Physiological, anatomical and growth responses of mango (Mangifera indica L.) trees to flooding

MISSING IMAGE

Material Information

Title:
Physiological, anatomical and growth responses of mango (Mangifera indica L.) trees to flooding
Physical Description:
viii, 143 leaves : ill., photos ; 29 cm.
Language:
English
Creator:
Larson, Kirk David, 1953-
Publication Date:

Subjects

Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1991.
Bibliography:
Includes bibliographical references (leaves 128-142).
Statement of Responsibility:
by Kirk David Larson.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001742452
notis - AJF5168
oclc - 26181239
System ID:
AA00003727:00001

Full Text













PHYSIOLOGICAL, ANATOMICAL AND GROWTH RESPONSES OF MANGO
(MANGIFERA INDICA L.) TREES TO FLOODING













By

KIRK DAVID LARSON


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1991


































In memory of Catherine Munn Fatherson
















ACKNOWLEDGEMENTS


Bruce Schaffer and Frederick Davies provided encouragement,

guidance with statistical design and analysis, excellent editorial

assistance, and financial support. I thank Peter Andersen, Donald

Graetz and James Syvertsen for editorial assistance, encouragement and

advice. The following people provided reviews of the respective

chapters: Jonathan Crane, Frederick Davies, Donald Graetz, Bruce

Schaffer, Charles A. Sanchez, George Snyder and James Syvertsen, Chapter

Three; Peter Andersen, Frederick Davies, Donald Graetz, Bruce Schaffer

and James Syvertsen, Chapter Four; Peter Andersen, Jonathan Crane,

Frederick Davies, Bruce Schaffer and James Syvertsen, Chapter Five;

Peter Andersen, Frederick Davies, Jack Fisher, Richard Litz, Terrance

Lucansky, Bruce Schaffer and James Syvertsen, Chapter Six; Peter

Andersen, Frederick Davies, Bruce Schaffer and James Syvertsen, Chapter

Seven.

Thomas Davenport, Jack Fisher, Richard Litz and Randy Ploetz

provided stimulating discussions, and generous use of equipment.

Leandro Ramos and Jack Fisher gave excellent instruction in plant

microtechnique. Charles Sanchez assisted with plant tissue analysis and

editorial reviews. Ramesh Reddy gave excellent advice regarding soil

analysis. Pablo Lara, Leonard Rippetoe, Lynette Eccles and Mary Jackson

provided technical assistance with various aspects of this study, as did

Mike Roessner, Charles Brunson and Glen Gillespie. I also acknowledge

the support of the secretarial staffs of the University of Florida











Tropical Research and Education Center and University of Florida Fruit

Crops Department. Zill's High Performance Plants and J.R. Brooks and

Son, Inc. generously provided some of the plant material used in this

study. I thank the Dade County AGRI-Council, the W.F. Ward family, H.E.

Kendall, Sr., the Dade County Women in Agriculture, and the Florida

Mango Forum for scholarships.

Most importantly, I thank my wife, Katherine Ann Whitson, for her

endless encouragement and support, friendship, sacrifice and patience

throughout the course of this study.

















TABLE OF CONTENTS

page
ACKNOWLEDGEMENTS........................................... iii

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

CHAPTERS

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

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

Effect of Flooding on Chemical, Physical
and Biological Processes in the Soil.............. 4
Flooding and Plant Mineral Nutrition................ 6
Flooding, Leaf Gas Exchange and Plant Water Status.. 8
Flooding and Plant Ethylene Evolution............... 11
Flooding and Plant Growth ........................... 13
Influence of Flooding on Plant Morphology
and Anatomy................................... 13

3. FLOOD-INDUCED CHEMICAL TRANSFORMATIONS IN CALCAREOUS
AGRICULTURAL SOILS OF SOUTH FLORIDA.................... 15

Introduction............................. .... ...... 15
Materials and Methods............................... 16
Results and Discussion.............................. 18
Conclusions........................................ 24

4. FLOODING, MINERAL NUTRITION AND LEAF GAS EXCHANGE
OF MANGO TREES.......................................... 36

Introduction....................................... 36
Materials and Methods............................... 37
Results .............. .... .... ................... 40
Discussion.......................................... 43
Conclusions........................................ 47

5. FLOODING, LEAF GAS EXCHANGE AND GROWTH OF MANGO
IN CONTAINERS...... .......... ........................ 57

Introduction ............................ .... ...... 57
Materials and Methods............................... 58
Results.............. ......... ...... .... .. ....... .. 62
Discussion......................... ... ... ... ...... 65
Conclusions......................................... 67


















page
6. FLOODWATER TEMPERATURE AND LENTICEL HYPERTROPHY
IN MANGO TRFES......................................... 77

Introduction....................................... 77
Materials and Methods............................... 78
Results .................... ....................... 81
Discussion............. ............................ 84
Conclusions......................................... 86

7. FLOODWATER DISSOLVED OXYGEN CONTENT, LENTICEL
HYPERTROPHY AND ETHYLENE EVOLUTION IN MANGO TREES...... 104

Introduction........................................ 104
Materials and Methods............................. 105
Results .......... ......... ........... ................ 110
Discussion......................................... 112
Conclusions........................................ 114

8. CONCLUSIONS............................................ 125

LITERATURE CITED .......................................... 128

BIOGRAPHICAL SKETCH ........................................ 143
















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


PHYSIOLOGICAL, ANATOMICAL AND GROWTH RESPONSES OF MANGO
(MANGIFERA INDICA L.) TREES TO FLOODING

By

Kirk David Larson

August, 1991


Chairman: Frederick S. Davies
Cochairman: Bruce Schaffer
Major Department: Horticultural Science (Fruit Crops)



Experiments were conducted to determine the effects of soil

flooding on the chemistry of calcareous soils, and on mineral nutrition,

leaf gas exchange, vegetative growth, anatomy, and ethylene evolution of

mango (Mangifera indica L.) trees. Concentrations of NH4+, Fe and Mn

were greater, but N03- and P were less, for anaerobically incubated than

for aerobically incubated calcareous soils. To determine the influence

of flooded soil chemical transformations on mineral nutrition and leaf

gas exchange, mango trees were grown with (+Fe) or without (-Fe) iron

fertilizer in calcareous soil, and flooded for 0, 10 or 20 days. Prior

to flooding, leaf gas exchange, total leaf chlorophyll content, and

foliar Mn and Fe concentrations were less for -Fe trees than for +Fe

trees. Chlorophyll, Fe, Mn, and leaf gas exchange increased in flooded

trees relative to nonflooded trees, and this increase was greatest for

the -Fe trees. Flooding reduced vegetative growth and increased the


vii
















shoot:root ratio. Rapid and simultaneous reductions in net CO2

assimilation and stomatal conductance for CO2 occurred with flooding,

yet substomatal CO2 concentration increased. After floodwaters subsided

there was a slow recovery of these physiological variables. The flood-

induced decline in leaf gas exchange was not accompanied by reduction in

leaf water potential. Hypertrophy of stem lenticels was frequently

observed in trees that survived flooding stress, but not in trees that

died as a result of flooding. If hypertrophied lenticels were covered

to impede gas exchange, the trees died within 2-3 days. Hypertrophy was

most rapid at floodwater temperatures of 300 C, but did not occur at 150

C, and was reduced when floodwater dissolved 02 content was increased to

15 ppm. Lenticel hypertrophy was characterized by a more spherical

shape of cells in the phellem and phelloderm, by development of

intercellular spaces in the phellem and lenticel filling tissue, and by

the production of additional phellem tissue adjacent to the lenticel

pore, resulting in a larger pore opening. Ethylene evolution from stem

tissue was greater with floodwater dissolved oxygen contents of 1-3 ppm

than with 02 contents of 6-15 ppm. The results of these investigations

indicate that although mango is not a hydrophytic plant, it does possess

certain adaptations that permit survival under flooded soil conditions.


viii
















CHAPTER 1
INTRODUCTION



The mango is one of the world's most widely planted fruit crops

(Anon, 1989a), and is grown in at least 87 countries (Bondad, 1980).

Cultivated for over 4,000 years (Purseglove, 1968), it is an important

food throughout the tropics and an increasingly significant item of

commerce in many tropical, subtropical and temperate areas (Subra, 1981;

Toohill, 1984). In recent years, production of mangos for local and

export markets has led to increased plantings throughout the Caribbean,

Central and South America, Africa, and parts of Australia and North

America (Subra, 1981; Toohill, 1984).

In many regions of the world, mangos are grown in either heavy

soils that impede internal drainage, or in low elevation areas that are

prone to periodic flooding as a result of intense rainstorms. In South

Florida (the major area of commercial mango production in the United

States), high land values have forced new mango plantings into low

elevation areas that are prone to seasonal flooding during the summer

and fall rainy season.

Although numerous studies have been conducted on flooding of fruit

trees (Andersen et al., 1984a, 1984b; Childers and White, 1942; Crane

and Davies, 1985; Davies and Wilcox, 1984; Davies and Flore, 1986a,

1986b, 1986c; Phung and Knipling, 1976; Syvertsen et al., 1983), very

little work has been done to quantify responses of mango to soil

flooding. The small amount of available information regarding mango











flood tolerance is conflicting, and based almost exclusively on field

observations. Some reports indicate that mango trees require good soil

drainage for adequate growth and yield (Alfonsi and Brunini, 1980;

Popenoe, 1920; Samson, 1986; Valmayor, 1962), whereas others indicate

that they are flood tolerant (Chandler, 1958; Jawanda, 1961; Young and

Sauls, 1981). Chandler (1958) noted that mangos tolerate shallow,

poorly aerated soils, but observed that young mango trees may be injured

by poor soil aeration. Mature mango trees appear to tolerate flooding

better than young mango trees (Chandler, 1958).

In Florida, the effect of flooding on mangos has been variable.

Established mango trees have withstood prolonged periods (months) of

flooding, whereas young trees planted in poorly drained areas often die

rapidly (K.Mitchell, J.R. Brooks and Son, Inc., personal communication).

Flooding of young bearing trees has resulted in tree decline and bloom

inhibition (C.Campbell, personal communication). Young and Sauls (1981)

reported that mangos on a sandy organic soil withstood flooding for at

least 2 months without harm. However, mangos did not perform well under

constant poor drainage (Young and Sauls, 1981).

The wide range of plant response to flooding and the conflicting

reports of mango flood tolerance indicate the need for definitive

studies on this subject. Such studies could assist in the development

of sound management programs for mangos in flood-prone areas, and would

provide a better basic understanding of fruit tree responses to

flooding.

This dissertation includes a literature review of the effects of

flooding on soil chemistry, and of plant responses to flooding (Chapter

2). The dissertation also encompasses five separate experiments that











were designed to determine the effects of flooding on: limestone soil

chemistry (Chapter 3); mango mineral nutrition (Chapter 4); net gas

exchange and vegetative growth of mango (Chapter 5); mango stem lenticel

hypertrophy (Chapter 6); and the relationship between floodwater oxygen

content, ethylene evolution and mango stem lenticel hypertrophy (Chapter

7). These five experiments were conceived, justified, executed, and

written as separate units during a three-year period. As such, there is

some redundancy between Chapter 2 (Literature Review) and the

introductions for each of Chapters 3 through 7.
















CHAPTER 2
LITERATURE REVIEW



Effect of Flooding on Chemical, Physical
and Biological Processes in the Soil



Flooding results in the displacement of 02 from the soil pores by

water and the consumption of available oxygen by obligate and

facultatively aerobic organisms. Thus, flooded soils may become nearly

devoid of 02 within a few hours (Ponnamperuma, 1984). The absence of 02

in flooded soils results in the predominance of facultative and obligate

anaerobes that use N03-, Mn4+, Fe3+, S042-, C02, N2, H+, and various

organic molecules as terminal electron acceptors in oxidative

phosphorylation and electron transport processes (Ponnamperuma, 1984).

After oxygen is consumed in a flooded soil, these constituents are

reduced, generally in a sequential manner, to N2, Mn2+, Fe2+, H2S, CH4,

NH4+ and H2, respectively (Rowell, 1981). These oxidized soil

constituents and their corresponding reduced forms comprise the redox

couples of a soil solution.

Redox potential (Eh) is a mixed potential of all the redox couples

in a soil solution, and is used to determine the oxidation-reduction

status of the soil (Gambrell and Patrick, 1978; Ponnamperuma, 1972,

1984). Typically, Eh of aerobic soil ranges between 300 and 800 my, and

that of anaerobic soil between -450 and 200 my (Ponnamperuma, 1972).

Determination of Eh is most useful in anaerobic soil solutions where the











high concentrations of redox-active ions and organic molecules, and the

high H+ exchange currents increase the stability of the redox potentials

(Bohn, 1971). However, variations among soil microsites in mineral

composition redoxx couples) and in bacterial or root populations

necessitate numerous measurements to ensure accurate assessment of soil

Eh (Rowell, 1981). In aerated systems, measurement of Eh is much less

useful; metal ions and organic compounds are oxidized and less soluble,

reducing the concentration of redox couples and the stability and

reproducibility of the measurements (Bohn, 1971).

Soil Eh is influenced by pH (Armstrong, 1975); therefore,

measurements are often converted to standard pH values (59 my per pH

unit) (Armstrong, 1975). Redox potential is also influenced by the

concentration of the specific redox couples in the soil solution, since

high concentrations of a redox couple can result in the buffering

(poise) of the Eh at a particular potential (Rowell, 1981). High levels

of soil organic matter tend to increase bacterial respiration rates,

resulting in a more rapid decrease in Eh (Rowell, 1981). Temperature

can also influence Eh by regulating bacterial respiration rates.

After flooding, the pH of an acid soil increases mainly due to

iron reduction, whereas the pH of an alkaline soil decreases due to C02

accumulation from bacterial respiration (Ponnamperuma, 1972, 1984).

Carbon dioxide reacts with water to form carbonic acid (H2C03), which

dissociates into H+ and HC03- ions (Buckman and Brady, 1968). In

alkaline soils, H+ ions neutralize OH- ions in the soil solution,

resulting in acidification of the soil solution. These flood-induced pH

changes can increase the exchange capacity of an acid soil, but decrease

that of an alkaline soil (Ponnamperuma, 1984).











The decomposition mineralizationn) of organic matter proceeds at a

much slower rate under anaerobic conditions than in the presence of

oxygen, and nitrification ceases, resulting in an increase in NH4+ in

the soil solution (Ponnamperuma, 1984). In addition to increases in

NH4+ concentration, flooding often results in an increase in other soil

nutrients due to an increase in the amounts of dissolved and suspended

nutrients, and solubilization of P, Si, Fe and Mn (Ponnamperuma, 1984).

Increases in the ionic strength of the soil solution often result in

displacement of K, Ca and Mg from the soil exchange complex, thereby

increasing the concentration of these elements in the soil solution

(Ponnamperuma, 1984). In porous soils, however, flooding can result in

nutrient losses by leaching (Ponnamperuma, 1984), and N03- disappears

rapidly from flooded soils by reduction, leaching or denitrification

(McGarity, 1961; Ponnamperuma, 1972, 1984; Stefanson and Greenland,

1970). Flooding also results in a reduction in soil temperature,

swelling of soil colloids and loss of soil structure, all of which can

adversely effect nutrient absorption and availability (Ponnamperuma,

1984).



Flooding and Plant Mineral Nutrition



For many plant species exposed to anaerobic stress, root decay

(Stolzy et al., 1975), and reductions in root metabolism (Labanauskas et

al., 1971; Trought and Drew, 1980) and root growth (Kozlowski, 1984)

often result in decreased plant nutrient concentrations. Although

nutrient uptake is generally inhibited, plant nutrient concentrations

are dependent on soil conditions as well as plant uptake responses











(Kozlowski and Pallardy, 1984). For example, in soils that are

moderately or severely deficient in P, flooding can result in increased

P solubility due to flood-induced shifts in soil pH. In such cases,

increased P availability may counteract the inhibition in P uptake

resulting from reduced root metabolism. Additionally, the reduction of

ferric Fe and manganic Mn to more soluble forms often makes these

elements more available to plants. Although plant nutrient

concentrations may increase with short-term flooding, long-term flooding

generally results in decreased concentrations of plant macronutrients,

due to degeneration of the root system (Kozlowski and Pallardy, 1984).

For some flood-tolerant species, morphological adaptations to

flooding may allow continued uptake of nutrients under flooded

conditions. In these plants, initiation of adventitious roots at the

surface of the submerged soil or in the floodwater permits continued

nutrient uptake in the most oxidized soil environment (Hook et al.,

1971; Kozlowski, 1984; Kozlowski and Pallardy, 1984). Additionally, the

development of aerenchymatous tissues in the roots and stems of some

flood tolerant plants can permit internal oxygen diffusion to the roots,

thereby maintaining root metabolic function (Armstrong, 1968; Coutts and

Armstrong, 1976). Provided with adequate nutrient availability in

flooded soils, flood-tolerant plants are often able to absorb sufficient

nutrients for maintaining plant growth. Thus, flooding often results in

decreased nutrient absorption in flood-intolerant plants, but increased

nutrient absorption in flood-tolerant species (Kozlowski and Pallardy,

1984).











Flooding, Leaf Gas Exchange and Plant Water Status



One of the earliest detected physiological responses of fruit

trees to flooding is a reduction in stomatal conductance (gs) (Andersen

et al., 1984a, 1984b; Crane and Davies, 1989; Davies and Flore, 1986a,

1986b, 1986c; Pereira and Kozlowski, 1977; Schaffer and Ploetz, 1989;

Schaffer et al., 1991; Smith and Ager, 1988; Syvertsen et al., 1983).

Although many woody plants exhibit a decrease in gg within 1 to 3 days

after flooding (Andersen et al., 1984a; Crane and Davies, 1989), the

duration of flooding required for decreases in gs can vary among

species. For example, gs of Citrus aurantium and Citrus lambhiri

seedlings decreased 4 and 8 days after flooding, respectively (Syvertsen

et al., 1983). Seasonal effects of flooding on gs have also been

observed. In general, flooding during the spring or summer results in

more rapid reductions in gs than flooding in the fall (Andersen et al.,

1984b; Olien, 1987).

Flood-induced decreases in g, result in reduced transpiration (E)

for a number of woody plant species (Crane and Davies, 1989; Davies and

Flore, 1986b; Kozlowski and Pallardy, 1984; Olien, 1989; Phung and

Knipling, 1976; Ploetz and Schaffer, 1987; Regehr et al., 1975; Sena

Gomes and Kozlowski, 1980; Smith and Ager, 1988; Sojka and Stolzy,

1980). For many species, reductions in gs are transitory, and gs

usually increases to that of nonflooded plants after the flooding stress

is relieved (Crane and Davies, 1988; Davies and Flore, 1986b; Joyner and

Schaffer, 1990; Schaffer et al., 1991; Smith and Ager, 1988). However,

for some species exposed to prolonged flooding, gs does not recover, or











recovers very slowly after plants are removed from flooding (Davies and

Flore, 1986b; Schaffer et al., 1991; Smith and Ager, 1988).

Various hypotheses have been suggested to account for reductions

in stomatal aperture under flooded conditions. In some studies, flood-

induced stomatal closure has been attributed to a reduction in leaf

water potential resulting from decreased water uptake by the roots

(Heincke, 1932; Kozlowski and Pallardy, 1984; Kramer, 1954) or

reductions in root hydraulic conductivity (Jackson et al., 1978; Kramer,

1952; Syvertsen et al., 1983; Willey, 1970), possibly as a result of

decreased cell membrane permeability (Bradford and Yang, 1981; Bradford

and Hsiao, 1982; Kramer, 1969; Slatyer, 1967). Under anaerobic

conditions, decreases in ATP and respiratory substrates, production of

toxic anaerobic respiration byproducts (i.e., acetaldehyde and ethanol)

in the plant, and production of toxic compounds in the soil may affect

membrane integrity and function (Drew, 1983). Andersen et al. (1984b)

attributed decreased conductance to water flow in roots and stems to

plugging of xylem vessels.

For most plants, early stomatal closure due to flooding occurs

without any reduction in plant water status (Andersen et al., 1984a;

Bradford and Hsiao, 1982; Davies and Wilcox, 1984; Davies and Flore,

1986a, 1986b, 1986c; Jackson and Hall, 1987; Pereira and Kozlowski,

1977; Smith and Ager, 1988). Reductions in root or stem hydraulic

conductivity (Andersen et al., 1984a; Davies and Wilcox, 1984; Davies

and Flore, 1986b; Syvertsen et al., 1983), reductions in K ion

concentration (Stolzy et al., 1975; Peaslee and Moss, 1966; Letey et

al., 1961a, 1961b), or hormonal factors (El-Betalgy and Hall, 1974;

Hiron and Wright, 1973; Reid and Bradford, 1984; Reid and Crozier, 1971;











Reid et al., 1969; Wright and Hiron, 1969) have been postulated to cause

flood-induced stomatal closure in the absence of a reduction in plant

water status. Reid and Bradford (1984) suggested that reduced cytokinin

and/or gibberellin synthesis in flooded roots, possibly in combination

with increased ABA levels, may be responsible for anoxia-induced

stomatal closure. Davies and Flore (1985) observed a reduction in gs

when root exudate from flooded blueberry plants was applied to

nonflooded plants, suggesting the presence of a translocatable substance

originating in the roots. Schaffer and Ploetz (1991), working with

approach-grafted avocado trees, also reported the presence of a

translocatable substance involved in stomatal regulation. However, no

universal mechanism for stomatal regulation in flooded plants has been

demonstrated.

The hypothesis that flood-induced stomatal closure limits net CO2

assimilation (A) is supported by studies in which gg, A and the internal

partial pressure of CO2 decreased simultaneously in flooded plants

(Davies and Flore, 1986a, 1986c). However, non-stomatal inhibition of A

also has been reported with flooding (Beckman et al., 1987; Bradford,

1983; Childers and White, 1942; Moldau, 1973). It has been postulated

that for g, to regulate A, reductions in stomatal aperture must occur

prior to reductions in A. In some flooding studies, however, A and gs

decrease simultaneously (Ploetz and Schaffer, 1989; Schaffer and Ploetz,

1989; Smith and Ager, 1988). Increases in internal partial pressure of

CO2 observed concomitant with decreases in A and gs (Schaffer and

Ploetz, 1989), suggest that, for some plants, flooding inhibits A more

than g, (Farquhar and Sharkey, 1982).











For blueberry, where stomatal closure appears to be regulating A

during the early stages of a flooding cycle, long-term flooding

eventually results in decreased carboxylation efficiency (Davies and

Flore, 1986b), thereby indicating a direct effect of flooding on the

photosynthetic apparatus. Decreased mesophyll conductance may be due to

reduced chlorophyll content or changes in carboxylation enzymes

(Kozlowski, 1982; Vu and Yelenosky, 1991), or to feedback inhibition due

to starch accumulation (Wample and Thornton, 1984). Beckman et al.

(1987) reported the occurrence of a translocatable photosynthetic

inhibitor in the xylem sap of flooded cherry trees. Application of

flooded tree xylem sap exudate to nonflooded trees resulted in a rapid

decrease in A, although gs was unaffected. With nonstomatal limitations

to CO2 assimilation, changes in A may regulate gs, since stomata

function to maintain a constant substomatal CO2 partial pressure (Wong

et al., 1979).

Stomatal and nonstomatal limitations to A resulting from flooding

may be species-dependent. The simultaneous decreases in A and g. often

observed in flooded plants (Davies and Flore, 1986c; Ploetz and

Schaffer, 1989; Schaffer and Ploetz, 1989; Smith and Ager, 1988) suggest

independent regulation of these two physiological processes.



Flooding and Plant Ethylene Evolution



Although ethylene is often produced in flooded soils (Crane and

Davies, 1986; Smith and Russell, 1969; Smith and Dowdell, 1974), and may

be an environmental source of ethylene in flooded plants, increased

ethylene synthesis has been observed in plants exposed to flooded soil











conditions (Bradford, 1981; El-Beltagy and Hall, 1974; Jackson and

Campbell, 1975, 1976; Kawase, 1972, 1976). Ethylene can induce many of

the symptoms commonly associated with plant flooding stress, such as

leaf epinasty (Bradford and Yang, 1980a; Jackson and Campbell, 1975,

1979), leaf senescence (El-Betalgy and Hall, 1974), adventitious rooting

(Drew et al., 1979; Kawase, 1971), aerenchyma development (Drew et al.,

1979, 1981), and stem and lenticel hypertrophy (Kozlowski, 1984; Reid

and Bradford, 1984). Anaerobiosis stimulates the production of the

ethylene precursor 1-aminocyclopropane-l-carboxylic acid (ACC) in the

roots of flooded tomato plants. However, anaerobiosis inhibits the

production of ethylene by the roots, since conversion of ACC to ethylene

is oxygen-dependent. The ACC moves in the transpiration stream,

resulting in increased ethylene synthesis in the shoots (Bradford and

Yang, 1980a, 1980b). The biosynthetic pathway of ethylene, from

methionine to S-adenosylmethionine (SAM) to ACC, has been well-

documented for tomato plants (Yang, 1980; Yang et al., 1982). However,

alternative precursors and pathways may exist in other plant species

(Jackson, 1985).

Other plant hormones may interact with ethylene or influence the

production of ethylene or ethylene precursors under flooded conditions.

Elevated auxin levels have been observed in the shoots of flooded plants

(Phillips, 1965; Wample and Reid, 1979), presumably due to reductions in

auxin transport or metabolism (Phillips, 1964; Reid and Bradford, 1984),

and increased auxin levels have been shown to stimulate ethylene

production (Abeles and Rubenstein, 1964; Yu and Yang, 1979). Although

cytokinins promote ACC and ethylene synthesis (Reid and Bradford, 1984),

levels of cytokinins have been shown to decrease in flooded plants











(Burrows and Carr, 1969). In contrast, flooding stress can promote ABA

production (Hall et al., 1977; Wright and Hiron, 1972), and ABA inhibits

ACC and ethylene synthesis (Reid and Bradford, 1984).



Flooding and Plant Growth



In certain plants, rapid stomatal closure maintains turgor and may

permit continued growth under flooded conditions (Regehr et al., 1975).

However, flood-induced inhibition of plant growth occurs in many species

(Andersen et al., 1984a; Davies and Wilcox, 1984; Kozlowski, 1984; Tang

and Kozlowski, 1982; Yu et al., 1969), partly due to stomatal closure

and reduced C02 assimilation (Phung and Knipling, 1976; Regehr et al.,

1975; Stolzy et al., 1964). In general, for most flood-intolerant

species, flooding adversely affects root growth and viability due to

root necrosis and pathogen infection (Kozlowski, 1984; Stolzy and Sojka,

1984). Flooding also curtails shoot elongation, and reduces leaf area

by reducing leaf initiation and leaf expansion, and by hastening leaf

senescence (Kozlowski, 1984). Root growth and survival is generally

more affected by flooding than is shoot growth, resulting in a reduced

root-to-shoot ratio (Kozlowski, 1984).



Influence of Flooding on Plant Morphology and Anatomy



Many plants adapt to flooded conditions by formation of

adventitious roots or enlarged lenticels that enhance internal oxygen

diffusion to the roots (Andersen et al., 1984a; Coutts, 1982; Hook and

Scholtens, 1978; Kozlowski, 1984; Kramer, 1983; Pereira and Kozlowski,











1977; Philipson and Coutts, 1978), or function as excretory organs for

the elimination of potentially toxic plant metabolites (Chirkova and

Gutman, 1972). Wetland plants often have greater root porosities due to

the separation and configuration of cortical cells, and exhibit greater

tissue porosity and aerenchyma development than mesophytic species

(Justin and Armstrong, 1987). Jensen et al. (1969) reported that

decreased soil oxygen and increased light and temperature were

associated with greater root porosities. Yu et al. (1969) found

increased root porosities in some plants subjected to flooding. In

wetland plants, internal oxygen transport to roots is often adequate for

root respiration (Barber et al., 1962; Conway, 1937; Teal and Kanwisher,

1966) as well as for diffusion into, and aeration of, anaerobic rooting

media (Armstrong, 1964, 1967, 1968, 1978; Hook et al., 1970, 1972).

Kozlowski (1984) observed that older trees generally tolerate

flooding better than seedlings or saplings. Older woody roots are more

tolerant of flooding than non-woody roots (Coutts, 1982; Coutts and

Philipson, 1978), probably due to the higher growth rates, respiration

rates and oxygen requirements of younger root tissues (Lahde, 1966;

Luxmore and Stolzy, 1972). This may partly explain why younger trees,

with proportionally more young root tissues, are sometimes more

sensitive to flooding than older trees.
















CHAPTER 3
FLOOD-INDUCED CHEMICAL TRANSFORMATIONS IN CALCAREOUS
AGRICULTURAL SOILS OF SOUTH FLORIDA



Introduction



Tropical fruit crops in Florida have an annual farm gate value of

over $40 million (Anon, 1987). These crops traditionally have been

grown on soils of the Krome very gravelly loam series (loamy-skeletal,

carbonatic, hyperthermic Lithic Rendoll) (Anon, 1989b; Burns et al.,

1965; Leighty and Henderson, 1958). This soil has a pH of 7.2 to 7.6,

and is excessively well-drained due to the porous limestone parent

material that is usually present at 18 cm or less below the soil

surface. Krome series soils occur in sites at elevations ranging from

2.4 to 4.3 m above sea level and may be even lower in some areas.

Although normally well-drained, low lying areas of this soil type are

prone to flooding during periods of high rainfall.

In recent years, urbanization has forced tropical fruit crop

production in Florida into areas characterized by the Chekika very

gravelly loam soil series (loamy-skeletal, carbonatic, hyperthermic

Lithic Udorthent) (Anon, 1989b; Leighty and Henderson, 1958). This soil

is derived from the same limestone parent material as the Krome soil but

has less soil development. The soil reaction and physical

characteristics of this soil are similar to, but more variable than,

those of the Krome soil series. Due to low elevation and shallow depth











to the water table, Chekika soils are subject to annual flooding in most

years. The high pH of both soils makes minor element nutrition of many

crops problematic and can be a major production cost (Davenport, 1983;

Malo, 1966; Schaffer et al., 1988).

Chemical transformations that occur in flooded soils have been

documented for several soil types (Armstrong, 1975; Gambrell and

Patrick, 1978; Gotoh and Patrick, 1972, 1974; Mahapatra and Patrick,

1969; Patrick and Mahapatra, 1968; Ponnamperuma, 1972, 1984; Reddy and

Patrick, 1983; Shapiro, 1958). However, flood-induced chemical changes

have not been documented for soils primarily composed of only slightly

altered limestone. This information is necessary for the development of

crop management practices which maximize production efficiency of

perennial crops grown on flood-prone limestone soils. The objective of

this study was to determine the chemical transformations that occur in

Krome and Chekika very gravelly loam soils in response to flooding.



Materials and Methods



Soils

Krome very gravelly loam soil was obtained from a sodded orchard

site with a long history of fruit tree cultivation. Chekika very

gravelly loam soil was obtained from a virgin tract of land. Both sites

had been recently plowed, but the Chekika site had never been cropped or

amended with fertilizers. Thus, both soil samples contained large

amounts of plant residues which had been incorporated during plowing.

Plant residue incorporation resulted in C:N ratios (w/w) of 13.9% :

0.58% (24:1) and 11.0% : 0.39% (28:1) for the Chekika and Krome soil











samples, respectively. The soil samples were air-dried at ambient (22-

340C) temperatures and passed through a 1.0-mm sieve.



Soil Redox Potential and pH

Soil redox potential (Eh) and pH of flooded Krome and Chekika

soils were monitored over a 35-day period. Three 200 g samples of each

soil were incubated in the dark at 220C in 0.47-L containers with

sufficient deionized H20 to create a 3.8 cm water column above the soil.

The containers were covered with Parafilm to prevent evaporation of the

water. Redox potential was determined using a calomel reference

electrode and three platinum-tip microelectrodes, as described by Stolzy

and Letey (1964). The microelectrode platinum tip was fused to a heavy

gauge brass alloy rod and the junction sealed in an epoxy resin (Bohn,

1971; Ponnamperuma, 1972). The microelectrodes were placed in the soil

to a depth of approximately 7 cm and measurements were recorded when a

stable millivolt reading was reached, usually within 5 min. The reading

was adjusted by the addition of +245 my to compensate for the potential

of the reference electrode. Redox potential was monitored daily during

the first week of submergence, at about 4-day intervals during the 2nd

and 3rd weeks, and at about weekly intervals thereafter. Concurrent

with Eh determinations, soil pH was monitored with a pH meter.



Chemical transformation studies

For each soil type, 25-g soil samples were placed in 160-ml serum

bottles. Samples were either aerobically incubated (nonflooded) at

field capacity moisture content (4 ml water/25 g soil), or anaerobically

incubated (flooded) in 50 ml water. For the flooded treatment, after











water was added, serum bottles were sealed with rubber septae and purged

three times with nitrogen gas to ensure anaerobiosis. Serum bottles

were incubated in the dark at 220C in covered trays with 1 cm of water

in the bottom to prevent the aerobic treatments from drying out. After

0, 1, 3, 5 and 7 weeks of incubation, 12 samples of each soil type, six

from each incubation treatment, were subjected to extraction with a

neutral 1N NH40Ac solution for determining concentrations of extractable

K, Fe, Mn, Mg and Ca (Thomas, 1982). Simultaneously, 12 additional

samples of each soil type, six from each incubation treatment, were

subjected to extraction with a 2N KC1 solution for determining

concentrations of NH4 N03- and P. Phosphorus concentrations were

determined after 0, 1, 3 and 7 weeks of incubation, but not after week

5. Thus, for all elements or compounds (with the exception of P), there

was a factorial arrangement of treatments (2 soils x 2 incubation

treatments x 5 extraction dates) that was replicated six times in a

completely randomized design. For the flooded treatments, a

proportionally smaller amount of extracting solution of greater

concentration was used to compensate for the dilution effect of the

floodwater. Concentrations of extractable K, Fe, Mn, Mg and Ca were

determined by flame atomic absorption spectrometry (Baker and Suhr,

1982) (Atomic Absorption Spectrophotometer Model 2380, Perkins-Elmer

Corp., Norwalk, CT). Concentrations of extractable NH+ and N03- were

determined colorimetrically using an autoanalyzer (Technicon

Autoanalyzer II, Technicon Instruments Corp., Tarrytown, NY) and

extractable P concentration was determined colorimetrically using a

recording spectrophotometer (Shimadzu Recording Spectrophotometer Model

UV-160, Shimadzu Scientific Instruments, Inc., Norcross, GA) (Anon,











1979). Data were analyzed by analysis of variance and linear and

nonlinear regression (SAS Institute, 1985).



Results and Discussion



Soil Redox Potential and pH

Immediately after flooding, Eh of both soils was +300 my (Fig. 3-

1A). Redox potential declined sharply to a minumum potential during the

first few days of flooding, then increased to a post-flood maximum

before finally decreasing to a stable potential, a pattern typical of

flooded soils (Ponnamperuma, 1972). Final Eh for both soils was -165

my. The greater initial decrease in Eh for the Chekika than for the

Krome soil may have been due to the lower initial concentration of redox

system components in the Chekika soil; specifically, lower

concentrations of extractable N03-, Mn and Fe (Figs. 3-2, 3-3)

(Ponnamperuma, 1972). These nutrients can buffer the redox system by

serving as electron acceptors under anaerobiosis (Patrick and Mahapatra,

1968). The greater concentration of redox system components in the

Krome soil apparently poised the redox system, preventing the Eh from

becoming as negative as the less fertile Chekika soil. For both soils,

Eh stabilization by day 21 indicated a reduced respiration rate,

possibly due to depletion of readily oxidizable organic matter,

depletion of electron acceptors, and the stabilizing effect of the

mobilized iron and manganese (Rowell, 1981).

Prior to incubation, pH was 7.9 for the Chekika soil and 7.5 for

the Krome soil (Fig. 3-1B). In general, the pH of both soils gradually

decreased until a stable pH of about 7.25 was reached on day 21. Soil











pH decreased with flooding since the carbonate system predominates over

the redox system in alkaline soils (Ponnamperuma, 1972). Bacterial

respiration leads to CO2 accumulation (Ponnamperuma, 1972; Russell,

1977), and consequent H2CO3 formation (Buckman and Brady, 1968). In

alkaline soils the dissociation of H2C03 into H+ and HC03" results in

acidification of the soil solution.



Chemical Transformation Studies

Ammonium nitrogen. Prior to incubation, concentrations of KC1-

extractable NH4+ were similar for both soils (Fig. 3-2). With time,

flooded soils developed higher NH4 concentrations than nonflooded soils

throughout the experiment. Regardless of soil type, NH4 generally

increased with flooding until week 5 and then slightly decreased.

Maximum concentrations were greatest for the Krome soil. Anaerobiosis

curtails microbial oxidation of NH4+ to NO3- (Patrick and Mahapatra,

1968; Ponnamperuma, 1972; Reddy and Patrick, 1983) but mineralization of

organic N to NH4+ continues, resulting in increased NH4+ concentrations.

For the nonflooded treatments, NH4+ of the Krome soil slightly decreased

over the course of the study, but increased for the Chekika soil up to

week 5 before decreasing at week 7.

Nitrate nitrogen. Prior to incubation, concentrations of KC1-

extractable N03- were greater for the Krome soil (23.4 yg/g soil) than

for the Chekika soil (4.2 pg/g soil) (Fig. 3-2). For both soils, N03-

decreased to nearly 0 ppm after one week of flooding. Nitrate reduction

occurs rapidly in anaerobic soils (Patrick and Mahapatra, 1968; Reddy

and Patrick, 1983) due to the presence of facultative organisms that

transform N03- to N-containing gases at low oxygen tensions











(Ponnamperuma, 1972). Denitrification occurs quickly because NO3- is

the first redox constituent to disappear from the soil following 02

depletion (Reddy and Patrick, 1983). For nonflooded soils, N03- content

generally increased over the course of the experiment, presumably due to

mineralization of organic matter and subsequent nitrification.

Manganese. Prior to incubation, concentrations of NH40Ac-

extractable Mn were approximately three times greater for the Krome soil

than for the Chekika soil (Fig. 3-3). For both soils, flooding resulted

in a steady increase in extractable Mn over time. With flooding,

maximum Mn concentrations were approximately 40 and 7 times greater than

the initial Mn concentrations of the Krome and Chekika soils,

respectively. The increase in extractable Mn with anaerobic incubation

is consistent with results of other studies that indicate an increase in

soluble Mn with flooding (Ponnamperuma, 1972). After NO3-, MnO2 is the

next redox system component likely to be reduced in an anaerobic soil.

Possibly because of the relatively low NO3- levels in the experimental

soils, large amounts of Mn became more extractable in both soils during

the first 7 days of flooding. Biological reduction of Mn4+ in MnO2 and

an increase in the concentration of water-soluble Mn2+ are two of the

principal transformations of Mn in flooded soils. For both nonflooded

soils, there was a gradual decrease in extractable Mn over the course of

the study.

Iron. Prior to incubation, concentrations of NH40Ac-extractable

Fe were similar for both soils (Fig. 3-3). For both soils, there was a

large increase in extractable Fe after one week of flooding, and maximum

Fe concentrations (week 5) were approximately 30 and 15 times greater

than the initial concentrations for the Krome and Chekika soils,











respectively. For both soils, aerobic incubation resulted in linear

increases in extractable Fe over time, but such increases were less than

with anaerobic incubation. Under anaerobiosis, the Fe3+ Fe2+ redox

couple is the next constituent of the redox system likely to be reduced

after Mn (Ponnamperuma, 1972). However, reduction of redox system

components does not always occur sequentially; often one redox component

is not completely reduced before the next most easily reduced component

begins to be reduced (Patrick and Mahapatra, 1968). The reduction of Fe

and its increased solubility is often one of the most important chemical

changes to occur in flooded soils (Ponnamperuma, 1972), and the increase

in extractable Fe observed in the present study is consistent with this

observation.

Magnesium, potassium, calcium and phosphorus. Prior to

incubation, concentrations of NH40Ac-extractable Mg were greater for the

Krome soil than the Chekika soil (50 and 30 pg/g soil, respectively

(Fig. 3-4). The concentration of Mg increased linearly with time in

both soil types, but the rates of increase were greater for the flooded

treatments.

Prior to incubation, concentrations of NH4OAc-extractable K were

greater for the Krome than for the Chekika soil (50 and 35 pg/g soil,

respectively (Fig. 3-4). In general, for both soils, K increased over

time regardless of incubation treatment.

Although there was no direct evidence for the presence of CaCO3 in

the experimental soils, levels of CaCO3 are presumably high since these

soils are derived from limestone (CaCO3) parent material. At day 0, the

concentration of NH40Ac-extractable Ca was greater for the Krome soil

(4.1 mg/g) than for the Chekika soil (3.2 mg/g) (Fig. 3-5). For both











soils, extractable Ca of flooded and nonflooded treatments generally

increased over the course of the experiment. However, the rate of

increase was greater for the nonflooded treatments. For both soils, a

test for homogeneity of the slopes of the regression lines for Ca

concentration in Fig. 3-5 showed that while the y intercept was

identical for flooded and nonflooded treatments, there was a significant

difference in the slopes of the lines (P < 0.01).

Prior to incubation, KCl-extractable P concentrations were more

than three times greater for the Krome soil (1.95 pg/g soil) than for

the Chekika soil (0.64 pg/g soil) (Fig. 3-5). For both soils, flooding

and nonflooding resulted in decreases in P, but the rate of decline was

more rapid for the anaerobic treatments.

With flooding, increased mobilization of NH4 Fe2+ and Mn2+

results in an increase in the ionic strength of the soil solution (Reddy

and Patrick, 1983). Mobilization of relatively large amounts of these

reduced cations can displace Mg2+, Ca2+ and K+ from the exchange

complex, resulting in increased levels of these cations in the soil

solution. Such enhancement of elemental solubility with flooding is

consistent with the changes in elemental extractability observed in the

present study. For example, although extractable Mg increased in both

soils regardless of incubation treatment, increases were greatest when

soils were flooded. Although the presence of MgCO3 was not determined

for soils of the present study, in flooded alkaline soils dissolution of

MgCO3 and the increased concentration of reduced cations in the soil

solution can lead to displacement of Mg2+ from the exchange complex,

thereby increasing its concentration in the soil solution.











Similarly, although there was no significant difference between

flooded and nonflooded treatments in extractable K, the trend toward

slightly higher extractable K concentrations in the flooded soils may be

due to displacement of K+ from the exchange complex by the increased

concentration of reduced cations in the flooded soil solution (Reddy and

Patrick, 1983).

Although concentrations of extractable Ca at week 7 were greatest

for the nonflooded treatments, there was a greater amount of extractable

Ca in the flooded treatments during the first 2 weeks of the experiment

(Fig 3-5). Displacement of Ca2+ from the exchange sites may account in

part for the reduction in extractable P in the anaerobic treatments at

that time (Fig. 3-5), because Ca2+ and P can react to form insoluble

calcium phosphate compounds. For most flooded soils, soluble P tends to

increase under flooded conditions (Patrick and Mahapatra, 1968; Shapiro,

1958), mainly due to the reduction of iron phosphate compounds (Reddy

and Patrick, 1983; Patrick and Mahapatra, 1968). In alkaline soils,

however, P does not usually increase because soluble P is controlled by

the calcium system (Reddy and Patrick, 1983). The decrease in

extractable P in nonflooded treatments prior to week 5 may be due to

microbial P metabolism, or the dissolution and subsequent precipitation

of Ca and P that occurred during collection, air-drying and subsequent

incubation of the samples at field capacity moisture content.



Conclusions



Despite relatively low native fertility, significant chemical

transformations occur in Krome and Chekika very gravelly loam soils











subjected to flooding. The increases in extractable Mn, Fe and Mg in

both flooded soils is consistent with increased elemental solubility

reported in other flooding studies. This increased solubility could

exacerbate leaching losses or result in nutrient toxicities, although in

the present study concentrations of extractable elements do not appear

to reach phytotoxic levels. Additional studies are needed to determine

if flooding can increase the assimilation of elements such as Fe and Mn

without harmful effects to the crop. In some mango growing areas, such

as South Florida, the ability exists to regulate flooding duration and

floodwater depth in flood-prone agricultural areas. Therefore, annual

flooding cycles should be evaluated for their potential to alleviate

minor element deficiencies in crops grown on these limestone soils.































Sa)


0-I r ,
I 1 I 41
*r 1 .4 J
M "n 0 :



r. 1 .




S0 "0 l

0 Cd pA

o a0 0H 0
v4 *rI V (
Sr I 0 % a) w
H r-I & 3
0 -I 0 4) 0a
00a *rI
C 0

0 4) to .

0 >r 0
o a1 w

*ri r-4




4O4 r. .0 0
ni --Ia) ( 0

















0 6 0i0 4o
60 a) o



4 *rv-i 4 4
Io OWp














cv.)
S 0^ g <-



Ylri P
+rl-
































13








00
/

^-a~~-a^^-'
?o




i i ~ ~ i


- I r T


0 0 0 0 0 0 0
0 00 0 0
I I I
(Aw) i3


m




5 o
0
0 0
I I

O 0


\ <
-I O


3/ /

C 00
000
,.


YLu a I


S N 0
r; (< r<


0



t
C
*-
0
c-
0
~O
-- ci,


0


i:


D/O/
j


L-YY





































OH
0 -I
)V
ao.


+ Md



>) U
cd






Wr4 4)

P. P
)0 4)w
xa 0
I10




0 r-
0a)
00



00)~

0 r oH

4a1
do o






L)U)
0a r-I


'0 A

;o
OH

I P4 0



ri4











pg NO3-/g soil


/g NH4+/g soil


I!os 6/_CON 6


0 0
14, c~q


0
0 0 0 0 0 0 0 0 0
C0 0 '- N, 0 -- CO (0 '- N4 0


l!os 6/+*HN 6q




































rz44
~I-I
14


93 0

P4J 1W-4
s-I







A 60
IVh 1 r4
I a



. 0 0




0o0
14



L140
0 0) 14
100
r.0 0H
0 r-I I






v 0



14~
4-h
%~V V
a c







o 'V
010
1V0
0 -I
0 O-H
(-4 0
1 44 CO


DW-














.g Mn / g soil

0
,- 00 to (0


\4


Go
0




S0\



0 0
II
NC








0
0
II



CO
00
C S


N
X
00


r..
o+ X

0 0



o o



N 0


ug Fe / g soil


\ -a


414c1\


0 0 0 0
0C (0 4 N
aoto4N


0 0
N
-4 i











\


.. d
0







+ !





a 0
0 0
3



0 -
Co

0--

\ 0


a/ a



/ Q


o I


x x
.o 4 o N

0 + 0 +
C I I f N

\ <


0 0 0 0
t* t) CN -


I!os 6 / aj 6Wr


Tco










S E
i F.


I!os 6 / u6 67/


































I



.~a)
0
-r4
to..






43 c



I~4~
00


o a







cow
b4.

goc


ow
ow
10


0 r-4

0 0
r4




u 4)
ro
0






Cr4












Ag Mg / g soil
0 0 0
N -t t-- C
I I I


4


usa


o







I
\\
\i
\ \P"
i,


,ug K / g soil


0 0 (D 10 0 0
S0 N '- ,- O It 0
4---- -.-I--I' -i--------
i .,\-.o




I 0!
CI
Sn







I I
+ 0




S0e
0 0

0 I

0'K


m IfR


II 0


o




I

0
CO


U)
0
10


CO4




hi


6 / 6S 67i


I!OS 6 / >1 6Br


0
d
11
I






0
+







0C
ui





o


I I I 0 i 0 I ; It
o 0 0 0 0 0 0 0 0 0 0 <
0 C N O 0 N N O) -


~"




a~ ob


clap







"4








































0




0
U





0

1.0


cor-4
.I14.44
MO



0



0 0






0
r-I

0-3




0 P-I
-r4



4O
ox
V






0 4

r1-4





-'4
Wzc













mg Co / g soil


CO (0
- --o
------1-


4t N
I I


N 0






N
o 0
o (
II 0
( n
CI.

a .
+


.-:

;o "
o 0
0 0

I ,

o <


\\
I
Iw\
\\


q
ON


u g P/g soil
N 9o
1- O


n

X

0
d
-o





-o
0
c
0

1
0
-o


.4 0
6 6


-to










i





i
4 -


/ /
/ t-


o

o
0d





0 11




o o
0-

I '0





I /

0 0
|


N 00 (0
Cj T^


CN 0 .4-
-- d d


i!os 6 / DO 6w1


\
_\
I \00
44 la


o
+ 'O
0


a


" o
(O

(0?




Sc 0
o +
r0

C |


I I
O

o










0

0

E


I


- i i i i -


C 0 CO ,o


-- --


r.


l!os 6/d 6r

















CHAPTER 4
FLOODING, MINERAL NUTRITION AND NET GAS EXCHANGE OF MANGO TREES



Introduction



In many woody plants, waterlogging results in reduced nutrient

uptake and decreased leaf nutrient content due to increased root

mortality and reductions in root metabolism, transpiration and hydraulic

conductivity (Kozlowski and Pallardy, 1984; Labanauskas et al., 1971;

Schaffer et al., 1991). In some cases, however, flooding can increase

foliar nutrient concentrations (Hook, et al., 1983; Labanauskas et al.,

1968; Labanauskas et al., 1966 ; Olien, 1989; Slowik et al., 1979).

This may be due, in part, to flood-induced reduction and solubilization

of relatively insoluble soil compounds that are then more available to

plants (Kozlowski and Pallardy, 1984; Ponnamperuma, 1972).

Mango (Mangifera indica L.) production in the United States occurs

almost exclusively on calcareous flood-prone soils, and deficiencies of

minor elements, particularly Fe and Mn, are common in these soils

(Davenport, 1983; Malo, 1965; Schaffer et al., 1988). Although flooding

decreased leaf gas exchange and vegetative growth, I have observed

slight increases in net CO2 assimilation of mango after short durations

of flooding (K.D. Larson, unpublished data). This may be due to

chemical changes that occur in flooded soil, since Fe and Mn are 10- to

50-fold more available when these calcareous soils are flooded. The

purpose of this study was to determine the effect of short-term flooding











on leaf nutrient concentration of mango trees grown in calcareous

limestone soil, and to determine the relationship between leaf nutrition

and gas exchange following short-term flooding.



Materials and Methods



Plant Material and Soil

'Peach' mango (Mangifera indica L.) trees were propagated from

seed and transplanted into a sand media in 3.8-liter containers. Trees

were trained to a single leader and fertilized monthly with 5 g of

granular fertilizer, and 1.8 g of chelated Fe (6% Fe) (Sequestrene 138,

Ciba Geigy Corp., Greensboro, NC 27419) applied as a soil drench. The

granular fertilizer contained: 1.0% nitrate N; 4.6% ammoniacal N; 1.682

water-soluble organic N; 0.72% water-insoluble N; 1.32% available P;

7.47% water-soluble K; 3.0% Mg; 0.50% Fe; 0.07% Mn; 0.07% Zn; 0.03% B;

and 0.03% Cu.

In June 1989, 60 uniform trees (mean height 0.5 m) were

transplanted into Krome very gravelly loam soil (loamy-skeletal,

carbonatic, hyperthermic Lithic Rendoll) in 11.3-liter containers and

grown outdoors. After transplanting, trees received 10 g of the

granular fertilizer described above at monthly intervals. Trees were

divided into 2 groups of 30 trees each immediately after transplanting.

One group received chelated Fe (+Fe) applied as a soil drench at the

rate of 5.0 g of chelate/tree/month for 7 months, while the other group

was given no chelated Fe (-Fe). After 7 months, visual symptoms of Fe

deficiency were observed in -Fe trees, but not in +Fe trees.











For all trees, stem diameter was determined 15 cm above the soil

line prior to flooding and 91 and 182 days after the imposition of

flooding treatments. Stem radial increase was calculated for each 3-

month period.



Flooding Treatments

On 29 March, 1990 all trees in each fertilizer treatment were

randomly exposed to one of three flooding treatments: 1) roots flooded

for 10 days (F10); 2) roots flooded for 20 days (F20); and 3) nonflooded

(NF control) plants. The design was a 2 X 3 factorial (2 Fe

fertilization rates X 3 flooding treatments) with 10 single-tree

replicates per treatment in a split plot design. Iron treatment was the

main plot and flooding treatment was the subplot. Flooding was

accomplished by submerging trees in plastic tubs in tapwater and

maintaining water levels ca. 10 cm above the soil surface. Nonflooded

trees were irrigated 2-3 times each week to maintain soil moisture near

container capacity. Ambient temperatures during the flooding period

ranged from 12 to 300 C, and flooded soil temperatures averaged 230 C.

Soil redox potential (Eh) and pH were monitored periodically for

six flooded trees from each Fe fertilization treatment, three trees from

each of the F10 and F20 treatments. Soil redox potential was monitored

at a soil depth of 15 cm using a Ag+/AgC1 reference electrode (Model

RC5, Jensen Instruments, Tacoma, Washington), an oxygen meter (Model

P5E, Jensen Instruments), and four platinum-tipped microelectrodes, as

described by Crane and Davies (1988). Soil pH was monitored with a pH

meter (Model 5995-30, Cole Farmer Instruments, Inc., Chicago, IL).











Leaf Mineral Nutrition and Chlorophyll Content

Three days prior to flooding and 82 days after flooding treatments

were imposed, total leaf chlorophyll content (Chl) was determined from

eight 0.32-cm2 leaf discs taken from 3 leaves at the mid-section of the

most recently matured vegetative growth flush on all trees as described

by Marini and Marini (1983). Two days after each chlorophyll

determination, the same leaves were harvested and prepared for nutrient

analysis as described by Schaffer et al. (1988). Foliar concentrations

of K, Ca, Mg, Mn, Fe and Cu were determined by atomic absorption

spectroscopy. Nitrogen was measured with a Kjeldahl apparatus, and N

and P concentrations were determined colorimetrically with an

autoanalyser (Technicon II, Tarrytown, NY).

For all trees, fertilizer applications were terminated for the

duration of the experiment once flooding treatments were imposed. After

flooding, all trees were watered as described for nonflooded trees.



Gas Exchange

Twelve days prior to flooding, one week after flooding was

imposed, and at about monthly intervals for 6 months thereafter, net CO2

assimilation (A) of single leaves was determined for all trees with a

portable infrared gas analyzer (Analytical Development Co, Hoddeson,

Herts., U.K.) as described by Schaffer and O'Hair (1987). All

measurements were made between 1100 and 1400 hrs (EST) when

photosynthetic photon flux (PPF) exceeded 500 ymol m-2 -1, which is

above light saturation for net CO2 assimilation of mango (Schaffer and

Gaye, 1989). Air flow into the leaf chamber was 375 ml min- C02

concentration in the chamber was 349 7 ymol mol-i, air temperature was











27.6 5.3 C, and mean vapor pressure was 1.2 kPa (ranging from 0.9 to

2.1 kPa). Gas exchange calculations were based on those described by

Jarvis (1971) and von Caemmerer and Farquhar (1981). Leaf gas exchange

varied little among similar aged leaves of individual mango trees (data

not shown). Therefore, one fully mature, sun-exposed leaf on the most

recently matured growth flush of each tree was used for gas exchange

determinations.

Data were analyzed by ANOVA and standard T-test.



Results



Soil Chemistry and Plant Growth

There was no effect of Fe fertilization on soil Eh or pH (P >

0.05). Therefore, Fe treatments were pooled for comparisons of Eh and

pH among treatments. Anaerobic soil conditions (< 200 mv)

(Ponnamperuma, 1972) developed one day after flooding (Fig. 4-1).

Thereafter, Eh decreased little until day 10 and was about -100 my by

day 14. Prior to flooding, soil pH was about 7.4 (Fig. 4-1), and

decreased to about 7.1 after 20 days of flooding.

There were no differences among treatments in stem radial growth

over the course of the experiment (data not shown).



Leaf Chlorophyll Content and Mineral Nutrition

Total leaf chlorophyll content decreased over time for both Fe

treatments, regardless of flooding treatments (Table 4-1). However,

this decrease was not significant for flooded, -Fe plants. For both Fe











treatments, the decrease was greatest for the nonflooded plants, and

least for plants that were flooded for 20 days.

Prior to flooding, mean Chl, and foliar Mn and Fe concentrations

were about 1.3 1.5-fold greater for the +Fe trees than for the -Fe

trees (Tables 4-1 and 4-2), whereas foliar K, Ca and Mg concentrations

were 1.5 1.7-fold greater for -Fe trees than for +Fe trees (Tables 4-3

and 4-4). Prior to flooding, there were no differences in foliar N, P

or Cu concentrations between -Fe and +Fe trees (Tables 4-3 and 4-4).

There was no difference in foliar Zn concentration among Fe or flooding

treatments at either sampling date (pooled mean for all treatments -

15.8 mg/kg; data not shown).

After imposition of flooding there were no significant

interactions among Fe fertilization and flooding treatment (P > 0.05)

with foliar concentrations of N, P, Mg, or Fe. Therefore, only main

effects due to Fe fertilization and flooding are presented for these

nutrients (Tables 4-2 and 4-3, respectively). Significant interactions

among Fe fertilization and flooding treatment with regard to Chl and

foliar Mn, K, Ca, and Cu concentrations (P < 0.05) are reported

separately within each Fe fertilization treatment.

There was a significant decrease in foliar Mn concentration

between sampling dates in the nonflooded, +Fe trees (Table 4-1). For

+Fe trees, foliar Mn concentration was unaffected by flooding, although

foliar Mn concentration tended to be higher with increase flooding

duration. In contrast, in the -Fe trees, there were significant

increases in foliar Mn concentration for the F10 and F20 treatments, but

foliar Mn concentration was unaffected in the nonflooded treatment. The











greatest increase in foliar Mn concentration occurred with the longest

flooding duration.

Foliar Fe concentration increased between sampling dates

regardless of Fe treatment (Table 4-2) or flooding (Table 4-3). Due to

the large variation in foliar Fe concentration of nonflooded plants at

the second sampling date, the increase in Fe concentration was not

significant for this treatment (Table 4-3). Almost three months after

flooding treatments were imposed, there was no difference in foliar Fe

concentration between +Fe and -Fe trees.

For both Fe treatments, flooding resulted in a significant

increase in foliar Mg. Magnesium concentration increased for all

flooding treatments, but the increase was greatest in trees exposed to

20 days of flooding (Table

4-2). Regardless of flooding treatment, foliar K and Ca concentrations

increased over time in the +Fe trees, but were stable (nonflooded and

F10 treatments) or decreased (F20 treatment) in the -Fe trees.

Although foliar N concentration decreased in the -Fe trees between

sampling dates (Table 4-3), there was no significant effect of flooding

on N (Table 4-2). For both Fe treatments there was a significant

increase in foliar P over time (Table 4-2) regardless of flooding

treatment (Table 4-3). For the +Fe trees, foliar Cu concentration

increased irrespective of flooding treatment, but increases tended to be

greater with increased flooding duration (Table 4-4).



Gas Exchange

Prior to flooding, mean net CO2 assimilation of the +Fe trees was

over 1.3 times greater than that of the -Fe trees (Figure 4-2).











Flooding resulted in rapid decreases in net CO2 assimilation, regardless

of Fe treatment, and recovery of net CO2 assimilation after flooding was

slow. Within each Fe treatment, there was no difference in net CO2

assimilation between F10 and F20 treatments. For the -Fe trees, 184

days after flooding was imposed, F20 treatment net C02 assimilation was

significantly greater than that of the control treatment. At the same

time, for the -Fe trees, net CO2 assimilation of the F10 and F20

treatments were similar to that of the +Fe trees.



Discussion



The slow decrease in Eh of flooded soil may have been due to

relatively cool ambient and soil temperatures, since at temperatures

below 250 C Eh decreases at a slower rate than at higher temperatures

(Ponnamperuma, 1972). The rate of decrease in flooded soil pH was

slower than that observed in previous flooding studies with Krome very

gravelly loam soil and may also have been affected by temperature

(Ponnamperuma, 1972).

Mango tree stem radial growth has been correlated with canopy

volume (unpublished data). Thus, the lack of difference among

treatments in stem radial growth indicate no difference in canopy

volume. Therefore, it is unlikely that differences in foliar nutrient

concentrations among treatments were due to dilution effects resulting

from vegetative growth differences.

To our knowledge, the critical foliar nutrient concentration

ranges for optimal growth and productivity of mango trees have not been

determined. However, Young and Koo (1971) working in Florida, and Gazit











(1969) working in Israel, reported foliar nutrient concentrations for

healthy trees grown in calcareous soils. In our study, foliar N and P

concentrations were similar to, but foliar Ca concentration was lower

than those reported by Gazit (1969) and Young and Koo (1971). For the

+Fe trees, foliar Fe concentration was similar to concentrations

previously reported for healthy mango trees (Gazit, 1969; Young and Koo,

1971). As in our study, Gazit (1969) found much higher foliar K

concentrations in iron deficient trees than in trees that were not Fe

deficient. Foliar Zn concentration (pooled mean for all treatments -

15.8 mg/kg) was only about 10 20Z of the mean foliar Zn concentration

in healthy trees in Florida and Israel (Gazit, 1969; Young and Koo,

1971). In our study, although visual symptoms of Zn deficiency were not

observed, the trees received minimal Zn fertilization and may have

suffered from incipient Zn deficiency. Foliar Mg concentration of the

-Fe trees was similar to that reported for healthy mango trees grown in

limestone soil in Florida (Young and Koo, 1971).

Similar to our observations, seasonal variations in mango Chl have

been reported previously (Schaffer and Gaye, 1989). Flooding stress

often results in reduced Chl (Trought and Drew, 1980; Wallihan et al.,

1961). Thus, the lack of a decrease in Chl between sampling dates for

flooded, -Fe trees may be due to the flood-induced increases in foliar

Mn and Fe concentrations observed for this treatment.

Increases in foliar Fe concentration have been observed for other

woody plants exposed to flooded soil conditions (Hook, et al., 1983;

Labanauskas et al., 1966, 1968; Olien, 1989; Slowik et al., 1979). This

was presumably due to the flood-induced reduction of insoluble Fe

compounds, making them more available to the plant (Ponnamperuma, 1972;











Kozlowski and Pallardy, 1984). Apparently, Fe deficiency resulted in a

stronger Fe sink and greater Fe uptake in the Fe-deficient trees than in

the nondeficient trees (Barber, 1979; Nye and Tinker, 1977; Pitman,

1965). The positive correlation between foliar concentrations of Fe and

Mn and leaf chlorophyll content observed for mango has also been

observed for many other species (Homann, 1967; Jacobsen and Oertli,

1956; Machold and Scholz, 1969; Spiller and Terry, 1980; Stocking, 1975;

Terry, 1980).

Although Mg is an essential component of the chlorophyll molecule,

Chl was not correlated with leaf Mg concentration (Tables 4-1, 4-2),

perhaps due to the fact that only 15 to 20% of total plant Mg is

associated with chlorophyll (Neales, 1956). Magnesium deficiencies are

common in South Florida limestone soils due to the high pH and Ca

saturation of the exchange complex. However, in flooded alkaline soils,

dissolution of MgCO3 and the increased concentration of reduced cations

(Fe2+, Mn2+) in the soil solution can lead to displacement of Mg2+ from

the exchange complex, thereby increasing its availabity (Reddy and

Patrick, 1983).

The greater preflood foliar concentrations of K, Ca and Mg for the

-Fe trees may have resulted from cation-anion balance effects (Kirkby,

1968). The -Fe trees had lower preflood foliar concentrations of Fe and

Mn, and therefore may have required greater foliar concentrations of K,

Ca and Mg to balance the negative charge of foliar anions. The

significant decreases in foliar K and Ca concentrations for the -Fe

trees exposed to 20 days of flooding may have also been due to ionic

balance effects, since 20 days of flooding resulted in large increases

in Fe, Mn and Mg for these trees.











Soil flooding frequently results in rapid decreases in NO3

concentration in the soil solution (Gambrell and Patrick, 1978;

Ponnamperuma, 1984), due to leaching and denitrification. Decreases in

soil N03-, and the inhibition of ion uptake and transport by roots under

anaerobic soil conditions (Trought and Drew, 1980) often result in

flood-induced decreases in foliar N concentrations (Hook et al., 1983;

Stolzy et al., 1975; Trought and Drew, 1980). With the flooded

treatments, foliar N concentrations were stable, possibly due to the

fact that only a small percentage of the N applied to the experimental

trees was N03--N, and flood-induced N losses from the soil were

therefore minimized.

Foliar P concentration often decreases with soil flooding (Herath

and Eaton, 1967; Hook et al., 1983; Kozlowski and Pallardy, 1984; Slowik

et al., 1979; Stolzy et al., 1975). However, in alkaline soils where

native P is not very soluble, flooding can increase P availability due

to reduction of insoluble P compounds (Kozlowski and Pallardy, 1984).

Thus, the lack of a decrease in foliar P for the flooded treatments may

be related to increased P availability under flooded conditions.

Additionally, to maintain an ionic balance, the observed increase in

concentrations of foliar cations, particularly Mg and Fe, in all

flooding treatments may have resulted in increased foliar concentration

of anionic species such as PO4- (Kirkby, 1968).

With flooding, the decrease in soil pH, the reduction of Fe and Mn

compounds in the soil and subsequent displacement of other cations from

the exchange sites, and the production of organic completing compounds

can result in increased solubility of Cu in the soil solution (Kozlowski

and Pallardy, 1984). For the +Fe, nonflooded trees, increased foliar Cu











concentration may be due to ionic balance effects (Kirkby, 1968), since

the concentration of foliar Mn (another cationic species) decreased for

this treatment. For the -Fe trees, the decrease, or lack of significant

increases in foliar Cu may also be related to cationic-anionic balance

effects, since -Fe trees tended to have large increases in concentration

of certain foliar cations.

Prior to flooding, the greater net CO2 assimilation for +Fe trees

than for the -Fe trees was probably due to higher Chl and foliar Fe and

Mn concentrations. Iron deficiency results in reductions in Chl,

photosynthetic electron transport, chloroplast protein content, RuBP

carboxylase/oxygenase activity, and reductions in the number of

chloroplast grana and stromal lamallae (Shetty and Miller, 1966; Spiller

and Terry, 1980; Stocking, 1975; Terry, 1980; Vesk et al., 1966).

Manganese deficiency results in fewer chloroplasts per cell and

chloroplasts with low chlorophyll contents (Homann, 1967). Prior to

flooding, greater Chl and foliar Fe and Mn concentrations contributed to

greater net CO2 assimilation in the +Fe trees. The increase in leaf gas

exchange for iron-deficient trees exposed to 20 days of flooding may be

related to the flood-induced increases in foliar micronutrient

concentrations, particularly Mn, and lack of a decrease in Chl.



Conclusions



Although previous studies have shown that flooding results in

transitory reductions in net gas exchange and vegetative growth of

containerized mango trees (K.D. Larson, unpublished data), the data from

this study indicate that short-term flooding of mango trees grown in







48


limestone soils results in significant increases in the concentration of

some foliar nutrients and can also result in increased net CO2

assimilation after floodwaters subside. Additional studies should be

conducted to determine the potential of using short-term flooding as a

management tool for reducing micronutrient deficiencies in mango trees

grown in limestone soils.
































4) P4
co a

e sPA 0
14 U

60 M








M 0r)
o %0




0 0
44 rl M
0 VO
41o






w 0

-4
4









0 44
O 0









4 W a


013 0 to
0.0
40 0




1







o0
-H
C4










Cu4


0

0




0
*a

44
lo 0

10 0
0 )
ao o




.-I V 0
M E

03 44


44
44 0 0)
0 -4







0 4 Trl
14 c3
4 C



A 00
AU

n V>
*44rl






V o




S *


K n 9!
u J a









pH


I -c
CL L.J

* w
em

em


C, U



U


O
(0
I-0
t- C/
0




C
O

0
L


- CN


w -- I I I I 10
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
LO 0 OnO
I I
(Aw) t3


I


I m m































0




o o
0 E

o -
L ) 0
a I
op.

0 0
a 4

S0




-r

0 +


-HV
0C 0



0o 0
as -H a
1
4S





0 0
0 0ra

















-04
60





V'40
00
0 (

4










00
9) 9


e4
4-1
c4)

sL
0)
00












4 C
0)






60
s a
4-4










O
0 r4




















CE
14 0)


















14
4.40

S0





4-1
or *
(0-







c
0
S-)



(0 *r E-



> +1 0












A (pmol C02 m-2 s-1)

r- t0 O ro N -


(r- S oZ O to- IoCu) N


(l-s 7-u 'Zoo loujrQ v


0
4-
t- ?
O


C
0
4-





N 0
a) Q.
E



r-




In


.- -0
0
0


S0
C
0

NO

E
L.
O



0
Q,















Table 4-1. Effect of flooding on total leaf chlorophyll
content (Chl) and foliar Mn concentration of 'Peach'
mango trees grown with (+Fe)or without (-Fe)
chelated Fe.


Chl Mn
(pg cm-2) (pg kg-)

Flood Pre Post Pre Post
trtz trty trtx trtw trtv

+Fe


NF 20.43 14.73 ** 65.10 15.20 c **
F10 19.11 13.75 ** 41.50 48.20 b NS
F20 18.58 14.07 ** 56.50 75.90 a NS

-Fe


NF 12.37 8.77 b ** 39.80 32.70 b NS
F10 12.13 10.22 ab NS 41.10 82.10 a *
F20 14.25 12.91 a NS 45.96 114.30 a **


Mean separation within rows by standard T-tests, n 10
trees; *, ** indicate significant differences (P < 0.05
or 0.01, respectively) between pretreatment and
posttreatment concentrations, NS indicates no
significance.

Mean separation within columns for each Fe treatment by
Duncan's Multiple Range Test (P < 0.05), n = 10 trees.
Absence of letters indicates no significance.

z NF nonflooded; F10 flooded for 10 days; F20 = flooded
for 20 days.

y Three days prior to imposition of flooding treatments.

x 82 days after imposition of flooding treatments.


One day prior to flooding.


v 84 days after imposition of flooding.






























4.)
O
01
0
04
101fr-t
PN
0

C4


0) 41
P 14
P1 4-1


0
ca


w
0 4

o: 0

r- 0



$04
U o
-) 0
01 4




01


O4 0
04-)
C *rI







001
0 0

a o

4-





Co


R0 0041




p o Q)

( CO




a. *) 4
i o


001.4J


Tl l
41CO )


0 4)




14 41





V 0
0) 0



uo
04

0)



-0 P
U H

0*)
4-O

0
4-1 g
HI 0

40 ri



ad V4-



O -i 0V
CO
0 -i-
) c








S00
0 (0

0 T' U
*a





4r a



0 0






Sd


O1
4) 0
r.00
44 0



0) 0
4-1 4J



4-4
o B





0E O
-ItS


4.1
(0 41
0 p
0 .41



041
044.


14 14



N
4i
0)14
r 4J-


-4
r1

























4' 4' *1
K K -4


Wo W Ec


4' 4
.4 *' 4


Co .4
Z -4'


In 0

(o rr



o r-
.-4


0 0U





0) oI
U' 44
0 f
rE -4


0
04






0


o
\0

o
hN
r4


41


0 4






4) 4




45
0 -
t w

04 P
I4)4
Q.4


'-,





1e4
40
14 -
UO






0 0
90




- V

r.4
o o

00 .


O C
S0

0 0 m

H0 a






S0.4
* 4
C 0l

a 4




(o 4

M ^1)















04' 0
4-i a)

Trl 4 4>
> at a












(B *k .0 C
,
S


in *
Go
o o

VH








0








o
gr4
4H .1

C


0













0
0


0
Wa




Co
Ie











a


C


04J






M c





ON
Q. 4J







04.



14



P 4J





W 41


-4

r3
:4-















Table 4-4. Effect of flooding on foliar K, Ca and Cu concentrations of
2-year-old 'Peach' mango trees grown with (+Fe) or without (-Fe)
chelated Fe.


K Ca Cu
(%) (%) (ig kg-1)

Flood Pre Post Pre Post Pre Post
trtz trty trtx trt trt trt trt

+Fe


NF 0.89 1.11 ** 0.738 1.135 a ** 4.20 12.30 *
F10 0.91 1.11 ** 0.764 0.985 b ** 3.70 17.90 *
F20 0.91 1.08 ** 0.643 0.992 b ** 3.90 19.10 **

-Fe


NF 1.60 1.45 a NS 1.164 1.072 a NS 4.00 7.10 b NS
F10 1.36 1.25 b NS 1.122 0.793 b NS 4.30 2.40 a **
F20 1.58 1.24 b 1.281 0.807 b ** 4.50 6.80 a NS


Mean separation within
indicate significant
between pretreatment


rows by standard T-tests, n 10 trees; and **
differences (P < 0.05 or 0.01, respectively)
and posttreatment concentrations within Fe


treatments, NS indicates nonsignificance.


Mean separation within columns within Fe treatments by Duncan's
Multiple Range Test (P < 0.05), n 10 trees. Absence of letters
indicates no significance difference.

y NF nonflooded; F10 = flooded for 10 days; F20 = flooded for 20 days.

x One day prior to flooding.


w 84 days after imposition of flooding treatments.
















CHAPTER 5
FLOODING, LEAF GAS EXCHANGE AND GROWTH OF MANGO TREES IN CONTAINERS



Introduction



Among the most rapid physiological responses of fruit trees to

flooding are a reduction in stomatal conductance (gs) and net CO2

assimilation (A), (Andersen et al., 1984a, 1984b; Crane and Davies,

1989; Davies and Flore, 1986c; Schaffer and Ploetz, 1989; Smith and

Ager, 1988; Syvertsen et al., 1983). Woody plants often exhibit a

decrease in A and gs within 1 to 3 days after flooding, although longer

flooding durations are required for reductions in growth (Andersen et

al., 1984a; Crane and Davies, 1989).

Tolerance of mango trees to flooding is not well known. Some

reports indicate that mangos require good soil drainage for adequate

growth and yield (Alfonsi, 1980; Samson, 1986), whereas other reports

indicate that they are flood tolerant (Chandler, 1958; Jawanda, 1961;

Young and Sauls, 1981). Therefore, experiments were initiated to

determine the physiological and growth responses of mango trees to

flooding using leaf gas exchange, leaf water potential, vegetative

growth and tree survival as stress indicators.











Materials and Methods



Flooding, Net Gas Exchange and Vegetative Growth (Experiment I)

Thirty, 4-year-old 'Tommy Atkins' mango trees, about 2.25 m in

height (15 on 'Turpentine' rootstock and 15 on an unknown seedling

rootstock) were grown outdoors in 57-liter containers with Krome very

gravelly loam soil (loamy-skeletal, carbonatic, hyperthermic Lithic

Rendoll). Trees were exposed to three treatments in October, 1989; 1)

nonflooded (control), 2) flooded for 14 days (14DF), or 3) flooded for

28 days (28DF). Five single tree replicates of each rootstock were used

for each experimental treatment. Thus, there was a 2 x 3 factorial

arrangement of treatments (2 rootstocks, 3 flooding treatments) that was

replicated five times in a split plot design, with rootstock as the

main-plot and flooding treatment as the subplot. Plants were flooded by

submerging the containers in plastic-lined metal tubs filled with tap

water. Water levels in the tubs were maintained about 10 cm above the

soil surface. Control trees were watered about twice a week to maintain

soil moisture near container capacity.

Diurnal air temperatures fluctuated between 22 and 34 C during the

experiment. Initially, soil temperatures were higher for the nonflooded

than the flooded treatments, but after 3 days mean soil temperatures at

5 cm depth were about 34 C for all treatments.

For flooded trees, soil redox potential (Eh) was monitored

periodically at a soil depth of 15 cm using a silver/silver chloride

reference electrode (Model RC5, Jensen Instruments, Tacoma, Washington),

an oxygen meter (Model P5E, Jensen Instruments) and four platinum-tipped

microelectrodes, as described by Crane and Davies (1988). The platinum











tips of the microelectrodes were fused to a 3 mm brass alloy rod and the

junction sealed in epoxy resin (Bohn, 1971; Mann and Stolzy, 1971).

Prior to use, accuracy of each microelectrode was checked by measuring

the electrical potential of pH-buffered quinhydrone solutions (Bohn,

1971).

For all trees, A, gs, transpiration (E), internal CO2

concentration (Ci) and leaf temperatures were determined with a portable

infrared gas analyzer (Analytical Development Co, Hoddeson, Herts.,

U.K.), as previously described (Schaffer and O'Hair, 1987). All gas

exchange determinations were made at photosynthetic photon flux > 500

pmol m-2 s-1, which is above light saturation for mango (Schaffer and

Gaye, 1989). Air flow into the leaf chamber was 375 ml min-I, C02

concentration was 346 + 9 ymol mol-1; air temperature was 29.8 + 5.5 C;

and mean vapor pressure was 1.1 kPa (ranging from 0.9 to 2.1 kPa). Gas

exchange calculations were based on those described by Jarvis (1971) and

von Caemmerer and Farquhar (1981). Repeated measurements showed little

difference in net gas exchange among leaves of similar age in individual

mango trees (data not shown). Therefore, one fully mature, sun-exposed

leaf on the most recent, mature growth flush of each tree was used for

gas exchange determinations. Measurements were made between 1000 and

1300 hr (EST) prior to flooding, at 2 and 5 days of flooding, and at

about weekly intervals for 65 days thereafter.

Stem radial growth was determined by measuring stem circumferences

10 cm above the graft union prior to flooding and 28, 56 and 84 days

after flooding was imposed. Shoot extension growth was determined on

two actively-growing, sun-exposed shoots per tree. Shoots were measured











prior to flooding and extension growth determined 84 days after flooding

was imposed.

Gas exchange and growth data were analyzed by ANOVA (P < 0.05).



Flooding and Leaf Water Potential (Experiment II)

Twelve 4-year-old 'Tommy Atkins' mango trees, six on 'Turpentine'

rootstock and six on a seedling rootstock, about 2.25 m in height, were

grown in 57-liter containers in Krome very gravelly loam soil. On 7

Dec., 1989, trees were subjected to two flooding treatments as described

in Expt. I: 1) nonflooded (control), and 2) continuously flooded for 14

days. Three single tree replicates of each rootstock were used for each

treatment. Thus, there was a 2 x 2 factorial arrangement of treatments,

replicated 3 times in a split plot design with rootstock as the main-

plot and flooding treatment as the subplot. Diurnal air temperatures

fluctuated between 10 and 24 C during the experiment, and flooded soil

temperatures averaged 16 C.

Prior to flooding and at biweekly intervals for 2 weeks

thereafter, A, gs, E and Ci were determined at 1000-1300 hr for each

tree. Gas exchange determinations were made as described in Expt I.,

except that temperature was 24.5 + 4.4 C, CO2 concentration was 345 + 6

pmol mol-1, and mean vapor pressure was 0.90 kPa (ranging from 0.42 to

1.40 kPa) in the leaf chamber. Immediately following gas exchange

determinations, mid-day (1200-1400 hr) leaf water potentials were

determined on three sun-exposed leaves of each tree with a pressure

chamber (Model 3000, Soil Moisture Equipment Corp., Sta. Barbara, CA) as

described by Scholander et al. (1965). Data were analyzed by a standard

t-test (P < 0.05).













Net Gas Exchange During the Early Stages of Flooding (Experiment III)

To examine the effects of flooding on gas exchange during the

early stages of flooding, ten, 1-year-old 'Peach' seedling mango trees,

about 60 cm in height, were grown in 7.5-liter containers in

peat:perlite (1:1 v/v). Five replicate plants were flooded on 6 Sept.,

1989 by submerging the containers in tap water in plastic tubs, and five

replicate plants were maintained nonflooded. Nonflooded plants were

maintained as described previously, but water levels were maintained

about 5 cm above the soil surface for flooded plants. The experimental

design was a randomized complete block, with one tree from each

treatment in each of five blocks. Determinations of A, gg, E and Ci

were made on two fully expanded, sun-exposed leaves of the most recent,

mature growth flush of each tree at 1300 hr, immediately before plants

were flooded. Thereafter, determinations were made at 1300 hr on day 1

of flooding, at 1000, 1300 and 1600 hr on days 2 and 3, and at 1000 hr

on day 4 of flooding. For all trees, gas exchange was monitored as

described in Expt. I, except for the following leaf chamber conditions:

CO2 concentration was 340 + 9 pmol mol-1, temperature was 31.4 5 C,

and mean vapor pressure was 0.94 kPa (ranging from 0.53 to 1.47 kPa).

Data were analyzed by a standard t-test (P < 0.05).



Flooding and Growth (Experiment IV)

One-year-old 'Peach' seedling mango trees about 70 cm in height

were grown in 11.5-liter containers in Krome very gravelly loam soil.

Fifty trees were divided into four categories on the basis of height and

basal stem diameters. Equal numbers of plants were randomly selected











from each size category and placed into five groups of 10 plants each,

so that there was no difference among the groups in mean height or stem

diameter. One randomly selected group was immediately harvested for

determination of leaf area, total shoot length, length of new shoot

growth flushes, and fresh and dry weights of leaves, new growth flushes,

total shoot and roots, and for calculation of shoot:root ratios. The

remaining groups were randomly assigned two treatments 1) flooded by

submerging the containers in plastic tubs; or 2) nonflooded (control),

both as described in Expt. III. Thus, two replicates of 10 sample trees

were assigned to each treatment in a randomized complete block design.

After 2 and 4 weeks of flooding, one group of plants in each treatment

(flooded and nonflooded) was harvested. Mean leaf area (portable leaf

area meter, Model LI-3000, LI-COR, Inc., Lincoln, NE 68504), total shoot

length and length of new shoot growth flushes were determined for trees

in all groups upon harvest. Plants were oven-dried at 60 C for 4 days

and mean leaf dry weight, total shoot dry weight, dry weight of new

shoot growth flushes, root dry weight, and root:shoot ratio were

determined. Data were analyzed by a standard t-test (P < 0.05).



Results



Experiment I

Within 3 days of flooding, mean Eh of the flooded soil was +26 mv,

indicative of anaerobic soil conditions (Gambrell and Patrick, 1978),

and soil pH was 7.4. After 21 days of flooding, Eh and pH had

stabilized at -150 my and 7.0, respectively.











There was no interaction between rootstock and treatment for any

of the variables measured (P < 0.05). Therefore, all 10 trees in a

given treatment were pooled for statistical analyses. Leaf wilting and

desiccation were observed for nine flooded trees within 3-4 days. With

the exception of these nine trees (five from the 14DF treatment, four

from the 28DF treatment), lenticel hypertrophy was observed on submerged

stems of all flooded trees within 5-7 days. The nine trees exhibiting

leaf dessication and shoot die-back were eliminated from the experiment

by day 14.

Within 2 days of flooding, A of flooded trees became negative, and

Ci, gs and E and were 109%, 62% and 75% that of the nonflooded trees,

respectively (Fig. 5-1). By day 58 (44 days after removal of plants

from flooding), A, gs and E of trees of the 14DF treatment were similar

to those of the controls, decreased on day 64 and recovered again by day

70 (Fig. 5-1). By day 70, A, gs and E of trees in the 28DF treatment

were lower, and Ci was higher, than that of either the controls or the

trees in the 14DF treatment (Fig. 5-1). Leaf temperatures of flooded

plants were 1-2 C higher than leaf temperatures of nonflooded plants

(data not shown).

Twenty-eight days after submergence, stem radial growth of trees

in the 14DF and 28DF treatments was similar, but was 40% and 59%,

respectively, of that of the nonflooded trees (Fig. 5-2). Fifty-six

days after submergence, stem radial growth of trees in the 14DF and 28DF

treatments was 31% and 46%, and by day 84 was 372 and 45%, respectively,

of that of the nonflooded trees. There was no difference among

treatments in shoot extension growth (data not shown).











After several weeks, infestations of bark-boring ambrosia beetles

(Xvloborus spp., Coleoptera:Scolytidae) were noted in all flooded trees.

This is often indicative of elevated ethanol concentrations in the xylem

sap (Cade et al., 1970). No infestations occurred on nonflooded trees.



Experiment II

Rootstock did not effect water potential or net gas exchange of

flooded and nonflooded trees (P < 0.05). Therefore, the six trees in

each treatment were pooled for statistical analyses. For flooded trees,

A decreased within 7 days of flooding (data not shown) and remained

lower than that of the nonflooded trees for the remainder of the

experiment. Flooding had no influence on leaf water potential, which

averaged 0.2 MPa, over the course of the experiment (data not shown).

There were no visual symptoms of plant stress, such as leaf dessication

or wilting, over the course of the experiment. Lenticel hypertrophy was

noted on the submerged stems of 5 of the 7 flooded trees after about 10

days of flooding.



Experiment III

Flooded trees had simultaneous decreases in A, E and gs, and an

increase in Ci, by the morning of the third day of flooding (Fig. 5-3).

In general, the difference in A between flooded and nonflooded plants

decreased over the course of day 3 before increasing by day 4.

Hypertrophied lenticels were observed at the floodline of stems of all

flooded trees after 4-5 days of flooding.











Experiment IV

Leaf area, leaf dry weight, total shoot length, length of new

growth flushes, dry weight of new growth flushes (data not shown) and

total shoot dry weight (Fig. 5-4) were similar for flooded and

nonflooded trees on any harvest date. Two and four weeks after

submergence, root dry weights of flooded trees were significantly

reduced, and were 76% (week two) and 66% (week four), respectively, of

the nonflooded trees (Fig. 5-4). The initial shoot:root ratio was 2.38.

After 2 weeks, shoot:root ratios of nonflooded and flooded trees were

2.72 and 3.80, respectively. By week 4, shoot:root ratios were 2.79 and

4.18 for nonflooded and flooded trees, respectively. There was no tree

mortality in any treatment, and hypertrophied lenticels were observed in

all flooded trees after 3-4 days of flooding.



Discussion



Greater tree mortality and a more rapid decrease in A were

observed in Expt. I than in a previous study of mango flooding

(unpublished data), or than in Expt. II. Although Expt. I and the

previous study were conducted under similar environmental conditions,

trees in the previous study had been recently transplanted into 57-liter

containers, whereas trees in Expt. I were maintained in the same

containers for 1.5 years, and were more root-restricted.

In Expt. II, A was not reduced until 7 days after submergence

(data not shown), and visible symptoms of flooding stress (i.e., leaf

desiccation, wilting) were not observed in any of the flooded trees.

The cooler temperatures and shorter days during this experiment may have











moderated the flooding response. Slower growth rates than in previous

experiments (data not shown) and reduced shoot and root respiration

rates (not measured) due to cooler temperatures may have also affected

the flooding response.

The stomatal response of flooded mangos in the present study (Fig.

5-1) differs from that of flood-tolerant species in which stomata reopen

after 1-2 weeks of flooding (Kozlowski and Pallardy, 1984). Flood-

induced stomatal closure has been attributed to decreases in leaf water

potential resulting from decreased water uptake by the roots (Heinicke,

1932; Kozlowski and Pallardy, 1984), possibly as a result of decreased

cell membrane permeability under anaerobiosis (Bradford and Hsiao,

1982). However, as in Expt. II, stomatal closure of flooded plants

occurred without reductions in leaf water potential (Andersen et al.,

1984b; Bradford and Hsiao, 1982; Kozlowski and Pallardy, 1984; Smith and

Ager, 1988).

In some flooding studies, decreases in A have been attributed

primarily to stomatal closure (Kozlowski and Pallardy, 1984). In the

present study, the increase in Ci, concomitant with decreased A and gs

suggested that non-stomatal factors may have been influencing carbon

assimilation (Farquhar and Sharkey, 1982).

Although a previous study showed a reduction of mango shoot

extension growth with long-term (110 days) flooding (K.D. Larson,

unpublished data), the shorter durations of flooding in Expt. I were

apparently insufficient to reduce shoot extension growth. However,

similar to previous observations (K.D. Larson, unpublished data) stem

radial growth was affected by flooding in the present study. Since

mango growth flushes are often restricted to certain shoots in one part











of the tree (Young and Sauls, 1981), stem radial growth measurements are

a more sensitive indicator of tree growth than are measurements of shoot

extension growth. An increased shoot:root ratio has often been observed

in flooded woody plants, reflecting a greater sensitivity of root than

shoot growth to flooding stress (Kozlowski, 1984). Inhibition of root

growth with flooding is characteristic of flood-intolerant species

(Kozlowski, 1984).

Lenticel hypertrophy occurs in several woody plant species

subjected to flooding (Andersen et al., 1984a; Kawase, 1981; Kozlowski,

1984) and is thought to enhance 02 diffusion to roots (Kozlowski, 1984),

or eliminate potentially toxic metabolites such as ethanol, acetaldehyde

or ethylene (Chirkova and Gutman, 1972). Development of hypertrophied

lenticels, however, does not necessarily confer flood-tolerance.

Andersen et al. (1984a) observed lenticel hypertrophy on submerged stems

of flood-intolerant Prunus persica. Profuse lenticel hypertrophy

occurred on stems of mango trees that survived flooding, but there was

little or no stem hypertrophy with mango trees that died. Furthermore,

when hypertrophied lenticels of flooded mango trees were covered, the

trees died within three days (K.D. Larson, unpublished data). Thus,

hypertrophied lenticels appear necessary for mango tree survival under

prolonged flooding, but their exact role in gas exchange or in the

elimination of metabolic end-products is not clear.



Conclusions



The ability of mango trees to survive prolonged flooding is

variable and may be related to environmental conditions and the







68



development of hypertrophied lenticels in individual trees. In previous

studies, mango trees survived flooding for 110 days or more. For trees

that survive flooding, the reductions in gas exchange and stem radial

growth, and the slow post-flood recovery with respect to gas exchange

and growth, indicate that this species is able to adapt to flooded soil

conditions, but is not highly flood-tolerant.


































14 0 -r-
0- 4H o
to 0
0 o







0 0
-I 44- 0
S-a) c1i
0 OD




0 *r 0 41
H 0-H 14
- 40 0 -




rIl 4 I
4O r- -H E-
) 004>




4c 0 0 ( o
S 0 I



















uL
400
Pr a) *



0N ) 0 4








4 0 0







-Ir-
s M
*r a j
o M<;
U 0 *
4-1 >'0
4-1 3 e a)

b1T g-
(3 0















o

II



*4 0


U /\
U4.

4N



^ yA


U,
^*4


I/I



*40
,,40



" O
-U 9


I U 1t i I
0 lO 0 0 00 <0 C
UO CN o r- LO
(-_s _w 00Ioww) s6 (-Ls -_w OZH IOww) 3


0

c -
o
i^
Y-!L


t 0







ii "

Y^
0 4 *



.



SN I


Is ---- 'I i -


0) LU T--0,-
(Q_8 _u ZOO l|wO ) V


o *
I
H

g/


."4.





0


0o 0f 0 r 0

(L-I0w Z0O |louww) !O


0

0
(0

0

cO


o E

CO4




0







0
0


0
cU

o0
0


o E


S


I

































in
141

CDV0
14J
00)


bO '-40




0 0)
44 4.)
v4 *W






4JG 0)4

0
Ac'!








.0
441 5)
400)P











0 04


14. '0 44)
aIP,

0 V4










r-4 U
4-r4










-H 4-1
Co










0 V
P4


















O0 o *-Q --4










Oco







D LLm 4 00

140







O



(WU) qlMO8Jb WJS ID!pDdI



































14 0 0 -
4.4 4.3 0
ow

*144.30
C6.4 v
v 04
coP

00
co 0 o
V~ .3 0 0
41.
o ~4J 0




0 0(



-H 04.3
V (r D Vb
0 a)0E






co 0 0 w


-H0 0 0H
00
C r4 o co
V~ 4)3 0 4
-H 0 r.









0 60
44~1r (A(



000V
4'00
H( U r'I (

C 4-44-g







0 w 0)
co 0 0.i-
4.3 0 4-4




0)06

0 0-V
S- 0 .
0hC O
(4 D







Erz









~c-l4

/~/

*E ~






I1/




J


-x-








//


`4
NZ


O 0 0 0 0 (0 -
0 0 LO
(ls ZUwZO lJoi) 86 (L-s _LU OQH IOLU) 3


~t' 4



/i/
r/


U 4


0o n


4
/
I
'N
-4


I-


\ /
I \








U
4-n


=-


<- 0N 0


C(0


0)


4c
C
0


0

0
SLL


(-low Zouj loww) !O


(0





c
C
0
O
o4-
0
L.

-u

0
0
*.L


-"o

4cc

II
4C


mI


(_S Z-_UJ Z0o IowlU) V


---~--------- ---


v



































440
000
0 04V
oa



0- 44 0)
aeo


.44
44. C
04C 0



0 0)

0044 4
V.) 0


w 9
000







044 CO
O4-4 4)4
0 0 4

0 0
ow

0)0

0" 0
Llnr

;o~r
00B
.r00



4H -%
S004
0 OC





0v



wDP












IM Aup looi / IM KJp looqs

(D LO O N 0
I I I I I I I


S O0
\

\
\




\
\













0D
I]


C,




-
-o


:3








- o
0

0



-0


(6) IM A/ip


I I I I


O






I
I

1


1I
I


O0

c)
0
0
0
r


O
c:
D7


I
















CHAPTER 6
FLOODWATER TEMPERATURE AND STEM LENTICEL
HYPERTROPHY IN MANGO TREES



Introduction



Lenticel hypertrophy intumescencee) has been observed on stems of

several woody plant species under flooded conditions (Andersen et al.,

1984; Angeles, 1990; Angeles, et al., 1986; Hook et al., 1970; Hook and

Scholtens, 1978; Kawase, 1981; Kozlowski, 1984; Sena Gomes and

Kozlowski, 1988; Tang and Kozlowski, 1984). Hypertrophic stem lenticels

may allow internal oxygen diffusion to flooded roots (Kozlowski, 1984),

or function as excretory sites for potentially toxic metabolites such as

ethanol and acetaldehyde formed in the roots during anaerobic

respiration (Chirkova and Gutman, 1972).

With mango, hypertrophic lenticels have been observed on stems of

trees that survived soil flooding, but not on trees that died as a

result of flooding stress (see Chapter Five, page 67). When

hypertrophic lenticels of flooded mango trees were sealed, trees died

within three days (K.D. Larson, unpublished data). Thus, it appears

that hypertrophic lenticels of mango are necessary for tree survival

under flooded soil conditions, although their specific role has not been

elucidated.

Lenticel hypertrophy results from increased phellogen activity,

cell enlargement, and cell elongation (Angeles, 1990; Kawase, 1981).











These growth processes are temperature dependent. For example,

phellogen activity of Ailanthus altissima, Fraxinus pennsylvanica, and

Robinia psuedacacia increased with increasing temperature (Borger and

Kozlowski, 1972), since cell growth involves temperature-dependent

enzymatic reactions. However, high soil and air temperatures also

typically increase flooding stress and reduce plant survival (Abbott and

Gough, 1985; Catlin et al., 1977; Childers and White, 1950; Crane and

Davies, 1989; Davies and Flore, 1986a; Davies and Wilcox, 1984;

Heinicke, 1932; Olien, 1987).

Anatomical and morphological development of hypertrophic stem

lenticels have been characterized for many flooded woody plant species,

including angiosperms (e.g. Salix nigra, Populus deltoides, Fraxinus

pennsylvanica, Quercus macrocarpa, and Platanus occidentalis), and

gymnosperms (e.g. Pinus ponderosa, Pinus resinosa, and Larix laricina

(Angeles, 1990; Angeles et al., 1986; Hook, et al., 1971; Hook and

Scholtens, 1978; Kozlowski et al., 1991). However, to our knowledge,

the influence of floodwater temperature on lenticel hypertrophy of trees

has not been reported. The objective of this study was to characterize

the changes in anatomy and morphology of stem lenticels of mango as

affected by floodwater temperature.



Materials and Methods



Eighteen, one-year-old 'Peach' seedling mango trees were grown in

a peat:sawdust:pine bark (1:1:1 v/v/v) media (pH 6.0) in 3.75-liter

plastic pots in a glasshouse. The trees were randomly divided into

three temperature treatments in March 1989, at which time average tree











height was 1.0 m. Pots were submerged in tap water in plastic

containers enclosed in temperature-controlled flooding chambers in a

laboratory (Gilreath et al., 1982). Soil, floodwater, and air

temperatures within the flooding chambers were maintained at either 15,

22.5, or 30 +20 C. Tap water adjusted to the specified temperature was

added daily to each container to maintain floodwater at a level 10 cm

above the soil surface. The foliated portion of the stems (upper 0.5 m

of the trees) protruded through the lids of the chambers. A 400-watt

mercury vapor lamp and four 100-watt incandescent light bulbs were

positioned above each chamber to expose the foliage to a 10-hour

photoperiod. Photosynthetic photon flux (PPF) in the upper part of the

tree canopy was approximately 750 pmol m2 1s, as determined with a

selenium photo cell attached to a Parkinson leaf chamber (Analytical

Development Co., Hoddeson Hertz., England). This PPF is above the light

saturation point for photosynthesis of individual mango leaves (Schaffer

and Gaye, 1989). Air temperature and relative humidity in the

laboratory ranged from 22 to 340 C, and 28 to 45Z, respectively.

Soil redox potential (Eh), pH, and dissolved oxygen content of the

floodwater were monitored for each plant prior to flooding, at daily

intervals during the first 4 days of flooding, and at 3-day intervals

thereafter. Redox potential was determined at a depth of 10 cm using an

oxygen diffusion meter (Model P5E, Jensen Instruments, Tacoma,

Washington), with a Ag+/AgCl reference electrode, and 4 platinum-tipped

microelectrodes (Crane and Davies, 1988). Soil pH was monitored with a

pH meter (Model 5995-30, Cole-Parmer Instruments, Chicago, IL), and

floodwater 02 content was monitored with an oxygen probe and meter

(Model 5946-10, Cole-Parmer Instruments, Chicago, IL).











Prior to flooding, and at 24-hour intervals for 13 days

thereafter, the development of lenticel hypertrophy was visually

determined for all trees. Hypertrophic lenticels were readily

distinguishable by a widening of the lenticel pore, and exposure of the

white parenchymatous filling tissue. Based on detection of this filling

tissue, hypertrophic lenticel density was determined daily in randomly

selected 1-cm2 stem sections within 1.5 cm of the floodline of each

tree. Data pertaining to the number of hypertrophic lenticels per unit

area were analyzed by non-linear regression using the Gompertz

Logistical Growth Model (Clow and Urquhart, 1974).

In conjunction with daily visual observations of hypertrophy, a

0.5 cm2 piece of bark, consisting of phloem, cortex, and periderm

tissues, was excised with a razor blade from an area of the stem within

1.5 cm of the floodline of five trees in each treatment. A sixth tree

in each treatment remained intact to determine whether wounding had an

effect on lenticel hypertrophy or tree survival. Excised tissues were

fixed in formalin acetic acid, dehydrated in alcohol, and embedded in

paraffin. Tissues were sectioned with a rotary microtome at a thickness

of 10 im, stained with safranin and fast green, and mounted on

microscope slides with Permount (Johansen, 1940). Photomicrographs of

representative tissue sections were taken with a Nikon Optiphot

Microscope with a Nikon AFX camera attachment.

Plants were removed from the flooding chambers after 13 days, and

the experiment was repeated. Two plants from the original 150 C

treatment were maintained submerged at 150 C for a total of 28 days, and

then were transferred to 300 C floodwater. For a given floodwater

temperature treatment, there were no differences between experiments in











regard to hypertrophic lenticel density or mean number of days required

for the development of lenticel hypertrophy (P > 0.05). Therefore, data

regarding hypertrophic lenticel density and mean number of days for

required for the development of hypertrophy were pooled for the two

experiments.



Results



Floodwater Oxygen Content and Soil Eh

Dissolved 02 content of the floodwater was initially 7.8 ppm for

all temperature treatments (Fig. 6-1A). An inverse relationship

occurred between temperature and oxygen content over time. Oxygen

content decreased most rapidly and was lowest for the 22.50 and 300 C

treatments, whereas the 150 C treatment consistently had the highest

oxygen content. Floodwater oxygen contents had stabilized by day 10,

with 1.1, 2.1, and 3.3 ppm 02 in the 30, 22.5, and 150 C treatments,

respectively.

For all treatments, soil Eh prior to submergence was approximately

+450 my (Fig. 6-1B). As with floodwater oxygen content, an inverse

relationship generally existed between temperature and soil Eh. By day

10, Eh of the 300, 22.50, and 150 C treatments was -14, +12, and +240

mv, respectively. By day 13, the soils of all treatments were anaerobic

(< 200 my) (Ponnamperuma, 1972); however, Eh of the 30 and 22.50 C

treatments was near -30 my, whereas that of the 150 C treatment was +160

my.

For all treatments, soil pH increased during flooding, but more so

for the 22.5 and 300 C treatments than for the 150 C treatment. After











13 days of flooding, soil pH was 6.8, 6.7, and 6.4 for the 300, 22.50,

and 150 C treatments, respectively.



Lenticel Morphology and Anatomy

Nonhypertrophic lenticels of mango are longitudinally oriented

and characterized by a relatively loosely structured, nonsuberized

filling tissue alternating with compact layers of thicker-walled, more

highly suberized cells (Figs. 6-2 6-5). The phellogen of

nonhypertrophied lenticels undergoes periclinal divisions to produce

radially arranged rows of cells in the phellem and phelloderm (Figs.

6-2 6-5).

For plants maintained at 300 and 22.50 C, lenticel hypertrophy was

first observed in a few plants on the fifth and sixth days of flooding,

respectively. However, the mean number of days of flooding that elapsed

until lenticel hypertrophy was observed was 6.6 0.8 and 8.1 0.7

(mean no. of days for 12 trees S.E.), for the 300 and 22.50 C

treatments, respectively. Lenticel hypertrophy was not observed after

28 days of flooding at 150 C. However, plants transferred to the 300 C

treatment after 28 days at 150 C exhibited lenticel hypertrophy within

three days.

For plants flooded at 300 C and 22.50, the development of

hypertrophic lenticels (number per cm2 of stem) followed a sigmoid

pattern (Fig. 6-6) defined by the Gompertz Logistical Growth Model (Clow

and Urquhart, 1974). Lenticel hypertrophy developed more rapidly at 300

C than at 22.50 C, as indicated by the steeper logarithmic growth phase

of the 300 C regression line between days 5 and 10. For the 300 C

treatment, virtually all stem lenticels at the floodline were











hypertrophied by day 10. Consequently, there was a reduction in the

rate of development of lenticel hypertrophy after day 10 for the 300 C

treatment. Thirteen days after flooding was initiated for the 300 C

treatment, adjacent hypertrophic lenticels were coalescing due to

pronounced hypertrophy. Although only about two-thirds of the lenticels

had hypertrophied by day 10 for the 22.50 C treatment (Fig. 6-6), the

rate of hypertrophy also appeared to be slightly decreasing for this

treatment by day 13 of submergence.

Figures 6-7 through 6-10 and Figs. 6-11 through 6-12 show the

development of lenticel hypertrophy for plants maintained at 22.50 and

300 C, respectively. Although there was no unaided visual evidence of

hypertrophy until 5 days after submergence at 300 C, some histological

changes proceeding hypertrophy were evident by day 3. Initial stages of

lenticel hypertrophy were characterized by a more spherical shape of

cells in the phellem and phelloderm (Figs. 6-8, 6-12), and by the

development of intercellular spaces in the phellem and lenticel filling

tissue (Fig. 6-12). As there was no evidence of tearing or mechanical

disruption of the tissues, development of intercellular spaces was not

considered to be an artifact of the sectioning procedure.

Later stages of hypertrophy were characterized by an increase in

phellogen activity and production of additional phellem tissue adjacent

to the lenticel pore, resulting in a larger pore opening (Figs. 6-9,

6-13). By day 6 of flooding, the increase in phellogen activity had

produced a phellem layer with a mean thickness of 94.6 8.7 pm and 76.9

6.9 pm for the 300 C and 22.50 C treatments, respectively (mean

phellem layer thickness of three representative sections from each of

three trees per treatment S.E.). In contrast, nonhypertrophic











lenticels had a mean phellem layer thickness of 48.4 7.1 pm (mean

phellem layer thickness of three representative sections from each of

three trees S.E.). Later stages of lenticel hypertrophy were also

characterized by the development of intercellular spaces in the cortex

(Fig. 6-14).

No effect of bark excision on hypertrophy or tree survival was

observed.



Discussion



The decreased solubility of oxygen in water with increasing

temperature, and increased microbial and root respiration apparently

resulted in rapid oxygen depletion and rapid reduction of soil Eh at 300

and 22.50 C. The higher Eh and greater floodwater oxygen content of the

150 C treatment indicate reduced 02 consumption at this temperature, and

decreased root 02 demand may have slowed the development of hypertrophic

lenticels in these trees. The higher Eh and oxygen content, and the

absence of lenticel hypertrophy at 150 C also suggest a critical

floodwater oxygen content (3 ppm) above which lenticel hypertrophy in

mango may be delayed or inhibited.

Development of intercellular spaces in adventitious roots of Zea

mays was stimulated by low partial pressures of 02 (Drew et al., 1979;

McPherson, 1939). However, Jackson et al. (1985) reported that

development of intercellular spaces in adventitious roots of Oryza

sativa was not promoted by an 02 deficit. Similarly, Wample and Reid

(1975) reported that flooding with either stagnant or aerated water

induced stem hypertrophy in Helianthus annuus. In previous experiments











with mango at temperatures of 25-300 C, lenticel hypertrophy was

inhibited by floodwater oxygen contents of 12-15 ppm, but no inhibition

was noted at a relatively low range of 02 contents (1-5 ppm) similar to

those in the present study (K.D. Larson, unpublished data), suggesting

an oxygen effect on lenticel hypertrophy.

The increased phellogen activity and more rapid hypertrophy that

were observed at 300 C for mango may reflect the tropical and

subtropical origin of this species (Mukherjee, 1985). Hypertrophy is a

growth process involving phellogen activity, cell enlargement, and cell

elongation (Angeles, 1990; Kawase, 1981). Since cell division and

growth involve temperature-dependent enzymatic reactions and metabolic

processes, acceleration of lenticel hypertrophy with increasing

temperatures would be expected. McPherson (1939) reported that high

temperatures increased the development of intercellular space in flooded

roots of Zea mays. Similarly, for the temperate-zone species Fraxinus

pennsylvanica, phellogen activity increased with each 5 C increase in

temperature over a range of 10 to 300 C (Borger and Kozlowski, 1972).

However, phellogen activity of two other temperate-zone species, Robinia

pseudacacia and Ailanthus altissima, generally increased as temperature

increased up to 250 C, but decreased at 300 C (Borger and Kozlowski,

1972). Borger and Kozlowski (1972) also reported that periderm tissue

developed in F. pennsylvanica seedlings grown at at 100 C, but did not

develop in A. ailanthus, again indicating differences among species in

phellogen activity and periderm development in response to temperature.

Flood-induced anaerobiosis stimulates ethylene production in some

plants (Drew et al., 1979; Wample and Reid, 1975), and an ethylene-

mediated increase in cellulase activity is a prelude to hypertrophy or











development of aerenchymatous tissue (Kawase, 1981). Ethylene

production is temperature-dependent, with the optimum temperature for

ethylene evolution at about 300 C (Yang, 1980). Thus, with mango, the

more rapid hypertrophy observed at 300 than at 22.50 C, and the

inhibition of hypertrophy at 150 C, may be due, in part, to temperature

effects on ethylene biosynthesis.

The development of hypertrophic lenticels may have aided tree

survival at 22.50 and 300 C. At 150 C, phellogen activity and cell

growth of mango were reduced and lenticel hypertrophy did not occur,

although hypertrophy occurred rapidly when the trees were transferred to

300 C. At 150 C, soil and plant respiration and flooding stress

presumably were reduced, and lenticel hypertrophy was inhibited, but at

higher temperatures, 02 consumption and plant stress increased, and

lenticel hypertrophy developed.



Conclusions



Although inhibition of lenticel hypertrophy in mango previously

was observed with floodwater oxygen contents of 12-15 ppm (K.D. Larson,

unpublished data), the small differences in floodwater 02 content among

treatments in the present study suggest that differences in hypertrophy

probably were due to temperature rather than 02 effects. Also, although

there was little difference in 02 content and Eh between the 22.50 and

300 C treatments, development of hypertrophied lenticels was slower at

22.50 C. This also suggests a response to temperature rather than to

oxygen. Thus, inhibition of lenticel hypertrophy at 150 C appears to be

due mainly to temperature-mediated reductions in respiration and plant






87



metabolic processes that regulate phellogen activity and cell growth.

Although various endogenous and exogenous factors (02 partial pressure,

ethylene) may influence hypertrophy of lenticels, lenticel hypertrophy

in mango appears to be a temperature-dependent response to flooding.

































r.
0
14
C: 0
0 V
*0r -* 0


w 0 ) 4.)
-4 0- 0 .
0 *

PO rl 4 4 A)

r i) 0)

40 00
4) '441

H *0 -r
0 O -4



ed 0 NrI

14 4 Cn
w.a'0 00





) C 14 r. v

V 0 0 m 0
o. 4a 0



-> 00 4 w
0 4e0
44 M 000

44 -H e M
*0 0 0 0 4)
0 04 -
4-4 0 > P H


0 'gOr M 4.


0 r-4 44
O *a X o

O e i


-4 -A c *-ir


I 4' > 0 .
%D


o






4)

E-4
o
o
V







00
00
00
co

M



4-1
r-4
410



V w
0)
0 0
60


c o
0 0

') *rl

00



Sr-(
o 0



















































00 CD -,* (" 0 0 0 0 0 0 0
0 0 0 0 0
r- M C, -


(wudd) .ualuoo ~O pDAIOSS(]


(AW) 43

































Figs. 6-2, 6-3. Non-hypertrophic stem lenticels of mango trees
flooded at 150 C. Photographs are representative of ten
sections from each of three trees; PH = phellem; PG =
phellogen; PD = phelloderm; C = cortex. 6-2) Lenticel prior
to flooding; 6-3) lenticel after 3 days of flooding.






91









I-.m:22jjm


































Figs. 6-4, 6-5. Non-hypertrophic stem lenticels of mango trees
flooded at 150 C. Photographs are representative of ten
sections from each of three trees; PH phellem; PG =
phellogen; PD phelloderm; C cortex. 6-4) lenticel after
6 days of flooding; 6-5) lenticel after 12 days of flooding.
For scale, see Fig. 6-2.