METABOLITES
OBJECTIVES
Research conducted in Section 3 established that Spartina AHerniflora
rapidly converted the absorbed atrazine to water soluble metabolites. The
objectives of Section 4 were: 1) to identify the metabolites produced and
2) to determine the pathway(s) followed in the metabolism of atrazine by
S. altez>ni flora.
MATERIALS AND METHODS
Growth and Treatment of Plants
S. alterniflora plants were collected from the salt marsh at Sapelo
Island, Ga., and maintained at Auburn University as described previously
(Section 3). In time course experiments, a uniform lot of vigorously growing
plants was divided into 10 groups of four plants each. Each group of plants
was placed in 300 ml of Hoagland's solution containing a 2.0 yCi of [14C]-
atrazine (24.9 yCi/mg). After 2 days, the plants were removed from the [14c]-
atrazine solution, the roots rinsed, and all but two groups were placed in
atrazine-free Hoagland's solution. The two groups not transferred to the
atrazine-free solution and two additional groups collected at 3, 8, 18, and
28 days after transfer to the atrazine-free nutrient solution were extracted
as described previously (Section 3). Briefly, the plants were separated into
roots and shoots which were ground separately and extracted with 80% methanol.
The extract was filtered and the filtrate was concentrated using a rotary
flash evaporator. The plant material remaining after filtration is referred
to as the 80% methanol-insoluble fraction. The concentrated extract was
diluted with distilled water and washed three times with chloroform. The
combined chloroform fractions were taken to dryness. The aqueous fraction
was also concentrated and the radioactive components isolated as described
below.
Isolation of Water-Soluble Atrazine Metabolites
Plants used specifically for the isolation of water-soluble metabolites
of atrazine were selected and placed in 1-L plastic cups, five plants per
2
Most of the material in this section has been submitted to the Journal
of Agriculture and Food Chemistry for consideration for publication. Copy-
right 1979, American Chemical Society. Printed by permission.
16
-------
cup, with their roots in 400 ml of Hoagland's solution. Each cup had 2.0 yCi
[14c]-atrazine anc' sufficient nonlabeled atrazine to give a 5 x 10~5 M
atrazine concentration. The plants were maintained in these cups for 2 days
after which the roots were rinsed with nonlabeled atrazine solution and
placed in a fresh atrazine-free Hoagland's solution for 18 days.
Approximately 1 kg of fresh shoot tissue from these plants was ground,
in 10-g portions, 3-min in a Waring Blender containing 80% aqueous methanol
(10 ml solvent/1 g tissue). After each grinding step, the extract was fil-
tered and the residue was placed in the extracting solvent for 1-2 days at
4 C until all the material was ground. The grinding, soaking, and filtering
steps for all the tissue was repeated two times. The combined extracts were
concentrated by flash evaporation at 37 C, dissolved in 400 ml water, and
washed with four 500-ml volumes of chloroform. Portions of the aqueous
fraction containing about 10? dpm were concentrated as before and dried by
lyoplvlization. The dried sample was dissolved in 50 ml of pyridine-acetate
buffer (pH 2.15) and applied to 2.5 x 95 cm column of AG 50W-X2 (200-400
mesh) cation-exchange resin. The resin and buffers were prepared as des-
cribed by Schroeder et al. (1962). The column was maintained at 15 C while
being washed with pyridine-acetate buffer gradient at 0.6 ml/min. The
first chamber of the gradient device contained 400 ml of 0.2 N buffer (pH
3.1), and the second and third chambers each contained 400 ml of the same
buffer solution (2N) at pH 5.0. Five-mi fractions were collected during
each column chromatography step. Adjacent fractions containing no radio-
activity were combined. Five major radioactive fractions were obtained from
the cation-exchange column; each was evaporated to dryness by flash evapora-
tion at 37 C, dissolved in minimum volume of water, and applied to a separate
2.5 x 95-cm column of AG 1x2 (200-400 mesh, acetate form) anion-exchange
resin, also maintained at 15 C. The column was eluted with an acetic acid
concentration gradient at a flow rate of 0.4 ml/min. The gradient device
contained 275-ml of water in each of the first two chambers, 275-ml of 0.5 N
acetic acid in chamber 3 and 275-ml of 0.38 N acetic acid in chamber 4.
The column was subsequently washed with 150-ml of 0.5 N acetic acid followed
by 150-ml/fractions each of 1 to 6 N acetic acid. Radioactive fractions
from this column were obtained as before and evaporated to dryness by flash
evaporation. The residue was washed from the flask with one 5-ml portion
and two 1-ml portions of absolute methanol which was then evaporated to
dryness under N2- The residue was then dissolved in few ml of water and
placed on a 1.5 x 95-cm column of Sephadex LH-20 which was washed with water
at the rate of 0.5 ml/min at 15 C. Each radioactive fraction from this
column was purified by TLC using silica gel G (250 ym thickness) on 20 x 20 cm
glass plates. The TLC plates were developed three times in benzene-ethyl
acetate-acetic acid-water (25:50:30:3, v/v/v/v) (Solvent System I), and
radioactive areas were located with a Berthold TLC Scanner. The silica gel
spots containing radioactivity were removed from the plate and washed five
times with methanol. The radioactive substances were then applied to
silica gel plates which were developed one time in n-butanol-acetic acid-
water (60:15:25, v/v/v) (Solvent System II). In most cases, further purifi-
cation of the radioactive substance was by dialysis followed by TLC, using
Solvent System II. A schematic outline of the column and thin-layer chroma-
tographic procedures is given in Figure 6.
17
-------
80% MEOH EXTRACT "j
*l «2 *3 «4 *» *6 »7 I
NR 37.0 T.S 29.6 4.4 21.8 NR J
CATION - EXCHANSE
CHROMATOGRAPHY
(PYRIOINE-ACETATE BUFFER)
SEPHAOEX COLUMN
CHROMATOGRAPHY
I4ZML
16) ML
30OML
OML
9OML
135 ML
800 ML
ANION-EXCHANGE
CHROMATOGRAPHY
(ACETIC ACID-HjO)
THIN-LAYER
CHROMATOGRAPHY
(BENZEN£:ETHYL ACETATE : ACETIC AC.:
H,0,J3:SO:50:5)X J
IBUTANOL: ACETIC Acio-Hto,
60:13:251X1
Figure 6. Outline of procedures used in the isolation and purification of
water-soluble metabolites of atrazine from S. alterniflora. The example is
given for fraction #6 from the cation-exchange column, but each fraction was
treated in a similar manner.
Characterization of Metabolites
For ami no acid analysis, a portion of each purified metabolite was
hydrolyzed under nitrogen in 200 yl of 6 N HC1 in a sealed ampule under
nitrogen at 110 C for 16 h. Each hydrolysate was diluted with distilled
water and evaporated to dryness several times under vacuum to remove the
HC1 and then treated for 2 h at room temperature with a mixture of 200 yl
of formic acid and 20 yl of 30% hydrogen peroxide (Lamoureux et al., 1973).
Amino acids were analyzed using a Beckman Amino Acid Analyser.
For analysis of the triazine portion of the purified substances from
the aqueous fraction of the 80% methanol extract, a portion of each sub-
stance was hydrolyzed under nitrogen in 200 yl of 6 N HC1 at 50 C for 8 h.
Hydroxy derivatives of atrazine and its metabolites were obtained. The
hydrolysates were compared by TLC using silica gel ^54 (250 ym thickness)
with standards of authentic hydroxy-atrazine and hydroxy derivatives of
N-dealkylated products of atrazine,using Solvent System II and 2-propanol-
28% ammonium hydroxide-water (80:10:10, v/v/v) (Solvent System III).
Derivatization for Mass Spectral Analysis
A portion of each isolated metabolite was suspended in 2.0 ml of absolute
18
-------
methanol and treated with diazomethane as described by Schlenk and Gellerman
(1960). After methylation, the product was dried under nitrogen and puri-
fied by TLC (Solvent System II). The major radioactive derivatives were
washed from the silica gel with absolute methanol and analyzed by mass
spectrometry.
RESULTS AND DISCUSSION
Translocation and Metabolism of Atrazine
Atrazine was readily absorbed and translocated by S. alterniflora.
After 2 days of continuous exposure to the radio!abeled herbicide, approxi-
mately 90% of the absorbed atrazine was present in the shoots (Section 3).
The extracts of S. alterniflora shoots were separated into chloroform,
aqueous, and 80% methanol-insoluble fractions. The chloroform fraction had
approximately 80% of the radioactivity after the initial 2-day exposure to
radioactive herbicide (Figure 7). Radioactivity in this fraction declined
rapidly during the first 5-6 days, followed by a slower decrease. There was
a corresponding increase in radioactivity of the aqueous fraction, suggesting
a precursor-product relation between components of the chloroform and aqueous
fractions.
100-
80-
>• 60
40
20
D D WATER-SOLUBLE FRACTION
O O CHLOROFORM-SOLUBLE FRACTION
A A INSOLUBLE FRACTION
10
15
TIME (DAYS)
20
25
30
Figure 7. Change in radioactivity with time in the aqueous, chloroform,
and insoluble fractions of an 80% methanol extract of shoots from S. alterni-
flora plants grown in [^C]-atrazine for 2 days and then transferred to an
atrazine-free nutrient solution for 28 days. (This figure is also shown
in Section 3 but is reproduced again here for easier access.)
19
-------
In smooth cordgrass, atrazine and its N-dealkylation products of the
chloroform phase were accompanied by relatively polar metabolites (Section
3). The amount of radioactivity in these polar metabolites increased slowly
with time, ranging from approximately 2% in plants treated for 2 days to
about 11% in s. altemiflora 28 days after transfer to the atrazine-free
solution. Similar metabolites were reported in the chloroform fraction from
sorghum. The chloroform-soluble atrazine metabolites in both sorghum and
S. alterniflora appear to be minor metabolites of the herbicide and do not
accumulate rapidly with time. Perhaps they are transitory intermediates that
are converted to water-soluble atrazine metabolites (Shimabukuro et al.,
1973).
Radioactivity in the 80% methanol-insoluble fraction of the s. altemi-
flora extract also increased slowly with time, reaching about 18% of the
total after 28 days in the atrazine-free nutrient solution (Figure 7). An
increase in radioactivity in the insoluble fraction of sorghum extracts
occurred with time but at a higher rate than in S. altemiflora (Lamoureux
et al., 1973), reaching ca 30% by the end of the experimental period. The
chemical form of the 80% methanol-insoluble metabolites of atrazine are
unknown.
Water-soluble Metabolites of Atrazine
Five to 8 days after exposure of Spartina to [^cj.atrazine, all but
about 3% of the herbicide was metabolized to either chloroform or water-
soluble substances, which appear to persist for a considerable length of
time (Figure 7). For example, after 28 days in atrazine-free nutrient
solution, radioactivity in the aqueous extract of [^c]_atrazine-treated
S. altemiflora represented ca 58% of the total. The resistance of this
marsh grass to atrazine is attributed to the rapid metabolism of the herbi-
cide to nontoxic substances (Section 3). Atrazine metabolism follows a
similar pattern to that in sorghum, which is also very resistant to the
herbicide but it is slower in S. altemiflora (Lamoureux et al. , 1973).
For example, extracts of sorghum plants treated as described for s. alterni-
flora contained over 60% of the radioactivity in the aqueous fraction at
the end of the 2-day exposure to the herbicide. Radioactivity increased
slightly thereafter but decreased after 20 days in the atrazine-free solution.
After passing the aqueous fraction of the S. altemiflora extract
through a series of cation-and am"on-exchange and Sephadex columns, it was
apparent that there are numerous water-soluble metabolites of atrazine most
of which appear to have a high turnover rate during the 28 days after ex-
posure to the herbicide. Five radioactive fractions (designated #2 through
#6) were obtained by passing the aqueous fraction through a cation-exchange
column (Figure 8), and 79% of the radioactivity placed on the column was
recovered. There was a shift from polar to less polar substances during
the first 8 days after transfer of the plants to the atrazine-free nutrient
solution. Fifty-two to 60% of the radioactivity accumulated in fraction 2
(fraction 1 contained no radioactivity) after the first 8 days, and changes
in the relative proportions of radioactivity in other fractions occurred
during the final 20 days of the experiment.
20
-------
»»
>-
K
>
•5
O
K
60
40
20
60
40
20
60
40
20
60
40
20
60
40
20
n
51.7% 30 DAYS
-
-
-
-
59.8%
17.6%
-------
TABLE 4. SUMMARY OF RESULTS OF THE ISOLATION OF SOME WATER-SOLUBLE METABO-
LITES OF ATRAZINE FROM S. ALTERWIFL0RA SHOOTS
Number & letter
designation*
Radioactivity (%) of
Fraction from
anion-exchange
coliimnt
Total aqueous
fraction-f
Compound
2A
2B§
2C
2D
2E
2F§
2G§
3A
4A
4B
5A
6A
30.7
5.3
24.6
8.6
15.1
6.8
8.7
100
28
72
100
100
11.9
2.0
9.5
3.4
5.8
2.6
3.4
7.6
8.6
22.3
4.7
22.8
1
-
2,3,3a
10
4#
-
_
12
13**
14,14a,15,15a,16
17
8,8a
* The numeral indicates the cation-exchange column fraction number. Fraction
1 contained no radioactivity. The letter designates the fraction(s) result-
ing when the cation-exchange column fraction was chromatographed on the
anion-exchange column.
t Values are percentages of the radioactivity in the parent fraction from the
cation exchange column.
t Values are percentages of the aqueous fraction (ca 3.5xl0^cpm) of the 80%
methanol extract.
u -A"^
§ These fractions were not processed further; f2-Hydroxyatrazine; 2-Hydroxy-
4-amino-6-isopropylamino-4-triazine.
Passage of each cation-exchange column fraction separately through an
anion-exchange column achieved further separation of the atrazine metabolites
(Figure 9). Each radioactive fraction from the anion-exchange column was
subsequently passed through a Sephadex LH-20 column, but little separation
of radioactive substances was achieved. This procedure only aided in the
purification process by removing some nonradioactive substances. Fractions
from the anion-exchange column containing relatively small amounts of
radioactivity (2B, 2F, and 2G) were not processed further. Each radioactive
component was further purified (or separated) by TLC, using two solvent
systems. A summary of the results obtained by these chromatographic procedures
is given in Table 4. TLC in two solvent systems indicated that fractions 2A,
2D, 2E, 3A, 4A, and 5A from the anion-exchange columns contained only a
22
-------
_l
2500
2375
2250
2125
2000
1875
1750
1625
1500
1375
1250
1125
1000
875
750
625
500
375
250
125
0
8.7
6.8
15.1
8.6
24.6
5.3
-
-
-
-
-
-
-
50.7
1
///i
UlL
Wt
U-U-
m
«4
*IO
•3'o
1
( 1 r— 1 |
•12 11°; nza *IT
• i i"yv^^ T™ .. . . i ,. _
" 28 (f/1 *I5 22Z2«8
* 0 *1 *4 **>
-------
2-hydroxy-4-amino- 6-ethylamino-s-triazine and 2-hydroxy-4,6-diamino-s-triazine
(Table 4).
Based on the previous research on atrazine metabolism in sorghum
(Shimabukuro et al., 1971), it was suspected that the remaining water-
soluble metabolites were intermediates in the glutathione conjugation path-
way. With the exception of compounds #4 and #13, each radioactive water-
soluble substance isolated was subjected to acid hydrolysis so that the
chemical nature of the s-triazine ring and possibly the conjugate groups
might be determined. The results of hydrolysis showed that the conjugated
water-soluble metabolites were nearly equally divided between derivatives
of atrazine and 2-chloro-4-amino-6-isopropylamino-s-triazine (Table 5).
2-Hydroxy-4-amino-6-ethylami no -s-triazine was obtained on hydrolysis of
only compound #15a, and no 2-hydroxy-4,6-diamino-s-triazine was detected
in any of the hydrolysates.
TABLE 5. CHEMICAL NATURE OF THE S-TRIAZINE RING IN THE INDIVIDUAL RADIO-
ACTIVE COMPONENTS FROM THE AQUEOUS FRACTION OF THE 80% METHANOL
EXTRACT OF S. ALTERNIFLORA GROWN IN ^-LABELED ATRAZINE SOLUTION
5-Triazine
Water-soluble compounds
isolated
2-hydroxy-atrazi ne
2-hydroxy-4-ami no-6-i sopropy1 ami no-
s-triazine
2-hydroxy-4-amino-6-ethylamino-s-
triazine
2-hydroxy-4,6-diamino-s-triazine
1, 2, 4, 8a, 10, 16
3, 3a, 8, 14, 14a, 15, 17
15a
Amino acid analysis of the hydrolysates did not reveal much about the
chemical nature of the conjugate portion of the water-soluble atrazine
metabolites. In each case, the usual protein amino acids were detected in
very low concentrations (< 5 nmoles), suggesting that the samples may have
contained small amounts of protein. However, in some cases, a particular
amino acid was present in higher relative abundances than the others and
was possibly part of the conjugate material. The hydrolysate of compound
#2, for example, contained high relative proportions of glutamic acid,
glycine, and an unidentified substance. Since glutathione is composed of
these two amino acids and cysteine, the results suggest that glutathione
conjugation occurs in Spartina as in other resistant plants (Shimabukuro
24
-------
et al., ]971). Compounds #8, #8a, #14, #16, and possibly #3a contained
glycine in higher relative proportions than "background" amino acids, sug-
esting that they may also be intermediates in glutathione conjugation as in
sorghum. A conjugated intermediate containing glycine would be expected
to occur early in the pathway (Lamoureux et al., 1973). Compound #14a con-
tained valine, the significance of which is not clear. Amino acids at
levels above background were not detected in the other samples.
Fraction 6A appeared to be a single radioactive substance through each
of the column and first thin-layer chromatographic steps (Figure 6). How-
ever, TLC of fraction 6A using solvent system II showed that it was an equal
mixture of two substances (Figure 10). Based on the hydrolysis data, they
differ by the chemical nature of the s-triazine, #8 being a derivative of
atrazine and #8a a derivative of 2-chloro-4-amino-6-isopropylamino-s-triazine,
Compounds #3, #14, and #15 appeared chromatographically pure according to
Solvent System I. As before, compounds #15 and #15a differed according to
the s-triazines, being derivatives of 2-chloro-4-amino-6-isopropylamino and
2-chloro-4-amino-6-ethylamino-s-triazine, respectively. The other two pairs
of radioactive substances did not differ according to the triazine portion
of the molecule, but did differ in the amino acid portion of the module
obtained on hydrolysis. For example, compounds #14 and #14a differed by
glycine and valine, respectively. The basis on which compounds #3 and #3a
were separated by TLC is not clear, (Table B).
SOLVENT SYSTEM A
*8.*8o
SOLVENT SYSTEM B
t8o
TIME (MIN)
Figure 10. TLC of fraction 6A from the anion-exchange column.
25
-------
As in sorghum (Shimabukuro et al., 1971; Lamoureux et al.,1973), atra-
zine is metabolized to a large number of substances in S. altevniflora. The
atrazine molecule may be simply hydrolyzed; it may be degraded through
dealkylation or both. However, the process leading to the largest number
of atrazine metabolites appears to be initiated by conjugation followed by
extensive chemical modification of the non-s-trrazine portion of the molecule.
The s-triazine portion of atrazine shuttled through the conjugation pathway
was present in numerous compounds, 16 of which were isolated and partially
characterized. The predominant chemical forms of the s-triazine ring among
these metabolites were the fully alkylated and 4-amino-6-isopropylamino
forms (Table 5). Although collectively the water-soluble substances (con-
jugation plus 2-hydroxy) represent the greatest number of atrazine metabolites
in Spartlna, nonconjugated chloroform-soluble metabolites represent the
quantitatively most important chemical species. For example, after 20 days
exposure to atrazine, ca 25% of the radioactivity from [^c] atrazine in
Spartina shoots was represented by only 3 compounds: 2-chloro-4-amino-6-
isopropylamino-s-triazine, 2-chloro-4-amino-6-ethylamino-s-triazine, and
2-chloro-4,6-diamino-s-triazine (Section 3). This combined with the ca 14%
of the total radioactivity represented by 2-hydroxy-atrazine and 2-hydroxy-
4-amino-6-iospropylamino-s-triazine amounts to ca 39% of the atrazine metabo-
lites represented by only 5 substances. The extent to which these substances
are conjugated after 30 days in the plant is not known. The final form of
a large portion of the s-triazine ring appears to be 80% methanol-insoluble
as suggested by extrapolation in Figure 7. The material appears to be
tightly linked to the structural components of the cells. Accumulation of
atrazine metabolites in this fraction of S. alterniflora is considerably
slower than in sorghum (Lamoureux et al., 1973). How incorporation of the
atrazine metabolites into this insoluble form relates to persistence in the
environment is not known.
SUMMARY
Spartina alterniflora plants were incubated 2 days in half-strength
Hoagland's solution containing 2.0 yCi ring-labeled atrazine and then trans-
ferred to an atrazine-free nutrient solution. Samples were taken 3, 8, 18,
and 28 days after transfer and the change in radioactivity in the water-
soluble atrazine metabolites as a function of time was determined. After
2 days, ca 90% of the radioactivity was present in the shoots. Radio-
activity in the aqueous fraction of S. alterniflora shoot extracts increased
for ca 8 days, after which the radioactivity remained relatively constant
representing ca 60% of the total radioactivity taken up by the plants.
The aqueous extract was fractionated using cation-and anion-exchange
and sephadex column chromatography followed by thin-layer chromatography
using two solvent systems. Some of the water-soluble metabolites of atrazine
were isolated and partially characterized. About half the water-soluble
metabolites contain fully N-alkylated triazine rings while the other half
contained the 4-amino-6-isopropylamino derivative. Only one contained the
4-amino-6-ethylamino derivative and no 4,6-diamino forms were detected.
2-Hydroxyatrazine and 2-hydroxy-4-amino-6-isopropylatnino-s-triazine were
identified as water-soluble metabolites of atrazine. Acid hydrolysates of
26
-------
the isolated metabolites contained low amounts of amino acids such as
glutamic acid, glycine, and valine,suggesting the glutathione conjugation
pathway of atrazine detoxification may be operative in S. alterniflora.
Atrazine is metabolized to a large number of substances by S. alterni-
flora some of which tend to slowly accumulate in the insoluble plant residue
with time.
27
-------
SECTION 5
EFFECT OF ATRAZINE ON UCA PUGNAX3
OBJECTIVES
The most conspicuous detritivore in the salt marsh is the fiddler
crab, Uoa pugnax. The objectives of the research in Section 5 were to
determine: 1) whether AAtrex was toxic to fiddler crabs and whether the
toxicity was due only to the active ingredient, atrazine; 2) the effect of
size and sex of the crab on atrazine toxicity; 3) whether the presence of
marsh soil affected toxicity; and 4) whether there were seasonal variations
in sensitivity of the crabs to atrazine.
MATERIALS AND METHODS
Atrazine Studies in the Field and in Microecosysterns
In these studies the effect of atrazine on fiddler crabs in a natural
salt marsh (field studies) were compared with those in microecosystems
(microecosystem studies). Both experiments were started in the middle of
June 1977 and terminated 10 weeks later.
Field Studies--
Twelve 1.8-m diameter 0.75-m tall metal cylinders were sunk 15 cm deep
in a uniform stand of Spartina alterniflora in a salt marsh near Sapelo
Island, Ga. The cylinders minimized the movement of fiddler crabs into and
out of these enclosures. Holes were cut in each cylinder at soil level and
then the holes covered with 0.6-cm mesh hardware cloth. The holes permitted
tide water to flow into and out of the enclosures, and the hardware cloth
prevented movement through the holes of all except the very small crabs.
The areas enclosed by cylinders were selected for uniformity of S. alterni-
flora stands and depth of flooding by the tides. At a time appropriate for
comparison with the microecosystem studies, three randomly selected enclosures
were sprayed with 0, 100, 1,000, or 10,000 ppm atrazine by applying the
appropriate concentrations of the 80% wettable powder, .AAtrex. The amount
of spray solution applied was sufficient to give 0, 0.05, 0.5, and 5 g
atrazine/m2. The herbicide was applied at a period of low tides and when
no flooding would be expected for at least 6 h after herbicide applications.
3
jd
permission.
o
Most of the material in this section is reprinted from.a.paper ac-
cepted for publication in Esjtuarie_s and is reproduced here with their
28
-------
Daily records were made on the appearance and general well being of plants
and animals in each of the treated areas and outside the enclosures. Daily
records were also made of the depths and durations of flooding and the
amounts of rainfall. Ten weeks after treatment, crab populations were esti-
mated using the method of Wolf et al. (1975).
Microecosystem Studies—
Each microecosystem used in these studies was constructed from 75-L
molded plastic, laundry tubs (described and illustrated in Section 10). Each
resultant container was 51 cm long by 51 cm wide by 97 cm deep and nad a
drain hole in the bottom. A 45-cm long 2.5-cm-diam plastic pipe, perforated
with 3-mm diam holes every 2 or 3 cm, was inserted through the drain hole
upward into the container. It made possible movement of water into and out
of the system but prevented any larger animals from escaping through this
port. Twelve of these containers were taken to the salt marsh for filling.
Approximately 25- by 25- by 20-cm-deep sections of soil on which a dense
stand of S. alteimiflora was growing were removed and fitted into the bottom
of the ecosystem containers. The spaces between the sections were filled
with soil also taken from the marsh. In securing these S. alterniflora-
bearing soil sections, the associated algae, soil microorganisms, and blood
worms were also automatically transferred into the container. The larger
animal components, fiddler crabs (Uoa pugnax) horse mussels (Geukensia
demissa) , and periwinkle snails (littorina ivrorata) were collected separately.
The ecosystem containers and animals were then transported back to the green-
house at Auburn, AL, for final assembly. Each ecosystem was attached to a
75-L seawater reservoir that was raised and lowered by a "tide machine" as
described by Everest and Davis (1977). Diagrams and additional details about
these systems are presented in Section 10.
As soon as the 12 systems were coupled with their reservoirs and tides
initiated, ca 15 randomly selected fiddler crabs were added to each system
giving a fiddler crab density comparable to that in the field. Additional
small crabs, both Uoa pugnax and Sesarma cinereum, may have been present in
the soil placed in the ecosystems. After all components had been added, the
systems had stabilized, and the S. alterniflora was growing well, word was
sent to Sapelo Island to spray atrazine at the prescribed rates on the 12
plots on the Island. One week after the herbicide was applied at Sapelo
Island, equal rates were applied to randomly selected microecosystems at
Auburn. Tidal frequency and depth and rainfall amounts were duplicated at
Auburn 1 week after they occurred at Sapelo Island. Ten weeks after herbi-
cide application, all fiddler crabs in each microecosystem were collected
and the number found was recorded.
Laboratory Studies.
Fiddler crabs,Uoa pugnax, were collected several times during the year
in the pristine marshes of Sapelo Island and Cabretta Island, GA. Crabs
were transported to Auburn, AL, and acclimated for 2 weeks to the new
environment before testing began. Laboratory temperature was 27 C and a
regime of 12 h light and 12 h darkness was maintained. Oven-dried leaves
29
-------
of S. alterniflora were provided as a food source.
To determine if age (as determined by size), sex, or the presence or
absence of marsh soil were important factors affecting atrazine toxicity,
the following experiments were performed in August 1977. Crabs were placed
singly in 144 plastic boxes (18 x 13 x 5 cm) with covers. Half of the 144
boxes contained approximately 1 inch of marsh soil. One hundred ml of atra-
zine at 0, 1, 100, or 1000 ppm of atrazine suspension was added to each box.
This was sufficient to cover the ventral surface but not the dorsal one.
AAtrex 80 (80% atrazine) was the formulation of atrazine employed, therefore,
the concentration of the formulated material used was actually 0, 1.25, 125,
and 1250 ppm. All solutions were prepared in Instant Ocean with a salinity
of 20%. Four different crab classes were exposed to each of the atrazine
concentrations with or without marsh soil. These were large males (carapace
width > 1.5 cm), large females ( > 1.5 cm), small males ( < 1.5 cm), or small
females ( < 1.5 cm). The experiment was replicated four times. Crabs were
observed 6, 12, and 24 h after atrazine addition and then every 24 h for
8 days. Toxicity to crabs was assessed in terms of percentage of death-free
days (DFD), percentage surviving, percentage stress-free days (SFD), and
percentage that passed the escape-response test (ER test). The percentage
death-free days was calculated by the formula: DFD(%) = (C-| + Cg + 03 . . .)
N~T~D
x 100 where: CT, C2, 03, etc. = the number of days each crab lived: N =
number of crabs tested; and D = duration of the experiment in days. The
percentage of stress-free days was calculated by the formula SFD(SK) =
(Si + $2 + S3. . .) x 100 where: S-j, S2, S3, etc. = the number of days
N X D
when each crab was neither dead nor showing obvious signs of stress; and
N and D are used as described above.
Locomotor response was determined for all live crabs at the end of
8 days by the methods of Ward and Busch (1976). For this locomotor test, a
40- by 25-cm shallow aluminum pan was used. Two lines were drawn across the
pan 5 cm from each end (30 cm between lines), and the pan lined with paper
towelling was moistened with sea water. The pan was placed inside a large
cardboard box having viewing ports which permitted observation of the crabs
without frightening them. One crab was placed in the end of each pan and
allowed to "habituate" before frightening the crab by waving an object through
an opening in the large box near the crab. The crab either "escaped" (crossed
far line in 15 sec) or had "no response" (did not cross far line within 15sec).
The above test was supplemented by another to ascertain whether the
additive in the formulated atrazine (AAtrex) or the atrazine itself was the
toxicant. In this study small males were exposed in the absence of soil to:
1) 1000 ppm pure atrazine, 2) 250 ppm of material added to atrazine to make
$
AAtrex, 3) 1250 ppm AAtrex, and 4) no chemicals.
All subsequent tests used only small males, no soil, and AAtrex as the
source of atrazine. Rates used are expressed in terms of the amount of atra-
zine present. These tests were otherwise conducted as described above and
were repeated in late August, September, and November 1977 and in March and
30
-------
August 1978. Atrazine concentrations were varied during these experiments
as attempts were made to ascertain the least effect level and the LD5Q
concentration.
Feeding Studies
Fiddler crabs were collected in early spring of 1977 in the marshes
near Sapelo Island, GA, and transported in marsh soil on ice to Auburn, AL.
In Auburn the fiddler crabs were divided into 84 groups of two or three crabs
each in such a manner that the weight of crabs per group was approximately
the same. One such group was placed in each of 84, 18- by 13- by 5-cm clear
plastic boxes. Each box contained one-half of an 8-cm diameter petri dish,
which served as a feeding station for the crabs. The floor of each box was
wetted with seawater initially and kept moist by rewetting as needed through-
out the 20-day experiment. The boxes were cleaned every 3 days. The lids
were kept on the boxes except when the crabs were being tended.
The 84 boxes were further divided into 3 groups which corresponded to
treatments. Each group (each group contained ca 60 crabs) was fed detritus
wetted with either 0, 10-6 M,or 10-4 M atrazine suspensions. Throughout the
20-day feeding period, daily observations were made of crab behavior and
mortality. All treatments were completely randomized, and a randomized
analysis of variance was performed on the data from the experiments. However,
before the analysis of variance was performed on the crab mortality, per-
centage data were converted by arcsin transformation (Steel and Torrie, 1970).
RESULTS AND DISCUSSION
Atrazine Effects in the Field and in Microecosysterns.
Studies in the field involved the effect of various atrazine concentra-
tions on the survival of crabs that happened to be trapped in the areas when
the 1.8-m diam cylinders were put in place. Similar studies in the micro-
ecosystems included 15 Uoa pugnax deliberately added to each system plus any
small crabs that were inside the soil mass when it was placed in the eco-
system containers.
Field Studies--
At harvest time the average number of fiddler crabs (Voa pugnax)/m^ for
the 10,000, 1000, 100, and 0 ppm treatments were 4, 67, 40, and 65, respect-
ively. Only the 10,000 ppm rate had a significant effect. Other crabs
(mostly Sesarma oinerewn and a few Eurytium and Panopeus spp) averaged 38,
21, 28, and 21/m2, respectively. Numbers of these crabs were not significantly
affected by the treatments.
Microecosystem Studies--
At harvest time average numbers of fiddler crabs/m2 for the 10,000,
1000, 100, and 0 ppm treatments were 8, 40, 36, and 24, respectively.
Sesarma einereum numbers/m2 Were 0, 4, 12, and 8, respectively. The
31
-------
variability between replications was too great for any of the differences
in averages to be significant. However, 2 weeks after atrazine application,
dead fiddler crabs were found in microecosysterns treated with 10,000 ppm
atrazine but not in microecosysterns receiving any other rate of treatment.
The observed kill by 10,000 ppm led to the other studies on atrazine toxicity.
Laboratory Studies
In the first experiment, August 1977, 1000 ppm atrazine (1250 ppm
AAtrex) was injurious or lethal to some crabs in both sexes and size classes
whether applied with or without marsh soil being present (Table 6). The
adverse effects of this concentration were shown by all paramenters used to
measure the crab response, but the percentages of stress-free days (SFD)
and percentages that survived and passed the escape response test (S & pass
ER test) seemed to be the most sensitive indicators of atrazine toxicity.
The effects of 100 ppm were less severe, and 1 ppm had no effect.
In the supplementary study, it was established that 1250 ppm AAtrex did
not differ in toxicity from 1000 ppm of atrazine alone and that 250 ppm of
the additive did not differ in toxicity from the untreated control.
The presence of marsh soil in the system decreased the toxicity, perhaps
because atrazine adsorption on the soil decreased the atrazine concentration
in water. The decreased toxicity was probably not apparent at the 1000 ppm
rate because the decrease in atrazine concentration was too small to have
an effect. Small males seemed to be more sensitive to atrazine than other
classes tested. Vernberg et al. (1974), working with a different species
of fiddler crab, u. pugilator, found that mercury-treated males died sooner
than females, and the rate of oxygen uptake by males was significantly lower
than by females. Since the uptake of mercury was the same in both sexes, it
was concluded that differences in metabolic rates and thus mercury toxicity
may reflect neuroendocrine differences. Unfortunately, their studies did not
involve different size classes of crabs. It is not certain whether different
rates of respiration have any effect on atrazine toxicity.
When the work was repeated in late August and again in September using
small males in a soil free environment, atrazine did not adversely affect
the crabs. The maximum atrazine concentration used, however, was 100 ppm
(data not included). In November the work was once again repeated, and there
was no adverse effect of atrazine even with 1000 ppm of atrazine (Table 7).
The last exoeriments, conducted in March and August, 1978, gave results
that confirmed the seasonal variations in sensitivity observed in 1977 (Tables
8,9). In the March experiment, only percentage stress-free days (SFD) was
affected by atrazine; whereas in August an increase in mortality was present
at higher atrazine concentrations, and adverse effects were again apparent
with as little as 100 ppm atrazine. Thus it is apparent that the toxicity
of atrazine to fiddler crabs is strongly influenced by the time of year when
the test is performed. Newell (1975) found that, in general, there was a
reduction in metabolism from February to July. Vernberg and Vernberg (1975),
working with the zoeae of u. pugnax and u. pugilator , found seasonal
32
-------
TABLE 6. EFFECTS OF ATRAZINE ON FOUR CLASSES OF FIDDLER CRABS IN SYSTEMS
WITH AND WITHOUT MARSH SOIL, AUG. 6, 1977 TEST*
Atr.
concn.
(ppm)
1
1
1
1
,000
,000
,000
,000
100
100
100
100
1
1
1
1
0
0
0
0
Sext
size
class
LM
LF
SM
SF
LM
LF
SM
SF
LM
LF
SM
SF
LM
LF
SM
SF
DFD,0^ Survive, %§
Soil Soil
69
12
34
53
100
100
100
100
100
100
100
100
100
100
100
78
34
41
81
84
69
50
69
100
100
100
100
100
100
100
100
50
0
0
25
100
100
100
100
100
100
100
100
100
100
100
100
75
0
0
25
50
0
25
50
100
100
100
100
100
100
100
100
Soil'0
0
0
0
0
100
100
100
100
100
100
100
100
100
100
100
100
25
0
0
25
25
0
0
25
50
100
100
75
100
100
100
100
SFD,%**
Soil
44
12
31
47
97
97
100
100
100
100
100
100
100
100
100
100
25
6
0
25
72
66
44
59
100
100
100
100
100
100
94
97
*Test continued for 8 days and was replicated four times
LM = large male, LF = large female, SM = small male, SF = small female.
DFD = death-free days expressed as a percentage of that possible.
§
Survive = percentage that lived 8 days.
u
S & PT = percentage that had survived and passed the escape-response test
on day 8.
SFD = stress-free days expressed as a percentage of that possible.
differences in survival at certain salinity-temperature combinations. In
all cases tested, zoeae of u. pugnax were more resistant to changes in
temperature and salinity in summer as compared to spring. Results with
U. pug-ilatop were more variable. The significance of these seasonal cycles
was not readily apparent, and it was concluded that field acclimation
phenomena influence the response of early-stage larvae. Teal (1959) found
that a 2-week acclimation period was adequate to acclimate u. pugnax to a
new thermal regime. Thus, it is concluded that seasonal differences in atra-
zine sensitivity that we observed cannot, be attributed to lack of acclimation or
to temperature differences since all crabs were acclimated for at least 2 weeks,
and all testing was conducted at the same temperature.
33
-------
TABLE 7. EFFECT OF ATRAZINE ON SMALL MALE CRABS IN SYSTEMS WITHOUT SOIL,
Nov. 7, 1977 TEST*
Atr. concn.
(ppm)
1,000
560
320
180
100
0
DFDf
(%)
99
91
90
90
95
97
Survived*
(%)
90
70
70
60
80
80
SFD§
(*)
100
100
100
100
100
100
*
Test continued for 30 days and was replicated 10 times.
DFD = death-free days expressed as a percentage of that possible.
*Survived = percentage that lived 30 days.
SFD = stress-free days expressed as a percentage of that possible.
TABLE 8. EFFECT OF ATRAZINE ON SMALL MALE CRABS IN SYSTEMS WITHOUT SOIL
March 20, 1978 TEST*
Atr. concn.
(ppm)
1,000
560
320
180
100
0
DFDf
(%)
100
TOO
100
100
100
100
Survived*
(*)
100
100
100
100
100
100
S & PT§
(*)
100
100
100
100
90
100
SFD#
(%)
0
0
3
17
28
99
*
Test continued for 9 days and was replicated nine times.
DFD = death-free days expressed as a percentage of that possible.
TSurvived = percentage that lived 9 days.
§
S & PT = percentage that survived and passed the escape-response test
on day 9.
#
SFD = stress free-days expressed as a percentage of that possible.
34
-------
TABLE 9. EFFECT OF ATRAZINE ON SMALL MALE CRABS IN SYSTEMS WITHOUT SOIL,
AUG. 1, 1978 TEST*
Atr. concn.
(ppm)
10,000
1,000
560
320
180
100
0
DFDf
(%)
23
54
37
53
42
54
100
Survived T
(«)
0
10
10
20
10
40
100
S & PT§
(%)
0
0
0
0
0
0
0
SFD#
(*)
0
0
0
0
0
0
0
Test continued for 9 days and was replicated 10 times.
DFD = death-free days expressed as a percentage of that possible.
^Survived = percentage that lived 9 days.
§
S & PT = percentage that survived and passed the escape-response test
on day 9.
SFD = stress-free days expressed as a percentage of that possible.
Feeding Studies.
The average percentage survival for four replications of crabs fed detritus
wetted with 0, 10-6, Or 10~4 M concentrations of atrazine solution were 85,
88, and 90, respectively. These values were not significantly different
according to Duncan's new multiple range test.
Although the detritus fed to the crabs was wetted with either 0, 10~6 M,
or 10-4 M atrazine solutions, the actual concentration in the detritus may
have been slightly higher or lower because both adsorption and absorption
of the atrazine were involved; some atrazine probably was metabolized by the
microflora during the course of the experiment.
Several investigators have reported adverse effects on behavior and
locomotion of fiddler crabs exposed to or fed pesticides (Odum et al., 1969;
Ward and Busch, 1976). Uoa pugnax fed detritus containing DDT and its
metabolites in concentrations ranging from 1.44 to 51.93 ppm developed im-
paired coordination and a sluggish behavior (Odum et al., 1969). An organo-
phosphorus insecticide (Temefos), at low concentrations, has also been shown
to adversely affect the ability of Uoa pugnax to respond to stimuli (Ward
and Busch, 1976). In contrast to those studies, Uca pugnax fed detritus
wetted with a 10-4 atrazine solution (22 ppm) did not exhibit any noticeable
differences in behavior or impairment in movement during the 20-day feeding
trial.
35
-------
The highest concentration of atrazine reported in waters of coastal
estuaries is 2.5 ppb (Anonymous, 1977). The lowest concentration at which
atrazine appears to have a detectable effect on the fiddler crab, Uoa pugnax,
from those areas is about 100 ppm and then only in late spring and summer.
Atrazine use is greatest during this period, and its maximum concentration
in coastal waters would be expected to coincide with the period at which
u. pugnax is most sensitive. From our data, we conclude that atrazine, even
if present at concentrations over 1000 times greater than expected as run
off from agricultural land, would have no significant direct effect on
u. pugnax of the size classes tested; however, effect of atrazine on the
larval stage was not investigated.
SUMMARY
Atrazine concentrations of 1000 ppm either killed or eliminated the
escape response ability (considered to be analagous to death) of Uoa pugnax
in laboratory experiments in August 1977. Adverse effects were observed at
concentrations as low as 100 ppm and the severity was dependent on size and
sex of the crab. However, in subsequent experiments, each with a new group
of crabs, the effects became smaller and smaller until a November experiment
when no deleterious effects were observed even at the 1000 ppm concentration.
Experiments in August 1978 confirmed the data obtained 1 yr earlier. Eco-
logical and physiological considerations of this seasonal variation in
response are discussed. Crabs fed for 20 days with detritus wetted with
10-4 M atrazine were not adversely affected. Crabs exposed to a single
application of 0, 100, 1000, or 10,000 ppm atrazine in the field and in
microecosysterns were adversely affected only by the 10,000 ppm rate. Toxicity
of atrazine to the larval stage of the crab was not investigated.
36
-------
SECTION 6
METABOLISM OF ATRAZINE IN SPARTINA ALTERNIFLOHA-DEIRUUS-
UCA PUGNAX FOOD CHAIN^
OBJECTIVES
One of the important food chains in the salt marsh starts with Spartina
alterniflora which, upon death, is decomposed to detritus, whichjn turn}is
consumed by fiddler crabs, Uca pugnax. Any toxicant absorbed by's. altevni-
flora may also pass along this food chain. The objective of this study was
to determine what happened to atrazine absorbed by s. altevniflora as it
passed along this food chain.
MATERIALS AND METHODS
Atrazine Metabolism During Conversion of Spartina to Detritus
This study used a model ecosystem that simulated the formation of
detritus from S. altevniflora in the salt marsh (Figure 11). Each system
consisted of a 12-cm diam buchner funnel containing marsh soil and connected
by Tygon tubing to a 1-L bottle which held 750 ml of seawater prepared from
Instant Ocean.^ The bottom of each funnel was covered by a nylon mesh filter
on which was placed first a 1.5-cm layer of washed sand and then a 1.5-cm
layer of soil collected in the marsh at Sapelo Island. The nylon mesh
filter and layer of washed sand prevented the movement of soil out of the
funnel when the reservoir bottles were raised and lowered to flood and drain
the soil surface. This system allowed for the quantification of atrazine
metabolites that were degraded, absorbed, and leached during the decomposition
of S. altemiflora.
S. alterniflora was collected from the marsh on Sapelo Island, GA,
transferred to Auburn, AL, and cultured in Hoagland's nutrient solution, and
transferred into six 1-L beakers containing ^C-labeled atrazine (24.9 uCi/mg),
i.e.,0.26 ppm or 1.2 x lO'^M. After 2 days the plants were removed from the
^c-atrazine solution, the roots rinsed, and the plants placed in atrazine-
4
Most of the material in this section is reprinted from the Journal of
Environmental Quality with their permission. The paper has not yet been
published.
Instant Ocean. Ward Natural Science Establishment Inc., Rochester,
NY 14601.
37
-------
anon DD n a
Enlarged view
Figure 11. Diagrammatic representation of experimental set-up used to make
detritus. Positions A, B, and C are the ones used to flood, drain, and
maintain a moist soil surface, respectively.
free Hoagland's solution. Three days later the leaves of the plants were
harvested, cut into 1-cm-long sections, and dried at 70 C to a constant
weight (34% of fresh weight).
The conversion of the dried S. altemiflora to detritus was studied in
six model ecosystems. About 5 g of the dried material was placed in each of
seven nylon mesh bags. One bag was placed on the soil surface in each sys-
tem and one was kept to determine the total radioactivity initially present
in the S. alterniflora leaves. The reservoirs were raised and lowered twice
daily during the 20-day study. The time between raising and lowering of
the reservoirs was adjusted so that the soil surface was flooded for
30 min twice each day. At other times the height of the reservoirs was
adjusted so that the soil surface was kept moist but not flooded. The
twice-daily complete wetting of the detritus and constant contact with moist
soil surface created conditions commonly found in the S. altemiflora
marshes. After 10 days the seawater was emptied from all reservoirs, saved
for herbicide and metabolite assay, and replaced with fresh seawater. After
20 days the bags of partially decomposed spartina, hereinafter referred to as
38
-------
detritus residue, were removed from the funnels, washed thoroughly with
deionized water to remove any material small enough to pass through the
0.01 mm holes in the nylon bags, dried to constant weight, and stored in
a freezer for later chemical and radiochemical analyses. The marsh soil
and sand from each buchner funnel, the seawater removed after 10 days, the
seawater removed at the end of the experiment, the water from the final wash
of the detritus residue, and the nylon bags and filters were also stored
under refrigeration for later analysis.
The procedures used to extract atrazine and atrazine metabolites from
the dried S. altemiflora and the detritus residue were essentially those
described in Section 3. The material was homogenized in a Waring blender in
80% (v/v) methanol and suction-filtered through a buchner funnel. The in-
soluble material was resuspended in 400 ml of 80% methanol, stirred overnight,
and filtered the next morning. The filtrates were combined, concentrated by
flash evaporation at 37 C, diluted with water, and washed with three 25-ml
portions of chloroform. Each fraction was brought to volume, and the amount
of radioactivity in each was determined by liquid scintillation spectrometry.
The insoluble fraction was dried at 70 C to constant weight, pulverized, and
50-mg portions radioassayed by suspending with Cabosil. Observed counts were
corrected for quenching, and the radioactivity was expressed as dpm per mg
of dry weight of the original S. alterniflora as percentages of the total
radioactivity in all samples.
The seawater removed from the reservoirs 10 and 20 days after the start
of the experiment was concentrated by flash evaporation, centrifuged at low
speed to remove insoluble material, and washed with chloroform. The water
and chloroform fractions were brought to volume and assayed for radioactivity.
The material removed by centrifugation was dried, and a 50-mg portion sus-
pended in a scintillation vial with Cabosil and radioassayed. Counts were
corrected and radioactivity expressed as described above. The water from
the final wash of the detritus was treated as described above except that it
was not centrifuged.
Any atrazine or atrazine metabolites which had been adsorbed to the
nylon bags or filters were removed by a 48-h extraction with 80% methanol.
The extracted material was concentrated, washed with chloroform, and radio-
assayed.
The soil was air dried and extracted by gently refluxing with 600 ml of
50% (v/v) methanol for 24 h. The methanol was filtered to remove suspended
soil, concentrated to near dryness by flash evaporation, and washed with
chloroform; and aliquots of the chloroform and water extracts were radio-
assayed.
Atrazine Metabolism by Fiddler Crabs Fed Detritus
The crabs used in these experiments were collected in early spring of
1977 in the marshes near Sapelo Island, GA, and transported in marsh soil on
ice to Auburn, AL. In Auburn the crabs were divided into 56 groups of two
or three crabs each in such a manner that the weight of crabs per group was
39
-------
approximately the same. One such group was placed in each of 56 clear
plastic boxes (18 x 13 x 15 cm). Each box contained one-half of an 8-cm-
diameter petri dish which served as a feeding station for the crabs. The
floor of each box was wetted with seawater initially and kept moist by re-
wetting as needed throughout the 20-day experiment. The boxes were cleaned
every 3 days. The lids were kept on the boxes except when the crabs were
being tended. This experimental set up was used in all of the crab feeding
experiments.
Two experiments, each using the same general procedure, were performed
to study atrazine metabolism in the mud fiddler crab. In both experiments
the atrazine concentrations used were not toxic to the crabs. In the fi* it
experiment, crabs were fed detritus residue (0.1 g/crab) made from the
leaves of spartina plants grown in ^4C-labeled atrazine. The detritus resi-
due used in the latter experiment was made by harvesting spartina leaves
from the marsh and converting them to detritus in a system similar to the one
described earlier. At the end of the 20-day feeding period, the crabs in
each group were sacrificed and stored under refrigeration. The crabs in
each group were analyzed by first grinding the crabs in a mortar with a pestle
in 80% methanol to pulverize their exoskeltons and were then homogenized in
a Waring blender. The extract was filtered, and the filtrate was concentrated
to near dryness and extracted with chloroform. The chloroform and water-
soluble fractions were brought to volume and assayed for radioactivity. The
insoluble portion was dried and a 50-mg portion suspended with Cabosil and
radioassayed. Observed counts were corrected for quenching, and the per-
centage of the total radioactivity in each fraction was determined.
RESULTS AND DISCUSSION
Atrazine Metabolism During Conversion of Spartina to Detritus
Table 10 summarizes the distribution of atrazine and atrazine metabolites
in the various components of the model ecosystem. The chloroform-soluble
extracts from plants, animals, and microorganisms exposed to atrazine have
been shown to contain predominantly unchanged atrazine and N-dealkylated
atrazine metabolites (Esser et al., 1975-and Shimabukuroet al., 1973), and the
water-soluble extracts to contain primarily hydroxy atrazine and glutathione-
conjugated metabolites of atrazine (Esser et al., 1975). The most conspicuous
feature of Table 10 is the relatively large amount of water-soluble atrazine
metabolites (78%) and small amount of chloroform extractable material (9%)
recovered. In the S. alterniflora leaves from which the detritus was formed,
about 38% of the total radioactivity was in the water-soluble fraction and
about 55% was in the chloroform-soluble fraction (Section 3). Thus during
the 20-day period, there was a decrease in the amount of chloroform-soluble
material (which contained any atrazine present) and an increase in the amount
of water-soluble atrazine metabolites. This is probably attributable to the
metabolism of atrazine by the microorganisms living in association with the
decomposing S. alterni-flora leaves.
During the 20-day experiment, the decomposing S. altem-i/lova was in-
undated twice daily with seawater thereby leaching atrazine and atrazine
40
-------
TABLE 10. AMOUNTS OF RADIOACTIVITY IN THE CHLOROFORM-SOLUBLE, WATER-SOLUBLE,
AND INSOLUBLE FRACTIONS IN PHYSICAL COMPONENTS OF THE SYSTEM
Physical
Component
Nylon bag
Nylon filter
Soil & sand
Water 10 day
Water 20 day
Final wash
Detritus
Total
Chloro-soluble
(dpm/
mg)f
1.02
.34
6.59
11.75
8.86
.28
4.72
33.60
(*)
52
30
16
9
8
16
6
9
Water-soluble
(dpm/
mg)t
.95
.81
33.36
11 7 . 80
97.25
4.85
23.90
278.90
(%)
48
70
84
91
92
84
33
78
Insol
(dpm/
mg)t
0.0
0.0
0.0
0.0
0.0
0.0
45.30
45.30
uble
U)
0
0
0
0
0
0
61
13
*
Total
(dpm/
mg)t
1.97
1.15
39.95
129.55
106.11
5.13
73.90
357.80
(%)
0.5
0.3
11.0
36.0
30.0
1.5
20.5
100.0
t
Total radioactivity recovered was about 90% of that initially supplied.
Expressed as dpm/mg of dried spartina leaves supplied to the system.
metabolites out of the decomposing leaves. About 2/3 of the initial total
radioactivity in the leaves was recovered in the two water samples (Water
10 day, Water 20 day) (the largest portion of this radioactive material was
soluble in water). A somewhat larger fraction of the total radioactivity was
recovered from the first 10-day than the second 10-day period (36.0% vs 30.0%),
and only a small amount was recovered in the final wash. This suggests that
at the end of the 20-day experiment, most of the readily soluble radioactive
material in the detritus had leached out. The amount of radioactivity ex-
tracted from the soil and sand layers was approximately 11% of the total
radioactivity recovered.
Weber et al. (1969) have shown that soils high in clay content and
organic matter are efficient in the adsorption of atrazine which reduces the
herbicidal toxicity of atrazine in such soils. Thus, in a marsh soil with
substantial amounts of organic matter and clay, a significant amount of
atrazine introduced into the marsh would be adsorbed on clay and organic
colloids,making it unavailable. In our study, we found that much of the
atrazine present in s. alterniflora was metabolized as the plant material
was converted to detritus. This metabolism plus adsorption of atrazine by
clays and organic matter would thus be expected to quickly detoxify atrazine
introduced into the food chain through S. altermiflora.
Atrazine Metabolism by Fiddler Crabs Fed Detritus
The percentage of [^C}-atrazine-derived radioactivity in the various
fractions extracted from the crabs and the detritus fed the crabs are given
in Tables 11 and 12.
41
-------
TABLE 11. AMOUNTS AND PERCENTAGES OF TOTAL RADIOACTIVITY IN THE VARIOUS
FRACTIONS EXTRACTED FROM RADIOLABELED DETRITUS RESIDUE* FED TO
FIDDLER CRABS AND IN THE FIDDLER CRAB THEMSELVES
Item
Chloroform-soluble
Water-soluble
Insoluble residue
(dpm) (% total) (dpm) (% total) (dpm) (% total)
Detritus
residue
Crabs
51,000
4,500
5.0
5.0
318,000
27,000
34.0
41.0
566,000
50,700
61.0
54.0
The detritus residue was derived from S. aUerniflora which was grown in
a solution containing ^^C ring-labeled atrazine. The amount of radio-
activity in the detritus residue was determined at the end of the 20-day
feeding period.
TABLE 12. AMOUNTS AND PERCENTAGES OF THE TOTAL RADIOACTIVITY IN THE VARIOUS
FRACTIONS EXTRACTED FROM DETRITUS RESIDUES* WETTED WITH RADIO-
LABELED ATRAZINE AND FED TO FIDDLER CRABS AND IN THE FIDDLER
CRABS THEMSELVES
Item
Chloroform-soluble
Water-soluble
Insoluble residue
(dpm) (% total) (dpm) (% total) (dpm) (% total)
Detritus
residue
Crabs
380,000
25,500
8.5
17.0
795,000
99,000
18.0
65.5
3,200,000
36,500
73.0
17.5
The detritus residue was derived from S. alterniflora leaves wetted with
14c ring-labeled atrazine. The amount of radioactivity in the detritus
residue was determined at the end of the 20-day feeding period.
The common result from the two metabolism studies is the increased
percent of the total radioactivity in the water-soluble fraction extracted
from the crabs compared to the detritus which they were fed. This increase
is due either to a selective absorption of water-soluble atrazine metabolites
by the crabs'intestinal mucosa or metabolism of chloroform-soluble atrazine
or atrazine metabolites to water-soluble metabolites by the crab or by the
flora inhabiting the crab's gut. Metabolism by the crabs seems probable
since Sesarma ainereum has been shown to metabolize atrazine (Section 7) and
because rats given atrazine orally have been shown to convert it to hydroxy-
atrazine (Bakke et al., 1972). Since fiddler crabs are filter feeders, no
42
-------
suitable method of introducing pure atrazine into the gut of the crab has
been developed. Furthermore, this would not eliminate the possibility of
atrazine metabolism by the microflora of the gut. Until a method of intro-
ducing pure atrazine into the crab's gut and the control of microfloral
activity is developed, the ability and extent to which fiddler crabs can
metabolize atrazine will remain undetermined. However, regardless of the
mechanism, it is apparent that the level of atrazine in the tissues of the
fiddler crab is lower than in the material on which it fed.
Concentrations of Atrazine and Atrazine Metabolites in the Detritus Food
Chain
Table 13 summarizes the amount and concentration of chloroform and
water-soluble radioactive material extracted from S. alterniflora leaves,
detritus residue derived from spartina leaves, and fiddler crabs fed such
detritus residue. As can be seen, there is a continuing decrease in the
concentration of chloroform-soluble atrazine or atrazine metabolites from
the leaves, to the detritus residue, and finally to the fiddler crab. Also,
there is a decrease in the concentration of water soluble metabolites; how-
ever, this decrease is not progressive as in the chloroform-soluble metabo-
lites. Thus,as atrazine is absorbed by the plant, decomposed to detritus
and detritus residue which is eaten by the fiddler.crabs, there is a marked
decrease in the concentration of atrazine and atrazine metabolites present.
Extrapolation of these results suggests that there would be a progressive
decrease in atrazine concentration from lower to higher trophic levels in
salt marshes exposed to atrazine.
TABLE 13. THE AMOUNTS AND CONCENTRATIONS OF WATER-SOLUBLE AND CHLOROFORM-
SOLUBLE ATRAZINE OR ATRAZINE METABOLITES IN SPARTINA LEAVES, IN
DETRITUS DERIVED FROM THE LEAVES, AND IN FIDDLER CRABS FED
DETRITUS DERIVED FROM THE LEAVES
Item Water-soluble Chioroform-soluble
Concentration Concentration
(ppm) (ppm)
Spartina leaves*
Detritus residue
Fiddler crabs
2.14
0.48
0.70
3.10
.085
.080
*
The values for s. alterniflora are from Section 3.
SUMMARY
Leaves from Spartina. altevni.flora plants grown with their roots in
solutions containing ^C ring-labeled atrazine were converted to detritus
in model ecosystems that simulated the salt marsh environment. Percentages
43
-------
of radioactivity in chloroform-soluble, water-soluble, and insoluble
materials in the leaves were 55, 38, and 7, respectively. Twenty days later
these values for detritus were 9, 78, and 13, respectively. Thus, there
was a decline in the percentage of radioactivity in the chloroform fraction
which contains atrazine and nontoxic metabolites and a concurrent increase
in the water-soluble fraction that contains only nontoxic metabolites. The
fiddler crab, when fed detritus labeled with atrazine, further decreased the
percentage of the chloroform-soluble atrazine or atrazine metabolites. Radio-
activity originally present in either atrazine or atrazine metabolites fed
to fiddler crabs was concentrated in the water-soluble extract from the
crabs,suggesting either selective absorption through the gut or metabo-
lism of the chloroform-soluble form(s) to water-soluble material by the
fiddler crabs or the crabs' enteric flora. Atrazine and atrazine metabolite
concentrations were reduced from lower (detritus) to higher trophic levels
(fiddler crabs) in salt-marsh microecosystem.
44
-------
SECTION 7
METABOLISM OF ATRAZINE BY SESARMA CINEPEUM
OBJECTIVES
Box crabs Sesarma cinereum, feed directly on the leaves of cordgrass,
Spartina alterniflora, and by so doing ingest toxicants taken up by 5.
alterniflora. The objectives of this investigation were: 1) to determine
whether altrazine present in S. altevniflora leaves was toxic to box crabs,
and 2) to compare the concentrations of atrazine and atrazine metabolites
present in leaves fed box crabs with those in the crabs and in crab feces.
MATERIALS AND METHODS
Box crabs and cordgrass, Spartina alterniflora , were collected from the
marsh on Sapelo Island, GA, and transported in marsh soil to Auburn, AL.
Each crab was washed in seawater and placed in an individual plastic con-
tainer approximately 20 cm x 20 cm x 18 c. Cordgrass plants were grown and
maintained in a controlled environment chamber as described in Section 3.
Uniform lots of four plants were transferred to six 1-L beakers containing
300 ml of Hoagland's nutrient solution, 0.2 ppm (1.2 x 10-6M),and 2 yCi of
ring-labeled [14C] atrazine (24.9 yCi/mg) (Ciba-Geigy, Greensboro, NC). After
2 days, the plants were removed from the atrazine solution, the roots rinsed
free of radioactive material, and placed in an atrazine-free Hoagland's
solution. After 3 days in the atrazine-free solution, leaves were harvested
and cut into 5 mm x 2 mm sections; and 100 mg of these sections were placed
in each of the plastic containers with the crabs (Daiber and Crichton, 1967).
Each container was covered with a transparent lid to prevent loss of water
by evaporation. The crabs were maintained in a growth chamber with 14 h of
diffused light (1.8 klux) at 22 C and a 10-h dark period at 20 C. Crabs were
removed from the containers daily and placed in the seawater (1/2 strength
Instant Ocean, Carolina Biological Supply House) to allow them to refill their
gill chambers. Feces were collected from the containers with a rubber police-
man, s. alterniflora leaf fragments were removed, rinsed with deionized water,
combined, and kept frozen until extraction and analysis. The crabs were then
returned to their containers with fresh s. alterniflora leaf sections. Two
experiments were conducted, the first with 50 crabs and the second with 68.
Most of the material in this section is reprinted from the Journal of
Environmental Quality with the permission of the Journal of Environmental
Quality. Volume and page numbers have not been established at this time.
45
-------
In each case, half the crabs were fed s. alterniflora leaf sections from
plants grown in atrazine-free solutions (controls) and half with leaves from
plants grown in solutions containing [^Q-labeled atrazine.
At the end of the 10-day feeding period, the box crabs were frozen and
homogenized in 80% aqueous methanol (10 ml/g of fresh weight of crab). The
homogenate was filtered, and the filtrate was concentrated by flash evapora-
tion at 37 C, diluted to 50 ml with water, and partitioned with chloroform.
The chloroform and aqueous fractions were concentrated separately, brought
to volume, and the radioactivity in each fraction was determined by liquid
scintillation spectrometry. The insoluble residue was refluxed for 2 h
with 25 ml of 0.5 N HC1 at 70 C, cooled, and filtered. The filtrate was
partitioned three times with 50 ml of ether, and combined washes were dried
over sodium sulfate. The aqueous phase of the hydrolysate was dried by
flash evaporation to remove the acid and then radioassayed. Radioactivity
in the insoluble residue remaining after refluxing was estimated by grinding
a portion of the material to a fine powder, suspending it in Aquasol (Beckman)
scintillation cocktail containing Cab-0-Sil (Beckman), and radioassayed as
before. Extraction and fractionation of various components of the S. alterni-
flora leaf sections and crab feces were conducted essentially as described
for box crab except that the 80% methanol-insoluble residue was not hydrolyzed.
All radioactive determinations were corrected for quenching.
Radiolabeled components of the chloroform fractions were separated by
thin-layer chromatography (TLC) using 20 cm x 20 cm glass plates coated with
a 250 ym-layer of silica gel HF-254 and activated for 1 h at 100 C. The
plates were developed twice in benzene-acetic acid (50:4, v/v), and the
metabolites were identified by comparing their RF values with those of
authentic standards (Ciba-Geigy Corporation, Greensboro, NC, U.S.A.). Atra-
zine and its N-dealkylation products were visualized on the TLC plates with
UV light (254 nm). Radioactivity in each component was determined by
scraping the radioactive areas from the plates and radioassaying.
Radioactive components of the aqueous fractions were separated by ion-
exchange chromatography. Portions of these fractions containing approxi-
mately 150,000 dpm were concentrated to dryness by flash evaporation, dis-
solved in 2 ml of 0.2 N (pH 2.1) pyridine-acetate buffer, and applied to a
1 x 80 cm water-jacketed column of AG 50 W-X2 Aminex resin (200-325 mesh)
(Bio-Rad Laboratories) at 15 C. The column was washed at 0.3 ml/min with a
pyridine-acetate buffer gradient developed from three chambers and the frac-
tions collected. The radioactivity in each fraction was determined. Prepara-
tion of the buffers and regeneration of the ion-exchange resin was described
by Schroeder et al. (1962). The first chamber of the gradient device con-
tained 300 ml of 0.2 N buffer at pH 3.1, and the second and third chambers
contained 300 ml of 2.0 N buffer at pH 5.0.
RESULTS AND DISCUSSION
Response of Box Crabs to Dietary Atrazine
To determine whether atrazine is toxic to box crabs, male and female
46
-------
box crabs of different sizes and stages of maturity were fed fresh s. alterni-
flora leaf strips either from plants grown in nutrient solution containing
atrazine or in atrazine-free solution. Forty-five crabs were used in pre-
liminary study, and in two subsequent experiments, 50 and 68 crabs were
tested. Preliminary work indicated that the crabs consumed about 20 mg of
cordgrass leaves per day per g body weight containing about 0.015 ppm atra-
zine and 0.046 ppm of its metabolites. At the end of the 10-day feeding
period, about 0.4 ppm atrazine and its metabolites accumulated in the body
of the crab and about 0.21 ppm was excreted with feces. During the feeding
period, 11.7% and 12.4% of the crabs died in the control and the treatment
groups, respectively. This was probably due to the stress conditions im-
posed on the crabs during the test. No dissimilar behavior was observed
between crabs fed atrazine-treated leaves and crabs in the control group.
Thus, it appears that atrazine, at the level present in the S. alterniflova
diet, was not toxic to the crabs.
Atrazine and Atrazine Metabolites in S. alterniflora
The plants used in this study were grown as described in Section 3
except that the plant material was collected 3 days after transferring the
plants from solutions containing ['4Q-atrazine to atrazine-free solution.
Leaf material used in the feeding experiments contained 56%, 39.5%, and 4.5%
radioactivity in the chloroform, aqueous, and 80% methanol-insoluble fractions,
respectively (Table 14). Although the S. alterniflora used for feeding was
stored at 4 C prior to use and a new supply of previously stored leaves were
given to the crabs daily over the 10-day feeding period, it was necessary to
know the changes that occurred in the relative proportions of radioactivity
in the plant material over a 10-day period (equivalent to 13 days after
transfer of the plants to an atrazine-free nutrient solution). Data in
Section 3 indicate that after 10 days the leaves contained 36%, 57%, and
7% radioactivity in the chloroform, aqueous, and 80% methanol-insoluble
fractions, respectively. This shows a slower shift in radioactivity from
the chloroform to aqueous fractions in the S. alterniflora compared to
crab (Table 14).
Metabolism of Atrazine and its Metabolites in the Box Crab
Chloroform Fraction—
The relative proportions of radioactivity in the various fractions of
the box crabs and its feces were quite different from those of the plant
material fed to the crabs initially and after 10 days incubation (Table 14).
The chloroform fraction of the crab and feces extracts contained 6.6% and
25% of the total radioactivity, respectively. The 5. alterniflora leaves
fed to the crabs contained 24% of the radioactivity as atrazine and after
10 days only about 13% (Section 3). Atrazine was 1.2% and 0.5% of the
radioactivity in the crab tissue and feces, respectively, at the end of the
feeding experiment (Table 15). It appears that atrazine is metabolized more
rapidly by the crab than by s. alterniflora since it represents only 1.7%
in the crab and feces after an equivalent period of time. It cannot be
determined unequivocally from these data whether atrazine metabolism occurs
47
-------
TABLE 14. RADIOACTIVITY IN CHLOROFORM, AQUEOUS, AND INSOLUBLE FRACTIONS OF
S. ALTERNIFLOM, CRABS, AND FECES*
Fractions
Chloroform
Aqueous
Insoluble
S. alterniflora
3 days?
56.0
39.5
4.5
13 days?
36.0
57.0
7.0
Crabsf
6.6
86.2
7.2
Feces
25.0
51.0
24.0
*Leaves from the S. alterniflora were fed to the crabs and feces collected
from the crabs.
^Each value is the mean of two experiments and is expressed as the percentage
of the total radioactivity in the 80% methanol extract.
$s. alterniflora leaves were collected 3 days after transferring the plants
from atrazine-containing solution to an atrazine-free nutrient solution.
Leaves were analyzed at harvest and after 10 days, the length of time that
feeding experiment lasted. The 13-day data are from Section 3.
mainly by enzymatic, chemical, or microbial action in the gut or in the crab
tissue after absorption.
Radioactive components other than atrazine in the chloroform fraction
of crab and feces extracts included the N-dealkylation products of atrazine
and relatively polar unidentified substances (Table 15). Both the box crab
and its feces contained the three expected N-dealkylation products; 2-chloro-
4-amino-6-isopropylamino-s-triazine, 2-chloro-4-amino-6-ethylamino-s-triazine,
and 2-chloro-4,6-diamino-s-triazine (Figure 12). 2-Chloro-4-amino-6-isopro-
pylamino-s-triazine was the principal N-dealkylation product found in the
crab and its feces after 10 days of feeding. Whether this is due to selective
absorption of this metabolite relative to others or preferential removal of
the ethyl group cannot be determined from these data. However, 2-chloro-4-
amino-6-isopropylamino-s-triazine is the principal dealkylation product of
most systems including S. alterniflora (Section 3). N-dealkylation appears
to be the principal way in which atrazine is degraded and partially detoxified
by soil fungi (Kaufman and Kearney, 1970), cotton, pea, and soybean (Shimabu-
kuro et al., 1966; Shimabukuro, 1967a) and seems to be a universal reaction
in higher plants, animals, and microorganisms (Shimabukuro et al., 1970).
N-dealkylation products represented a quantitatively minor amount of the
atrazine metabolites in the crab and feces after the 10-day feeding period
(4 and 11% of the total radioactivity, respectively), and it would appear
that N-dealkylation may represent a relatively important pathway in the
metabolism of atrazine in box crabs. However, interpretation of the impor-
tance of this pathway in crabs is complicated by the fact that in addition
48
-------
TABLE 15. RADIOACTIVITY IN ATRAZINE, N-DEALKYLATION PRODUCTS, AND UNINDENTI-
FIED POLAR COMPONENTS OF THE CHLOROFORM FRACTION OF EXTRACTS FROM
S. ALTERNIFLOEA LEAVES*
Atrazine or
metabolite
Percentage
S. alterniflora
of total t
Crabs
Feces
Atrazine
2-chloro-4-amino-6-
i sopropyl ami no-s-tri azi ne
2-chloro-4-amino-6-
ethyl ami no-s-tri zai nes
23.7
20.4
4.2
1.2
2.2
1.4
0.5
7.2
2.0
2-chloro-4,6-diamino-
s-triazine
unidentified
2.8
4.8
0.4
1.4
1.6
13.8
The s. alterniflora was grown in [^C]-atrazine solution. These leaves were
fed the crabs and feces collected from the crabs.
Each value is the mean of two experiments with three replications each and
is expressed as a percentage of the total radioactivity in the 80% methanol
extract.
to atrazine, polar metabolites (see below) of the aqueous fraction can also
originate from the N-dealkylation products. That is, N-dealkylation may
proceed at a higher rate than is apparent because the products can be
rapidly converted to water-soluble substances which do not accumulate in the
chloroform fraction.
In addition to atrazine and its N-dealkylation products, the chloroform
fraction of crab and feces extracts contained unidentified, relatively polar
radiolabeled substances. N,N-bis(4-ethylamino-6-isopropylamino-s-triazine-2)-
cystine was identified in the chloroform fraction of extracts of atrazine-
treated sorghum (Shimabukuro et al., 1973), and an unidentified substance(s)
with similar TLC migration properties was detected in S, alterniflora (Section
3). Similar chloroform-soluble substances comprised 20% of the radioactivity
in the chloroform fraction from crab and 55% of those in feces (Table 15).
Since the s. alterniflora leaves fed to the crabs contained only 8.6% of
these substances, it appears that reactions leading to the production of polar
metabolites from atrazine or its N-dealkylation products were occurring in
the gut of the box crab. It cannot be determined if these substances in the
crab were formed and absorbed from the gut or were also produced in the crab
tissue.
49
-------
ABC
ATRAZINE
ISOPROPYL AM1NO-
ETHYLAIilNO-
»** DIAMINO-
. I
o
a
-------
the box crab and converted to water-soluble substances, possibly via the
hydroxylation or conjugation reactions. However, the dechlorination reaction
is not considered important in detoxification of atrazine in those animals
tested (Shimabukuro et al., 1971).
Although the relative amounts of radioactivity in the aqueous fraction
of cordgrass remained relatively constant between 10 and 20 days after trans-
ferring the plants from a nutrient solution containing ^C-labeled atrazine
to an atrazine-free solution (Section 3), considerable changes in the radio-
labeled components within this fraction occurred during this time
(Section 4). Passing portions of the aqueous fractions from crab, feces,
and- the S. alterniflora fed to the crabs through a cation-exchange column
showed that the relative proportions of the components differed among
samples (Figure 13). From previous studies (Section 4), it was anticipated
that approximately 42% of the radioactivity would accumulate in the first
two column fractions after 10 days incubation under the feeding experi-
mental conditions. While this appeared to be true for the feces, radio-
activity was more evenly distributed between the 6 column fractions for
the crab extract, with about 28% in fraction 3. This further suggests that
reactions unique to the crab involving atrazine, N-dealkylation products,
and possibly intermediates in the conjugation pathways may have occurred.
Studies of column fractions of cordgrass extracts showed that each fraction
is a mixture of two to several individual atrazine metabolites (Section 4).
Of the metabolites isolated from S, alterniflova leaves, about half
yielded amino acids such as glycine and glutamic acid on hydrolysis, sug-
gesting a possible link to glutathione conjugation (Section 4). In excised
sorghum leaves, the glutathione conjugation pathway operates almost exclu-
sively (Lamoureux et al., 1972), and the contribution of this pathway to the
detoxification of atrazine by sorghum leaves may approach 87% (Lamoureux
et al., 1973). Glutathione conjugation is the initial reaction leading to
mercapturic acid biosynthesis, a pathway recognized as a means for detoxifi-
cation and excretion of foreign compounds in mammals (Boyland and Chasseaud,
1969), birds (Wilt and Leeuwaugh, 1969), and insects (Cohen et al., 1960).
Almost complete detoxification of atrazine and the abundance of the water-
soluble metabolites (86%) suggests that, as in other animals, the metabolism
of this herbicide in crabs may occur via the glutathione conjugation pathway.
SUMMARY
The metabolism of atrazine [2-chloro-4~(ethylamino)-6-(isopropylamino)-
s-triazine] in box crabs (Sesarma cinevewn} was determined. Leaves of
smooth cordgrass (Spartina altemiflora Loisel) collected from plants grown
for 2 days in nutrient solution containing f^C]-atrazine, followed by 3 days
in an atrazine-free nutrient solution, were fed to box crabs for 10 days.
No significant effects of atrazine on the behavior or survival of crabs were
found. At the end of the 10-day feeding period, box crabs and their feces
were extracted with 80% methanol, and the extracts were concentrated and
partitioned with chloroform. Radioactivities in the chloroform, aqueous,
and 80% methanol-insoluble fractions (remaining crab material) of the extract
were 7%, 86%, and 7% for crabs and 25%, 51%, and 24% for feces, respectively.
51
-------
Only 1.2% and 0.5% of the total radioactivity in the crab and feces, respec-
tively, was atrazine, compared to 24% in the s. altemiflora used as a food
source. This indicates that atrazine is metabolized in the crab. The accum-
ulation of water-soluble metabolites in the crab suggests that, as in other
animals, glutathione conjugation or a comparable pathway is responsible for
the almost complete degradation and detoxification of atrazine in these
organisms.
52
-------
SECTION 8
EFFECT OF ATRAZINE ON MARINE DIATOMS7
OBJECTIVES
Since algae, in particular the marine diatoms, make a major contribution
to the food supply in the S. alterniflova salt marsh ecosystem, it is im-
portant to know the effect of atrazine on this group of plants. The objec-
tives of this investigation were to determine the effect of atrazine on:
1) cell numbers, 2) chlorophyll synthesis, 3) carbon fixation, and 4) species
diversity of marine diatoms common in the S. alterniflor-a salt marsh when
grown in monocultures, microecosysterns, and in the field.
MATERIALS AND METHODS
Culture Studies
Guillard's (1962) f/2 nutrient solution with 26.5 ppm NH4C1 adjusted to
a salinity of 20% with Instant Ocean and buffered with 500 ppm of tris buffer-
HC1 at pH 7.4 was used. Culture conditions were 14-h days with 4 klux of
light at 25 C and 10-h nights at 22 C. Cultures were constantly shaken at
160 cycles/min. Cultures of Thalassiosira fluviatilis Hustedt and Nitzsehia
sigma Grun. were isolated from marsh soil samples from the bank of Cabretta
Creek, Cabretta Island, GA, by the methods of Pringsheim (1946) as modified
by Trainor (1978).
Effects of a 7-day exposure to 0, 10"?, 10~6, and 10"5 M atrazine on
a mixed culture of T. fluviatilis and N. sigma were investigated. There were
14 replications for each concentration, and cultures were maintained in
1.5- x 12.5-cm culture tubes. Cultures were initiated by taking 1 ml from
an actively dividing culture and adding it to 4 ml of nutrient solution
containing sufficient atrazine to give the desired final concentration.
After the 7-day exposure period, chlorophyll content was determined for
three randomly selected cultures from each atrazine concentration as des-
cribed by Yentsch and Menzel (1963). Another three randomly selected tubes
from each concentration were used to determine cell numbers with a hemocy-
tometer using the methods of Guillard (1973). Each of the remaining tubes
had 2 ml of the appropriate atrazine solution containing a constant amount
of 14c (supplied as ^co2) added. The 8 remaining tubes of each concentration
Most of this material is from a paper accepted for publication in
Estuaries and is reproduced here with their permission.
53
-------
were divided into two sets of four each, and one set was held in the dark
for 2 h and the other in the light for 2 h. After the 2-h incubation period,
the cells were separated by filtration, washed, and the fixed 14c measured
with a liquid scintillation spectrometer.
Microecosysterns
In February 1977 and again in April 1977, atrazine effects on edaphic
algae were monitored in model ecosystems at Auburn, AL. A 10-cm layer of
soil obtained from the creekbank zone of Cabretta Creek was placed in 35-cm
diam plastic tubs. Tidal action was simulated by the raising (high tide)
or lowering (low tide) of 9-L buckets containing 7.6 L of water adjusted
to a salinity of 20% with Instant Ocean. The plastic tubs were connected
to the moveable buckets by rubber tubing. Water in the buckets was con-
tinually aerated by bubbling air. The "tide machine" employed for raising
and lowering the water reservoirs was designed by Everest and Davis (1977)
and was capable of simulating tidal flow with a high degree of accuracy.
Ten fiddler crabs, Uaa pugnax, were added to each system to prevent blue-
green algal mats from becoming the dominant vegetation.
Atrazine was introduced into four of the eight water reservoirs at a
concentration of 10~5 M (2.2 ppm or 0.16 g/m?). Systems were flooded with
herbicide-containing seawater "twice" daily for 5 consecutive days. Water
in the reservoirs was changed following the last herbicide application,
and the system was allowed to cycle for several days during which time the
edaphic algae were sampled. The motile algal population, primarily diatoms,
was sampled by placing 10 sections of double layer lens paper, 2.5 by 2.5 cm,
on the soil surface of each system at low tide (Eaton and Moss, 1966). Fol-
lowing removal of the lens paper, the nonmotile and/or nonsurface edaphic
algae were sampled for chlorophyll analysis with 2.3 cm diameter PVC coring
tubes. All algal samples were held in icebox coolers until initiation of
the various assay procedures. Core samples were frozen until the chloro-
phyll was to be extracted (Gallagher, 1971).
Chlorophyll was measured fluorometrically in 3 of the 10 lens
paper samples by the methods of Yentsch and Menzel (1963). Core samples
were sectioned horizontally with a thin spatula into two segments, 0 to 2 mm
and 2 to 5 mm, and chlorophyll was determined similarly.
Diatom cell number was assayed by boiling three of the lens paper
samples from each microecosystem in nitric acid and then identifying the
cleaned diatoms (Patrick and Reimer, 1966). Species diversity (H1) and the
structure of diatom communities (SIMI) were computed with methods des-
cribed by Sullivan (1975). In the February experiment, 100 values from
each of the four replicates for each of the two treatments were enumerated.
Species diversity (H1) was calculated and the results pooled for presentation.
Diatom community structure (SIMI) was calculated with the pooled data for
the 400 enumerated values. In the April experiment, one system was selected
at random from each of the two treatments and 100 values enumerated. Results
of species diversity determinations were tested for statistical differences
54
-------
using the Student T test. No statistical tests were applied to the SIMI
values. Photosynthesis rates were determined by placing four of the 10 lens
paper samples from each microecosystem on 3- by 3-cm glass slides. The lens
paper was spread as evenly as possible, and a few drops of distilled water
added to prevent desiccation. The lens paper samples were placed in an air
tight chamber (Darley et al., 1976) and incubated for 30 min in the light
(3 reps) or dark (1 rep) with a known amount of 1^C02- Light intensity was
4 klux from cool white fluorescent lamps, and the temperature was 25 C. At
the end of the exposure period, samples were exposed to HC1 fumes to remove
any unfixed 14c, placed in scintillation vials with cocktail, and vortexed
prior to determining ^C incorporation by liquid scintillation spectrometry.
Carbon fixation estimates were obtained by the formula of Darley et al.
(1976).
Field Studies (Tubs)
Plastic tubs 36 cm in diam and 29 cm deep were filled to a depth of
15 cm with surface soil obtained from the area previously identified. The
tubs were partially buried in the soil so that both the inside and outside
soil surfaces were at the same level. A 3-cm diameter hole in the side of
the tubs at ground level provided tidal water movement into and out of the
systems. Tubs were positioned on a level area of creekbank marsh in Southend
Creek, Sapelo Island, GA, in June 1977. Edaphic algal populations were
allowed to habituate to their new surroundings for 7 days prior to herbicide
application. For herbicide application, the tubs were removed from the
creekbank and placed on high ground. The systems were then flooded twice
daily for 5 consecutive days with either 0 or 10~5 M atrazine solution in
7.6 L on nonfiltered water collected from the nearby creek. Each treatment
was replicated four times. Flooding periods were from 1.5 h before to 1.5 h
after high tide as given in the tide tables for the Savannah River entrance
to the marsh. After the 5-day treatment period, the tubs were returned to
their original positions along the creekbank, and 10 u. pugnax were added
to each system.
Edaphic algal productivity was measured as described for the micro-
ecosystems. Samples were taken both inside and outside of the tubs to
determine not only the effects of atrazine but also any effect induced by
the tubs. Cell number, species diversity, and SIMI values were determined
as previously described. A total of 900 values were enumerated: 300 from
each of the three collection zones. Chlorophyll levels were not determined
in this experiment.
Field Studies (Metal Cylinders)
To help insure the continued presence of fiddler crabs in test plots,
12 aluminum cylinders 1.8m in diam and 80 cm high were pressed 15 cm into
the soil surface, leaving 75 cm above ground. A single 5-cm diameter hole,
covered with 2 thicknesses of hardware cloth, at ground level permitted
normal exchange of tidal water and prevented the movement into and out of
the system of all but the smallest fiddler crabs. Three of the 12 plots
were chosen at random for each atrazine concentration.
55
-------
Atrazine was applied as a spray in 500 ml of water at concentrations
of 0, 100, 1000, and 10,000 ppm (0.0, 0.05, 0.5, and 5.0 g/m2), respectively.
The 0.5 g/m2-rate is similar to the 0.22-0.45 g/m2 used for weed control in
corn. The methods of Van Raalte et al. (1974) were modified for determining
14c-uptake by edaphic algae following addition of atrazine. Cores of sur-
face soil were taken with PVC coring tubes as previously described. Effects
of the sides of the cylinders were minimized by taking cores at least 25 cm
from them. The top 0.5 cm of each core was placed in an air-tight 75-ml
light or dark jar and inoculated with 10 ml of distilled water containing
a known amount of NaH^cOs. Salinity was adjusted to 20%, with NaCl and
the pH adjusted to 8 with NaOH. Samples were shaken vigorously to expose
the maximum surface area of each core to incident radiation and incubated
in afternoon sunlight for 2 to 4 h in a shallow-water bath. Samples were
killed with 0.5 ml of concentrated HC1. The samples were frozen to aid in
cell rupture prior to lyophilization; 10 ml of concentrated nitric acid
were added to each dried sample, and the samples were allowed to digest
overnight. A 0.1-ml aliquot was pipeted into scintillation vials containing
15 ml of Aquasol and radioassayed for 10 min or 10,000 cpm with a liquid
scintillation spectrometer.
RESULTS
Culture Studies
Atrazine, at the highest concentration tested, 10-5 M, significantly
reduced the chlorophyll level, rate of photosynthesis, and cell number of
N. sigma and T. fluviatilis following a 7-day exposure (Figure 14), while
the lowest concentration, 10-7 M, had no significant effect on any paramanter
assayed. The results with the intermediate concentration, 10-6 M, were
variable. There was no effect on chlorophyll content for either species,
a reduction in cell numbers only in T. fluviatilis, and a reduction in
photosynthesis rate in both cultures.
Mi croecosys terns
In the first experiment (February 1977), primary productivity was
reduced from 191.4 to 29.0 mg C/m2/h in microecosysterns flooded for 5 days
with 10-5 M atrazine (Table 16). Chlorophyll content was also reduced in
the 0-2 and 2-5 mm soil layers. Cell numbers and chlorophyll content of
the surface algae (lens paper harvested) were not affected. In the April
experiment, primary productivity was again reduced. Chlorophyll content
was reduced in the surface algae but not in the soil segments (Table 16).
There was no clearly identifiable effect on diatom species diveristy in
either study (Table 17), but atrazine increased the numbers of Cymatosir-a
belgioa in both experiments (Figure 15). Atrazine had little effect on
community structure in either experiment as evidenced by SIMI values
(Table 17).
56
-------
-T. Fluviatilis
Chlorophyll Cell number Photosynthesis Chlorophyll
HOp content content
a
• -N. sigma
Cell number Photosynthesis
o
IOO
90
80
1 70
c
o
u
,_ 60
o
£ 50
u
o>
°- 40
30
20
10
-
.
•
-
-
-
0
0
a
'
<
l(57
a
d
'
i66
b
Ft3
|.:-:;'J
ICT
0
-
0
PW
o'
b
10*"
c
r™i
l-':i
l-"'i
io5
a
0
a
•n*
6'
c
166I05
a
•y?
:'•
?':;j
"•.-.•
v
•" .
;;-:
0
a
-_.
•>"<
•'..;
.- ••
;..;-
••;.
'•;•:•:
•'\
"';
10'
o
f.
'I
Ob
b
I • I
1 1
t \
I0b
a
•.:"
::;'
-:"•
0
a
•':•:
•"_;
10'
a
I0b
b
I0"b
a
0
^
10'
b
IOb
c
rrm
IC)^
Atrazine concn. (M)
Figure 14. Effect of 7 days of exposure to various atrazine concentrations
on the chlorophyll content, cell number, and rate of photosynthesis of
laboratory cultures of Thallasiosira fluviatilis and Nitzsoh-La sigma. Within
a given measurement, columns having the same letter (a, b, or c) are not
significantly different at the 5% level according to Duncan's new multiple
range test.
Field Studies (Tubs)
Five days of treatment by flooding with 10~5 M atrazine solutions did
not significantly reduce the number of cells counted 1 or 7 days after com-
pletion of treatment; the actual counts for the controls and treated areas
averaged 2.81 x 108 and 2.85 x lQ8/m2, respectively, on day 1 and 13.5 x
108 and 7.1 x 108, respectively, on day 7. Cell numbers inside the en-
closures did not vary significantly from outside the enclosures. Rates of
carbon fixation (mg C/m2/h) were measured 7 and 18 days after treatment.
Carbon fixation was significantly decreased by the treatment at both of
these sampling times. The average fixation rates for the controls versus
the treated areas were 577 and 155, respectively, on day 7 and 334 and 161,
respectively, on day 18. Average rates of photosynthesis were six times
lower outside the containers than for the controls inside the containers
7 days after initiation of the experiment.
57
-------
TABLE 16. EFFECTS OF ATRAZINE ON PHOTOSYNTHESIS, CELL NUMBERS, AND
CHLOROPHYLL CONTENT OF DIATOMS IN MICROECOSYSTEMS
Feb.
Apri 1
Herb.
Concn
(M)
0
10-5
0
10-5
Photo.
mg C per
mZ/h
191.4 a*
29.0 b
283.4 a
30.3 b
Cells
per
m2x!08
12.2 a
11.2 a
28.6 a
6.5 a
Chlorophyll
Surface
(mg/m2)
2.5 a
2.1 a
4.8 a
1.6 b
0-2 mm
(mg/m2)
77 a
43 b
87 a
62 a
2-5 mm
(mg/m2)
54 a
35 b
61 a
69 a
Means in the same column for the same date followed by a common letter
are not statistically different at the 5% level as determined by Duncan's
new multiple range test.
35
30
25
u
c
.E 20
0
T3
£ 15
o>
u
Of
°- 10
5
0
1 Cymatosira belgica
2 Melosira sp.
i
-
—
-
:
•
:
:
i
:
•
2
:
4
3
5
n
0
2
3
"
4
1
.
.,
5
fl
3
2
4
3 Thalassiosira fluviatilis
4 Nitzschia sp.
5 Navicula sp.
3
2
:
|:|
|
1
P;
4 ?i
3
1
5
1
5
i 3
Izlfi
"71
:j
i
:):
:|:
•i;
5
2
1,4
X
:•:
Control Atrazine Control Atrazine Control Atrazine Outside
February April
Microecosystems
-Field Plots-
Figure 15. Percentage dominance for each of the five most common diatom
species in atrazine-treated and control areas in microecosystems and in
the field.
58
-------
Atrazine did not significantly affect species diversity (H1) in the
microecosystems or in the field (Table 17). However, atrazine did consis-
tently increase the number of C. belgica (Figure 15). Atrazine appeared to
cause an increase in Naviaula sp. in the field but not in the microecosystems,
There were larger numbers of C. belgiaa and Navioula sp. in the untreated
controls inside than outside the tubs. In the microecosystems, the matrix
of similarity values (SIMI) between atrazine-treated and untreated systems
was 0.838 for the February experiments and 0.906 for the March experiments.
A value of 1 results when two communities support the same species in the
same relative abundance and 0 when the communities are completely different.
In the field, when atrazine-treated areas inside the tubs were compared
with untreated areas outside the tubs, the SIMI-value was 0.786 again, indi-
cating that atrazine did not greatly alter community structure. However,
when untreated areas inside the tub were compared with untreated areas
outside the tub, the SIMI-value was 0.506. This relatively low value indi-
cates that the presence of the tub had a significant effect on community
structure which in turn means that some other means should have been devised
to determine the effect of atrazine in the salt marsh. The use of larger
containers and a longer delay time after transfer of the soil into the
container might alleviate this problem.
TABLE 17. DIATOM SPECIES DIVERSITY (HABITS/INDIVIDUAL) AND NUMBER OF
SPECIES (S) FOR ATRAZINE-TREATED AND UNTREATED MICROECOSYSTEMS
AND IN THE FIELD*
Value Herb. Microecosystems
Field (tubs).
February
April
Inside
Outside
H1
H1
n
S
C
o
No
Vpc
No
Voc
I ci>
2.99 + 0.42
Q Rl + n AA
16.20 + 1.50
7i nn 4- fi ^n
C. 1 . UU T D . OU
3.80
? A?
25.00
Oo no
3.85 + 0.53
3 A.9 + 0 ~\7
24.70 + 7.50
99 oc 4. T en
3.66 + 0.17
24.70 + 2.50
Values for the microecosystems in the February experiments are averages of
four replications (100 values each) +_ 1 standard deviation (SD), and for
the April experiments are for one replication with 100 values used. Values
for the field plots are averages of three replications (100 values each)
+_ 1 SD. None of the H' values for the atrazine-treated plots vary signifi-
cantly from the companion control plots (p=0.05).
59
-------
Field Studies (Metal Cylinders)
The studies in the metal cylinders involved much higher rates of
atrazine than were used in other studies. Carbon fixation was significantly
reduced by 0.05 and 0.5 g/m2 16 days after treatment but not 26 days after
treatment. The 5.0-g/m2 rate inhibited carbon fixation through the 42-day
assay but not on the 67-day assay. Assays were continued on all plots
through the 108th day. Even the lowest rate used, 0.05 g/m2 (100 ppm), is
several hundred times greater than would be expected in the runoff from
treated fields (Anonymous, 1977).
The degree of inhibition of photosynthesis, chlorophyll production,
and cell numbers in cultures on N. sigma and T. fluviatilis by atrazine are
in close agreement with values obtained for other algal types. Davis et al.
(1976) found that prometryn, another s-triazine, reduced 02 production in
Chorella pyrenoidosa by 94% after a 40-min exposure at 5 x 10~5 M. The
ability to maintain chlorophyll synthesis in T. fluviatilis and cell division
and chlorophyll production in N. sigma with significantly reduced photosyn-
thesis rates (atrazine at 10-6 M) poses interesting questions. It would
appear that these two benthic alga species may be able to maintain themselves
with reduced light (if atrazine addition is considered to be analagous to a
reduction in light intensity). Research by others partially supports our
findings.
Admiraal (1977) reported a reduction in doublings/day for N. sigma
when grown under 16-h days as compared to 8-h days; photosynthesis rates
and chlorophyll production were not determined. In our experiments with
N. sigma, cell number was decreased about 50% when photosynthesis was
reduced about 33%. Similarly, Cyalotella meneghiniana , a freshwater diatom,
has the same chlorophyll level and photosynthesis rate when grown under
3 or 30 klux (Jorgensen, 1964). If T. fluviatilis and N. sigma can maintain
chlorophyll synthesis and cell division in the equivalent of reduced light,
their optimal dominance should be in the winter. Williams (1962) has pre-
viously reported a winter maximum for N. sigma in a Georgia salt marsh.
Williams also found maximum photosynthesis rates at approximately 50% of
full sunlight, in winter and summer, further suggesting that at least some
members of the community may be saturated at less than 100% of full sunlight.
Carbon fixation by untreated controls in the microecosystems (Table 16)
were similar to previously reported values; e.g., Pomeroy (1959) estimated
that winter production in a Georgia salt marsh was 150 mg C/m2/h at low
tide, and Darley et al. (1976) reported a summer value as high as 244.7
mg C/m2/h for creekbank samples from the same area. Estimates obtained in
the April microecosystem experiment were somewhat higher (283.4 mg C/m2/h)
than expected. Possible explanations for these high values include increased
productivity due to greenhouse effects, an alteration of light quality and
intensity inside the plastic tubs used as microecosystems, or addition of
algal cells to the soil surface from the water reservoirs. Estimates of
cell numbers in microecosystems were in close agreement with those for the
field. Thus these microecosystems appeared to give reasonable values for
cell numbers and primary productivity even though species diversity was
60
-------
adversely affected. The value for cell numbers in April nontreated micro-
ecosystems (Table 16) was higher than expected. This was due to a bloom in
one system of T. fluviatilis, a small diatom which could be present in
large numbers and still not significantly alter chlorophyll level or rate
of photosynthesis.
Chlorophyll estimates in microecosystems were also similar to reported
values of Sullivan and Daiber (1975 b),who estimated a yearly average of
100 mg Chl/m2 in the top 1 cm of a Delaware marsh. The nonatrazine- treated
systems in this study contained ca 135 mg Chl/m^ in the top 0.5 cm of soil.
The presence of high rates of photosynthesis in the lens paper samples
which contained only a small percentage of total chlorophyll indicates that
there may be a large population of photosynthetically inactive algae living
just beneath the surface. Examination of this layer with a light microscope
verified the presence of numerous living cells. Heterotrophy would be the
only means of survival for this community, but Parley et al . (1979) have
found that the surface community could only obtain up to 1.0% of its carbon
by this means; no work has been undertaken to ascertain heterotrophic poten-
tial of subsurface algae.
The effects of atrazine in the field were less severe than in the cul-
ture work or microecosystems. This could have been due to the higher summer
temperatures with correspondingly higher rates of microbial degradation of
atrazine and increased volatilization. However, the dilution of the applied
atrazine by tidal flux may have been more important. The nearly six-fold
increase in primary productivity observed inside the test enclosures as
compared to adjacent nonenclosed controls is possibly explainable on the
basis of decreased sunlight intensity and/or duration inside the containers.
Pomeroy (1959) has reported an inhibition of photosynthesis by the summer
sun, but this has not always been confirmed in other studies (Gallagher,
1971; Van Raalte et al., 1976). It is possible that the shade-adapted dia-
toms produced more chlorophyll and had a correspondingly greater ability to
assimilate
In the field as in the microecosystems, c. belgiaa was the most domi-
nant diatom species, and the addition of atrazine appeared to increase its
numbers. Although only 100 values were enumerated per sample in determining
species diversity, the low coefficient of variation (always < 15%) between
replicate samples would indicate that sufficient values were enumerated.
The species diversity values, both for the field plots and the microeco-
systems, agree well with the data of Sullivan (1975) who reported H1 -values
of 4034 to 4688 in a Delaware salt marsh. The degree of similarity between
the atrazine- treated microecosystems and the nontreated systems indicates
either (1) that only a few species can survive in this artificial environ-
ment and that the resultant population is stable even when exposed to
atrazine or (2) (less likely) the severity of atrazine effects on species
diversity is less in the winter than the summer. Similarity values were
not computed for microecosystem diatom populations versus those in the field
because of known differences in taxa throughout the year (Williams, 1962).
61
-------
The primary objective of this study was to evaluate the effects of
atrazine on the salt marsh edaphic algae and, based on these findings,
estimate safe levels of atrazine for this ecosystem. Several methods have
been proposed for designating acceptable levels in the aquatic environment;
only two will be discussed here.
The American Society for Testing and Materials Standards (1964) has
devised a system whereby the ISQ level obtained for diatoms in laboratory
cultures exposed to the pollutant for 7 days is multiplied by a safety
factor of 0.3. For this study, the 150 for N. sigma and T. fluviatiHs
was computed to be 4.36 x 10"15 M atrazine; 4.36 x 10'6 M x 0.3 = 1.3 x 10-6
M (0.28 ppm).
EPA has sometimes used a general "rule of thumb" of a 10-fold safety
factor below the least effect level when this level is well-known. The
least effect level (I-j) for algae grown in culture was computed to be about
5 x 10~7 M atrazine in these studies. With a 10-fold safety factor, accept-
able levels would be 5 x 10~8 M (10 ppb). The authors consider this to be
a more acceptable value than the 0.28 ppm calculated by the first method.
One study (Anonymous, 1977) has reported atrazine concentrations in the range
from 0.0 to a maximum of 2 ppb in water in tributaries of the Chesapeake Bay.
It has been postulated that atrazine might be responsible for declines in
aquatic vegetation in the Chesapeake Bay. Results from this study do not
support this hypothesis.
SUMMARY
A 10~5 M (2.2 ppm) concentration of atrazine significantly reduced the
rate of photosynthesis, chlorophyll content, and cell numbers in unialgal
cultures of Nitzchia sigma Grun. and Thalass-iosira fluviatilis Hustedt iso-
lated from a salt marsh habitat. Results with lower atrazine concentrations
indicated an ability to maintain chlorophyll production and cell division
with reduced photosynthesis. The effects of a 10~5 M concentration of
atrazine in unialgal cultures were also evident in microecosystems and in
the field at the same concentration. Severity of atrazine effects was less
in the field than in microecosystems or cultures. Cell number and produc-
tivity of the diatoms from nonatrazine treated microecosystems agreed well
with field data and previously published data. Diatom species diversity
was not affected by 10~5 M atrazine in microecosystems or in the field but
the number of Cymatosira belgiea was increased. Diatom populations in
atrazine-treated versus nontreated microecosystems were very similar (SIMI
value = 0.906). Results were less conclusive in the field but the trend
was toward a lower level of similarity (0.838). Based on the least effect
level of atrazine to diatoms, the maximum safe level for atrazine in the
salt marsh is estimated to be 10 ppb.
62
-------
SECTION 9
ATRAZINE RESIDUES IN SALT-MARSH ECOSYSTEM COMPONENTS:
A COMPARISON OF FIELD AND MICROECOSYSTEM RESULTS
OBJECTIVES
The ultimate test of how well an investigator has succeeded in building
a microecosystem that faithfully reproduces conditions in the field is to
conduct an experiment in the field and in the microecosystem and compare
results. The objectives of this investigation were: 1) to measure persis-
tence of atrazine in various compartments in the salt-marsh ecosystem and
2) to compare the values obtained in the field with those in the micro-
ecosystem.
MATERIALS AND METHODS
Herbicide Application
Plots on the salt marsh on Sapelo Island, GA, and microecosysterns at
Auburn, AL, were sprayed with atrazine suspensions in amounts and concen-
trations sufficient to give atrazine application rates of 0.0, 0.05, 0.50,
and 5.00 g/m2. The atrazine concentrations used were 0.0, 100, 1000, and
10,000 ppm, respectively. Rates of application were replicated three times
and were completely randomized. Atrazine residues in various ecosystem
components were determined 10 weeks after herbicide application as described
below. In addition to determining the amounts of atrazine remaining, the
experiment was designed to test how well the results in the microecosysterns
matched those in the field. In order to enhance the parallelism between
the microecosystems and the field plots, herbicide application were made in
microecosysterns 1 week after they were made in the field. This made it
possible to duplicate variables observed in the field such as rains or
height and periodicity of tides 1 week later in the microecosystems. Samples
were taken for atrazine residue determination in the microecosystems 1 week
after they were taken in the field.
Establishing Field Plots
Twelve 1.8-m diam cylinders 105 cm tall were sunk 25 cm deep in soil
along a tidal creek in the salt marsh on Sapelo Island. Each cylinder had a
5-cm diam hole cut in the side at soil level. The hole was covered with
8-mm mesh hardware cloth. This port allowed free movement of tidal water
into and out of the enclosures but prevented movement of all of the large
fiddler crabs into and out of the area.
63
-------
Establishing the Microecosysterns
Each microecosystem was constructed from two 75-L molded plastic
laundry tubs which were taken to Sapelo Island for filling with soil and
biological components as described in Section 5. Additional details about
these microecosysterns are presented in Section 10 and by Everest and
Davis (1977).
The ecosystem containers were returned to Auburn, placed on greenhouse
benches, and the drainage pipes for the containers were connected to water
reservoirs by means of 1.8-cm diameter rubber hoses and standard plumbing
fittings. The water reservoirs (also 75-L laundry tubs) were filled with
one part natural seawater (enough to supply an inoculum of associated micro-
organisms) and nine parts synthetic seawater. Each water reservoir con-
tained 37.9 L of seawater (20 ppt). At least 2 weeks were allowed to lapse
before any experiments were initiated to allow the systems to stabilize.
Sufficient demineralized water was added every 3 days to compensate for
water lost by evaporation and not added by simulated rainfall.
A "tide machine" was designed that mimicked tidal frequency, duration on
the marsh, and rate of flow onto and off the marsh. The "tide machine" had
three main components: (1) a supporting frame, (2) a mechanical drive system,
and (3) an electrical timing system. The maximum depth of seawater above
the soil level in each microecosystem was 17.5 cm; and during each 24 h and
50 min period, two tidal cycles were completed, each consisting of 2 h of
flooding and 10 h and 25 min of no flooding. Additional details concerning
the construction and function of this system are given in Section 10 and by
Everest and Davis (1977).
Three days after the tides were initiated and soil had settled in place,
mud fiddler crabs (Uaa pugnax), periwinkle snails (Littorina -Lrrorata), and
ribbed mussels (Geukensia demissa) were placed in the systems at densities
comparable to those observed on the plots on Sapelo Island. A 2-week delay
was imposed between the introduction of the animals into the systems and
herbicide treatment to insure that the animals had survived and had estab-
lished normal movement and feeding behavior. At this time personnel on
Sapelo Island were told to apply the herbicide when conditions become
suitable (no rain and no tides that would flood the area within 6 h after
application). The herbicide was applied the third week in July on Sapelo
and 1 week later at Auburn.
The water in the microecosystem reservoirs was emptied and replaced by
fresh synthetic seawater 6, 20, 37, 70, and 73 days after the herbicide
was applied.
Sampling Procedure for Atrazine Residue Determination
Sampling was done 10 weeks after herbicide application. Each sample
from each field plot or microecosystem was kept separate through the final
atrazine residue determination. Spartina alterniflora was cut off at the
64
-------
soil surface and divided into stems longer than or shorter than 0.5 m. Fresh
and dry weights were determined, and then 500- to 1000-g subsamples were
drawn, ground, and stored in a deep freeze until assayed. Fiddler crabs
samples consisted of all crabs collected. The collected crabs were counted,
thoroughly washed, weighed, and stored frozen in a deep freeze. Periwinkle
snail samples consisted of all snails found on the s. altevniflora or on the
soil surface. They were prepared as described for the fiddler crabs. Horse
mussels were collected, counted, washed thoroughly, and weighed; their flesh
was removed and stored frozen in a deep freeze.
Soil samples were taken from the 0-1, 1-10, and 10-25-cm layers'from
four randomly selected areas in each plot or microecosystem. Soil samples
were taken before the soil was disturbed by any other sampling procedure.
A soil surface sample was made by carefully removing a 1-cm thick layer from
each of four ca 100 cm^ areas. Polyvinyl coring tubes were inserted into
these areas to a depth of 25 cm (24 cm + 1 cm removed initially). The
coring tubes were then removed, and the contents were divided into the soil
present in the 1-10 cm layer and the 10-25 cm layer. Sufficient tubes were
used to give composite samples of 200-300 g for each soil layer. The soil
was then freeze-dried, ground, and stored in a deep freeze until analyzed.
Water samples (500 ml) were taken from the microecosysterns that were
treated at the rate of 0.5 g/m^ when the water was changed in the reservoirs
(6, 20, 37, 70, and 73 days after herbicide application).
Atrazine Residue Determination
Atrazine residues were determined by gas-liquid chromatography. Water
samples were washed with CH2Cl25 taken to dryness, and the extracts analyzed
directly. The other materials were dried, ground to a powder, extracted by
refluxing for 1 h with 90% acetonitrile, made to volume; an amount equivalent
to 10 g of the sample was removed and a known amount of 14c-label added to
the extract. The acetonitrile solution was concentrated by flash evaporation,
washed several times with CH2C12 and hexane, and the extract passed through
an aluminum oxide column to isolate the atrazine. Recovery efficiency was
measured by liquid scintillation radioassay of the column eluate. Atrazine
concentration in the eluate was determined by a Hewlett-Packard (HP) gas
chromatograph (model 5710 A) equipped with a 1.22 m x 2 mm glass column
packed with 2% OV-101, a HP Nitrogen-Phosphorus detector, and an HP 3380 A
data system.
RESULTS AND DISCUSSION
Residues in Soils
Atrazine residues found in soils 10 weeks after herbicide application
are given in Table 18. The variability between replications was so large
and the residue levels so low that there is no certain evidence that either
soil from the field or microecosystems treated at the 0.05 or the 0.50 g/m^
rates had a measurable level of atrazine remaining. It is apparent that the
amount, if any, must be no more than 0.1 ppm. Field plots treated with
65
-------
TABLE 18. AVERAGE CONCENTRATIONS OF ATRAZINE IN VARIOUS SOIL LAYERS IN THE
FIELD AND IN MICROECOSYSTEMS 10 WEEKS AFTER HERBICIDE APPLICATION
Atrazine Location
rate
g/m2
0.00
0.00
0.05
0.05
0.50
0.50
5.00
5.00
Field
Microeco
Field
Microeco
Field
Microeco
Field
Microeco
Soil layer samples (cm)
0-1
(ppm)
0.03 + 0.07*
0.06 + 0.15
0.05 + 0.03
0.11 + 0.18
0.06 + 0.02
0.00 + 0.01
1.20 + 0.39
0.00 + 0.03
1-10
(ppm)
0.05 + 0.08
0.09 + 0.19
0.03 + 0.05
0.00 + 0.02
0.02 + 0.02
0.00 + 0.01
0.77 + 0.33
0.01 + 0.03
10-25
(ppm)
0.00 + 0.03
0.00 + 0.00
0.04 + 0.11
0.00 + 0.07
0.02 + 0.02
0.00 + 0.02
0.25 + 0.27
0.01 + 0.03
0-25
(ppm)
0.02f
0.03
0.04
0.00
0.02
0.05
0.48
0.01
Values are given ;+ N-l standard deviation. All samples were spiked with
known amounts of radiolabeled atrazine. When the amount added was sub-
tracted from the value determined for the sample, the result was sometimes
negative. Occasional averages were slightly negative. Negative values
were used in calculating the standard deviation, but any apparently negative
averages are presented as 0.
The value for the 0-25 cm layer is calculated from the values for the three
layers (0-1 cm, 1-15 cm, 15-25 cm) involved.
5.0 g/m2 had 1.20, 0.77, and 0.25 ppm in the 0 to 1-, 1- to 10-, and 10- to
25-cm layers, respectively. The average was 0.48 ppm in the top 25 cm.
If one assumes a specific gravity of 1.2 for the soil, then this 25-cm layer
contained 144 mg of atrazine or 2.88% of that applied. The microecosysterns
averaged only 0.01 ppm which is not significantly different from the control.
It is not known why the microecosysterns lost more atrazine than the field
plots. It is possible that it was easier for the tidal flux to remove the
atrazine from the small surface area (0.26 m2) of the microecosystems than
from the relatively larger area (2.5 m?) in the field plots. No doubt areas
of very slow movement occurred in the field plots since all of the water
had to go into or out of one comparatively small hole.
Residues in 5. alterniflora
Atrazine residues in s. alterniflova are presented in Table 19. As
for the soil, atrazine residues in plants grown in the field or in micro-
ecosystems treated with 0.05 or 0.50 g/m2 of atrazine were too low and too
variable to be significantly greater than the controls. The value of 0.50 ppm
for S. alterniflora > 0.5-m-tall harvested in the field is higher than one
would expect in view of the other data. We have no basis for explaining
this one relatively large value.
66
-------
TABLE 19. AVERAGE CONCENTRATIONS OF ATRAZINE IN TWO SIZE CLASSES OF S.
ALTERNIFLORA (<0.5MAND>0.5M TALL) IN THE FIELD AND IN
MICROECOSYSTEMS 10 WEEKS AFTER HERBICIDE APPLICATION
Atrazine
Rate
(g/m2)
0.00
0.00
0.05
0.05
0.50
0.50
5.00
5.00
Location
Field
Mi croeco
Field
Mi croeco
Field
Mi croeco
Field
Mi croeco
Size of Plant
< 0.5 m
(ppm)*
0.00 +
0.04 + 0.13
0.11 + 0.23
0.02
0.16 + 0.20
0.04 + 0.05
21.58 + 9.85
16.80 + 6.59
Sampled
> 0.5 m
(ppm)*
0.04 + 0.06
0.00 + 0.03
0.50 + 0.23
0.07 + 0.07
0.09 + 0.11
0.51 + 0.46
12.83 + 2.46
21.07 + 7.73
*
Values given are averages +_ the N-l standard deviation. When no standard
deviation is given, the material from all three plots were combined in
order to get one sample large enough to assay.
The amounts of S. alterniflora harvested from these plots did not vary
significantly either between rates of atrazine applied or between the eco-
systems and the field. Average dry weight yields were 45.5 g/m2 for s.
altemiflora < 0.5 m tall and 231.2 g/m2 for plants > 0.5 m tall. Using
these values for S, altepniflora yields and the concentration of atrazine
found in subsamples from S. alterniflora < or > 0.5 , the total amount of
atrazine in S. altemiflora harvested from field plots treated with 5.0
g/m2 of atrazine was 3.9 mg/m2 or ca 0.08% of that applied. For the micro-
ecosystems the average of the atrazine residues was ca 5.6 mg/m2 or ca 0.11%
of that applied. The higher concentration of atrazine in the S. alternifloT
from the microecosystems than from the field is probably because in the
field each tidal change brought in essentially atrazine-free water whereas
new water was not placed in the microecosystem reservoirs until 6 days after
herbicide application.
Residues in Animals
Atrazine residues in periwinkle snails, mussels, and fiddler crabs
from plots treated with 5.0 g/m2 of atrazine are given in Table 20. None
of the animals from plots receiving lower rates of treatment contained a
significant amount of atrazine except for snails in the plots treated with
0.50 g/m2 atrazine which contained 0.45 and 0.34 ppm from the field and
the microecosystems, respectively. Atrazine concentration was 15-20 times
higher for snails from the microecosystems receiving 5.0 g/m2 atrazine than
for snails from plots in the field treated at the same rate. It is possible
that part of this difference is due to snail migration between plots in the
67
-------
TABLE 20. AVERAGE CONCENTRATIONS OF ATRAZINE IN PERIWINKLE SNAILS, HORSE
MUSSELS, AND FIDDLER CRABS IN THE FIELD AND IN MICROECOSYSTEMS
10 WEEKS AFTER HERBICIDE APPLICATION
Atrazine
rate
(g/m2)
5.00
5.00
Location
Snails
(ppm)*
Field 0.45
Microeco 7.76
Organism
Mussel|
(ppm)
0.00
3.49
Crabs
(ppm)*
0.31
*
Values are for one composite sample. There was insufficient material to
any single plot to make a reliable assay of the atrazine present.
field whereas this was only remotely possible between microecosystems. How-
ever, the atrazine concentration in mussels was also much greater in the
microecosystems than in the field, and these organisms did not migrate out
of their plots. It seems possible that the greater concentration of atrazine
in the microecosystem than in the field is (as was mentioned above) because
atrazine-containing water flushed back and forth into and out of the micro-
ecosystems whereas the tide brought in essentially atrazine-free water in
the field. Perhaps some food organisms or organic material containing
atrazine also moved back and forth into and out of the microecosystems.
There were not enough crabs in the field plots receiving the highest atra-
zine concentration to analyze the atrazine in their bodies. Therefore, it is
not known whether the crabs from the microecosystems also contained more
atrazine than those from the field.
Residue in Water
In order to estimate the amount of atrazine removed in tidal water,
periodic samples were taken from the three reservoirs attached to ecosystems
treated with atrazine at the rate of 0.50 g/m2. The samples were taken just
prior to replacement of the water with fresh seawater. The concentrations
of atrazine present in these samples are given in Table 21. After 6, 20,
37, and 70 days, the accumulated atrazine discarded when the water was
changed amounted to 42, 49, 50, and 50% of that applied, respectively.
After harvesting was completed, an additional assay was made. During this
3-day period, an additional 0.1% of that applied had moved into the water.
It is difficult to apply these results to what might be expected in the
field. If atrazine is accidentally spread on a small area of the marsh, it
would seem reasonable to guess that about half of it would be washed away by
the tide in 1 week. The rate of disappearance would be expected to be
higher than that found for the microecosystems; each tidal invasion in the
field would bring essentially atrazine-free water, whereas fresh water was
introduced in the microecosystems only after 6 days. However, if the
source of the atrazine contamination to the marsh is the tide itself, it
68
-------
TABLE 21. AVERAGE AMOUNTS OF ATRAZINE COLLECTED IN WATER RESERVOIRS ATTACHED
TO MICROECOSYSTEMS TREATED WITH 0.50 g/m? (0.13 mg/system) ATRAZINE
Collection period
(days) Concn (ppm)
0 -
6 -
20 -
37 -
70 -
6
20
37
70
73
1.440 + 0.114
0.235 + 0.026
0.035 + 0.003
0.007 + 0.002
0.003 + 0.008
ing in 37.9 L
54.57
8.91
1.33
0.265
0.114
% of applied
41.98
6.85
1.02
0.11
0.09
Values given are average + the N-l standard deviation. The average value
for the 0- to 6-day sample from the untreated controls was 0.00165 j^ 0.0008.
would be expected to diminish in
tidal water decreased.
the marsh as atrazine concentration in the
also be exercised in applying the data accrued for atrazine
in soil, plants, and animals to situations found in the field.
high residues found in these components from areas treated
have no relevance to any ordinary happening in the field. The
Care must
concentrations
The relatively
with 5.00 g/m2
5.00 g/m^-rate is 10-20 times the rate of application that is used for weed
control in corn (WSSA Herbicide Handbook Committee [1979]) and was included
only to insure that measurable atrazine concentrations could be found in
some of the ecosystem components after 10 weeks. To achieve the lowest
atrazine treatment level used, the areas were sprayed with suspensions con-
taining 100 ppm atrazine. Tidal water contamination would not be expected
to be as much as 0.1 ppm (Chesapeake Bay Research Consortium, 1977). How-
ever, two conclusions are possible. First, if a small section of the marsh
is accidentally contaminated with atrazine, atrazine would be rapidly ex-
ported in the tide, and unless the contamination rate was extremely high,
residue levels should approach 0 within 3 months. Second, ordinary levels
of atrazine contamination would not result in any atrazine carryover from
one growing season to the next.
Atrazine levels found in the microecosystems were similar to those in
the enclosed plots in the field. Many of the differences observed are ex-
plainable in terms of the fact that each tidal flush in the field involved
essentially atrazine-free water; whereas in the microecosystems, the water
running off the systems was replaced by atrazine-free water only at irregular
intervals. The first change was 6 days after atrazine application.
SUMMARY
Within 3 months after atrazine application the total amount of atrazine
remaining in the soil, s. alterniflora, snails, fiddler crabs, and horse
mussels was less than 3% of that applied. At the lower rates of application
69
-------
(0.05 and 0.50 g/m2) the amounts of herbicide remaining were often below
the sensitivity of the assay system. Although both the tests with micro-
ecosystems and in the field confirmed the rapid disappearance of atrazine
from the treated areas, there were significant differences in results be-
tween the two systems. Considerably more atrazine remained in the soil of
systems in the field than in the microecosysterns. This was perhaps because
the enclosures around the field plots permitted tidal movement out only
through one relatively small hole. This may have resulted in areas of
stagnation such that atrazine was not swept away nearly as vigorously as in
the microecosysterns. Contrariwise, more atrazine was found in the S. alterni-
fLora, snails, and mussels in the microecosystems than in the fields. This
was probably because each new tide in the field introduced new water,whereas
the water in the reservoirs was not changed until after 6 days.
70
-------
SECTION 10
MICROECOSYSTEMS
LITERATURE REVIEW
Over the last two decades, there has been a growing interest in the
ecosystem concept in biology. Natural ecosystems are often quite large and
variable. The most difficult problem in working with ecosystems is to estab-
lish boundaries between them. In nature, ecosystems tend to blend or grade
into one another^ making any boundary which is established purely artificial.
Researchers have developed a new technique to aid in the study of eco-
systems. The technique involves enclosing a small portion of an ecosystem
to isolate this functional unit from the rest of the biosphere. This de-
liberately isolated ecosystem is known as a microcosm or a microecosystem
(Booth, 1977). There are several advantages of using microecosystems in
ecosystem studies. Their small size and isolation from the surrounding
environment make possible the study of the effects of very expensive or
highly toxic materials and the more precise measurement of responses. Fur-
thermore, precise control of environmental conditions such as temperature,
gas composition, or light intensity is more feasible. The microecosystem
approach also makes possible replicated studies using very similar ecosystems,
which is seldom feasible in nature.
Ideally, each microecosystem should include: one or more primary pro-
ducers which serve as a food source for one or more primary consumers, one
or more trophic levels of carnivores, and decomposers. Furthermore, popu-
lation densities and soil, air, and water amounts and composition should be
comparable to those in nature. The systems should be self-sustaining for
long periods of time without the need to supply more organisms, feed, or
fertilizer.
Most of the microecosystems which have been used in research have been
aquatic rather than terrestrial and they have rarely approached the ideal
microecosystem described above. They have usually been quite simple in
design, containing only two trophic levels and a very limited number of kinds
of macroscopic organisms.
9
Several portions of this section are reprinted from Pages 167-171 of
the 2nd ed of Research Methods in Weed Science. Copyright Sout. Weed Sci.
Soc. Used by permission": ' ~~
71
-------
Cooke (1969) gave an excellent review of the literature dealing with
aquatic microecosystem usage. He pointed out the parameters that could be
measured or controlled, and the benefits and disadvantages of using micro-
ecosystems in a wide variety of investigations. Beyers (1963) subjected
microbial populations in glass containers to different environmental regimes
and determined the effect that these modifications had on diurnal metabolism.
Abbott (1966, 1967, 1969) used similar systems to determine the effects of
nitrate and phosphate enrichment. Taub (1976) investigated the effect of
some algicides, insecticides, and heavy metals in aquatic microecosysterns
containing bacteria, algae, and grazing microorganisms. He showed that
shifts in species composition occurred with increased populations of re-
sistant forms.
Other aquatic and aquatic-terrestrial microecosysterns have been designed
and used to study the accumulation and effects of various pesticides in the
environment (Isensee et al., 1973; Metcalf et al., 1971; Sanborn, 1974;
Yu et al., 1975). These studies demonstrated that some pesticides tended to
accumulate in organisms in the higher trophic levels, while others had a low
potential for bioaccumulation. Results from these studies have shown the
value of microecosysterns for predicting the fate of pesticides in the envi-
ronment.
MICROECOSYSTEMS USED
During the course of our investigations, several different types of
microecosystem were designed and used to determine the fate and effects of
atrazine in a salt marsh. The microecosysterns employed varied in size,
complexity, and mode of operation.
The microecosystem used to study the metabolism of atrazine in Spartina
alterniflora-detritus-Uca pugnax food chain (Section 6) was the smallest and
simplest in design. This system consisted of a Buchner funnel containing
a nylon mesh liner covered with a 1.5-cm layer of sand on top of which was
a 1.5-cm layer of marsh soil containing the associated microflora. Each
funnel was connected to a small water reservoir by plastic tubing. The
tidal flux was obtained by the raising and lowering of the water reservoirs
(Figure 11, Section 6). This microecosystem made it possible to follow the
conversion of dried S. alterniflora to detritus, and the fate of 14c-atrazine
in the S. alterniflora during this conversion. Quantitative assessment of
the various metabolites of atrazine produced would not be feasible in the
field. There were several ways in which these systems did not match con-
ditions in the salt marsh. Since the microecosystems did not contain grazing
organisms, the microflora populations present in each system were not re-
duced by normal feeding activity. Probably the most important variation
from the field was the closed seawater system. The same seawater, except
for one change after a 10-day period, was used to flush the systems during
each tidal cycle. This did not bring about dilution comparable to the
tides and may have resulted in the gradual accumulation of toxic materials.
Daily changes of the seawater or the design of a more complex chemostat
system could have been used to overcome this weakness. The small size
of the systems limited the amount of detritus that could be produced.
72
-------
The microecosystems used in the greenhouse to study the effect of atra-
zine (Section 8) on marine diatoms were intermediate in size and complexity.
Each system consisted of a plastic tub containing a 10-cm layer of creek
bank soil, associated diatoms, and other microorganisms. These tubs were
connected to individual water reservoirs (plastic buckets) by rubber tubing
fastened to glass tubes inserted into rubber stoppers fitted into holes in
the tubs and buckets. There were several benefits in using these micro-
ecosystems. The tubs and buckets used were inexpensive and relatively heat
and sunlight stable. The soil-containing tubs were yellow-orange in color
so that they did not reflect or absorb significant amounts of heat or light.
Also, the tubs and buckets were small and relatively lightweight (even when
full) so filling in the field and transporting them to the greenhouse was
easy. As with any design, there were weaknesses or limitations discovered
while using these systems. Since the systems contained no fiddler crabs to
graze on the algae or disturb the soil surface and thus produce turbid water,
the systems developed unnatural blooms of blue-green algae. Fiddler crabs
had to be added to the systems to prevent such blooms. The seawater reser-
voirs were not large enough to give adequate tidal dilution. This was es-
pecially apparent after the herbicide treatments when the systems were re-
covering from the atrazine effects. This problem might have been overcome
by changing the seawater in the reservoirs daily after the herbicide treat-
ments to simulate natural tidal dilution. The single most annoying problem
was the splitting of the tubs and buckets where the rubber stoppers were
inserted into them. The use of silicone caulking material or cement to
hold the stoppers in place or some alternative connectors might have allevi-
ated this problem.
The microecosystems used in the field to study atrazine effects on marine
diatoms were much simpler (Section 8). Each microecosystem consisted of a
36-cm-diam plastic tub filled to a depth of 15 cm with soil collected in the
immediate vicinity. The tubs were partially buried in the soil so that both
the inside and outside soil surfaces were at the same level. A 3-cm-diam
hole in the side of the tub at the soil surface could be closed during
atrazine treatment or left open to allow normal tidal movement onto or off
of the soil in the tub. The ecosystems were placed on the creek bank in
the field and thus apparently varied from the natural situation only by the
presence of the container walls. The systems were allowed to stabilize for
7 days prior to atrazine treatment. Effects of atrazine on photosynthesis,
cell numbers,and cell kinds were compared within the tubs and also between
the untreated areas inside and outside the tubs. Algal numbers did not vary
significantly from the outside. However, the kinds of algae did,and the
rate of photosynthesis inside the tubs was six times greater than that out-
side. At high tide fiddler crabs could easily crawl out of the tubs. The
loss of these grazers, disturbance of the soil when filling the tubs, and
shading by the walls of the tubs may have caused the observed variations.
The use of larger containers and a longer delay time after transfer of the
soil into the containers might decrease the observed differences.
The largest and most complex microecosystem used in these investigations
was partially described in Sections 7 and 9. The microecosystem containers
and water reservoirs were constructed from 75-L molded plastic tubs. The
73
-------
microecosystem containers were connected to their individual reservoirs by
means of rubber tubing. The microecosystems contained marsh soil and associ-
ated microorganisms, s. alterniflora, fiddler crabs, ribbed mussels, and
periwinkle snails. The water reservoirs containing seawater (salinity-20ppt)
were connected to a device called a tide machine which was designed to raise
and lower the reservoirs9thus causing water to flow onto and off of the soil
in the microecosystems.
The tide machine had three main components: (1) a supporting frame
(Figure 16), (2) a mechanical drive system (Figure 17) and (3) an electrical
timing system (Figure 18). The supporting frame was 4.8 m long by 1.2 m wide
and supported on eight 2.1-m legs. The frame was constructed from 5.1-cm
galvanized pipe. A 5.1-cm solid steel rotating shaft supported by pillow
block bearings ran lengthwise through the center. A large, 80-tooth cog-
wheel was attached to the end of the steel shaft. Twelve, 3-mm-diam aircraft
cables were attached to the shaft,and the free ends were attached to the
water reservoirs.
Figure 16. Arrangement of the water reservoirs on the tide machine.
74
-------
3R~ TR1 ! ™ |
1 I 'vfJ3 ' '
TRT CR2 u£
J^ooll 1
iV.OO'il
LSft CM
1
, a!?po o^t • ot'CR?*" i
• ^sS~? f "*• ^i!is^
LV2 CR1 Down
Bottom |
°vM ° ° \\
1ST CR2
Rl
aViAA.o
°TIVW° £.-2
4_ Field ,Arm ..
"V—
CR2
Tinner circuit
Motor Forward
relay circuit
Motor Reverse
relay circuit
Motor and
Speed Control
circuit
Figure 17
Figure 18
Figure 17. Arrangement of mechanical drive and electrical timing system of
the tide machine: 1. variable speed control; 2. digital interval timer;
3. 80-tooth cogwheel; 4. chain; 5. 30:1 right angle speed reducer; 6. 28-tooth
cogwheel; 7. coupling; 8. 1/15 HP AC/DC gear motor.
Figure 18. Schematic diagram of electrical system of the tide machine.
The mechanical system consisted of a variable speed, reversible, 1/15
H.P., right angle gearmotor connected directly to a 30:1 right angle speed
reducer bearing a 28-tooth cogwheel. Drive was transmitted from this small
cogwheel to the larger cogwheel on the steel shaft by a chain (Figure 17).
The electrical system consisted of a synchronous, motor driven, automatic
reset timer, and a series of adjustable limit switches and reversal relays.
When the system was activated, the turning of the gearmotor resulted in
rotation of the shaft which wound the aircraft cable around the shaft and
raised the water reservoirs. As they continued to rise, water flowed from
the reservoirs into the microecosystems and over the soil surfaces. When the
tidal seawater reached the desired height, a limit switch and relay reversed
the motor and permitted the unwinding of the cable and the lowering of the
water reservoirs. When the reservoirs descended to the desired point (low
tide), a limit switch and relay reversed the motor and simultaneously, the
the system was shut off by the timer. After a predetermined time interval ,
75
-------
the timer started the motor and a new cycle was initiated. By controlling
the rate and distance of rise and fall, the tide machine permitted the simu-
lation of tidal water movement onto and off of the marsh at the rate and
with the duration of a typical tidal cycle.
These more complex microecosystems with the associated tide machine
permitted a close simulation of the natural situation. Mortality of plants
and animals in the microecosystem was low,and normal feeding and animal
movements occurred in these systems after a short acclimatization period
(usually 2 weeks). These systems were used in several studies over a period
of 3 years and, once the initial design and sampling problems were resolved,
the systems functioned well with routine maintenance. However, some diffi-
culties were encountered.
Animal death or escape from the microecosystems was a problem in the
early stages of development of these systems. Initially, fiddler crabs
tended to move to the drain opening at the bottom of each system and suffo-
cate in the trapped water between tidal cycles. A perforated plastic pipe
was introduced into each drainage port and this pipe prevented entrapment.
In some preliminary experiments some of the periwinkle snails escaped by
moving up and over the sides of the microecosystem containers. This problem
appeared to be due to excessive concentrations of snails in the systems.
When the snail numbers were reduced to naturally occurring levels, snails
seldom left the systems.
Insect infestations of S. alterniflora are rarely a problem in a salt
marsh apparently due to the presence of some control agent or agents. Ap-
parently the control agent or agents was missing from the microecosystems ,
as some cordgrass plants were lost due to a heavy scale infestation. Two
different scale insects, Greenisoa palustris Dodds and Haliaspis spart-inae
(Cmst.), fed heavily on the stems and leaves of the infested plants. Care-
ful washing of the plants, hand removal of the scale insects, and good
sanitation procedures controlled this problem. Daily observation of the
systems was necessary to control the insects.
A problem was encountered in recovering fiddler crabs from the systems
for analysis. Most of the components of the microecosystems were readily
sampled. However, the burrowing and evasive behavior of the crabs made
sampling them extremely difficult without disturbing the soil surface. It
was found that if a small trench was dug along one side of the container,
water remained in these depressions during low tide and the crabs hid in
these pools. This facilitated their capture.
Electrical and mechanical malfunctions did present problems at times.
Despite routine maintenance, mechanical and electrical components deterio-
rated rather rapidly because of the use of salt water. It was necessary
to continually watch for corrosion and rust and replace components periodi-
cally. Since the tide machine had to function throughout an experiment,
replacement parts for every electrical and mechanical component were pur-
chased. The design was such that a faulty part could be replaced within
a few minutes.
76
-------
Probably the most serious limitation of this design was that it did
not include a mechanism to simulate rain. The S. alterniflora leaves were
washed twice each week with deionized water to remove accumulated salt and
to simulate the frequent rains common in the salt marsh,and deionized water
was added to the reservoirs at periodic intervals to maintain relatively
constant levels of salinity. Approximately monthly replacement of the sea
water in the reservoirs helped to maintain healthy systems, perhaps by
removing accumulating toxic materials.
SUMMARY
The use of microecosystems made a major contribution to these investi-
gations. Many would have been impossible without them. However, it must
always be remembered that in the final analysis many of the findings re-
sulting from studies using microecosystems need to be confirmed in the
field. Furthermore, the novice proposing to use the microecosystem approach
should realize that development of a microecosystem that closely simulates
a natural ecosystem requires a thorough knowledge of the relationships among
the components of the system, and a good understanding of the relative num-
bers and biomasses of the living components. He also must be aware of the
many cycles (seasonal, diurnal, tidal, etc.) which may affect activities of
the organisms under study. Once developed, the systems make possible better
replicated and more accurately monitored experiments with a higher level of
environmental control than can be achieved in the natural setting.
77
-------
REFERENCES
Abbot, W. Microcosm studies on estuarine water. I. The replicability of
microcosm. J. Water Pollut. Control Fed., 38:258-273, 1966.
Abbot, W. Microcosm studies on estuarine water. II. The effect of single
doses of nitrate and phosphate. J. Water Pollut. Control Fed., 39:113-122,
1967.
Abbot, W. High levels of nitrate and phosphate in carboy microcosm studies.
J. Water Pollut. Control Fed., 41:1748-1751, 1969.
Admiraal, W. Influence of light and temperature on the growth of estuarine
benthic diatoms in culture. Mar. Biol., 39:1-9, 1977.
American Society for Testing and Materials Standards, Part 23. In: Tenta-
tive method of test for evaluating inhibitory toxicity of industrial waste
waters. Am. Soc. Test. Mater., Philadelphia, PA., pp. 517-526, 1964.
Anonymous. Assessing nonpoint source pollution: A detailed study of a
rural watershed in the coastal plain of Maryland. Chesapeake Bay Center
for Environmental Studies. Smithsonian Institution, Washington, DC,
59 pp., 1977.
Bakke, J.E., J.D. Larson, and C.E. Price. Metabolism of atrazine and
2-hydroxyatrazine by the rat. J. Agric. Food Chen., 20:602-607, 1972.
Balicka, N., and H. Bilodub-Pantera. The influence of atrazine on some
soil bacteria. Acta. Microbiol. Pol., 13:149, 1964.
Beyers, R.J. The metabolism of twelve aquatic laboratory microecosystems.
Ecol. Monogr., 38:199-122, 1963.
Booth, G.M. Using model ecosystems to predict the environmental behavior
of pesticides. Am. Biol. Teach., 39:275-278, 1977.
Boyland, E., and L.F. Chasseaud. The role of glutathione and glutathione
s-transerases in mercapturic acid biosynthesis. Adv. Enzymol. Relat. Areas
Mol. Biol., 32:173-219, 1969.
Chesapeake Bay Center for Environmental STudies. Assessing nonpoint source
pollution: A detailed study of a rural watershed in the coastal plain of
Maryland. Chesapeake Bay Center for Environmental STudies, Smithsonian
Institution, 39 pp., 1977.
78
-------
Christian, R.R., K. Bancroft, and W.W. Wiebe. Resistance of the microbial
community within salt marsh soils to selected perturbations. Ecology, 59:
1200-1210, 1978.
Cohen, A.J., J.N. Smith, and H. Turbert. Comparative detoxification. 10.
The enzymatic conjugation of chloro compounds with glutathione in locusts
and other insects. Biochem. J., 90:457, 1964.
Cooke, G.D. Aquatic laboratory microecosystems and communities. In: The
structure and function of freshwater microbial communities. John Cairn,
Jr., ed. Virginia Polytechnic Institute and State University, Research
Division, Blacksburg, VA, pp. 47-85, 1969.
Correll, D.L., J.W. Pierce, and T.L. Wu. Studies of the transport of atra-
zine and Arochlor from minimum till corn fields in Chesapeake Bay tidal
waters. Proc. Northeast. Weed Sci. Soc. 32:21-32, 1978.
Cotnoir, L.J. Marsh soils of the Atlantic coast. In: Ecology of halo-
phytes. R.J. Reimold and W.H. Queens, eds. Academic Press, NY, pp. 441-
447, 1974.
Couch, R.W., and D.E. Davis. Effect of atrazine bomacil, and diquat on
d402-fixation in corn, cotton, and soybeans. Weeds, 14:251-255, 1966.
Daiber, F.C., and C. Crichton. Tide marsh ecology and wildlife. Annual
Pittman-Robertson Report on Delaware Board of Game and Fish Commissioners.
Project W-22-R-2. Job #4, 1967.7.
Davis, D.E., H.H. Funderburk, Jr., and N.G. Sansing. The absorption and
translocation of Cl4-labeled simazine by corn, cotton, and cucumber.
Weeds, 7:300-309, 1959.
Davis, D.E., C.G.P. Pillai, and B. Truelove. Effects of prometryn, diuron,
fluometuron, and MSMA on Chlorella and two fungi. Weed Sci., 24:587-593,
1976.
Darley, W.M., E.L. Dunn, K.S. Holmes, and H.G. Larew, III. A 14C method for
measuring epibenthic microalgal productivity in air. J. Exp. Mar. Biol.
Ecol., 25:207-217, 1976.
Darley, W.M., C.T. Ohlman, and B. Wimpee. Utilization of dissolved organic
carbon by natural populations of epibenthic salt marsh diatoms. J. Phycol.
15:1-5, 1979.
Eaton, J.W., and B. Moss. The estimation of numbers and pigment content in
epipelic algal populations. Limnol. Oceanogr., 11:584-595, 1966.
Edwards, A.C., and D.E. Davis. Effects of herbicides on the Spavtina salt
marsh. In: Ecology of halophytes, R.J. Reimold and W.H. Queen, eds.
Academic Press, NY, pp. 531-545, 1974.
79
-------
Eno, C.F. The effects of simazine and atrazine on certain of the soil micro-
flora and their metabolic processes. Soil Crop Sci. Soc. Fla. Proc.,
22:49-56, 1962.
Esser, H.O., G. Dupuis, E. Ebert, C. Vogel, and G. Marco. S.-triazines.
In: Chemistry, degradation, and mode of action of herbicides. P.C. Kearney,
and D. D. Kaufman, eds. Marcel Debber, Inc., NY, pp. 129-308, 1975.
Everest, J.W. Predicting phosphorus movement and pesticide action in salt
marshes with microecosystems. Ph.D. Dissertation. Auburn University,
Auburn, AL, 88 pp., 1978.
Everest, J.W., and D.E. Davis. Microecosystems for herbicides research and
other studies. In: Research methods in weed science, 2nd Ed., B. Truelove,
ed. South. Weed Sci. Soc., Auburn, AL, pp. 167-171, 1977.
Gallagher, J.L. Algal productivity and some aspects of the ecological
physiology of the edaphic communities of Canary Creek tidal marsh. Ph.D.
Dissertation, University of Delaware, Newark, DE, 120 pp., 1971.
Gallagher, J.L. Effect of an ammonium nitrate pulse on the growth and
elemental composition of natural stands of Spartina alte-miflora and
Juncus roemevianus. Am. J. Bot. 62:644-648, 1975.
Gallagher, J.L., R.J. Reimold, R.A. Linthurst, and W.J. Pfeiffer. Aerial
productivity, mortality, and mineral accumulation—export dynamics in
Spartina altemiflora and Junaus roemerianus plant stands in a Georgia salt
marsh. Ecology (In press), 1979.
Guillard, R.R.L. Division rates. In: Handbook of phycological methods:
Culture methods and growth measurements. J.R. Stein, ed. Cambridge Univ.
Press, NY, pp. 289-311, 1973.
GuiTlard, R.R.L., and J.H. Ryther. Studies of marine planktonic diatoms.
I. Cyelotella nana. Hustedt and Detonula aonfervacea (Cleve) Gran.
Can. J. Microbiol., 8:229-239, 1962.
Gunther, F.K., ed., and J.D. Gunther, asst. ed. The triazine herbicides.
Residue Rev. 32. Springer-Verlag, NY, 413 pp., 1970.
Haines, E.B. Relation between the stable carbon isotope composition of
fiddler crabs, plants, and soils in a salt marsh. Limnol. Oceanogr.,
21:880-883, 1976.
Haines, E.B. The origins of detritus in Georgia salt marsh estuaries.
Oikos, 29:254-260, 1977.
Hall, J.K., M. Pawlus, and E.R. Higgins. Losses of atrazine in runoff water
and soil sediment. J. Environ. Qual., 1:172-176, 1972.
80
-------
Hamilton, R.H., R.S. Bandurski, and W.H. Reusch. Isolation and characteri-
zation of cyclic hydroxamate from "Lea mays. Cereal Chem., 39:107-113, 1962.
Hanson, R.B. Nitrogen fixation (acetylene reduction) in a salt marsh amended
with sewage sludge and organic carbon and nitrogen compounds. Applied
Environ. Microbiol., 33:846-852, 1977.
Isensee, A.T., P.C. Kearney, E.A. Woolson, G.E. Jones, and V.P. Williams.
Distribution of alkyl arsenicals in model ecosystems. Environ. Sci.
Techno!., 7:841-845, 1973.
Jensen, K.I.N., G.R. Stephenson, and L.A. Hunt. Detoxification of atrazine
in three gramineae subfamilies. Weed Sci., 25:212-220, 1977.
Jorgensen, E.G. Adaptation to different light intensities in the diatom
Cyalotella meneghin-iana Kutz. Physiol. Plant., 17:136-145, 1964.
Kaufman, D.D., and P.C. Kearney. Microbial degradation of triazine herbi-
cides. In: The triazine herbicides, Gunther, F.K. ed., and J.D. Gunther
asst. ed. Springer-Verlag, NY, pp. 235-266, 1970.
Lamoureux, G.L., R.H. Shimabukuro, R.H. Swanson, and H.R. Frear. Metabolism
of 2-chloro-4-ethylamino-6-isopropyl-s-triazine (atrazine) in excised sorghum
leaf sections. J. Agric. Food Chem., 18:81-86, 1970.
Lamoureux, G.L., L. E. Stafford, and R.H. Shimabukuro. Conjugation of
2-chloro-4,6 bis (alkylamino)-s-triazines in higher plants. J. Agric. Food
Chem., 20:1004-1010, 1972.
Lamoureux, G.L., L.E. Stafford, R.H. Shimabukuro, and R.C. Zaylskie. Atra-
zine metabolism in sorghum: Catabolism of the glutathione conjugate of
atrazine. J. Agric. Food Chem., 21:1020-1030, 1973.
Metcalf, R.L., G.K. Sangha, and I.P. Kapoor. Model ecosystems for the
evaluation of pesticide biodegradability and ecological magnification.
Environ. Sci. Techno!., 5:709-713, 1971.
Newell, R.C. Factors controlling metabolic capacity adaption in marine
invertebrates. In: Physiological ecology of estuarine organisms, F.J.
Vernberg, ed. University of South Carolina Press, Columbia, SC, pp. 112-
127, 1975.
Odum, E.P. The role of tidal marshes in estuarine production. The conser-
vationists, New York State Conserv., Albany, NY, 15:12-15, 1961.
Odum, E.P. Fundamentals of ecology. W. B. Saunders Co., Philadelphia, PA,
574 pp., 1974.
Odum, E.P., and A.A. de la Cruz. Particulate organic detritus in a Georgia
salt marsh-estuarine ecosystem. In: Estuaries, G.H. Lauff, ed. Am. Assoc.
Adv. Sci. Publ. 83, Washington, DC, pp. 383-388, 1967.
81
-------
Odum, W.E., G.M. Woodwell, and C.F. Wurster. DDT residues absorbed from
organic detritus by fiddler crabs. Science, 164:576-577, 1969.
Patrick, R., and C.W. Reimer. The diatoms of the United States. Vol. 1.
Acad. Nat. Sci., Philadelphia, PA, Monograph No. 13, 688 pp., 1966.
Pimentel, D. Ecological effects of pesticides on nontarget species. U.S.
Govt. Printing Office, Washington, DC, 220 pp., 1971.
Pomeroy, L.R. Algal productivity in salt marshes of Georgia. Limnol.
Oceanogr., 4:386-397, 1959.
Pomeroy, L.R., L.R. Shenton, R.D.H. Jones, and R.J. Reimold. Nutrient flux
in estuaries. In: Nutrients and eutrophication: The limiting nutrient
controversy. Proc. Symp. on nutrients and eutrophication, G.E. Lichens, ed.,
Vol. 1., Am. Soc. Limnol. Oceanogr. Lawrence, KS, pp. 274-291, 1972.
Pringsheim, E. Pure cultures of algae. Cambridge Univ. Press, 119 pp., 1946.
Reimold, R.J. Mangals and salt marshes of eastern United States. In: Eco-
systems of the world: Wet coastal ecosystems, V.J. Chapman, ed. Elsevier
Scientific Publishing Co., Amsterdam, The Netherlands, pp. 157-166, 1977.
Robinson, D.E., and D.W. Greene. Metabolism and differential susceptibility
of crab grass and witch grass to simazine and atrazine. Weed Sci., 24:500-
504, 1976.
Round, R.E. Benthic marine diatoms. Oceanogr. Mar. Biol. Annu. Rev.,9:83-
139, 1971.
Sanborn, J.R. The fate of selected pesticides in the aquatic environment.
EPA 660/3-74-025, National Environmental Research Center, U.S. Environmental
Protection Agency, Corvallis, OR, 81 pp., 1974.
Schroeder, W.A., R.T. Jones, J. Cormick, and K. McCalla. Chromatographic
separation of peptides on ion-exchange resins. Separation of peptides from
enzymatic hydrolysates of the a, 0, y> chains of human hemoglobins. Anal.
Chem., 34:1575, 1962.
Shimmel, S.M. The effect of grazing on epibenthic algae in a Georgia salt
marsh. MS Thesis., University of Georgia, Athens, GA, 44 pp., 1979.
Shimabukuro, R.H. Significance of atrazine dealkylation in root and shoots
of pea plants. J. Agric. Food Chem., 15:557-562, 1967a.
Shimabukuro, R.H. Atrazine metabolism and herbicidal selectivity. Plant
Physiol., 42:1269-1276, 1967b.
Shimabukuro, R.H., D.S. Frear, H.R. Swanson, and W.W. Walsh. Glutathione
conjugation: An enzymatic basis for atrazine resistance in corn. Plant
Physiol., 47:10-14, 1971.
82
-------
Shimabukuro, R.H., R.E. Kadunce, and D.S. Frear. Dealkylation of atrazine
in mature pea plants. J. Agric. Food Chem., 14:392-395, 1966.
Shimabukuro, R.H., G.L. Lamoureux, D.S. Frear, and J.E. Bakke. Pesticide
terminal residues. Supplement to Middle East Tech. Univ. 0. Pure Appl.
Chem., pp. 323-342, 1971.
Shimabukuro, R.H., H.R. Swanson, and W.C. Walsh. Glutathione conjugation:
Atrazine detoxification mechanism in corn. Plant Physiol., 46:103-107, 1970.
Shimabukuro, R.H., W.C. Walsh, G.L. Lamoureux, and L.E. Stafford. Atrazine
metabolism in sorghum: Chloroform-soluble intermediates in the N-dealkyla-
tion and glutathione conjugation pathways. J. Agric. Food Chem., 21:1031-
1036, 1973.
Smalley, R.J. The role of two invertebrate populations, Littovina in-orata
and Orchelimum fidicinium, in the energy flow of a salt marsh ecosystem.
Ph.D. Dissertation, University of Goergia, Athens, GA, 126 pp., 1959.
Steel, R.G.D., and J.H. Torrie. Principles and procedures of statistics.
McGraw Hill Book Co., Inc., NY, 481 pp., 1970.
Sullivan, M.J. Diatom communities from a Delaware salt marsh. J. Phycol.,
11:384-390, 1975.
Sullivan, M.O., and F.C. Daiber. Light, nitrogen, and phosphorus limitation
of edaphic algae in a Delaware salt marsh. J. Exp. Mar. Biol. Ecol., 18:79-
88r 1975.
Taub, F.B. Demonstration of pollution effects in aquatic microcosms. Int.
J. Environ. Stud., 10:23-33, 1976.
Teal, J.M. Distribution of fiddler crabs in Georgia salt marshes. Ecology,
39:185-193, 1958.
Teal, J.M. Respiration of crabs in Georgia salt marshes and its relation to
their ecology. Physiol. Zool., 32:1-14, 1959.
Teal, J.M. Energy flow in the salt marsh ecosystem of Georgia. Ecology,
43:614-624, 1962.
Thompson, O.C., B. Truelove, and D.E. Davis. Effects of triazines on energy
relations of mitochondria and chloroplasts. Weed Sci., 22:164-166, 1974.
Trainor, F.R. Introductory phycology. John Wiley & Sons, Inc., MY, 525 pp.,
1978.
Van Raalte, C.D., W.C. Stewart, I. Valiela, and E.J. Carpenter. A C
technique for measuring algal productivity in salt marsh muds. Bot. Mar.
17:186-194, 1974.
83
-------
Van Raalte, C.D., I. Valiela, and J.M. Teal. Production of epibenthic salt
marsh algae. Light and nutrient limitation. Limnol. Oceanogr., 21:862-872,
1976.
Vernberg, F.J., and W.B. Vernberg. Adaptations to extreme environments. In:
Physiological ecology of estuarine organisms. F.J. Vernberg, ed. Univ. of
South Carolina Press, Columbia, SC, pp. 165-180, 1975.
Vernberg, W.B., P.J. DeCoursey, and J. O'Hara. Multiple environmental factor
effects on physiology and behavior of the fiddler crab, u. pugilator. In:
Pollution and physiology of marine organisms. F.J. Vernberg, ed. Academic
Press, NY, pp. 381-425, 1974.
Ward, D.V., and D.A. Busch. Effects of Temefos, an organophosphorus insecti-
cide, on survival and escape behavior of the marsh fiddler crab Uoa pugnax.
Oikos, 27:331-335, 1976.
Weber, J.B., S.B. Weed, and T.M. Ward. Adsorption of s-triazines by soil
organic matter. Weed Sci., 17:417-421, 1969.
Williams, R.B. The ecology of diatom populations in a Georgia salt marsh.
Ph.D. Dissertation. Harvard University, Cambridge, MA, 146 pp., 1962.
Wilt, J.G., P. Leeuwaugh. Mercapturic acid formation and enzyme-catalyzed
conjugation with glutathione in pigeons. Biochim. Biophys. Acta., 177:329,
1969.
Wolf, P.L., S.F. Shanholtzer, and R.J. Reimold. Population estimates for
Uaa pugnax (Smith, 1870) on the Duplin estuary marsh, Georgia, USA,
Crustaceana (Leiden), 29:79-91, 1975.
WSSA Herb. Handbook Comm. Herbicide handbook of the Weed Science Society
of America, 4th Ed. Weed Science Society of America, Champaign, IL, 479 pp.,
1979.
Yentsch, C.S., and D.W. Menzel. A method for the determination of phyto-
plankton chlorophyll and phaeophytin by fluorescence. Deep-Sea Res., 10:221-
231, 1963.
Yu, Ching-Chieh, G.M. Booth, and J.R. Larson. Fate of the triazine herbicide
cyanazine in a model ecosystem. J. Agric. Food Chem., 23:1014-1015, 1975.
84
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-79-m
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Atrazine Fate and Effects in a Salt Marsh
5. REPORT DATE
November 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D.E. Davis, J.D. Weete, C.G. Pillai, F.G. Plumley,
J.W. Everest, B. Truelove, and A.M. Diner
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Department of Botany and Microbiology
Auburn University Agricultural Experiment Station
10 PROGRAM ELEMENT NO.
IEA615
11 CONTRACT/GRANT NO.
R803835
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Gulf Breeze, FL 32561
13. TYPE OF REPORT AND PERIOD COVERED
Final July 9,1979-March 1,1979
14. SPONSORING AGENCY CODE
EPA/600/4
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Components of the Spartina alterniflora salt marsh were exposed to atrazine individu-
ally, in microecosystems, and in the field, to determine its effects on salt marsh
components and its fate in the salt marsh. Components studied were S_. alterni flora;
horse mussel, Geukensia demissa; periwinkle snail, Littorina irrorata; box crab,
Sesarma cinereum; fiddler crab, Uca pugnax; diatom spp, including Nitzschia sigma and
Thallassiosira fluyiatilis; detritivores, soil, and tidal water. Only algae were
affected by possible contaminant concentrations (0.01 ppm) in seawater in the marsh.
:S. alterniflora was fairly tolerant but 0.1 ppm decreased growth slightly. Adult
U_. pugnax at their most sensitive stage may have been slightly affected by 100 ppm.
S_. cinereum was unaffected when fed leaves from S_. alterni flora grown in nutrient
solution containing 0.6 ppm atrazine. Conversion to detritus was unaffected when
S.. alterni flora leaves were wetted with 0.26 ppm atrazine solution. No effects on
snails or mussels were detected when the marsh was sprayed with 5 g/m^ atrazine.
Atrazine was metabolized by S_. alterni flora, S_. cinereum, U_. pugnax, and detritivores,
Three months after atrazine application to the marsh the total herbicide remaining
in the soil, S_. alterni flora, L_. irrorata, U_. pugnax, and G_. demissa was less than 3%
of that applied. Seventy days after atrazine application to microecosystems, 50% had
removed in tidal water; 42% within the first 6 days.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Herbicide
Marshes
Aquatic animals
Aquatic biology
Atrazine
Food chains
Spartinia alterniflora
Littorina" irrorata"
SesarmaTinereum, Uca
pugnax, Nitzschi' sigma,
Thai!assiosira fluviati1•
diatoms, detritivores,
S-triazines, non-target
Gpsvstorm
06/C
06/T
08/A
18. DISTRIBUTION STATEMENT
Release to public
Unclassified
21. NO. OF PAGES
84
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |