Effect of gibberellic acid and 2,4-dichlorophenoxyacetic acid on waterhyacinth, Eichhornia crassipes (Mart.) Solms


Material Information

Effect of gibberellic acid and 2,4-dichlorophenoxyacetic acid on waterhyacinth, Eichhornia crassipes (Mart.) Solms
Eichhornia crassipes
Physical Description:
viii, 98 leaves : ill. ; 28 cm.
Joyce, Joseph Clarence, 1948-
Publication Date:


Subjects / Keywords:
Water hyacinth -- Control   ( lcsh )
Herbicides   ( lcsh )
Gibberellic acid   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1982.
Includes bibliographical references (leaves 87-97).
Statement of Responsibility:
by Joseph Clarence Joyce.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000319001
notis - ABU5851
oclc - 09260001
System ID:

Full Text






I dedicate this work to my mother


(1928 1967)

whose life-long hope and ambition was that I receive a

complete and quality collegiate education.


This research was partially supported by a cooperative agreement

with United States Department of Agriculture, ARS Number 58-7B30-0-177.

The grant was administered as a project of the University of Florida,

Institute of Food and Agricultural Sciences, Center for Aquatic Weeds.

I am extremely grateful to Dr. William T. Haller, my major

professor, for his advice and encouragement throughout this study.

Appreciation is also extended to the members of my committee: Dr. Daniel

E. Canfield, Jr.; Dr. Leon A. Garrard; Dr. Jerome V. Shireman; and

Dr. Shirlie H. West; and to a former committee member, Dr. Thai K. Van,

for their interest and constructive criticisms.

I also wish to express my thanks to Dr. Palakurthi Suresh C. Rao for

the use of his carbon oxidizer without which a major portion of this

study would not have been possible. Similarly, I would like to thank

Dr. George Bowes for the use of his scintillation counter.

My sincere thanks are extended to the U.S. Army Corps of Engineers,

Jacksonville District, who afforded me with the opportunity and finan-

cial support to pursue this degree. In particular, I would like to

thank my supervisors, Mr. Gail G. Gren, Mr. Girlamo DiChiara, and

Mr. James D. Hilton and the members of the Natural Resource Management

Section, James M Dupes, James T. McGehee, and Larry T. Taylor for their

patience and support. I want to thank Eddie Knight for his assistance in

conducting the field portion of this study and to Betty Hyman and

Jerrine Hamm for their assistance in preparing this manuscript.

Sincere thanks go to Jim Cobb, Ken Langland, and Daniel Thayer for

their assistance and advice throughout this study.

I am most grateful to my wife Pamela T. Joyce and sons, Joseph

C. Joyce, Jr. (JJ) and Christopher T. Joyce (Ty) for their support,

patience, and sacrifice.



ACKNOWLEDGEMENTS.................................................. iii

ABSTRACT .................................................................................. vii

INTRODUCTION...................................................... 1

LITERATURE REVIEW................................................. 4

Gibberellic Acid............................................. 4
History ................................................. 4
Chemical and Physical Characteristics ................... 5
Relative Potency of Gibberellins......................... 8
Toxicology............................ .......... ..... 9
Biosynthesis and Metabolism.............................. 10
Sites of Synthesis............................ .......... 11
Transport....................... ........................ 12
Mode of Action....... ................................. 13
Interaction with Other Plant Growth Substances.......... 20
Commercial Applications.................................. 22
Effects on Waterhyacinths................................ 24

2,4-Dichlorophenoxyacetic Acid............................... 25
History ................................................. 25
Physical and Chemical Characteristics................... 27
Toxicology.............................................. 28
Persistence in the Environment.......................... 29
Mode of Action........................................... 31
Translocation............................ .......... ..... 35
Metabolism by Plants..................... .......... ..... 36
Efficacy of 2,4-D on Waterhyacinths..................... 37

PART 1. SMALL PLOT EVALUATIONS................................... 40

Introduction................................................ 40
Materials and Methods....................................... 41
Results and Discussion...................................... 43
Summary ................................. .................... 53

PART 2. TRANSLOCATION EVALUATIONS................................ 54

Introduction................................................ 54
Materials and Methods....................................... 54
Results and Discussion...................................... 57
Summary........................................... .......... 70

PART 3. FIELD EVALUATIONS........................................ 71

Introduction................................................. 71
Materials and Methods........................................ 72
Results and Discussion....................................... 75
Summary...................................................... 80

DISCUSSION.......................................................... 82

CONCLUSIONS......................................................... 86

LITERATURE CITED................................................... 87

BIOGRAPHICAL SKETCH................................................. 98

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




December 1982

Chairman: William T. Haller

Major Department: School of Forest Resources and Conservation

Effects of combinations of gibberellic acid (GA3) and

2,4-dichlorophenoxyacetic acid (2,4-D) on waterhyacinths (Eichhornia

crassipes (Mart.) Solms) were evaluated to determine if GA3 increased

waterhyacinth sensitivity to 2,4-D under field conditions, and if

GA3 increased translocation of 14C labeled 2,4-D in waterhyacinths grown

in growth chambers. Effects of GA3 on waterhyacinth sensitivity to

2,4-D were evaluated in two phases. In the first phase, small bulbous

leafed waterhyacinths were grown outdoors in 70-liter containers and

treated during three growing seasons with combinations of GA3 at 0.0,

23.5, 47.0, 94.0 and 188 g/ha and 2,4-D at 0.00, 0.28, 0.56, 1.12, and

2.24 kg/ha. The second phase was conducted during late summer in a

dense population of mature, non-bulbous leafed waterhyacinths in the

St. Johns River, Florida, with combinations of GA3 at 0.0, 23.5, 47.0,

and 94.0 g/ha and 2,4-D at 0.00, 0.56, 1.12, 2.24, and 4.48 kg/ha.

Effects of treatment rates were recorded as percent change from initial

biomass and initial number of plants. Effects of GA3 on the transloca-

tion of 14C labeled 2,4-D were monitored through time for levels of
14C per milligram of plant tissue and percent of total 14C translocated

to separate plant parts.

Regression analysis indicated the lack of significant interaction

between GA3 and 2,4-D in terms of increased efficacy of 2,4-D above

routine application rates of 2.24 kg/ha. Additive effects of 2,4-D and

GA3 were suggested, however. Costs analysis of various combinations of

GA3 and 2,4-D indicated that addition of GA3 in order to lower rates of

2,4-D would increase waterhyacinth control costs by over 300.0 percent.

Translocation of 14C-labeled 2,4-D to meristematic waterhyacinth

tissues was not increased due to pretreatment with 100 mg/1 GA3.

Increased translocation to leaves other than 2,4-D treated leaves was

suggested. Use of GA3 to significantly reduce rates of 2,4-D used to

control waterhyacinths under field conditions was not justified from either

an increased efficacy or economic standpoint.



Man's attraction and search for botanical species which are unique

in structure, food potential, floral characteristics, productivity or

ability to survive in specific environmental situations have resulted in

the introduction of numerous plants outside of their native range. Such

introductions may result in an exotic species becoming established in

a habitat optimum for growth in the absence of its naturally limiting

environmental factors or organisms. Once freed of these naturally

limiting conditions, certain exotic species may proliferate to such an

extent that the population interferes with man's intended use of its


The waterhyacinth, Eichhornia crassipes (Mart.) Solms, a perennial,

floating aquatic plant, is an example of such an exotic species. The

waterhyacinth is generally considered to be a native of Brazil from

where it has spread to nearly all sub-tropical and tropical regions of

the world where conditions favor its growth (Penfound and Earle, 1948;

Bock, 1966). Since its introduction into the United States in the mid

1880's, the waterhyacinth has proliferated to such an extent that it has

created serious water resource management problems such as 1) inter-

ference with commercial and recreational navigation, 2) reduction in

water conveyance capabilities of streams, rivers, and man-made flood

control structures and waterways, 3) adverse modification of fish and

wildlife habitat, and 4) threatening public health by providing habitat


for disease vectors (Webber, 1897; Penfound and Earle, 1948; Hitchcock

et al., 1949; Seabrook, 1962; Anonymous, 1965; Bock, 1966; Buker, 1982.)

Prior to 1946, waterhyacinth control operations consisted of

mechanical operations utilizing various conveyor, chopper or shredder

machines; containment apparatus such as floating booms, fences or traps;

pusher boats to assist the plants to salt water; and the use of various

inorganic chemicals such as sodium arsenite (Webber, 1897; Brown et al.,

1946; Penfound and Earle, 1948; Hitchcock et al., 1949; Bock, 1966;

Wunderlich, 1967; Buker, 1982). Large-scale control of waterhyacinths

was revolutionized in the mid-1940's by the discovery of the herbicidal

properties of 2,4-dichlorophenoxyacetic acid (2,4-D). Since its

development, various formulations of 2,4-D have been routinely utilized

for control of waterhyacinths (Anonymous 1946; Brown et al., 1946;

Hildebrand, 1946; Hitchcock et al., 1949; Penfound and Earle, 1948).

Currently, dimethylamine salt is the only formulation of 2,4-D labeled

by the United States Environmental Protection Agency (EPA) for use in

public waters which are either flowing and/or potable water supplies.

During 1981, governmental agencies in Florida applied approximately

63,700 kg of 2,4-D to public waters for control of nuisance aquatic

vegetation, the majority of which was waterhyacinths (Dupes and Mahler,


Increased public awareness, spurred by the environmental movement

and adverse publicity associated with the military's use in Vietnam of

the defoliant, "Agent Orange", which was a combination of 2,4-D and

2,4,5-trichlorophenoxyacetic acid (2,4,5-T), has caused various regula-

tory agencies to more closely scrutinize labeling of new herbicides and

to review existing registrations of herbicide products currently


utilized in agriculture, and more specifically, aquatic environments.

In order to reduce public concern over use of chemical control agents

and to gain more effective control of pest species, two basic approaches

have been pursued; i.e., (a) the use of alternate control methods such

as biological agents, mechanical harvesting systems, and/or environmental

manipulation and (b) modification of herbicide application techniques,

equipment, and/or the use of various chemical adjuvants in order to

reduce the amount of herbicide actually introduced into the environment.

This study was designed to determine if the quantity of 2,4-D

routinely utilized for large scale control of waterhyacinths could be

reduced by simultaneously applying a naturally occurring plant growth

substance, gibberellic acid. The study was divided into three major

parts. Part one involved small plot evaluations of the efficacy of

various rates and combinations of 2,4-D and gibberellic acid for

waterhyacinth control. The objective was to determine the presence of

an interaction or synergistic effect between 2,4-D and gibberellic acid

which would increase the sensitivity of waterhyacinths to 2,4-D. Part

two involved growth chamber evaluations utilizing radioactive-labeled

2,4-D and unlabeled gibberellic acid. The objective was to ascertain if

the cause of increased sensitivity, if present, to 2,4-D was due to an

increase in the quantity of 2,4-D translocated to various locations

within the plant when also treated with gibberellic acid. Part three

involved field application of a combination of 2,4-D and gibberellic

acid under routine operational conditions. The objective of this final

phase was to determine if a specified combination of gibberellic acid

and a reduced rate of 2,4-D would be efficacious in a large-scale,

operational waterhyacinth control program.


Gibberellic Acid

Hi story

Gibberellins are defined by Phinney and West (1961) as substances

possessing the same or similar carbon skeleton as gibberellic acid (GA3)

and that are biologically active in stimulating cell division, cell

elongation, or both in plants. The earliest known description of the

effects of gibberellins was in 1809 by Konishi, a semi-literate Japanese

rice farmer, who described rice plants which grew excessively tall

(Stowe and Yamaki, 1957). The diseased plants could not support them-

selves and eventually died due to parasitic action (Yabuta, 1935;

Salisbury and Ross, 1978). Salisbury and Ross (1978) indicate that as

early as the 1890's the Japanese were referring to these symptoms as the

"bakanae" ("foolish seedling") disease. The disease was determined to

be caused by a fungus, Gibberella fujikuroi (sexual stage) and Fusarium

moniliforme (asexual stage). The active compound was isolated and iden-

tified in the 1930's by Yabuta and Hayashi, Japanese pathologists, who

named it gibberellin (Yabuta, 1935; Stowe and Yamaki, 1957; and Russell,

1974). Japanese scientists were interested in the pathological aspects

of gibberellins rather than physiological impacts (Salisbury and Ross,

1978). Thus, even though the first gibberellins were isolated in the

1930's, Western scientists did not become interested in the physiologi-

cal effects of gibberellins until the early 1950's due to (1) the



preoccupation with indoleacetic acid (IAA) and synthetic auxins,

(2) a lack of contact with the Japanese, and (3) World War II

(Salisbury and Ross, 1978). Stuart and Cathey (1961) reported that by

1961 approximately 9 compounds had been isolated which exhibited both

gibberellin-like activity and structure. This number increased to 19 by

1965 (Russell, 1974), 29 in 1970 (Lang, 1970), 38 by 1973 (Russell,

1974), 44 by 1974 (Barendse, 1974) and to over 50 by 1978 (Salisbury and

Ross, 1978). These gibberellins have been isolated from fungi, algae,

ferns, mosses, and many higher plants (Barendse, 1974; Russell, 1974;

Salisbury and Ross, 1978). Sircar and Kindu (1960), Bhanja and Sircar

(1966), and Sircar et al. (1973) reported the presence of four

gibberellin-like substances in the shoot and root extract of

waterhyacinths. Russell (1974) attributes this exponential rise in the

number of known gibberellins to the initiation of a systematic search

for growth promoting compounds which was aided, as observed by Lang

(1970), by the simultaneous development of highly effective techniques

for the separation and identification of naturally occurring compounds.

Both Barendse (1974) and Russell (1974) indicated that due to the

limited number of plants examined, the total number of gibberellins

and their derivatives is unknown.

Chemical and Physical Characteristics

Gibberellins are isoprenoid compounds and are classified as

diterpenes (Lang, 1970; Barendse, 1974; Salisbury and Ross, 1978). All

gibberellins have either 19 or 20 carbon atoms which are arranged in

either a four or five ring system, and all have at least one carboxyl

group (Barendse, 1974; Reeve and Crozier, 1974; Russell, 1974; and


Salisbury and Ross, 1978). Each compound is abbreviated GA, with a

subscript such as GA1, GA2, etc., to distinguish the different com-

pounds. All could be called gibberellic acid, however, due to the com-
mercial availability of GA3, it has been studied more extensively than
the other forms and is commonly referred to as gibberellic acid (Stuart
and Cathey, 1961; Salisbury and Ross, 1978).

Gibberellic acid (GA3) is a white to pale yellow, crystalline
powder, with a molecular weight of 346.37, empirical formula
C19H2206 and a chemical configuration of


I \



(Abbott Laboratories, 1962; Barendse, 1974; Lang, 1970; Reeve and

Crozier, 1974; Salisbury and Ross, 1978).


According to Abbott Laboratories (1962), gibberellic acid solubility

in various solvents is as follows:

Solvent Mg/ml Solvent

Dimethyl Formamide 450

Ethyl Alcohol 200

Methyl Acetone 180

Diacetone Alcohol 120

Isopropyl Alcohol 80

Acetone 40

Tap Water 5

Since gibberellic acid is readily soluble in numerous compounds

including water, it is readily formulated for various commercial applica-

tions (Abbott Laboratories, 1962). The potassium, sodium, and ammonium

salts of the acid are more soluble than the acid alone due to the buf-

fering of the aqueous solutions (Abbott Laboratories, 1962). Stuart and

Cathey (1961) indicated that these salts of gibberellic acid were equally

as active on the growth of pea and cucumber seedlings as the acid.

Abbott Laboratories (1962) indicated that dried gibberellic acid

powder is stable indefinitely; however, its aqueous solution is rela-

tively unstable at room temperatures with a half-life of one month.

Stability can be increased by storing the solution at low temperatures

or by utilizing sterile water. Gibberellic acid solutions prepared with

anhydrous solvents are extremely stable; therefore, commercial con-

centrates usually utilize non-phytotoxic organic solvents.

Commercially, gibberellic acid is produced by culturing

Gibberella fungus in a liquid culture medium from which the acid is


extracted. The process utilized is similar to that used in the produc-

tion of antibiotics (Borrow et al., 1955; Marth et al., 1956; Russell,


Relative Potency of Gibberellins

The standard method of assessing the potency of a gibberellin is by

various bioassay techniques (Bailiss and Hill, 1971; Reeve and Crozier,

1974). Bailiss and Hill (1971) conducted an extensive review of gib-

berellin bioassay techniques and listed 33 test systems which were

based on such diverse processes as coleoptile, leaf sheath, epicotyl,

hypocotyl and radicle growth, bud dormancy, seed germination, induction

of a-amylase synthesis, leaf expansion and senescence, and flower and

cone induction. Interpretation of bioassay data should be done

cautiously because individual bioassays exhibit a great deal of

species and even variety specificity (Reeve and Crozier, 1974). As

noted by Reeve and Crozier (1974), the barley aleurone and cucumber

hypocotyl tests only exhibit response to a limited number of

giberellins, whereas the dwarf rice bioassay responds to almost all


Based on the overall assessment of the relative responses obtained from

barley aleurone a-amylase synthesis, dwarf pea growth, lettuce and

cucumber hypocotyl growth, and Tan-ginbozu drawf rice microdrop bio-

assays, Reeve and Crozier (1974) assessed the relative activities of 38

gibberellins. This ranking indicates that the highest activities are

provided by GA1, GA3, GA7, and GA32. Good responses were also induced

by GA5, GA6, GA26, and GA37 but not of the same order of magnitude. Other


GA's, such as GA9, GA10, GA23, and GA24, exhibited species specificity by
being highly active in some bioassays yet induced poor responses in

others. A ranking of the overall relative activity, indicated that

GA3 was the most active compound (Stuart and Cathey, 1961; Reeve and

Crozier, 1974). Consistently low activity was exhibited by GA8, GAll,

GA12, GA13,, GA14, GA17, GA21, GA25, GA27, GA28, GA29, GA33, and GA34.

In all bioassay systems, GA26 was inactive at all concentrations

tested. Barendse (1974) indicated that many of the weaker gibberellins

were isolated from immature seeds, and it is not certain if (1) they are

also present in the growing plant, or (2) they are merely by-products

or intermediates for interconversion during biosynthesis of

more active gibberellins.


Gibberellic acid has been evaluated for toxicological effects in

rats, mice, guinea pigs, rabbits, dogs, cats, and chickens (Peck et al.,

1957; Warden and Schaible, 1958; Kimura et al., 1959). The compound was

shown to be asymtomatic and free of pathologic changes in subacute toxi-

city studies in mice and subchronic toxicity studies in dogs and rats

(Kimura et al., 1959). Subacute studies showed gibberellic acid to be

tolerated by mice at 2 g/kg, intravenously for 5 days, and at

1 g/kg, subcutaneously for 14 days (Abbott Laboratories, 1962). Studies

of acute intravenous toxicity of gibberellic acid in mice yielded an

LDO of 4.2 g/kg, an LD50 of 6.3 g/kg, and an LD100 of 8.7 g/kg (Peck et

al., 1957). Peck et al. (1957) indicated that signs of toxicity were

nonspecific and no deaths and only minimal signs of toxicity were

observed after the oral administration of 25.0 g/kg to mice. Based on


their studies, Peck et al. (1957) stated that gibberellic acid presents

no apparent hazard either to the individual who uses the material for

agricultural purposes or to the individual who consumes products on

which gibberellic acid or its salts have been used.

Biosynthesis and Metabolism

Much of the information available to date on gibberellin biosyntheis

has come from studies utilizing cultures of Gibberella fujikuroi (Saw.)

Wr. Barendse (1974) noted that although the pathway of biosynthesis in

higher plants is less well known, studies involving higher plants

suggest that the pathway follows the same scheme as found in Gibberella

fujikuroi. Birch et al. (1958) confirmed the fact that GA3 has a

diterpenoid nature by feeding radioactive-labelled acetate or mevalonate to

cultures of Gibberella fujikuroi. By examining relative amounts and

positions of the label, they were able to suggest a biosynthesis pathway

which was later confirmed (Salisbury and Ross, 1978).

The basic pathway of biosynthesis of gibberellic acid described by

Birch et al. (1958), Barendse (1974), and Salisbury and Ross (1978)

results in the following sequence of compounds: acetyl coenzyme A;

mevalonic acid; isopentenyl pyrophosphate; geranylgeranyl pyrophosphate,

a 20-carbon compound which serves as the donor for all gibberellin car-

bon atoms (Salisbury and Ross, 1978); copalyl pyrophosphate; kaurene;

kaurenol; kaurenal; kaurenoic acid; and GA12 aldehyde which is the first

true gibberellane ring system. This aldehyde of GA12 is converted

either directly to other gibberellins or to GA4, a 19-carbon compound,

which is interconverted to other gibberellins (Barendse, 1974; Salisbury

and Ross, 1978).


Once formed, gibberellins can be readily converted to bound

inactive forms in which they are stored or translocated for release at

the proper location and physiological time (Lang, 1970; Barendse, 1974;

Salisbury and Ross, 1978). Glucosides are the most prominent known

bound form of gibberellin (Barendse, 1974; Salisbury and Ross, 1978).

Salisbury and Ross (1978) indicate that other unidentified bound forms

are also known to exist, some of which appear to be stable protein-

gibberellin conjugates.

Sites of Synthesis

Major sites of gibberellin synthesis are developing seeds, apical

buds, young leaves, and root tips (Barendse, 1974; Salisbury and Ross,

1978). Lockhart (1957) implicated the shoot tip as the site of gib-

berellin synthesis by restoring growth in decapitated pea seedlings by

applying GA3. The diffusion technique of Jones and Phillips (1964)

later confirmed these results and also identified root tips as sites of

synthesis. The diffusion technique is based on the following principle:

if the amounts of diffusible gibberellins obtained in an agar block over

a period of time exceeds the amounts of extractable gibberellins in the

same organ, active biosynthesis or conversion occurs in the organ.

Barendse (1974) reviewed the role of plant roots in gibberellin synthe-

sis and reported that root tips were the site of conversion of inactive

forms of gibberellin synthesized in apical buds to more active forms.

Salisbury and Ross (1978) stated that repeated excision of parts of the

root system markedly reduced the amount of gibberellin in plant

foliage which may partially explain why root pruning inhibits shoot

growth. Developing seeds are generally considered to be sites of


synthesis based on observations that they contain large concentrations

and many types of gibberellins and the fact that accumulation of gib-

berellins is inhibited by growth retardants (Baldev et al., 1965;

Barendse, 1974; Salisbury and Ross, 1978).


Gibberellin-like substances have been isolated from both phloem
and xylem (Audus, 1972). Hoad and Bowen (1968) isolated gibberellins in

seive tube sap of several species which indicated phloem transport.

Exogenously applied GA3 follows a distribution pattern within the plant

and rate of movement typical of substances moving within the phloem

(McComb, 1964; Chin and Lockhart, 1965). Barendse (1974) cited various

studies which documented xylem transport of gibberellins in a number of

species including sunflower, peas, grapes, birch, maple, apple, and pear.

According to Salisbury and Ross (1978), the transport pathway from young

leaves into the stem below is uncertain, but does not involve vascular

transport because young actively growing leaves import but rarely export

through either xylem or phloem. Kato (1958a) conducted translocation

studies with pea stems and indicated that gibberellic acid does not show

a pattern of polar translocation. Clor (1967) confirmed these findings

utilizing tritium-labeled GA3. However, these experiments utilized

concentrations of GA3 which far exceeded physiological levels and may

have affected normal movement of the growth substance (Jacobs and

Kaldeway, 1970). Subsequent investigations by Jacobs and Kaldeway

(1970) and Jacobs and Pruett (1973) utilizing physiological levels of

GA3 revealed that GA3 exhibits strong basipetal polar movement in Zea

mays roots and Coleus petioles. As with auxins, cortex and pith are


believed to be involved. Thus, the evidence indicates that gibberellins

and their conjugates are readily transported in the entire conductive

system of plants; however, the physiological significance of this

transport has not been determined (Barendse, 1974).

Mode of Action

The literature contains numerous reviews of the physiological,

morphological, and biochemical mechanisms of action of gibberellins.

Addicott (1970) stated that gibberellic acid has probably been tested

more widely for its effects on higher plants than any other naturally

occurring substance. However, contradictory results were frequently

obtained from seemingly similar experiments. Addicott (1970) and Low

(1974) explain the differing results in terms of physiological con-

ditions of the experimental material, i.e., age, size, nutrient and

light availability, temperature, species, and type of tissue studied.

All of these factors have been shown to alter the level of other endoge-

nous hormones in the experimental tissues and thus change the plant's

sensitivity to the exogenous gibberellic acid (Low, 1974). As an

example, Goyal and Baijal (1980a and 1980b) reported different responses

between varieties of the same species of rice. The following discussion

of the mode of action of gibberellic acid presents the most commonly

accepted responses.

Typical morphological effects of sensitive species to gib-

berellin treatments include germination of dormant seeds (Stuart and

Cathey, 1961; Chen, 1974), growth of dormant buds (Shafer and Monson,

1958), stimulation of flowering and inflorescence size (Stowe and

Yamaki, 1957; Stuart and Cathey, 1961; Weaver, 1972, Krishnamoorthy,


1974), prevention or delay in flowering (Stowe and Yamaki, 1957; Weaver,

1972), leaf heterophylly (Stowe and Yamaki, 1957; Stuart and Cathey,

1961; Israelstam and Davis, 1979), chlorosis (Stowe and Yamaki, 1957;

Ende and Koornneef, 1960; Weaver, 1972; Sarma and Hussain, 1979), inhi-

bition or no effect on root growth (Brian et al., 1955; Kato, 1958b),

delay and/or acceleration of senescence (Brian et al., 1955; Stowe and

Yamaki, 1957; Stuart and Cathey, 1961; Addicott, 1970; Weaver, 1972;

Valdovinos, 1974; Salisbury and Ross, 1978); vernalization (Stowe and

Yamaki, 1957; Stuart and Cathey, 1961; Weaver, 1972), stem elongation

(Brian et al., 1955; Marth et al., 1956; Stowe and Yamaki, 1957;

Phinney and West, 1961; Stuart and Cathey, 1961; Audus, 1972; Weaver,

1972; Low, 1974; Watson, 1982) and increases in shoot to root weight

ratio (Stowe and Yamaki, 1957; Stuart and Cathey, 1961; Weaver, 1972;

Low, 1974).

Effects on fresh and dry weights reported for gibberellic acid

treatments are varied. Morgan and Mees (1956) reported increased vege-

tative growth of a majority of the species tested but no increases in

net yield. Ende and Koornneef (1960) reported a 25 percent increase in

stem length in tomato plants treated with gibberellic acid; however,

there was no significant difference in plant weights when compared to

untreated plants. Stuart and Cathey (1961) noted increases in yield of

dry matter in winter pasture grasses but no increase in total weight of

Eucalyptus. Stowe and Yamaki (1957) reviewed the effects of gibberel-

lins on fresh and dry weights of plants and concluded that gibberellin-

induced elongation does not always cause a parallel increase in dry

weight and, in fact, may result in a decrease depending upon the species.

Marth et al. (1956) investigated the effects on the weight of soybean


plants after foliar application of gibberellic acid. After one week

both the fresh and dry weights were significantly higher; however, after

two weeks there was no difference between treated and control plants.

Physiological effects of the gibberellins which account for the

change in growth, metabolism, and morphology of plants are as varied

and as contradictory as those reported above for the morphological

effects. These effects, however, can be grouped into the following

categories: cell division; cell elongation; osmotic potential; respira-

tion through enzyme metabolism; carbohydrate and lipid metabolism; and

changes in membrane permeability.

The exogenous application of gibberellins has been shown to produce

a pronounced increase in cell division in the subapical meristem of

Hyoscyamus niger and Samolus parviflorus (both of which are rosette

species), Phaseolus vulgaris (a non-rosette), and various other species

(Sachs and Lang, 1957; Weaver, 1972; Shininger, 1974). In a review by

Shininger (1974), increased cell division was noted in 13 of 21 species

treated with gibberellins. Salisbury and Ross (1978) hypothesized that

increased cell division by gibberellins may be caused by an increase in

the number of sites on the chromosome where DNA and RNA synthesis can

occur by unmasking initiation sites for DNA and RNA synthesis. Nitsan

and Lang (1966) demonstrated that elongation of lentil epicotyls in

response to gibberellic acid required DNA synthesis. Nakamura and

Takahashi (1973) also reported gibberellic acid enhanced DNA synthesis.

The response of a given cell to divide or elongate appeared to depend

upon its age or stage of development. Younger cells respond by dividing

while older cells respond by elongation only (Marth et al., 1956; Mann,

1974; Shininger, 1974).


Shininger (1974) reported gibberellin caused cell elongation in 9 of

21 species tested and both cell elongation and increased cell division

in 4 of the 21 species. The physiological mode of action for cellular

elongation results from numerous inter-related processes; however, the

sequence in which these processes occur remains uncertain. Gibberellins

have been shown to increase hydrolysis of starches, fructosans, and

sucrose molecules (Paleg, 1965; Audus, 1972; Kaufman, 1974; Salisbury

and Ross, 1978). Paleg (1960) demonstrated an increase in a-amylase

content of barley endosperm treated with gibberellic acid. Chen and

Park (1973) also demonstrated gibberellic acid enhanced amylase syn-

thesis and reported enhanced synthesis of proteins and RNA in Avena

fatua seeds. Weaver (1972) cited studies which attributed increased

enzyme activity to increased synthesis; however, Kaufman (1974) attri-

buted increased enzyme activity in barley aleurone to increased release

of enzymes rather than increased synthesis. Cleland et al. (1968)

reported increased levels of activity of cell wall hydrolases in Avena,

i.e., cellulase, hemicellulase, and pectinase, which has been shown to

result in increased plastic extensibility of the cell wall within one

hour after treatment with gibberellic acid (Adams et al., 1973; Montague

et al., 1973). The above conditions have been shown to result in

changes in osmotic potential at the cellular as well as whole plant

level. Ende and Koornneef (1960) reported a higher osmotic potential in

tomato plants; however, Castro (1976, cited in Castro and Rossetto,

1979) reported lower osmotic potential in tomatoes treated with gib-

berellic acid. Other plants in which lower osmotic potentials have

resulted from gibberellin treatment include lettuce (Stuart and Jones,

1977), sunflower (de la Guardia and Benllock, 1980), cucumber (Katsumi,


et al., 1980), and cotton (Castro and Rossetto, 1979). Castro and

Rossetto (1979) reported that gibberellic acid treatment lowered the

osmotic potential of cotton plants such that it interfered with aphid

feeding. Thus, gibberellic acid increased activity of hydrolytic enzy-

mes results in conditions leading to increased cellular growth by (1)

increased cellular osmotic concentration which permits water to enter

the cell more rapidly thus diluting the sugars and causing cell expansion

(Kato, 1956; Audus, 1972; Cleland et al., 1968; Salisbury and Ross,

1978; de la Guardia and Benllock, 1980); (2) increased cellular plasti-

city which allows the cell walls to stretch in response to the change in

osmotic potentials (Addicott, 1970; Stuart and Jones, 1977; Molz and

Boyer, 1978); and/or (3) increased availability of hexose molecules which

provide energy for respiration and the formation of pectin and

hemicellulose, the cell wall maxtrix polysaccharides and cellulose, the

microfibril fraction of the cell wall (Cleland et al., 1968; Kaufman,

1974; Salisbury and Ross, 1978).

Kaufman (1974) and more recently Rappaport (1980) reviewed control

points for the physiological activity of gibberellins. Figure 1, as

taken from Kaufman (1974), provides a schematic representation of mecha-

nism of gibberellic acid in growth metabolism at the cellular level as

discussed above. This scheme has the following basic features:

(a) Under natural conditions photosynthetically produced sugars

function as the substrate for tissue growth.

(b) Sugars enter the tissue at rates determined by transport

mechanisms and membrane permeability and become incorporated into a

pool of simple or phosphorylated sugars.

(c) Some of the saccharides are converted to storage carbohydrates

(starches and fructans); others are utilized in respiratory metabolism

or enter metabolic pathways leading to synthesis of cellular components.

(d) Respiratory cellular energy (ATP) is used by endergonic processes

such as active transport and synthetic pathways.

(e) Cell wall components and properties are adjusted in order to

allow for cellular growth.

(f) Protein synthesis is stimulated in order to allow for growth.

Many of these proteins are specific enzymes needed for catalyzation of

growth metabolism.

(g) The asterisks in Figure 1 represent possible control points at

which gibberellins are thought to regulate key metabolic pathways.

The above discussion indicates that most workers have concentrated

on metabolic pathways to explain gibberellin actions. An increasingly

popular explanation is in terms of alterations of membrane permeability

(Rappaport, 1980). Wood and Paleg (1972) were the first to clearly

demonstrate that gibberellic acid can influence the permeability of

model membranes. Wood and Paleg (1972) proposed that effects on sub-

cellular membrane permeability were due a biophysical alteration of one

or more of the membrane components through some sort of bond formation.

Involvement of endoplasmic reticulum and increased enzyme synthesis

was summarized by Mann (1974) and further confirmed with electron

microscopy by Pyliotis et al. (1979). However, as noted by Kaufman

(1974), no one mechanism has been ellucidated for gibberellins and

regulatory sites of this hormone appear to occur at several points in

plant growth metabolism.




L --p

1. -

cn cc
,. c U
>- ,, ,-

d -p

-N- -
C C..)

/ L %
I ,l
*E- I'

*, z C-)
0 D 0 0
m uj

U3 u) >-
0 (- 0

Q: 0

Q. 3:
0- =












o W


C.. --

= 0




Interaction with Other Plant Growth Substances

Interaction or synergism between gibberellins and other plant

growth substances, auxins in particular, has been evaluated in a variety

of plant species and experimental conditions. Nitsch and Nitsch (1956)

evaluated effects of various combinations of IAA and GA3 on oat

coleoptiles and the first internode and demonstrated less than additive

growth responses in all but one combination of the two substances. Kato

(1958b) noted increased growth response of GA and IAA in cucumber over

IAA alone; however, since the response was less than additive, no

synergism was implied. Marth et al. (1956) commented that wounding

of several species prior to application of GA3 increased the

GA3 response, suggesting synergism with wound hormones. Stowe and

Yamaki (1957) cited increased responses to combinations of IAA and gib-

berellin which were more than additive in pea epicotyls. Ng and Audus

(1964 and 1965) demonstrated the requirement for an unidentified endoge-

nous substance from Avena internodes in order to induce a synergistic

interaction between GA3 and applied IAA or 2,4-D on Avena stem segments.

Audus (1972) reported a three-fold increase in sensitivity of rice

leaves to IAA when also treated with GA3. Pieterse et al. (1980) and

Pieterse and Roorda (1982) reported a tenfold increase in the activity

of 2,4-D on Eichhornia crassipes when applied in combination with GA3.

The observed interactions between gibberellins and applied natural

and synthetic auxins have revolved around two possible mechanisms (1)

an alteration of the endogenous auxin levels, or (2) an alteration of

the rate of translocation of auxins. Numerous studies have reported

enhanced biosynthesis of IAA gibberellin (Galston and Purves, 1960;

Paleg, 1965; Varga and Humphries, 1974; Maheshwari et al., 1980).

However, others have reported that observed synergistic interactions

were due to an auxin-sparing reaction wherein applied gibberellins inhi-

bited or reduced the concentration of auxin degrading enzymes such as

IAA oxidase (Brian and Hemming, 1958; Kogl and Elema, 1960, cited in

Weaver, 1972; Galston and Purves, 1960; Sarma, 1979).

Ashton (1959) postulated that the reason 2,4-D was more effective as

a herbicide in plants that were actively growing was due to an increase

in translocation; therefore, the growth promoting properties of GA might

increase the effectiveness of 2,4-D. To test this hypothesis red kidney

bean plants were pretreated with 100 mg/l of the potassium salt of GA3

prior to the treatment of a single primary leaf with radioactively

labeled 2,4-D. Twenty-four hours after 2,4-D treatment the amount of

2,4-D in the whole plant was higher than plants not receiving GA3

treatment. This effect disappeared after 72 hours and was shown not to

be due to a reduction in the breakdown of 2,4-D by GA3. Similar results

were reported with 2,4-D by Basler (1959) and 2,4,5-T by Basler (1974)

utilizing Phaseolus vulgaris. Pilet (1965) reported pretreatment of

Lens culinaris epicotyl segments with GA3 increased the uptake and velo-

city of movement of applied IAA out of apical growing regions. Basler

(1974) demonstrated that increased translocation of 2,4,5-T by

GA3 appeared to be specific to 2,4-5-T, since the translocation of

labeled sucrose-3H and glycine-14C were relatively unaffected by

GA3 after 4 to 24 hours post treatment. Pieterse and Roorda (1982) also

hypothesized that increased sensitivity-of waterhyacinths to 2,4-D

when combined with GA3 was due to increased translocation. The only


study discussed above which either suggested or investigated a possible

mechanism for GA3-enhanced translocation of auxin compounds was that of

Basler (1974). Utilizing cycloheximide, a protein synthesis inhibitor,

it was shown that continuous protein synthesis was needed to maintain

high rates of 2,4,5-T translocation in bean seedlings. The simultaneous

treatment of 2,4,5-T, GA3 and cyclohexamide negated the enhancement of

translocation of 2,4,5-T by GA3. In an unrelated study utilizing corn

(Zea mays) seedlings, which does however demonstrate the interrela-

tionship of various plant growth substances, synergism was demonstrated

between GA3 and fluridone, a herbicide which blocks carotenoid synthesis

(Devlin et al., 1980). The exact mechanism of synergism was unknown;

however, Devlin et al. (1980) suggested that since cartenoids are pre-

cursors of abscisic acid (ABA) and since ABA and GA had been shown to be

mutally antagonistic (Corcoran, 1974), reduced ABA levels caused by

the absence of carotenoids might allow greater expression of

GA3 activity.

Audus (1972) summarized the interaction of plant hormones by stating

that there may be no underlying interdependence of gibberellins and

auxin-like compounds because (1) gibberellins will produce different

metabolic responses in different tissues, and (2) growth or develop-

mental process may be subject to the regulatory action of a balance of

several hormones whose points of action may be quite independent but

appear to interact due to the mutual association of each component to

the total growth system of a given species.

Commercial Applications

The numerous physiological and morphological effects of gibberellins

have led to investigations into potential commercial uses. Initial


applications of gibberellins to seeds, soil, or growing plants for the

purpose of increasing crop yields generally have not produced

encouraging results. In comprehensive field trials by Morgan and Mees

(1956), treatment with gibberellic acid failed to increase yields in

wheat, potatoes, turnips, carrots, peas, runner beans, lettuce, celery,

black currants, kale, and corn. During these trials, vegetative growth

of most of the treated plants was stimulated but no increase in crop

yield occurred. Marth et al. (1956) conducted a similar series of

trials utilizing 42 different species of plants and also reported

increases mainly in vegetative growth. Stuart and Cathey (1961)

reviewed the applied aspects of gibberellins and reported that gib-

berellins are applied to plants cultivated for their flowers in order

to (1) replace the requirement for cold temperatures for flowering;

(2) accelerate flowering; and (3) enlarge and extend the lasting quality

of inflorescences. Current routine commercial uses of gibberellins, as

reported by Salisbury and Ross (1978), include increasing the size and

distance between Thompson seedless grapes, increasing the rate of the

malting processes in breweries, increasing stalk length and

crispness of celery, delaying senescence in various fruits, and as an

aid in fruit set. Abbott Laboratories, Inc., markets gibberellic acid

for the treatment of turf grasses in order to initiate growth and pre-

vent color change during cold stress. Abbott Laboratories has also

registered gibberellic acid under the brand name Pro-Gibb (EPA

registration numbers 275-20, 275-15, and 275-12) for use on various

crops. Rates of application under these registrations vary from

1.1 g/ha to obtain uniform bolting and increase lettuce seed production


to 208 g/ha to increase sucrose yield in sugar cane. No routine commer-

cial use of gibberellins in aquatic environments could be found.

Stuart and Cathey (1961) observed that gibberellins often induce

maximum responses when growth conditions are adversely affected by

temperature, nutrition, light or other environmental factors, and postu-

lated that under these conditions the synthesis of endogenous gib-

berellins and gibberellin inhibitors may be retarded permitting the

greater response from applied gibberellins. The major limiting factors

to the use of gibberellins have been the additional cost, the failure to

increase yields of major crops, and the fact that the only method of pro-

ducing them is through the culture of the Gibberella fungus (Salisbury

and Ross, 1978). Stuart and Cathey (1961) stated that future commercial

applications will depend upon additional data concerning methods and

timing of application and the interaction or synergism with other endo-

genous and applied plant growth regulators. Based on this review, this

statement still appears to apply to gibberellin technology.

Effects on Waterhyacinths

Pieterse et al. (1976) reported that the formation of bulbous or

"float" type waterhyacinth leaves could be inhibited by growing the

plants in low concentrations (0.03 mg/1) of GA3 in water under

greenhouse conditions. The ratio of petiole length and the greatest

circumference was approximately 1.0 at 0.00 mg/l GA3, 4.20 at 0.03 mg/l

GS3 and 12.4 at 1.00 mg/l GA3. The general growth pattern of leaves

following treatment with low concentrations of GA3 was characterized as

similar to morphological responses of plants growing in dense stands

or rooted in the shallow hydrosoil as described by Penfound and Earle


(1948) and Center and Spencer (1981). GA3 also markedly inhibited vege-

tative reproduction and induced profuse flowering. Watson et al. (1982)

reported similar effects on leaf morphogenesis, vegetative repro-

duction, and flower production when GA3 (0.00 to 0.05 mg/1) was applied

to waterhyacinth roots. However, at concentrations greater than 0.05

mg/l GA3, higher rates of production of inflorescences and daughter roset-

tes were observed. Watson et al. (1982) also reported an increase in

stolon elongation at GA3 concentrations up to 0.03 mg/l. However, above

0.05 mg/l, GA3 caused a decrease in stolon length and suppressed stolon

production at 1.0 mg/l. Foliar application of GA3 at a rate of 6.0 g/ha

was also shown to inhibit the formation of float-type leaves although

apparently not as dramatically as when GA3 was applied to the roots

(Pieterse et al., 1980; Pieterse and Roorda, 1982).

2,4-Dichlorophenoxyacetic Acid


The chemical, 2,4-dichlorophenoxyacetic acid (2,4-D), is a systemic

herbicide which is routinely used for the control of broadleaf weeds

(Weed Science Society of America, 1979). The synthesis of 2,4-D was a

result of research initiated in the 1880's by Julius Sachs, a German

botanist, who suggested the presence of plant organ-forming substances

(Salisbury and Ross, 1978). The existence of naturally occurring

compounds which stimulated plant growth was first demonstrated by

Seubert in 1925 (Weaver, 1972). In 1926, Went developed a procedure for

quantitative isolation of growth-promoting compounds which stimulated

their isolation and identification. Went was also the first to

call these compounds auxins (Weaver, 1972; Salisbury and Ross, 1978).


Auxin has become a general term for a group of compounds which typically

induce elongation of shoot cells (Brian et al., 1955; Weaver, 1972) and

they typically produce physiological responses similar to indoleacetic

acid (IAA), a naturally occurring growth substance (Brian et al., 1955;

Weaver, 1972; Black and Buchanan, 1980). The auxin discovered by Went

was indoleacetic acid, IAA (Salisbury and Ross, 1978) and is considered

to be the most common auxin occurring in higher plants.

IAA was found to be relatively unstable and a search was begun for

synthetic compounds of similar chemical constitution and growth-promoting

activity. In 1942, work was begun with a series of substituted pheno-

xyacetic acids of which 2,4-D is a member (Zimmerman and Hitchcock,

1942). Two,4-D is considered to be a synthetic auxin because it causes

growth reactions similar to the naturally occurring indole auxins;

however, it is more active and persists in the plant for a longer

period of time than IAA (Van Overbeek, 1964). With the advent of World

War II the idea developed that auxins might be used in high con-

centrations to kill enemy crops or limit crop yields (Norman, 1946),

and the initiation of tests to study the potency of these new

compounds including 2,4-D was begun. Many of these studies were con-

ducted in 1944 and 1945 as a part of the activities of the Special

Projects Division, Chemical Warfare Service at Camp Detrick in Fredrick,

Maryland (Norman, 1946). Publication of this early work was delayed due

to wartime security policies (Norman, 1946); however, the June, 1946

issue of the Botantical Gazette contained 18 papers describing the

results of the Camp Detrick studies. Commercial application of

2,4-D began soon after early trials such as those conducted by Blackman

(1945) which demonstrated that broad-leaved weeds growing in grain


fields could be selectively killed without injury to adjacent cereal

crops. Herbicidal effects of 2,4-D were described by F. D. Jones in

U. S. Patent No. 2,390,941 and this patent was assigned to the American

Chemical Paint Company (Weed Science Society of America, 1979). Black

and Buchanan (1980) credit recognition of the herbicidal properties

of 2,4-D as the catalytic discovery which led to development of many

chemically related herbicides and to development of weed control as

a scientific discipline.

Physical and Chemical Characteristics
Commercial concentrates of 2,4-D are generally formulated as salts

or esters. The pure acid is a white, odorless crystal with a molecular

weight of 221 and a molecular formula of C8H6C1203 (Weed Science Society

of America, 1979). According to the Weed Science Society of America

(1979), the solubility of 2,4-D acid formulation in various solvents is

as follows:
Solvent g/lOO1g solvent

Acetone 85.0

Diesel oil and kersosene 0.10 to 0.35

Ethanol, 50 percent 10.3

Ethyl alcohol, 95 percent 130.0

Isopropanol 31.6

Water 0.09

The dimethylamine salt formulation is extremely soluble, 300 g/lOOg of

water, soluble in alcohols and acetone, but insoluble in kerosene and

diesel oil. The butoxyethanol ester of 2,4-D is insoluble in water, but

soluble in most organic solvents.


The pure acid formulation of 2,4-D is prepared by combining

2,4-dichlorophenol and monochloroacetic acid. The salts are formulated

by addition of amines or inorganic hydroxide to the pure acid. Esters

are produced by reaction of 2,4-D with the appropriate alcohols.


Effects of various formulations of 2,4-D have been evaluated for

a wide range of organisms in both the aquatic and terrestrial food

chains. Studies have shown that toxicity of 2,4-D varies with the

formulation utilized in the evaluations (Davis, 1970; Duke, 1971; Weed

Science Society of America, 1979; Halter, 1980). Acute oral toxicity

(LD50) of the various formulations of 2,4-D has been reported to be

300 to 1,000 mg/kg for rats, guinea pigs, and rabbits (Weed Science

Society of America, 1979). Hansen et al. (1971) reported that rats fed

from 0 to 1250 mg/kg 2,4-D acid for two years exhibited no significant

effect on growth rate, survival rate, organ weights, or hematologic

values. Ninety-three percent of dogs fed from 0 to 500 mg/kg 2,4-D acid

in their diet for two years were clinically normal with no 2,4-D related

effects noted (Hansen et al., 1971). In a critical review of the

effects of 2,4-D on the aquatic environment, Halter (1980) reported no

effects of 2,4-D acid on phytoplankton at rates as high as 300 mg/1.

Following a large-scale application of the BEE formulation at a rate of

49.7 kg/ha (a.e.), Whitney et al. (1973) reported no difference in

plankton populations following herbicide treatment. A review of the

effects of both the DMA and BEE formulations on benthic invertebrates

following routine aquatic weed control operations indicates no adverse

effects due to the herbicide, other than those caused due to habitat


changes (Halter, 1980). Davis and Hardcastle (1959) reported a 48 hour

LC50 of 375 and 350 mg/1 of the DMA formulation for bluegills (Lepomis

macrochirus) and largemouth bass (Micropterus salmoides) respectively.

However, LC50 values for bluegills as low as 2.1 mg/l have been reported

for the liquid BEE formulation and 34.5 mg/l for granular BEE (Hughes

and Davis, 1965). Halter (1980) indicates that 2,4-D is essentially

non-toxic to waterfowl. Duke (1971) reported avoidance by mosquitofish

(Gambusia affinis) of water treated with the BEE formulation of 2,4-D.

Mosquitofish sought water free of 1.0 and 10.0 mg/l 2,4-D, but did

not seek water free from 0.1 mg/l (Duke, 1971). Some formulations may

cause skin irritation in humans, but no characteristic symptoms of

poisoning are documented for humans (Weed Science Society of America,

1979). Moore (1974) reported that electron microscope assay procedures

for herbicide and membrane interactions indicated that animals and algae

lack biochemical mechanisms (specific binding sites) to respond to

2,4-D. Tolerances for residues resulting from the aquatic application

of 2,4-D in food and raw agricultural commodities have been established

at 0.10 mg/l for potable water (21 Code of Federal Regulations 123.100,

dated December 16, 1975) and 1.0 mg/l for fish and shellfish (40 Code of

Federal Regulations 180.142, dated December 9, 1975).

Persistence in the Environment

In terrestrial situations, 2,4-D undergoes microbial breakdown in

warm, moist soils in one to four weeks. The actual rate of decom-

position depends upon the temperature, moisture, organic matter, and

other soil characteristics (Hemmett and Faust, 1969; Weed Science

Society of America, 1979; Halter, 1980). Halter (1980) reviewed


thirty-four papers concerning the persistence of 2,4-0 in water under

both laboratory and field conditions and concluded (1) under laboratory

conditions, 2,4-D repeatedly decomposed in water in periods of hours to

days; (2) under some warm water field conditions, 2,4-D has repeatedly

been shown to be reduced to non-detectable levels (low ppb range) in

closed water bodies in approximately one month; and (3) persistence of

2,4-D at extremely low levels may be encouraged by water movements in

lakes, reservoirs, and streams. Joyce and Sikka (1977) randomly moni-

tored 2,4-D levels in a large riverine system for seven months in con-

junction with routine waterhyacinth control operations utilizing 2.24

to 4.48 kg(a.e.)/ha 2,4-D DMA and reported 2,4-D levels from non-

detectable to 1.3 pg/l. Smith and Isom (1967) reported similar residue

levels in large freshwater reservoirs in the Tennessee Valley in con-

junction with Eurasian watermilfoil (Myriophyllum spicatum L.) treatments

at 44.8 to 112.0 kg(a.e.)/ha 2,4-D BEE. Schultz (1973) reported that

2,4-D DMA persists in the hydrosoil at the mg/1 level for about one

month, whereas Smith and Isom (1967) documented the persistence of 2,4-D

BEE in hydrosoil at 58.8 mg/kg 10 months after treatment. Hemmett and

Faust (1969) demonstrated that biodegradation of 2,4-D follows zero-

order kinetics, with the oxidation rate independent of substrate

(2,4-D) concentration. The rate was dependent upon (1) period of time

in which the system has acclimatized to 2,4-D; and (2) the natural con-

dition of the aquatic environment. The various formulations of 2,4-D

also do not persist or bioaccumulate in fish (Schultz, 1973; Whitney et

al., 1973; Sikka et al., 1977; Halter, 1980), blue crabs, Callinectes

sapidus (Joyce and Sikka, 1977) and benthic invertebrates (Whitney et

al., 1973; Halter, 1980). Hildebrand (1946) during the first documented


application of 2,4-D to waterhyacinths noted "no adverse effects to

water fauna" during and after the application of 2,4-D. After over

30 years of continuous use in aquatic environments, monitoring continues

to indicate no adverse effects to water fauna due directly to 2,4-D

(Smith and Isom, 1967; Whitney et al., 1973; Moore, 1974).

Mode of Action

The mode of action of 2,4-D has been studied more than any other

herbicide (Ashton and Crafts, 1973; Mullison, 1982). Most reviews of the

mode of action of 2,4-D indicate that it affects almost every biological

activity of a plant (Brian, 1964; Carns and Addicott, 1964; Kiermayer,

1964; Wort, 1964a; Ashton and Crafts, 1973; and Mullison, 1982).

However, the primary mechanism and site of action has not been clearly

established (Ashton and Crafts, 1973, Weed Science Society of America,

1979; Black and Buchanan, 1980; Mullison, 1982). Van Overbeek (1964)

and Black and Buchanan (1980) suggested that the growth of a plant is

regulated by rhythmic fluctuations in levels and locations of

plant growth substances. This fluctuation is interrupted by 2,4-D and

orderly plant development is altered. Immature cytoplasm is pre-

vented from maturing and mature cytoplasm reverts back to an immature

physiological state (Van Overbeek, 1964). Therefore, 2,4-D can be

effective throughout the life of susceptable species, but is especially

effective during immature stages when endogenous levels of growth

hormones are highest and the plant is actively growing (Black and

Buchanan, 1980).

Morphological and physiological responses by plants to 2,4-D depend

upon the sensitivity and physiological condition of the treated species


(Ashton and Crafts, 1973; Mullison, 1982) and the rate and type of for-

mulation applied (Weaver, 1972). At low application rates, responses are

typical of those caused by plant growth substances and may induce

rooting, blossom set, ripening of fruit, and delaying preharvest drop

(Wort, 1964b; Weed Science Society of America, 1979). At higher appli-

cation rates, 2,4-D, typically causes epinasty or downward twisting

and bending of stems and petioles and curling of leaves (Cardenas et

al., 1968; Weaver, 1972; Black and Buchanan, 1980; Mullison, 1982).

Young leaves cease expanding due to cell elongation, photosynthesis is

reduced, and chlorosis may occur (Turkey et al., 1945; Van Overbeek,

1964; Aston and Crafts, 1973). As the herbicide is translocated through

the plant, mature parenchyma cells tend to first swell and then divide

radially more rapidly producing callus tissue and root primorida which

results in the blockage of phloem tissue and the cessation of assimilate

transport in the phloem (Turkey et al., 1945; Van Overbeek, 1964; Ashton

and Crafts, 1973). Meristem activity is inhibited and new organ or

lateral bud growth may occur (Weaver, 1972). Growth of mature or

primary roots is inhibited and roots may lose the ability to take up

water and salts (Van Overbeek, 1964; Wort, 1964a; Cardenas et al., 1968;

Mullison, 1982). The progression of these responses ultimately leads to

the withering, collapse, and death of 2,4-D sensitive species due to a

combination of new and irregular leaf and root growth and inadequate

nutrition because of phloem blockage (Ashton and Crafts, 1973). At high

application rates, 2,4-D functions as a contact herbicide, does not

translocate throughout the plant and thus may not completely kill

meristematic tissue (Ashton and Crafts, 1973).


The physiological response of plants to 2,4-D has been attributed to

the inhibition and/or stimulation of respiration, blockage of pro-

toplasmic streaming, alteration of DNA transcription and/or RNA

translocation, and interference with enzyme regulatory systems (Brian,

1964; Wort, 1964a; Ashton and Crafts, 1973; Yamada et al., 1974;

Mullison, 1982). During the mid-1950's numerous papers appeared which

linked auxin action with nucleic acids (Ashton and Crafts, 1973). West

et al. (1960) reported an increase in the protein and RNA content of

cucumber stem tissue when treated with herbicidal concentrations of

2,4-D. This led to the assumption that the cytochemical basis of 2,4-D

action was a stimulation of nuclear activity and a reversion to meriste-

matic metabolism (Chrispeels and Hanson, 1962). It was later shown that

this increase in RNA was primarily in the ribosomal fraction which was

also accompanied by an increase in a messenger type RNA (Key and

Shannon, 1964; Key, 1964). Shannon et al. (1964) suggested that 2,4-D

induced protein synthesis and excess nucleic acids would preclude normal

cell function and that this was the biochemical basis for the herbicidal

action of 2,4-D. Chen et al. (1972) experimented with a 2,4-D tolerant

wheat species and a 2,4-D sensitive cucumber species and demonstrated

that as 2,4-D concentrations increased, the RNA content of wheat showed

a net decrease, whereas the cucumber RNA content increased by over 200

percent. Chen et al. (1972) postulated that the ability to resist

alteration of RNA species by a plant was the basis for the selectivity

of 2,4-D. Ashton and Crafts (1973) summarized the effects of 2,4-D on

nucleic acid and protein content of susceptable species and indicated

that this was the main biochemical reaction to 2,4-D. This increase in

RNA was attributed to an inhibition of the synthesis of ribosomal RNase


which prevented RNA catabolism. The presence of high levels of ribo-

nucleases in other resistent grasses has furthered this hypothesis

(Ashton and Crafts, 1973). Wort (1964a) reviewed the effects of 2,4-D

on cellular enzymes and documented effects on the activity of 16 enzymes

including amylase, ascorbic acid oxidase, IAA oxidase, invertase,

phosphorylase, and proteolytic enzymes. These changes were postulated

to be caused by changing (1) the cellular conditions such as pH or

hydration under which enzymatic progress occurs; (2) the supply of

material for enzyme formation; and/or (3) the supply of energy necessary

for endergonic reactions through alteration of ATP production (Wort,

1964a). In support of these latter two mechanisms, Mostafa and Fang

(1971) hypothesized seven sites where 2,4-D either inhibits or regulates

respiratory breakdown of glucose thus increasing or decreasing the

concentration of various metabolities. Various studies have documented

inhibition of the Hill reaction and oxidative phosphorylation; however,

Ashton and Crafts (1973) consider these effects to be of secondary


Much of the discussion surrounding 2,4-D mode of action deals with

which response is the cause and which is the effect of auxins and auxin-

like substances (Audus, 1972). Morgan and Hall (1962) documented that

2,4-D treated cotton plants exhibited an increased release of ethylene,

a gaseous plant hormone which can cause epinasty. Ashton and Crafts

(1973) attributed the initial epinastic effects to cell elongation and

not to an ethylene effect; however, subsequent cell divisions and cell

proliferation are perhaps caused by ethylene, resulting in phloem

blockage and eventually death.


Ashton and Crafts (1973) summarized this issue by stating, "the mode

of action of the chlorophenoxy compounds must consist of a great number

of structural and biochemical reactions revolving around the central

theme of prolonged abnormal growth with failure of those changes charac-

teristic of maturity and senescence. In no other way may the great

number and diversity of structural and metabolic changes be reconciled."

(Ashton and Crafts, 1973, page 284).


Polar salts of 2,4-D are absorbed more readily by the roots of

most species, whereas ester formulations are more readily absorbed

by leaves (Weaver, 1972; Weeds Science Society of America, 1979).

Foliar-absorbed 2,4-D is transported polarly within the phloem with

assimulated sugars, and root-absorbed 2,4-D moves upward in the xylem

during transpiration (Audus, 1972; Weaver, 1972). Movement is towards

rapidly growing tissues, such as developing flowers and meristematic

shoots and roots (Audus, 1972; Weaver, 1972; Ashton and Crafts, 1973).

Thus, thorough distribution of a herbicide, such as 2,4-D, is dependent

upon the active movement of foods in the plant. However, not all

species absorb and translocate 2,4-D at the same rate. Fang (1958)

demonstrated with radioactive labeled 2,4-D that the rate of transloca-

tion of 2,4-D was slower in peas and tomatoes than in bean plants.

Morphological characteristics that prevent 2,4-D absorption and translo-

cation and the plant's ability to conjugate or metabolize 2,4-D have

also been suggested as mechanisms of selectivity by tolerant species

(Fang, 1958; Mullison, 1982).

Metabolism by Plants

A herbicide has been defined as a compound which deranges the phy-

siology of a plant over a period long enough to kill it (Van Overbeek,

1964). Thus, those plants which can more effectively metabolize or

inactivate the herbicide generally exhibit less sensitivity to the

herbicide. Numerous studies indicate that plants exhibit varying abili-

ties to either metabolize or inactivate 2,4-0 once it has entered the

plant (Weintraub et al. 1952; Fang, 1958; Crafts, 1964; Van Overbeek,

1964; Ashton and Crafts, 1973; Feung et al., 1978). The most common

mode of metabolism appears to be decarboxylation and hydrolysis to

free phenol (Crafts, 1964; Ashton and Crafts, 1973). One of the first

investigations into the metabolism of 2,4-D by plants was conducted by

Weintraub et al. (1952) utilizing radioactive 2,4-D with 14C in either

the carboxyl, ethylene, or ring positions. These studies indicated that
14C02 was readily released from the carboxyl position and was not

released from phenol ring positions. Weintraub et al. (1952) also

documented the presence of a wide variety of metabolities. Ashton and

Crafts (1973) suggested that a plant's ability to oxidize carboxyl and

ethylene carbon atoms of 2,4-D correlated with its tolerance to 2,4-D.

Other mechanisms of 2,4-D metabolism or inactivation by plants include

(1) ring hydroxylation followed by oxidation of hydroxyls to car-

boxyls with ring splitting (Crafts, 1964; Ashton and Crafts, 1973; Feung

et al., 1978); (2) completing with proteins (Fang, 1958; Crafts, 1964;

Van Overbeek, 1964); (3) completing with amino acids followed by ring

hydroxylation (Feung et al., 1978); (4) sugar conjugation (Feung et al.,

1978); and (5) ring hydroxylation followed by sugar conjugation (Feung

et al., 1978; Weed Science Society of America, 1979).


Efficacy of 2,4-D on Waterhyacinths

The first recorded use of 2,4-D for the control of waterhyacinths

appeared to be in April 1945 near Tampa, Florida (Hildebrand, 1946).

Hildebrand (1946) applied 0.059 to 0.125 percent (by volume) 2,4-D solu-

tions to 0.006 ha plots and reported nearly complete control. Seale and

Allison (1946) evaluated five 2,4-D esters, four 2,4-D amine salts, and

five inorganic salts of 2,4-D and reported that the ester formulations

were the most effective; however, at higher rates (> 2.24 kg/ha in

934 1/ha aqueous solution) the amine and inorganic salt formulations

were equally effective. Seale and Allison (1946) reported the first

effective aerial application of 2,4-D to waterhyacinths. Hitchcock et

al. (1949) conducted extensive evaluations at various rates (2.24 to

17.92 kg/ha) and spray volumes (56.0 to 168 1/ha) and concluded

that the principal limiting factors affecting 2,4-D efficacy for

waterhyacinth control were (a) concentration and rate of application,

(b) rate of delivery of spray solution, (c) stage of growth and

development, and (d) atmospheric conditions. Application of 2,4-D

to mature waterhyacinths with daughter plants attached by stolons

resulted in death of the parent plants; however, daughter plants

survived (Hitchcock et al., 1949). Singh and Muller (1979b) used

radioactively labeled 2,4-D to confirm these observations by

demonstrating maximum transport of labeled 2,4-D to daughter

plants from the parent up to the two-leaf stage of the daughter plants

and no translocation after the daughter plants reached the four-leaf

stage. Singh and Muller (1979a) also demonstrated that radioactively

labeled 2,4-D when applied to a single waterhyacinth leaf was

transported (a) in small amounts (23.5 percent of total applied); (b) at

a slow rate (maximum amount in six days); and (c) almost entirely

towards the newly developing leaves (20.8 percent of total applied).

Singh and Muller (1979a) used radioactively labeled 2,4-D to

demonstrate that three hours after spraying single waterhyacinths at a

rate of 0.75 kg/ha in a spray volume of 800 1/ha, 53.3 percent of the

total sprayed solution was in the water culture medium. It was also

shown that waterhyacinths grown in culture medium containing labeled

2,4-D can absorb and translocate sufficient 2,4-D from treated water to

result in their death if the 2,4-D concentration is above 1.0 mg/l.

Thus, Singh and Muller (1979a) suggested that immediately after spraying

2,4-D in the field, the upper surface layer of water, if exposed, might

contain concentrations of 2,4-D which would be readily available for

uptake by waterhyacinth roots. Others have suggested that 2,4-D which

is added to the water surface from spray drift or runoff has an added

effect and may be a factor in explaining why waterhyacinths grown in

small containers in greenhouse experiments appear to be more sensitive

to lower rates of 2,4-D as compared to field studies (Hildebrand, 1946;

Hitchcock et al., 1949; Koch et al., 1978).

Koch et al. (1978) reported an aerial application rate of 4.5 kg/ha

in spray volumes as low as 15 1/ha in the White Nile River, Sudan. This

rate was deemed effective due to the low relative humidity and high tem-

peratures in southern Sudan. The reliability of applications was also

deemed to be reduced at spray volumes <100 1/ha. Public agencies

responsible for aquatic plant control in Florida routinely apply 2,4-D

to waterhyacinths at rates of 2.24 kg/ha in aqueous spray volumes of 467

to 934 1/ha for ground applications and up to 4.48 kg/ha in spray vol-

umes of 56.7 1/ha for aerial applications. The U.S. Army Corps of


Engineers, Jacksonville District, applies these latter rates of 2,4-D

under a control program which maintains the waterhyacinth population at

a minimum non-problematic level (Joyce, 1977). This approach has

reduced by 50 percent the quantity of herbicide utilized annually on the

St. Johns River, Florida, for waterhyacinth control (McGehee, 1982).



Pieterse et al. (1980) reported a ten-fold increase in the sen-

sitivity of waterhyacinths to 2,4-D due to a synergistic effect of 2,4-D

and gibberellic acid. These investigations were conducted under

greenhouse conditions in 200 x 100 x 50 cm concrete reservoirs. Plants

were first treated with an atomizing spray system with concentrations of

2,4-D (amine salt) ranging from 0 to 1000 g/ha at a volume rate of

200 1/ha. GA was then applied to control plants and plants which

received 0, 50, 100, and 200 g/ha of 2,4-D. Concentrations of

GA3 utilized were 0, 2, 4, 6, and 8 g/ha at a volume rate of 200 1/ha.

Results of these experiments indicated that combinations of GA and 2,4-D

at 6 to 8 g/ha and 100 g/ha, respectively, caused death of the plants

within one week. The same response was noted in plants which received

only 1000 g/ha 2,4-D. Pieterse and Roorda (1982) reported similar

results when the 2,4-D and GA were applied simultaneously in the same

solution at an extremely low volume rate of 40 1/ha.

Based on results of these two studies, it was hypothesized that (1)

such a large reduction in the quantity of 2,4-D might lower the costs of

control programs even though costs of GA is relatively high (Pieterse et

al., 1980) and (2) a decrease in the 2,4-D concentration would lower the

risk to nearby native vegetation or crops (Pieterse and Roorda, 1982).

The objective of this part of the overall study was to evaluate this


reported synergism of GA and 2,4-D for the control of waterhyacinths in

an outdoor environment in order to eliminate the "greenhouse" effect

reported by Hitchcock et al. (1949) and Koch et al. (1978).

Materials and Methods

Waterhyacinths used in this study were collected from Lake Ocklawaha

near Palatka, Florida, and a small tributary stream of the St. Johns

River in Jacksonville, Florida. Plants were maintained in outdoor pools

prior to experimental use or placed directly into the experimental con-

tainers after field collection. Regardless of initial source, all plants

were allowed to remain in the experimental containers for a period of

three to four days prior to initial weighing and treatment. Experiments

were conducted during the 1980, 1981, and 1982 growing seasons.

Initial experiments were conducted outdoors in full sunlight in

approximately 70-liter metal barrels lined with polyethylene bags. All

barrels were filled with tap water to within 2.5 cm of the top. Eight

grams of commercially prepared 20-20-20 soluble plant nutrients and

micronutrients were dissolved in the barrels to yield a calculated

nitrogen concentration of 20 to 22 mg/l (approximately equivalent to 10.0

percent Hoagland's solution). Solution levels were maintained as

necessary by addition of tap water. Experiments were set up in three

replications of five to six plants each, depending upon size and weight of

the plants. Waterhyacinth plants used in the experiments were selected

on the basis of appearance, freedom of disease, and uniformity in size.

Prior to placement into barrels, all dead plant material, flower

spikes, and daughter plants were removed. Plants were allowed to drain

and were lightly shaken by hand in order to remove excess water from the


roots prior to weighing. Initial fresh weights were obtained on a

Mettler balance and recorded to the nearest 0.0 g. During the first five

series of treatment replications, 45 individual plants were sampled for

determination of the percent dry weight which was used for calculation

of initial dry weights. The percent dry weight was consistent (4.72

percent) and always within reported values (Penfound and Earle, 1948;

Bock, 1966; Westlake, 1963; Knipling et al., 1970).

Treatments were conducted in a completely randomized block design.

Treatments consisted of the simultaneous application of various com-

binations of gibberellic acid (GA3, Eastman Co.) and 2,4-D (Union Carbide

Corporation) at the following rates 0.0, 23.5, 47.0, 94.0, and 188 g/ha

GA3 and 0.0, 0.28, 0.56, and 1.12 kg/ha 2,4-D. During each

series of treatments, an additional treatment was made at a rate of 2.24

kg/ha 2,4-D to simulate routine operational field application rates.

Each treatment rate was replicated three times per treatment series and

all treatment series were replicated at least twice as separate trials

(blocks). One additional treatment series in which the 2,4-D con-

centration was held constant and GA3 concentration was varied between 0.0

to 188 g/ha was also included in the analysis. Treatment volumes

were approximately 934 1/ha. All applications were made with a hand-

held sprayer from a height of 30 cm.

Each treatment series was observed at least twice weekly for evidence

of treatment effects and to adjust the water level in the barrels.

Plants were harvested after 14 to 17 days. Efficacy evaluations con-

sisted of counting the number of viable plants as determined by the

presence of a living meristem (Seale and Allison, 1946), removing necrotic


tissue, and weighing the plants to the nearest 0.0 g in order to determine

post-treatment biomass. Periodically, plants were placed in a drying

oven at 65 C for 72 hours to determine dry to fresh weight ratios. Dry

to fresh weight ratios obtained from subsamples were used to calcu-

late final dry weights of all treated plants.

Results were analyzed by general linear regression procedures. The

resulting analysis of variance was used to conduct a trend analysis for

regression of the response of the mean percent change in biomass and

number of plants caused by the main effects of GA3 and 2,4-D and their

interaction (Chew, 1977). Means of the percent change in biomass and

number of plants due to discrete treatment levels were also compared

using the Waller-Duncan procedure. The Waller-Duncan procedure was

employed in lieu of other multiple comparison methods because it has the

advantage of using the observed overall F-value in the calculation of

the least significant difference (LSD). This characteristic provides a

mechanism for accounting for both the comparisonwise and experimentwise

Type I error rates, the lack of which has been a major criticism of the

standard Duncan Multiple Range Test (Chew, 1977).

Results and Discussion

Tables 1-1 and 1-2 present the analysis of variance of mean percent

changes in dry weight and number of waterhyacinths, respectively, due to

treatment with combinations of GA3 and 2,4-D excluding the 2.24 kg/ha

2,4-D treatment. Analysis of Tables 1-1 and 1-2 indicates that the

variability between trials was highly significant (a=0.01) and accounted

for more of the variability than the differences between replications

within trials. The overall response of waterhyacinths to GA3 and 2,4-D

Table 1-1.

Analysis of variance of the effects of GAg and 2,4-D on
waterhyacinths utilizing general linear models procedures
(percent change in weight as the dependent variable).

Sources of variation d.f s.s m.s. f

TREATMENT 19 348888.1 183625.3 22.64 **
GA3 4 85621.6 21405.3 2.64 *
Linear 1 59705.7 59705.7 7.36 **
Quadratic 1 257.5 257.5 0.03 N.S.
Other 2 25658.4 12829.3 1.58 N.S.
2,4-D 3 3319331.0 1106443.7 136.43 **
Linear 1 2513039.7 2513039.7 309.86 **
Quadratic 1 744276.0 744276.0 91.77 **
Residual 1 62015.3 62015.3 7.65 **
Interaction (GA x 2,4-D) 12 83929.7 6994.14 0.86 N.S.
Linear x Linear 1 2672.8 2672.8 0.33 N.S.

ERROR 196 515763.2 2631.4
Trial (Treatments) 52 421723.1 8110.1 12.42 **
Reps. (Trials x Treatments) 144 94040.1 653.1

TOTAL (Corrected) 215 4004644.5

N. S.

Significant at 0.01
Significant at 0.05
Not significant

Table 1-2.

Analysis of variance of the effects of GA3 and 2,4-D on
waterhyacinths utilizing general linear models procedures
(percent change in number as the dependent variable).

Sources of variation d.f s.s. m.s. F

TREATMENT 19 8125895.2 427678.7 32.59 **
GA3 4 317602.8 79400.7 6.05 **
Linear 1 222983.1 222983.1 16.99 **
Quadratic 1 17977.1 17977.1 1.37 N.S.
Other 2 76642.6 38321.3 2.92 N.S.
2,4-D 3 7680433.9 2560144.6 195.10 **
Linear 1 6210423.2 6210423.2 473.27 **
Quadratic 1 1390888.5 1390888.5 106.00 **
Residual 1 79122.2 79122.2 6.03 *
Interaction (GA x 2,4-D) 12 12785.5 10654.9 0.81 N.S.
Linear x Linear 1 13380.1 13380.1 1.02 N.S.

ERROR 196 1108430.3 5655.3
Trial (Treatments) 52 682368.0 13122.5 4.44 **
Reps. (Trials x Treatments) 144 426062.3 2958.8

TOTAL (Corrected) 215 9234325.5

N. S

Significant at 0.01
Significant at 0.05
Not significant


treatments was linear with respect to GA3 and both linear and quadratic

in response to 2,4-D. There also existed a significant undescribed

relationship above the quadratic function for the effects of 2,4-D.

There was no significant interaction between GA3 and 2,4-D (a=0.05).

Thus, 2,4-D treatment levels had a much greater influence over the

response observed for waterhyacinths in the barrel studies than did com-

binations of GA3 and 2,4-D.

Tables 1-3 and 1-4 present summaries of regression analyses and

regression coefficients for the treatment levels of GA3 and 2,4-D,

excluding the 2.24 kg/ha 2,4-D treatment, on the percent change biomass

and number of plants, respectively. These analyses demonstrated that

(1) the response to GA3 was linear, (2) the negative response to 2,4-D

was both linear and quadratic, and (3) the interaction coefficient for

GA3 and 2,4-D was insignificant, suggesting that at best the effect of

GA3 was only additive. The resulting regression models for the response

of waterhyacinths to GA3 and 2,4-D accounted for 82.9 percent of the

variability of mean change in biomass (Table 1-3) and 85.1 percent of

mean change in number of plants (Table 1-4). In order to account for an

additional portion of the variability of the response to treatments,

pretreatment weights of the waterhyacinths was included in the model but

found to be insignificant.

Tables 1-5 and 1-6 provide another representation of the results in

terms of minimum and maximum values observed, treatment means, standard

error of the mean, and Waller-Duncan groupings for percent change in

biomass and number of plants. Figures 1-1 and 1-2 provide a graphical

representation of the percent change in biomass and number of plants,

respectively. The Waller-Duncan comparisons detected significant

Results of regression analysis to determine the
of waterhyacinths grown in barrels to treatment
2,4-D. The general form of the equation is Y =
b2x2 + b3(xl)2 + b4(x2)2 + b5(xl)(x2) where Y =
change in biomass due to treatment and b = the
regression coefficients for each independent val
The coefficient of determination for the regres
(R2) is 0.829 and the regression analysis of val
F-value is 203.65 (5, 210 OF, P = 0.0001).

with GA3 and
bo + bixi +
the percent

riable (X).
sion model
ri ance

Parameter Coefficient (b) T* Prob. > 1 T

Intercept (bo) 230.91 26.16 0.0001

GA (bi) -0.39 -1.84 0.0675

2,4-D (b2) -739.96 -22.12 0.0001

(GA)2 (b3) 0.0002 0.24 0.8135

(2,4-D)2(b4) 414.54 15.13 0.0001
(GA)x(2,4-D) (b5) 0.13 0.91 0.3663

*T-statistics for the null hypothesis that the coefficient = 0.

Table 1-3.

Table 1-4.

Results of regression analysis to determine the response
of waterhyacinths grown in barrels to treatment with GA3 and
2,4-D. The general form of the equation is Y = bo + bixi +
b2x2 + b3(xl)2 + b4(x2)2 + b5(xl)(x2) where Y = the percent
change in number of plants due to treatment and b =
the regression coefficients for each independent variable (X).
The coefficient of determination for the regression model
(R2) is 0.851 and the regression analysis of variance
F-value is 239.32 (5, 210 DF, P = 0.0001).

Parameter Coefficient (b) T* Prob. > T

Intercept (bo) 440.26 35.14 0.0001

GA (bl) -1.11 -3.70 0.0003

2,4-D (b2) -1072.92 -22.60 0.0001

(GA)2 (b3) 0.0023 1.62 0.1070

(2,4-D)2 (b4) 567.72 14.61 0.0001

(GA)x(2,4-D)(b5) 0.28 1.43 0.1549

*T-statistics for the null hypothesis that the coefficient = 0.

Table 1-5.

Mean percent change from initial dry weight of water-
hyacinths grown in barrels and treated with combinations
of GA3 and 2,4-D.

Treatment Rate Waller Mean N Min. Max. Std
2,4-D GA3 Duncan Error
(kg/ha) (g/ha) Grouping1

0.00 23.5 A 287.70 6 170.74 353.41 25.52
0.00 94.0 AB 248.41 9 117.56 392.78 31.36
0.00 0.0 BC 220.80 27 88.40 374.05 16.04
0.00 47.0 C 180.82 9 107.59 241.53 18.10
0.00 188.0 C 180.00 9 95.17 274.32 17.39
0.28 0.0 D 45.13 24 -38.13 220.00 13.08
0.28 23.5 D 39.81 6 -24.37 91.60 15.46
0.28 94.0 D 17.55 6 -1.95 44.44 6.70
0.56 0.0 E -25.53 27 -85.02 85.84 9.60
0.28 47.0 EF -41.88 6 -93.63 16.49 18.51
0.56 23.5 EF -52.02 6 -86.96 3.67 16.08
0.56 47.0 EFG -65.94 9 -98.28 -10.96 9.15
0.28 188.0 FG -77.34 6 -99.29 -54.95 6.22
0.56 94.0 FG -81.76 9 -99.71 -61.84 4.08
1.12 0.0 FG -82.90 24 -100.00 -52.30 2.84
0.56 188.0 G -94.25 9 -100.00 -69.97 3.34
1.12 47.0 G -97.82 6 -100.00 -89.70 1.69
1.12 94.0 G -98.96 6 -100.00 -97.38 0.52
1.12 23.5 G -99.72 6 -100.00 -98.99 0.18
2.24 0.0 G -99.89 27 -100.00 -97.80 0.08
1.12 188.0 G -100.00 6 -100.00 -100.00 0.00

Means with the
K ratio =100).

same letter are not significantly different (a=0.05,

Table 1-6.

Mean percent change from initial number of waterhyacinths
grown in barrels and treated with combinations of GA3
and 2,4-D.

Treatment Rate Waller Mean N Min. Max. Std
2,4-D GA Duncan Error
(kg/ha) (g/ha) Grouping

0.00 0.0 A 444.78 27 240.00 750.00 23.54
0.00 94.0 A 432.59 9 340.00 583.33 24.11
0.00 23.5 A 429.44 6 360.00 516.67 25.83
0.00 188.0 B 340.00 9 220.00 460.00 27.49
0.00 47.0 B 318.52 9 240.00 400.00 19.01
0.28 0.0 C 177.64 24 40.00 433.33 20.07
0.28 23.5 CD 146.67 6 116.67 183.33 12.85
0.28 94.0 D 116.11 6 60.00 220.00 22.15
0.56 0.0 E 45.06 27 -80.00 216.67 16.12
0.28 47.0 EF 33.33 6 -60.00 160.00 35.65
0.56 23.5 EFG -6.11 6 -83.33 120.00 32.16
0.56 47.0 EFGH -13.33 9 -80.00 60.00 18.86
0.28 188.0 FGH -23.33 6 -80.00 60.00 22.16
0.56 94.0 GHI -39.47 9 -80.00 0.00 10.26
1.12 0.0 GHIJ -54.86 24 -100.00 0.00 5.41
0.56 188.0 HIJ -68.89 9 -100.00 40.00 15.32
1.12 94.0 IJ -87.22 6 -100.00 -60.00 6.58
1.12 47.0 IJ -93.33 6 -100.00 -80.00 4.22
1.12 23.5 IJ -93.33 6 -100.00 -80.00 4.22
2.24 0.0 IJ -97.78 27 -100.00 -60.00 1.63
1.12 188.0 J -100.00 6 -100.00 -100.00 0.00

1 Means with the
K ratio =100).

same letter are not significantly different (a=0.05,


0- -0

0 f0
U 0

m a N

LD z 0 0

O- m o
iT Z

I 04


o 0 r-
in In u I

In I-

" 00000
z o'- o. Ow c

0 0000 4-J 0-
Z4r- 2 .-

TT o-0

ST 4-

0- r-0

oo In < aCo-a -
4 -o
E N)

I In -r c

Icu a

^^ d dddd dddd dd'- '-'--cWi- u

U ')
o 0I
o 0

o 0
u 0)

U) (..-

S-0 -

Tn 4- C-

4- .

0 O,
ST- (-,1
I I o
S^T '-4-C'

4-)) 0 .

i I I
- U 3-c
B 0 ce c cS

Of m ,- -

T4 -

0 0

o- o y -COoIt-C- ) It Nvoc 0- --o----v--c- (-
oo- -

S-00000000000000-0 i-
I 'T'c


differences (a=0.05) between the GA3 treatment levels at 2,4-D rate of

0.00 kg/ha; however, the differences were not in proportion to the rate

of GA3. It was also observed that when 2,4-D rates were held constant,

the addition of GA3 increased the mean response of waterhyacinths and

this response was frequently significant particularly at 2,4-D rates of

0.28 and 0.56 kg/ha. This observation supports the observation that

the effect of GA3 on waterhyacinth sensitivity to 2,4-D recorded for

this series of observations was additive rather than synergistic.


Based on regression analysis, the response of waterhyacinths grown

outdoors in 70-liter containers to treatments with combinations of 2,4-D

and GA3 does not indicate a synergistic effect or interaction between

GA3 and 2,4-D either in terms of the percent change from initial biomass

or a percent change from the initial number of plants. However, at lower

rates of 2,4-D, the response does appear to be additive. These findings

are not in agreement with results reported for the effects of GA3 and

2,4-D on waterhyacinths by Pieterse et al. (1980) and Pieterse and

Roorda (1982). However, differences in experimental design and environ-

mental controls (greenhouse versus non-greenhouse, open air situations)

may account for the lack of agreement.



Various studies have shown increases in rates and amounts of

translocation of radioactive labeled auxins and 2,4-D when plants

were pretreated with GA (Ashton, 1959; Basler, 1959; Pilet, 1965;

Basler, 1974). The studies utilized either whole, immature plants or

stem segments of bean plants. Singh and Muller (1979b) reported that

asulam and amitrole have higher rates of translocation in waterhyacinths

than 2,4-D, and this may be the reason they are more efficacious against

waterhyacinths. Pieterse et al. (1980) and Pieterse and Roorda (1982)

suggested that increased translocation was the mechanism by which

waterhyacinths exhibited a ten-fold increase in sensitivity to 2,4-D

when also treated with GA. The objective of this portion of the

overall investigation was to determine if increased translocation of

2,4-D in waterhyacinths was the mode of action of increased sensitivity

of waterhyacinths to 2,4-D when also treated with gibberellic acid.

Materials and Methods

Waterhyacinths for this study were collected from a small tributary

stream of the St. Johns River in Jacksonville, Florida. Plants were

selected for uniformity of size (approximately seven leaf stage)

and apparent freedom from disease and insect damage. All plants were

allowed to remain in the growth chamber for 3 days prior to treatment.

A total of 32 waterhyacinths were placed in individual one-liter

beakers which were filled with 5 percent Hoagland's solution to

within 2 cm of the top. Nutrient solution volumes were replenished

approximately every 48 hours. Temperature in the growth chamber was

maintained between 16 (night) and 32 (day) C. Light was supplied by

eight flourescent and eight incandescent bulbs which provided 330

micro-Eiensteins/M 2/sec at plant height. Based on the results of

Patterson and Duke (1979), the plants could be expected to photosynthe-

size at approximately 78 percent of the rate expected at full sunlight.

The photoperiod simulated a 14-hour day and a 10-hour night. Prior to

treatment with the radioactive labeled 2,4-D, one-half or 16 of the

plants were individually removed from their beakers, the roots were

shielded with plastic, and the foliage was sprayed to the point of

"runoff" with 100 mg/1 aqueous solution of the potassium salt of gib-

berellic acid (Eastman Company). The plants were replaced in the

growth chamber, and the GA was allowed to dry on the plants prior to

the application of 2,4-D.

Fifty milligrams of 14C ring-labeled 2,4-D, specific activity

940 uCi/mM, was dissolved in 5 ml of ethanol. One ml (10 mg of 2,4-D)

of this solution was then added to 9 ml deionized water to yield a

1,000 mg/l solution with an activity of 4.25 uCi/ml. A total of 0.216

uCi was applied by placing 51 ul of the 1000 mg/l labeled 2,4-D solu-

tion in three separate 17 Ml droplets on the lamina of a single leaf of

each of the plants, according to methods described by Singh and Muller

(1979a and 1979b). Treated leaves were either the fourth or fifth leaf

from the outside of the rosette of leaves, depending upon which was

most horizontal in orientation to minimize droplet movement. Treated

leaves were marked with small gummed labels for ease of identification.

Plants were placed at random in the growth chamber.

Four plants treated with GA3 and 2,4-D and four treated with 2,4-D

only were harvested at 1, 3, and 6-day intervals. The above procedure

was replicated for the six-day treatment series only. The presence of

necrotic tissue, daughter plants, and flower spikes was noted. Plants

were then separated into the following fractions: treated lamina,

petiole of treated leaf, individual leaves, daughter plants, meristem

and youngest leaf, and roots. Plant parts were placed in paper bags

and placed in a forced-air drying oven for 72 h at 65 C. Dried plant

parts were weighed separately to the nearest 0.01 mg and wrapped in a

low-ash, unscented, ungummed cigarette papers. Weighed samples were

combusted in a Packard TRICARB sample oxidizer to collect 14C02 for

assaying radioactivity. Oxidizer sample burn time was set at 1.5

minutes; however, all samples were completely combusted within 0.5

minute. Six milliliters of OXISORB 2 and 10 ml of OXIPREP (New England

Nuclear Ltd., Boston (USA)) were used to absorb the 14C02 and serve as

a scintillation fluor, respectively. Radioactivity was assayed by a

PACKARD TRICARB scintillation spectrometer. Samples were counted for

10 min and counts per minute (cpm) were converted to disintegrations

per minute (dpm) based on a linear regression obtained from a quench

curve. The quench curve was obtained by oxidizing blank samples of

varying weights. After combustion, known levels of radioactivity were

added to the scintillation fluor and the counting efficiency was

obtained. All results were reported as dpm/mg of plant material and


as percent of total translocated material present in a given plant


Mean dpm/mg of plant material and mean percent of total

translocated 14C-labeled material per plant part per harvest day were

calculated. Mean dpm/mg and mean percent of translocated 2,4-D of

plants receiving pretreatment with gibberellic acid were compared

with means from untreated plants. A simple t-test for determining the

presence of significant differences between two means was used (Walpole

and Myers, 1918). Prior to comparison of the mean percentages of

translocated radioactivity, an arcsine transformation was performed in

order to normalize the distribution (Sokal and Rohlf, 1969).

Results and Discussion

In general, results presented below indicate that, without regard

to gibberellic acid treatment, movement of 14C labeled 2,4-D follows

a pattern of polar movement towards rapidly growing tissues such as

meristematic shoots and newly emerging daughter plants as previously

documented for waterhyacinths and other species (Audus, 1972; Ashton and

Crafts, 1973; Singh and Muller, 1979b).

Tables 2-1 and 2-2 summarize the results of the movement through

time of the 14C-labeled 2,4-D within treated waterhyacinths expressed

as dpm/mg dry plant material and percent of total amount of translo-

cated radioactivity, respectively. Tables 2-3, 2-4, 2-5, and 2-6 pre-

sent a detail analysis of the results on a dpm/mg basis for the plants

harvested on days 1, 3, 6 (trial 1), and day 6 (trial 2), respectively.

Tables 2-7, 2-8, 2-9, and 2-10 present a similar analysis of the

results on a percent of total amount of translocated 14C-labeled 2,4-D.

Table 2-1. Effects of gibberellic acid (100 mg/1) on the translocation
of 14C-labeled 2,4-D in waterhyacinths; (data expressed as
mean dpm/mg dry weight).

Day 1 Day 3 Day 6 Day 6
(Trial 1) (Trial 1) (Trial 1) (Trial 2)

Treated Petiole
(minus lamina)
With GA 94.58 131.41 71.19 94.92
Without GA 88.50 121.45 146.80 100.16

Other Leaves
With GA 2.51 5.89 6.00 5.08
Without GA 1.86 3.86 9.93 7.42

Meristem and Youngest Leaf
With GA 15.05 19.34 14.39 14.96
Without GA 12.50 27.68 35.87 15.35

Daughter Plants
With GA 15.50 17.29 9.74 15.50
Without GA 17.74 21.66 32.71 55.51

With GA 1.85 4.24 5.06 3.94
Without GA 2.03 3.25 10.44* 3.71

Total Translocated
With GA 13.59 18.23 14.71 14.45
Without GA 12.60 15.08 20.77 18.52

** Significant at a=0.01
* Significant at a=0.05

Table 2-2.

Effects of gibberellic acid (100 mg/1) on the translocation
of 14C-labeled 2,4-D in waterhyacinths; (data expressed as
mean percent of total 14C translocated).

Day 1 Day 3 Day 6 Day 6
(Trial 1) (Trial 1) (Trial 1) (Trial 2)

Treated Petiole
(minus lamina)
With GA 60.74 51.21 43.11 44.06
Without GA 60.64 52.64 33.50 36.83

Other Leaves
With GA 11.27 15.39 24.29 17.25
Without GA 8.05** 13.16 17.42* 21.49

Meristem and Youngest Leaf
With GA 19.47 15.37 10.59 6.57
Without GA 19.29 18.00 18.15* 6.08

Daughter Plants
With GA 3.73 11.91 12.29 19.70
Without GA 5.70 11.13 19.07 29.43

With GA 4.77 6.16 9.71 12.42
Without GA 6.33 5.06 11.85 8.16

** Significant at a=0.01
* Significant at a=0.05

Table 2-3.

Effects of gibberellic acid (100 mg/1) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment; (means expressed as dpm/mg dry

the translocation
the first day post

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

Total Transl ocated
With GA
Without GA

94.58 78.69
88.50 81.63









13.59 10.13
12.60 10.01

116.68 11.40
93.10 3.50

















Table 2-4. Effects of gibberellic acid (100 mg/1) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment; (means expressed as dpm/mg dry

the translocation
the third day post

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

Total Transl ocated
With GA
Without GA

131.41 59.26 167.99 36.08
121.45 81.24 175.91 19.82



19.34 10.17
27.68 21.77

17.29 11.18
21.66 14.85



18.24 6.94
15.08 12.08

















Table 2-5. Effects of gibberellic acid (100 mg/1)
of 14C-labeled 2,4-D in waterhyacinths
treatment in trial 1; (means expressed

the translocation
the sixth day post
dpm/mg dry weight).

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

Total Translocated
With GA
Without GA

71.19 54.85
146.82 92.26





9.74 5.25
32.71 16.69



13.59 10.13
12.60 10.01

102.89 15.85
219.54 37.85

















Table 2-6. Effects of gibberellic acid (100 mg/1) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment in trial 2; (means expressed as

the translocation
the sixth day post
dpm/mg dry weight).

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

Total Translocated
With GA
Without GA

94.92 66.22 149.93 18.78
100.16 67.56 136.29 17.73





15.50 10.13
55.51 10.88



14.45 11.10
18.52 9.79

















Table 2-7.

Effects of gibberellic acid (100 mg/i) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment in trial 1; (means expressed as
amount of translocated 14C).

the translocation
the first day post
percent of total

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

60.74 58.78
60.64 58.89

11.28 10.95
8.05 7.10

19.47 15.53
19.29 17.76




















Table 2-8.

Effects of gibberellic acid (100 mg/i) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment in trial 1; (means expressed as
amount of translocated 14C.

the translocation
the third day post
percent of total

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

51.21 42.59
52.64 44.88



15.37 14.37
18.00 14.70




















Table 2-9.

Effects of gibberellic acid (100 mg/1) on
of 14C-labeled 2,4-D in waterhyacinths on
treatment in trial 1; (means expressed as
amount of translocated 14C).

the translocation
the sixth day post
percent of total

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

43.11 35.94
33.50 30.57

24.28 23.59
17.42 14.53






















Effects of gibberellic acid (100 mg/1)
of 14C-labeled 2,4-D in waterhyacinths

treatment in trial 2; (means
amount of translocated 14C).


the translocation
the sixth day post
percent of total

Std. Prob.
N mean min. max. Error > T

Treated Petiole
(minus lamina)
With GA
Without GA

Other Leaves
With GA
Without GA

Meristem and
Youngest Leaf
With GA
Without GA

Daughter Plants
With GA
Without GA

With GA
Without GA

44.06 35.31
36.83 27.26

17.25 11.24
21.49 15.95



19.70 10.36
27.43 13.70



Table 2-10.

















Tables 2-3 to 2-10 contain the actual alpha levels at which the dif-

ference in the treatment means of plant parts harvested on the same day

may be considered significant; however, in the discussion which follows,

alpha levels greater than 0.05 are not considered to be "statistically

significant" (Sokal and Rohlf, 1969).

For ease of discussion, results are discussed in terms of treat-

ment means of individual plant parts through the three harvest


The petioles of treated leaves consistently contained the highest

14C-labeled 2,4-D activity on both a dpm/mg and percentage of

total translocated radioactivity, as would be expected due to the close

proximity and connection to treated lamina. No clear pattern of

distribution due to pretreatment with GA3 was evident on either

dpm/mg or percentage basis. Results obtained for day 6 (trial 1)

highlight the lack of correlation of GA3 treatment with levels of

14C-labeled 2,4-D translocation. On day 6, trial 1, plants which did

not receive GA3 pretreatment contained over twice as much radioactivity

as plants receiving GA3; however, on a percentage basis GA3 pretreatment

plants contained more 14C activity. Due to the variability of the data

these differences are not considered significant but are noted merely to

demonstrate the lack of consistency in the results.

Mean dpm/mg and mean percentage of translocated 14C in the

meristems and youngest leaves did not reflect any significant

differences due to GA3 treatment in plants harvested on days 1 and 3.

However, on day 6, trial 1, meristems of plants which did not receive

GA3 pretreatment had a significantly higher percentage (18.15)

of the total amount of translocated material at the a=0.014 level than

meristems of plants which did receive GA3 pretreatment (10.59 percent).

On a dpm/mg basis, the same pattern of distribution occurred; however,

it was not significant at the a=0.05 level.

Daughter plants which were produced after initiation of the treat-

ment period consistently contained higher amounts of 14C activity on

both a dpm/mg and percentage basis when the parent plant did not receive

GA3 pretreatment. However, due to the variability in the data as

reflected by the standard error term, none of the differences in the

means were significant at the a=0.05 level. Six-day, trial 2, plants

yielded similar results but were also non-significant at the a=0.05


A significant difference in mean dpm/mg was noted for roots of six-

day, trial 1, plants, i.e., 5.06 dpm/mg for the plants with GA3 and

10.44 dpm/mg for the plants without GA3 (a=0.04). A similar trend was

also observed on a percentage of total translocated radioactivity for

roots of six-day, trial 1, plants, however, the difference between

treatment means was not significant at the a=0.05 level. Significant

differences were not observed for roots of six-day, trial 2, plants on

either a dpm/mg or percentage basis.

Results obtained for leaves other than the leaves which received the

direct application of 14C-labeled 2,4-D provided the only indication of

increased translocation of 2,4-D due to GA3 pretreatment. Leaves of

both the one-day and six-day (trial 1) plants which were treated with

GA3 contained significantly higher (a=0.004 and a=0.021, respectively)


mean percentages of the total translocated 14C-labeled 2,4-D than plants

which did not receive GA3 pretreatment. No significant differences were

noted on a dpm/mg basis.

An analysis of the cumulative amount of 14C-labeled 2,4-D translo-

cated from treated lamina to plants taken as a whole revealed no

apparent significant differences (a=0.05) due to the treatment of

waterhyacinths pretreated with gibberellic acid.


Results of the 14C-labeled 2,4-0 translocation experiments,

while highly variable, suggest that under growth chamber conditions,

pretreatment of waterhyacinths with 100 mg/1 gibberellic acid (1) does

not increase the translocation of 2,4-D to meristematic tissues, and

(2) may increase the movement of 2,4-D to previously emerged leaves on a

percentage of total radioactivity basis. This suggestion may appear to

be somewhat in conflict with other studies which have reported signifi-

cant increases in translocation of auxins and/or 2,4-D due to treatment

with GA3 (Ashton, 1959; Basler, 1959; Pilet, 1965; Basler, 1974);

however, none of these studies were conducted with rosette species such

as waterhyacinths. It is possible, as suggested by Audus (1972), that

there may be no interdependence of gibberellins and auxin-like compounds

because gibberellins have been shown to produce different responses in

different plant species and tissues within the same species.



Various public agencies routinely utilize 2,4-D for the control of

waterhyacinth in the United States. One such agency, the U.S. Army

Corps of Engineers, Jacksonville District, annually utilizes approxi-

mately 8,200 kg of 2,4-D at rate of 2.24 kg/ha in a 934 1/ha aqueous

solution to control waterhyacinths on the St. Johns River, Florida

(McGehee, 1982). Pieterse and Roorda (1982) reported a ten-fold

enhancement of 2,4-D sensitivity of waterhyacinths when the plants were

simultaneously treated with GA3 at 6 to 8 g/ha under greenhouse

conditions. It was also suggested by Pieterse and Roorda (1982) that

such a large reduction in the amount of 2,4-D would lower the risk of

damage to nearby crops or vegetation and may be attractive from an eco-

nomical point of view due to the large reduction in the amount of 2,4-D

required to control waterhyacinths. In order to test this possibility,

the following investigations were conducted (1) the treatment rates

tested in the small plot evaluations (Part 1) were repeated in a natural

stand of dense waterhyacinths; (2) a selected rate of 2,4-D and GA3 was

evaluated under actual waterhyacinth control operational conditions; and

(3) an economic analysis of the use of GA3 in conjunction with 2,4-D was

performed utilizing the assumption that GA3 would reduce the amount of

2,4-D required for waterhyacinth control by a factor of 10.


Materials and Methods

Field applications were conducted on Lake Dexter, one of a chain of

lakes located on the St. Johns River, 6.7 km southeast of Astor, Florida.

The waterhyacinths appeared free of any disease, but did exhibit evi-

dence of moderate feeding by waterhyacinth weevils, Neochetina spp. The

plants were growing in a large stand of Nuphar luteum which prevented

movement of the waterhyacinth mat during the treatment period.

Field evaluations were conducted in two phases. In phase one, the

experimental plots were 9.2 m x 3.0 m. Three experimental plots were

established in each of 17 separate 46.0 m x 3.0 m transects such that

each of the 51 9.2 m x 3.0 m plots were separated by an untreated plot.

The untreated plots served as buffer areas between treated plots.

Prior to treatment, three random 0.33 sq m samples were taken from the

plots to be treated. Plants within the 0.33 sq m samples were counted,

allowed to drain of excess water, and weighed to the nearest 0.05 kg in

order to determine pretreatment biomass and number per sq m. In phase

two, experimental plots were laid out as three separate 0.40 ha-plots

each separated by an untreated strip of waterhyacinths. Pretreatment

biomass determinations were not made because the efficacy of treatments

was based on visual evaluations of individual plants to obtain a propor-

tion of dead plants per plot.

In phase one, the 51 individual plots were treated by use of an

airboat equipped with a tank-mix spray system calibrated to deliver a

spray volume of 467 1/ha. Each plot was treated twice in order to

obtain a spray volume of 934 1/ha. Applications were made with a fixed

boom equipped with a single Delavan Type-D20 flooding nozzle. The boom


was adjusted to approximately 45 cm above the plant canopy such that the

swath width was equal to 3.0 m. Treatments consisted of simultaneous

applications of combinations of gibberellic acid (Asgrow Florida

Company, EPA Accession No.08728) and 2,4-D (Union Carbide Corporation

EPA Registration No. 264-2AA) at the following rates: 0.0, 23.5, 47.0,

and 94.0 g/ha gibberellic acid and 0.56, 1.12, and 2.24 kg/ha 2,4-D. An

additional 2,4-D treatment at a rate 4.48 kg/ha was made due to the

higher biomass present in the field plots compared to the small plot

evaluations discussed in Part 1 of this study.

In phase two, three 0.40 ha-plots were treated by an airboat which

contained a tank mix spray system calibrated to deliver 934 1/ha through

a hand-held spray gun equipped with a Delavan Type DFA Dela-foam nozzle.

The application was made by a spray crew employed by the U.S. Army Corps

of Engineers, Jacksonville District. The crew was instructed to treat

each plot in the same manner in which they conduct routine control

operations. Treatments consisted of 0.84 kg/ha 2,4-D; 0.84 kg/ha 2,4-D

plus 94.0 g/ha gibberellic acid; and 2.24 kg/ha 2,4-D. These rates of

2,4-D and gibberellic acid were chosen based on results of the small

plot evaluations (Part 1) and results of phase one, above.

All treatments were examined weekly for evidence of treatment

effects. At the conclusion of 24 days of phase one, the plants in the

plots treated with 2.24 kg/ha and 4.48 kg/ha 2,4-D appeared to exhibit

100 percent control, and the decision was made to complete the efficacy

evaluation on day 25. Phase one efficacy evaluations consisted of har-

vesting the plants in three random 0.33 sq m samples from each treatment

plot, counting the number of viable plants, removing obvious necrotic


tissue, and weighing the remaining plant material to the nearest 0.05 kg

in order to determine post treatment biomass. Results of phase one were

analyzed for the mean percent change in fresh weight and mean percent

change in number of plants from initial pretreatment levels. The means

of the percent change by treatment were analyzed for the presence of

significant differences utilizing Waller-Duncan procedure for the com-

parison of multiple means. The Waller-Duncan procedure was employed

because it has the advantage over other such methods in that the

observed F value is used in the calculation of the LSD (least signifi-

cant difference). The use of the observed F value provides a method of

accounting for both the comparisonwise and experimentwise Type I error

rates (Chew, 1977).

Phase two efficacy evaluations consisted of visual assessment of the

presence of viable meristematic tissue in the treated plots (Seale and

Allison, 1946). At the conclusion of 22 days post-treatment, plants

treated with 2.24 kg/ha 2,4-D appeared to exhibit near 100 percent

control and the decision was made to complete the efficacy evaluation on

day 24 post-treatment. Thirty random plants on three transects through

the treated plots were examined and the proportion of dead plants per

plot calculated. Proportions of dead plants per treatment plot were

analyzed for significant differences utilizing a method described by

Walpole and Myers (1978).

An economic evaluation of the use of 2,4-D was performed by calcu-

lating the costs of converting the waterhyacinth control operation con-

ducted by the U.S. Army Corps of Engineers on the St. Johns River,

Florida, to a control program utilizing various combination of GA3 and


lower than normal rates of 2,4-D. No costs were included for labor,

conversion of the spray equipment to allow the use of GA3, or for

increased storage and transportation of GA3.

Results and Discussion

The mean pretreatment biomass (fresh weight) of the Lake Dexter

waterhyacinth population was 21.98 kg/sq m (standard error 0.67) and the

mean pretreatment number of plants per sq m was 70.76 (standard error

2.92). These means are within the ranges reported by Center and Spencer

(1981) for mature stands of waterhyacinths in a North-Central Florida

lake during August. At the conclusion of phase one, mean fresh weight

and mean number of plants -per sq m of the control plots were not signi-

ficantly different from the pretreatment levels (a=0.05) which indicates

that the plants were physiologically mature and had become space-limited

as reported by Richards (1980) and Center and Spencer (1981).

Tables 3-1 and 3-2 summarize the results of the phase one eval-

uations of the 17 treatment combinations of GA3 and 2,4-D in terms of

mean percent change in fresh weight and mean percent change in number

due to treatment, respectively.

Comparisons of Table 3-1 and 3-2 reveals that within 2,4-D rates of

0.56 and 1.12 kg/ha and any rate of GA3, the mean percent change

in fresh weight per sq m was always greater than the mean percent change

in number of plants per sq m. This was a reflection of the morphologi-

cal response of the waterhyacinths to the treatments. At the lower rate

of 2,4-D, the older leaves became necrotic at the base and readily

separated from the plant, such that the only remaining viable tissue was

the meristem, youngest leaves, and roots. However, the plants remained

Table 3-1.

Mean percent change in fresh weight of waterhyacinths/sq m
in Lake Dexter, Florida, treated with combinations of
gibberellic acid and 2,4-dichlorophenoxyacetic acid.

Treatment Rate Waller- Mean n Min. Max. Std
2,4-D GA Duncan Error
(kg/ha) (g/ha) Grouping1

0.00 23.5 A 32.11 3 15.87 44.45 8.48
0.00 94.0 A 25.86 3 9.59 51.68 13.05
0.00 47.0 A 21.76 3 10.75 37.01 7.87
0.00 0.0 B 1.15 3 -7.12 7.43 4.32
0.56 94.0 C -81.33 3 -83.98 -78.99 1.45
0.56 47.0 C D -82.33 3 -88.20 -74.80 3.96
0.56 0.0 C D E -86.44 3 -89.13 -82.74 1.91
0.56 23.5 C D E -88.15 3 -88.85 -87.77 0.35
1.12 0.0 C D E -94.25 3 -95.48 -93.16 0.67
1.12 47.0 D E -96.61 3 -97.43 -95.49 1.60
1.12 94.0 E -97.44 3 -97.99 -96.93 0.31
1.12 23.5 E -99.05 3 -99.77 -98.60 0.36
2.24 23.5 E -99.23 3 -99.79 -98.46 0.40
2.24 0.0 E -99.37 3 -99.66 -98.91 0.23
2.24 94.0 E -99.50 3 -100.00 -98.59 0.45
2.24 47.0 E -99.68 3 -100.00 -99.47 0.16
4.48 0.0 E -100.00 3 -100.00 -100.00 0.00

1Means with the same letter are not significantly different (a=0.05).

Table 3-2.

Mean percent change in the number of waterhyacinths/sq m
in Lake Dexter, Florida, treated with combinations of
gibberellic acid and 2,4-dichlorophenoxyacetic acid.

Treatment Rate Waller- Mean n Min. Max. Std
2,4-D GA Duncan Error
(kg/ha) (g/ha) Grouping1

0.00 0.0 A 3.47 3 -15.66 15.48 9.67
0.00 94.0 A 1.63 3 -16.13 12.50 8.95
0.00 47.0 A 1.28 3 -9.26 20.00 9.41
0.00 23.5 A -12.23 3 -39.60 1.59 13.69
0.56 47.0 B -51.53 3 -67.69 -36.00 9.15
0.56 0.0 B -54.43 3 -57.89 -51.11 1.96
0.56 23.5 B -55.66 3 -68.37 -45.28 6.77
0.56 94.0 B D -64.78 3 -80.00 -51.85 8.21
1.12 0.0 B D -67.00 3 -76.56 -59.62 5.01
1.12 47.0 B D -70.58 3 -73.77 -68.75 1.60
1.12 94.0 C D -82.65 3 -82.82 -82.35 0.14
1.12 23.5 C D -85.12 3 -93.51 -72.97 6.22
2.24 0.0 C -94.90 3 -95.45 -94.19 0.37
2.24 23.5 C -96.12 3 -98.86 -94.12 1.42
2.24 94.0 C -96.57 3 -100.00 -90.91 2.85
2.24 47.0 C -99.16 3 -100.00 -98.59 0.43
4.48 0.0 C -100.00 3 -100.00 -100.00 0.00

1Means with the same letter are not significantly different (a=0.05).


viable and retained the ability to reinfest the plot. For this reason,

the percent change in number would appear to be a more realistic

estimator of efficacy.

According to the Waller-Duncan multiple range test, there was no

significant difference in the mean change in weight or number between

the three treatment rates of GA3 (2,4-D rate = 0.00 kg/ha). There was a

significant increase in the mean weight change of the three levels of

GA3 (2,4-D rate = 0.00 kg/ha) when compared to the control plot; how-

ever, the GA3 treatments did not affect the mean number of plants per

sq m when compared to the control plot. Tables 3-1 and 3-2 indicate that

there were no significant (a=0.05) increases in the efficacy of any

fixed rate of 2,4-D when combined with any of the four levels of GA3.

Table 3-3 presents the results of the large scale application of 2,4-D

and gibberellic acid to a dense stand of waterhyacinths. Twenty-four days

post-treatment, plot 3, which received 2.24 kg/ha 2-4-D, and 0.0 g/ha GA,

had the highest percentage (80.40) of dead waterhyacinth plants per

transect. The next highest percent dead plants was plot 2 which received

0.84 kg/ha 2-4-D, and 94.1 g/ha GA, with 34.70 percent, followed by plot

1 which received 0.84 kg/ha 2-4-D, and 0.0 g/ha GA, with 24.20 percent

dead plants per transect. The proportions of dead plants in plots 1 and

2 were not significantly different from each other at a=0.05; however,

the proportion of dead plants in plot 3 was significantly different from

plots 1 and 2 at a=0.05. A visual inspection of the phase two plots was

conducted 63 days post-treatment. Quantitative evaluations were not

possible due to disturbance of the plots by wind, currents, and the

24-day post-treatment sampling procedure. However, the visual evalua-

tion did indicate that the waterhyacinth populations in plot 3 had been

Table 3-3.

Effects of large-scale operational application of 2,4-D
and gibberellic acid (GA3) to waterhyacinths; 24 days

Treatment Rate Proportion of Dead
Plants per Plot
Plot No. 2,4-D GA
(kg/ha) (g/1a)

1 0.84 0.0 23/95 = 0.242 b

2 0.84 94.1 33/95 = 0.347 b

3 2.24 0.0 74/92 = 0.804 a

Proportions followed by the same letter are not
at a=0.05.

significantly different


(1) reduced to a non-problematic level and replaced by a monoculture of

water lettuce, (Pistia stratiotes L.) and (2) plots 2 and 3 still con-

tained problematic levels of waterhyacinth and were in need of


Table 3-4 presents a cost comparison for use of a combination of

GA3 and 2,4-D for control of waterhyacinths on the St. Johns River,

Florida, as conducted by the U.S. Army Corps of Engineer's Jacksonville

District. Based on an average of 3642 ha of waterhyacinths controlled

per year, the normal application rate of 2,4-D, 2.24 kg/ha, results in

an annual herbicide cost of $27,096. With the use of GA3 at 8.0 g/ha

and a ten-fold reduction in the amount of 2,4-D required for an equiva-

lent level of control (0.224 kg/ha), as suggested by Pieterse and Roorda

(1982), the annual herbicide costs would be $35,947 or 32.7 percent

higher than the current rate. Based on the results of the field appli-

cation at Lake Dexter, Florida (Table 3-2), the lowest combination of

rates of GA3 and 2,4-D which would provide a level of control (number

per sq m) not significantly different from 2.24 kg/ha, 2-4-D was 23.5

g/ha and 1.12 kg/ha, respectively. This option would result in an

annual herbicide cost of $111,117 or 310.1 percent higher than the

current rate.


The results of the field test of various treatment rates of 2,4-D

and GA3 indicated that on an operational basis, GA3 does not enhance the

effect of 2,4-D on water hyacinths to a significant degree. The

additional costs of using GA3 and 2,4-D at the relative levels suggested

by Pieterse and Roorda (1982) are not justified from an economic stand-

point at the current market prices of GA3 and 2,4-D.

Table 3-4.


Costs comparison for the use of a combination of GA3 and
2,4-D compared to normal application rates of 2,4-D for the
control of waterhyacinths on the St. Johns River, Florida.

Option A: Normal rate of 2,4-D (2.24 kg/ha)

1982 2,4-D costs= $3.32/kg x 2.24 kg/ha = $7.44/ha

Annual costs = $7.44/ha x 3642 h/yrl= $27,096/yr

Option B: 8.0 g/ha GA3 and 0.10 of normal application rate of
2,4-D (0.224 kg/ha)2

1982 2,4-D costs = $3.32/kg x 0.224 kg/ha = $0.75

1982 GA3 costs3= $1.14/g x 8.0 g/ha = $9.12

Annual costs = $9.87/ha x 3642 ha/yr = $35,947/yr

Option C: 23.5 g/ha GA3 and 1.12 kg (a.e.)/ha4

1982 2,4-D costs = $3.32/kg x 1.12 kg/ha = $ 3.72/ha

1982 GA3 costs = $1.14/g x 23.5 g/ha = $26.79/ha
Total $3U.blha

Annual costs = $30.51/ha x 3642 ha/yr = $111,117/yr

1. Current bid price (McGehee, 1982).
2. Pieterse and Roorda (1982).
3. Current market price (Asgrow Florida, Inc.).
4. Lowest rates which produced a level of control which was not
significantly different from 2,4-D at 2.24 kg/ha (see Table 3-2).


Results of this investigation are in conflict with findings of

Pieterse (1979) and Pieterse and Roorda (1982) who reported a ten-fold

enhancement of the effects of 2,4-D when simultaneously treated with

gibberellic acid. However, differences in experimental design and

environmental controls probably account for the lack of agreement.

Pieterse and Roorda (1982) conducted their studies in concrete reser-

voirs in a heated greenhouse, utilized an atomizing-type spray apparatus,

and applied spray volumes of 40 1/ha or 4 ml/sq m. This study was con-

ducted outdoors in plastic-lined metal containers using a hand-held non-

atomizing sprayer and in an undisturbed infestation of waterhyacinths in

the St. Johns River using an airboat spray system. Spray volumes were

934 1/ha or 93.4 ml/sq m.

Hitchcock et al. (1949) reported quicker killing of individual

waterhyacinths when sprayed with an atomizer as compared to similar

rates of 2,4-D sprayed on undisturbed plants in the field. Hitchcock

et al. (1949) and Koch et al. (1978) reported increased 2,4-D efficacy

under greenhouse conditions due to more effective wetting and the fact

that spray solution which does not fall directly on the target plants

may be trapped on the water surface of the experimental containers,

allowing greater absorption by plant roots than would be the case under

natural conditions. These observations also provide a partial explana-

tion for the field trials in Lake Dexter (Part 3) not producing the same

level of efficacy observed in the small plot evaluations (Part 1).


The physiological state of plants in each separate evaluation also

contributed to the lower level of activity in the field plots. The

plants grown in plastic-lined metal containers (Part 1) were young,

rapidly growing specimens, whereas, the plants treated in Lake Dexter,

Florida (Part 3) were mature, spaced-limited specimens. This condition

was evident since the controls in Part 3 exhibited insignificant growth

in terms of fresh weight and number of plants produced per sq m during

the study. Addicott (1970) and Low (1974) supported these observations

by explaining the differing results from seemingly similar experiments

with gibberellins in terms of physiological conditions of the experi-

mental material, i.e., age, size, nutrient and light availability,

temperature, species, and type of tissue examined.

The apparent lack of interaction or synergism observed in this study

is also partially explained by the above referenced comments of Addicott

(1970) and Low (1974). However, numerous other researchers reported

that synergistic interactions between gibberellins and auxin-like com-

pounds were due to an auxin-sparing reactions wherein the applied gib-

berellins inhibited or reduced endogenous concentrations of auxin

degrading enzymes (Brian and Hemming, 1958; Kogl and Elema 1960, cited

in Weaver, 1972; Galston and Purves, 1960; Sarma 1978). Such mechanisms

in which levels of endogenous auxins are increased by GA3 could possibly

account for supposedly increased sensitivity of plant tissues to reduced

concentrations of auxins or 2,4-D applied with gibberellins. The phy-

siological state of the plant and the environmental conditions of the

experiment could then determine the level of production of other endoge-

nous plant growth substances, as suggested by Low (1974). The magnitude

of this production could result in either additive or synergistic effects.


The nature of response observed in the 14C-labeled 2,4-D transloca-

tion studies can be partially explained by a recent study by Mulligan

and Patrick (1979). GA3 was shown to promote the transfer of 14C and

32P-labeled photosynthates to the site of GA3 application rather than to

competing "sinks" such as roots and meristematic tissues. This increase

in transfer away from normal "sinks" to the site of GA3 application was

shown not to be caused by increased photosynthesis rates, increased

assimilate export rate from "sources", nor by altering the mobilizing

ability of other competing sinks. Mulligan and Patrick (1979) provided

evidence that GA3 was not acting on any transfer process remote from its

point of application but was acting locally. A similar relationship in

waterhyacinths could account for the suggestion of increased transloca-

tion to GA3 treated petioles in this study. The data also suggested

reduced translocation to meristematic sinks in GA3 treated plants;

however, due to variability of the data, significance was not

demonstrated consistently.

The costs analysis of the use of GA3 in conjunction with lower than

normal rates of 2,4-D indicates that its use is not justified economi-

cally. However, a substantial increase in 2,4-D prices, a reduction in

GA3 costs, or a documented environmental concern over the quantity of

2,4-D applied to public waters could alter this analysis. Pending the

development of more environmentally compatible and efficacous

herbicides, the most prudent way to reduce the quantity of 2,4-D used in

waterhyacinth control and thus herbicide expenditures is to begin

control operations before the plants reach problematic levels (Hitchcock

et al., 1949) and maintain a low level of plants through a regular

patrol system to prevent reestablishment (Seale and Allison, 1946).


This concept is the essence of current maintenance control concepts for

waterhyacinth control on large riverine systems (Joyce, 1977).

As a final summary to this discussion, the following comments are

offered. Despite years of intensive research, the role of growth

substances in the life of the intact growing plant appears far from

clear. In fact, a new concept appears to be emerging in plant growth

substance theory. Trewavas (1981) stated, "Those who work in the area

will be only too familiar with the often-confusing contradictions, the

apparently endless and puzzling interactions and the plain uncertainties

of supposedly established facts. Even the outline of a simple physiolo-

gical mechanism of control for any growth substance in the intact plant

cannot be deduced with any certainty" (Trewavas, 1981, page 203). The

entire concept of plant "hormones" as substances which have localized

biosynthesis, control physiological and biochemical events by changing

their concentration, and cause actions at a distance from their synthe-

sis stems from mammalian hormore theory and was challenged by Trewavas

(1981). Much of the confusion surrounding plant growth substances was

blamed on the prevailing trend of explaining the actions of these

substances in terms of mammalian hormonal theory. Trewavas (1981)

suggested that growth substances might represent, instead, a form of

cell-to-cell interaction or communication and that the varying responses

to substances such as gibberellins by different species and tissues can

be explained by differing sensitivity of plants to gibberellins. This

concept may explain the variation in responses of waterhyacinths to

GA3 and 2,4-D under environmental and physiological conditions

referenced and investigated in this study.


Evaluation of effects of combinations of GA3 and 2,4-D on

waterhyacinths in this study indicate that

1. Under field conditions existing during this study, there was no

significant synergism between 2,4-D and GA3 in terms of increased effi-

cacy of 2,4-D. At best, the response of waterhyacinths to combinations

of GA3 and 2,4-D was additive. However, the use of these compounds

under other conditions may yield differing results.

2. Pretreatment of waterhyacinths with 100 mg/l GA3 does not signi-

ficantly increase the translocation of 14C labeled 2,4-D to meristematic

waterhyacinth tissues on either a dpm/mg or percent of total 14C

translocated basis. The data suggest a possible increase in transloca-

tion to leaves other than the leaf receiving direct 2,4-D treatment.

3. Costs analysis of utilizing GA3 in conjunction with 2,4-D in

order to lower rates of 2,4-D used to control waterhyacinths on an

operational basis indicated that the addition of GA3 was not economi-

cally justified. However, the use of these compounds under differing

field conditions and/or a significant change in the cost of 2,4-D or

GA3 may alter this situation.


Abbott Laboratories. 1962. Gibberellic acid: growth promoter for the
malting industry. Tech. Bull. No. 360. 22 pp.

Adams, P. A., P. B. Kaufman and H. Ikuma. 1973. Effects of gibberellic
acid and sucrose on the growth of oat (Avena) stem segments.
Plant Physiol. 51:1102-1108.

Addicott, F. T. 1970. Plant hormones in the control of abscission.
Biol. Rev. 45:485-524.

Anonymous. 1946. Plant growth regulators. Special Products Division,
Chemical Warfare Service. Science. 103:469-470.

Anonymous. 1965. Expanded project for aquatic plant control. Commission
on Public Works, Wash., D. C. report. 145 pp.

Ashton, F. M. 1959. Effect of gibberellic acid on absorption, transloca-
tion, and degradation of 2,4-D in red kidney bean. Weeds.

Ashton, F. M. and A. S. Crafts. 1973. Mode of Action of Herbicides.
John Wiley and Sons, Inc., New York. 504 pp.

Audus, L. J. 1972. Plant Growth Substances. Leonard Hill, London.
402 pp.

Bailiss, K. W. and T. A. Hill. 1971. Biological assays for gibberellins.
Bot. Rev. 37:437-479.

Baldev, B. A., A. Lang, and A. 0. Agatep. 1965. Gibberellin production
in pea seeds developing in excised pods: effect of growth retardant
Amo-1618. Science. 147:155-157.

Barendse, G. W. M. 1974. Biosynthesis, metabolism, transport and
distribution of gibberellins, pp. 65-80. In Krishnamoorthy, H. N.,
ed., Gibberellins and Plant Growth. John-Wiley and Sons, New York.

Basler, E. 1959. The effects of gibberellic acid on the translocation,
metabolism and toxicity of 2,4-D in bean plants. Proc. South. Weed
Conf. 14:171.

Basler, E. 1974. Abscisic acid and gibberellic acid as factors in
the translocation of auxin. Plant and Cell Physiol. 15:351-361.

Bhanja, A. and S. M. Sircar. 1966. Gibberellins from the root of
waterhyacinth (Eichhornia crassipes (Mart.) Solms). Sci. Cult.

Birch, A. J., R. W. Richards, and H. Smith. 1958. The biosynthesis
of gibberellic acid. Proc. Chem. Soc. 192-193.

Black, C. C. and G. A. Buchanan. 1980. How herbicides work:
the phenoxyacetic acids and related herbicides. Weeds Today.

Blackman, G. E. 1945. Plant growth substances of selective weed
killers: a comparison of certain plant-growth substances and other
selective herbicides. Nature. 155:500-501.

Bock, J. H. 1966. An ecological study of Eichhornia crassipes with
special emphasis on its reproductive biology. Ph.D Thesis,
University of California, Berkeley, CA. 175 pp.

Borrow, A., P. W. Brian, V. E. Chester, P. J. Curtis, H. G. Hemming, C.
Henchan, E. G. Jeffreys, P. B. Lloyd, I. S. Nixon, G. L. F. Norris,
and M. Radley. 1955. Gibberellic acid, a metabolic product of the
fungus Gibberella fujikuroi: some observations on its production
and isolation. J. Sci. Food and Agr. 6:340-348.

Brian, R. C. 1964. The effects of herbicides on biophysical processes
in the plant, pp. 357-386. In Audus, L. J., ed. The Physiology and
Biochemistry of Herbicides. Academic Press, New York.

Brian, P. W. and H. G. Hemming. 1958. Complementary action of gib-
berellic acid and auxins in pea internode extension. Ann. Bot.
(London) 22:1-17.

Brian, P. W., H. G. Hemming, and M. Radley. 1955. A physiological
comparison of gibberellic acid with some auxins. Physiol. Planta.

Brown, C. A., Q. L. Holdeman, and E. S. Hagood. 1946. Waterhyacinth
and 2,4-D. Louisiana Agric. Exp. Stat. Ann. Rept. 74 pp.

Buker, G. E. 1982. Engineers vs. Florida's green menace. The
Florida Historical Quarterly. 413-427 pp.

Cardenas, J., F. W. Slife, J. B. Hanson and H. Butler. 1968.
Physiological changes accompanying the death of cocklebur
plants treated with 2,4-D. Weed Sci. 16:96-100.

Carns, H. R. and F. T. Addicott. 1964. The effects of herbicides
on endogenous regulator systems, pp. 343-356. In Audus, L. J, ed.
The Physiology and Biochemistry of Herbicides. Academic Press,
New York.

Castro, P. R. C. Effects of growth regulators in tomato (Lycopersicon
esculentum m.11.). 1976. Ph.D. Dissertation. S. Paulo Univ.,
Piracicaba, Brazil. 148 pp. (cited in Castro and Rossetto, 1979).

Castro, P. R. C. and C. J. Rossetto. 1979. The influence of growth
regulators on aphid infestation in cotton. Turrialba. 29:75-77.

Center, T. D. and N. R. Spencer. 1981. The phenology and growth of
waterhyacinth (Eichhornia crassipes (Mart.) Solms) in an eutrophic
north-central Florida lake. Aquat. Bot. 10:1-32.

Chen, L. G., C. M. Switzer, and R. A. Fletcher. 1972. Nucleic acid
and protein changes induced by auxin-like herbicides. Weed Sci.

Chen, S. S. C. 1974. Role of gibberellins in dormancy and seed germina-
tion, pp. 91-99. In Krishnamoorthy, H. N., ed. Gibberellins and
Plant Growth. John-Wiley and Sons, New York.

Chen, S. S. C. and W. Park. 1973. Early actions of gibberellic acid
on the embryo and on the endosperm of Avena fatua seeds. Plant
Physiol. 52:174-176.

Chew, V. 1977. Comparisons among treatment means in an analysis of
variance. Agricultural Research Service, United States Department of
Agriculture. ARS/H/6. Available from Univ. of Florida, Gainesville,
FL. 64 pp.

Chin, T. Y. and J. A. Lockhart. 1965. Translocation of applied
gibberellin in bean seedlings. Am. J. Bot. 52:828-833.

Chrispeels, M. J. and J. B. Hanson. 1962. The increase in ribonucleic
acid content of cytoplasmic particles of soybean hypocotyl induced
by 2,4-dichlorophenoxyacetic acid. Weeds. 10:123-125.

Cleland, R. E., M. Thompson, D. L., Rayle, and W. K. Purves. 1968.
Difference in effects of gibberellins and auxins on wall exten-
sibility of cucumber hypocotyls. Nature. 219:510-511.

Clor, M. A. 1967. Translocation of tritium-labelled gibberellic acid
in pea stem segments and potato tuber cylinders. Nature. 214:

Crafts, A. S. 1964. Herbicide behavior in the plant, pp. 75-110. In
Audus, L. J., ed. The Physiology and Biochemistry of Herbicides.
Academic Press, New York.

Corcoran, M. R. 1974. Gibberellin antagonists and antigibberellins,
pp. 289-332. In Krishnamoorthy, H. N., ed. Gibberellins and Plant
Growth. John NfTley and Sons, New York.

Davis, F. S. 1970. Review of toxicology, persistence and mobility
of phenoxy herbicides within the environment. Interagency
Research Advisory Committee, Aquatic Plant Control Program. 18 pp.

Davis, J. T. and W. S. Hardcastle. 1959. Biological assay of herbicides
for fish toxicity. Weeds. 7:397-404.

de la Gauardia, M. D. and M. Benllock. 1980. Effects of potassium
and gibberellic acid on stem growth of whole sunflower plants.
Physio. Plant. 49:443-448.

Devlin, R. M., M. J. Kisiel, and A. S. Kostusiak. 1980. Enhancement
of gibberellic acid sensitivity in corn (Zea mays) by fluridone
and R-40244. Weed Sci. 28:11-12.

Duke, T. 1971. The effects of 2,4-D formulations on estuarine
organisms. Environmental Protection Agency, Gulf Breeze Laboratory
Technical Report. 17 pp.

Dupes, J. M. and M. J. Mahler. 1982. Public funded aquatic plant control
operations in Florida. Aquatics. 4:6.

Ende, Van den J. and J. Koornneef. 1960. Gibberellic acid and osmotic
pressure. Nature. 186:327.

Fang, S. C. 1958. Absorption, translocation and metabolism of
2,4-D-1-C14 in pea and tomato plants. Weeds. 6:179-186.

Feung, C. S., S. L. Loerch, R. H. Hamilton, and R. G. Mumma. 1978.
Comparative metabolic fate of 2,4-dichlorophenoxyacetic acid in
plants and plant tissue culture. J. Agric. Food Chem. 26:1064-1067.

Galston, A. W. and W. K. Purves. 1960. The mechanism of action of
auxin. Ann. Rev. Plant Physiol. 11:239-276.

Goyal, A. K. and B. D. Baijal. 1980a. Effect of gibberellic acid on
RNase activity and RNA contents at early seedling stage in certain
rice (Oryza sativa L.) genotypes. Indian J. Agric. Res. 14:111-114.

Goyal, A. K. and B. D. Baijal. 1980b. Response of certain rice
varieties to gibberellic acid at early seedling stage. Acta Bot.
Indica. 8:37-40.

Halter, M. T. 1980. 2,4-D in the aquatic environment, pp. 88-182. In
Literature Reviews of Four Selected Herbicides: 2,4-D, Dichlobenil,
Diquat and Endothall. Municipality of Metropolitan Seattle, Seattle.

Hansen, W. H., M. L. Quaife, R. T. Habermann, and 0. G. Fitzhugh. 1971.
Toxicol. Appl. Pharmacol. 20:122-129.

Hemmett, R. B., Jr. and S. D. Faust. 1969. Biodegradation kinetics of
2,4-dichlorophenoxyacetic acid by aquatic micro-organisms, pp.
191-207. In Gunther, F. A., ed. Residue Reviews. Vol. 29.
Springer-Verlog, New York.

Hildebrand, E. M. 1946. Herbicidal action of 2,4-dichlorophenoxyacetic
acid on the waterhyacinth, Eichhornia crassipes. Science.

Hitchcock, A. E., P. W. Zimmerman, H. Kirkpatrick, Jr. and T. T. Earle.
1949. Waterhyacinth: its growth, reproduction, and practical control
by 2,4-D. Contrib. from Boyce Thompson Inst. 15:363-401.

Hoad, G. V. and M. R. Bowen. 1968. Evidence of gibberellin-like
substances in phloem exudate of higher plants. Planta. 82:22-32.

Hughes, J. S. and J. T. Davis. 1965. Comparative toxicity to bluegill
sunfish of granular and liquid herbicides. Proc. Ann. Conf. South-
east Game and Fish Comm. 16:319-323.

Israelstam, G. F. and E. Davis. 1979. An analysis of the effects of
gibberellic acid on the leaflet structure of dwarf cultivars of
pea (Pisum sativum). Can. J. Bot. 57:1089-1092.

Jacobs, W. P. and H. Kaldeway. 1970. Polar movement of gibberellic
acid through young Coleus petioles. Plant Physiol. 45:539-541.

Jacobs, W. P. and P. E. Pruett. 1973. The time-course of polar movement
of gibberellin through Zea roots. Amer. J. Bot. 60:896-900.

Jones, R. L. and I. D. J. Phillips. 1964. Agar-diffusion technique
for estimating gibberellin production by plant organs. Nature.

Joyce, J. C. 1977. Selective maintenance control plan, St. Johns
River, Florida. In Proc. Research Planning Conf. on the Aquatic
Plant Control Program. U.S. Army Corps of Engineers Waterways
Experiment Station. Misc. Paper. 77-3:45-48.

Joyce, J. C. and H. C. Sikka. 1977. Residual 2,4-D levels in the St.
Johns River, Florida. J. Aquatic Plant Manage. Soc. 15:76-82.

Kato, J. 1956. Effect of gibberellin on elongation, water uptake and
respiration of pea-stem sections. Science. 123:1132.

Kato, J. 1958a. Nonpolar transport of gibberellin through pea stem and
a method for its determination. Science. 128:1008-1009.

Kato, J. 1958b. Studies on the physiological effect of gibberellin II.
Physiol. Planta. 11:10-15.

Katsumi, M., H. Kazama, and N. Kawamura. 1980. Osmotic potential of the
epidermal cells of cucumber hypocotyls as affected by gibberellin
and cotyledons. Plant and Cell Physiol. 21:933-937.

Kaufman, P. B. 1974. Relationship between gibberellins and metabolism,
pp. 225-238. In Krishnamoorthy, H. N., ed. Gibberellins and Plant
Growth. John WTley and Sons, New York.

Key, J. L. 1964. RNA synthesis and the control of expansive growth of
excised soybean hypocotyl. Plant Physiol. Abstr. 39:xxx.

Key, J. L. and J. C. Shannon. 1964. Enhancement by auxin of ribonucleic
acid synthesis in excised soybean hypocotyl tissue. Plant Physiol.

Kiermayer, 0. 1964. Growth responses to herbicides, pp. 207-233. In
Audus, L. J., ed. The Physiology and Biochemistry of Herbicides.
Academic Press, New York.

Kimura, E. T., P. R. Young and K. Staniszewski. 1959. Gibberellic
acid: toxicologic and pharmacologic studies. J. Am. Pharm. Assoc.
Sci. Ed. 48:127-9.

Knipling, E. B., S. H. West, and W. T. Hailer. 1970. Growth
characteristics, yield potential, and nutritive content of water-
hyacinths. Proc. Soil and Crop Sci. Soc of FL. 30:51-63.

Koch, W., G. Harris, K. B. El Tigani, F. R. Hamza, M. Obeid, M.
Akusha, V. Leffler, and Th. Hafliger. 1978. Investigations on
the chemical control of Eichhornia crassipes (Mart.) Solms. in
the Sudan, pp. 415-427. Proc. EWRS 5th Symp. on Aq. Weeds.

Kogl, F. and Elema, J. 1960. Wirkungs-bezichungen zwischen Indole
3-essigsaune und gibberellinsaurene. Naturwiss. 47:90 (cited by
Weaver, 1972).

Krishnamoorthy, H. N. 1974. Role of gibberellins in juvenility,
flowering, and sex expression, pp. 113-143. In Krishnamoorthy, H. N.,
ed. Gibberellins and Plant Growth. John Wiley--and Sons, New York.

Lang, A. 1970. Gibberellins: structure and metabolism. Ann. Rev.
Plant Physiol. 21:537-570.

Lockhart, J. A. 1957. Studies on the organ of production of the natural
gibberellin factor in higher plants. Plant Physiol. 32:204-207.

Low, V. H. K. 1974. Role of gibberellins in root and shoot growth,
pp. 101-114. In Krishnamoorthy, H. N., ed. Gibberellins and Plant
Growth. John Vfley and Sons, New York.

Luis, A. G. and J. L. Guardiola. 1981. Effect of gibberellic acid on
ion uptake selectivity in pea seedlings. Planta. 153:494-496.

Maheshwari, R., C. Shailini, K. Veluthambi, and S. Makadevan. 1980.
Interaction of gibberellic acid and indole-3-acetic acid in the
growth of excised Cuscuta shoot tips "in vitro." Plant Physiol.

Mann, J. D. 1974. The mechanism of action of gibberellins, pp. 238-287.
In Krishnamoorthy, H. N., ed. Gibberellins and Plant Growth. John
Wiley and Sons, New York.

Marth, P. C., W. V. Audia, and J. W. Mitchell. 1956. Effects of
gibberellic acid on growth and development of plants of various
genera and species. Bot. Gaz. 118:106-111.