Title: Water management effects on photosynthate distribution, physiology, and nutritive value of perennial peanut
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00103063/00001
 Material Information
Title: Water management effects on photosynthate distribution, physiology, and nutritive value of perennial peanut
Physical Description: x, 128 leaves : ill. ; 29 cm.
Language: English
Creator: Mansfield, Charles W., 1954-
 Subjects
Genre: bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: by Charles W. Mansfield.
Thesis: Thesis (Ph. D.)--University of Florida, 1990.
Bibliography: Includes bibliographical references (leaves 122-127).
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00103063
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 001575943
notis - AHJ9790
oclc - 22913692

Full Text











WATER MANAGEMENT EFFECTS ON PHOTOSYNTHATE
DISTRIBUTION, PHYSIOLOGY, AND NUTRITIVE VALUE
OF PERENNIAL PEANUT











BY

CHARLES W. MANSFIELD


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


1990














ACKNOWLEDGMENTS

The pursuit of higher education originally came about by mutual

consensus between my wife and me. We thank God for having sustained us

and our family in the venture working as He does in marvelous ways

through others. Personally, I am forever indebted to my wife for her

support and to my children who have been extremely patient.

I am also greatly indebted to the chairman of my committee, Dr.

Jerry Bennett, who provided excellent leadership and advice pertaining

to correct and ethical science. He is an outstanding person of the

highest integrity in all matters. Dr. Ken Boote also provided much

expertise related to laboratory analyses and techniques. The other

members of my committee, Dr. Lynn Sollenberger, Dr. David Wofford, and

Dr. George Bowes, not only provided excellent instruction at the

classroom level, but also advice related to the many aspects of doing

research and collecting data. Dr. Quesenberry was instrumental in

orienting me in my initial communications with the Agronomy Department,

and counseling throughout the time that I was studying. Finally, I

would like to thank Mr. Neil Hill for his expert assistance in helping

get the work done.

United States Department of Agriculture grant 86-CRSR-2-2846,

Genetics, Physiology, Ecology, and Utilization of Tropical Forage

Legumes, provided funds for purchase of the LI-COR 6200 Portable

Photosynthesis System used in this project. Dr. Ken Albrecht, my








original contact and initial chairman of the committee, was instrumental

in setting up this research opportunity for me before moving to another

institution. Also, Dr. David Baltensperger provided valuable expertise

before going to another institution. There are many others, who I have

failed to mention, that have been of great assistance during the time of

study, and it was through the support and effort of all that the pursuit

of higher education became a reality.














TABLE OF CONTENTS

ACKNOWLEDGMENTS ............................................ ii

LIST OF TABLES ............................................... v

LIST OF FIGURES ........................................ ..... vii

ABSTRACT .................................................... ix

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

LITERATURE REVIEW ........................................... 4
Photosynthate Distribution in Plants..................... 4
Physiological Responses to Water Deficits ................. 7
Effect of Water Stress on Forage Nutritive Value .......... 15

MATERIALS AND METHODS .................... ....................... 18
Establishment of Perennial Peanut (1986) ................. 18
Treatments and Data Collection (1987 and 1988) ............ 19
Leaf Carbon Exchange Rate and Leaf Nitrogen Concentration 26
Laboratory Analyses ................................... 28

RESULTS AND DISCUSSION ............... ............ ........ ..... 29
Rainfall and Irrigation Amounts and Distribution .......... 29
Photosynthate Accumulation and Distribution ............... 38
Physiological Measurements ............................. 68
Plant Nitrogen Concentration and in vitro Digestibility ... 93

SUMMARY AND CONCLUSIONS ........................................ 104

APPENDIX ................................................... 109

LITERATURE CITED ........................................... 122

BIOGRAPHICAL SKETCH ......................... ............... .. 128














LIST OF TABLES


Table page

1. Monthly rainfall for 1987 and 30-year mean monthly
rainfall for Gainesville, Florida.............................30

2. Rainfall and irrigation specific to each harvest cycle
as well as before and after the perennial peanut growing
season in 1987 ............................................ 32

3. Monthly rainfall for 1988 and 30-year mean monthly
rainfall for Gainesville, Florida..............................34

4. Rainfall and irrigation specific to each harvest cycle
as well as before and after the perennial peanut growing
season in 1988 ............................................ 37

5. Above-ground, rhizome, total, and forage dry matter
distribution and leaf area index (LAI) of perennial peanut
as affected by water management treatments in 1987.............41

6. Above-ground, rhizome, total, and forage dry matter
production by perennial peanut averaged across water
management treatments in 1987.................................50

7. Seasonal dry matter accumulation and distribution as
affected by water management treatment in 1987................52

8. Above-ground, rhizome, total, and forage dry matter
distribution and leaf area index (LAI) of perennial peanut
as affected by water management treatments in 1988............55

9. Above-ground, rhizome, total, and forage dry matter
production by perennial peanut averaged across water
management treatments in 1988.................................66

10. Seasonal dry matter accumulation and distribution as
affected by water management treatment in 1988.................67

11. Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular
to ambient CO, concentration (Ci/Ca) of perennial peanut
as affected by water management treatment on 18, 26 and 29
May, and 8 and 10 June 1987.................................... 76








Table Dage

12. Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular
to ambient CO2 concentration (Ci/Ca) of perennial peanut
as affected by water management treatment on 27 and 28
August, and 7, 22 and 29 Oct. 1987............................79

13. Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular
to ambient CO2 concentration (Ci/Ca) of perennial peanut
as affected by water management treatment on 12, 16, 20
and 31 May, 1 June and 8 July 1988............................86

14. Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular
to ambient CO, concentration (Ci/Ca) of perennial peanut
as affected by water management treatment on 17 and 24
Oct. 1988.................................................90

15. Nitrogen concentration of perennial peanut forage, and
leaf, stem and rhizome plant components (averaged across
water management treatments) in 1987..........................94

16. In vitro organic matter digestibility of perennial peanut
forage, and leaf and stem components (averaged across
water management treatments) in 1987..........................96

17. Nitrogen concentration of perennial peanut forage, and
leaf, stem and rhizome plant components (averaged across
water management treatments) in 1988..........................97

18. Nitrogen concentration of perennial peanut forage, and
leaf and stem components as affected by water management
treatment on 25 July 1988.................................. 99

19. In vitro organic matter digestibility of perennial peanut
forage, and leaf and stem components (averaged across
water management treatments) in 1988.........................101

20. In vitro organic matter digestibility of perennial peanut
forage, and leaf and stem components as affected by water
management treatment on 25 July 1988.........................102















LIST OF FIGURES


Figure

1. Daily rainfall distribution during the growing season
for perennial peanut in 1987 at Gainesville, Florida.........

2. Daily rainfall distribution during the growing season
for perennial peanut in 1988 at Gainesville, Florida.........


page


..31


..35


3. Above-ground dry matter accumulation of perennial
peanut as affected by water management treatment on
15 May, 11 June, 23 July, 8 Sep., and 2 Nov. 1987.
Different letters above bars within a harvest date
indicate differences at P<0.05 according to Duncan's
New Multiple Range Test..............................

4. Root length density of perennial peanut as affected
by water management treatment on 19 June 1987. Root
length density of rhizomes is reported separately at
the 50-mm depth since they occur as a rhizome mat
just beneath the soil surface..........................

5. Above-ground, rhizome, and total dry matter of
perennial peanut (averaged across all water management
treatments) on 15 May, 11 June, 23 July, 8 Sep.,
and 2 Nov. 1987.....................................


........ 39





........43




........47


6. Rhizome nonstructural carbohydrate concentration
of perennial peanut as affected by water management
treatment on 15 May, 11 June, 23 July, 8 Sep., and
2 Nov. 1987. Different letters above bars within a
harvest date indicate differences at P1O.05 according
to Duncan's New Multiple Range Test...........................48

7. Above-ground dry matter accumulation of perennial
peanut as affected by water management treatment on
19 May, 8 June, 25 July, 12 Sep., and 1 Nov. 1988.
Different letters above bars within a harvest date
indicate differences at P<0.05 according to Duncan's
New Multiple Range Test.................................... 53

8. Gravimetric soil water content on 8 June 1988.................59








Figure page

9. Above-ground, rhizome, and total dry matter of
perennial peanut (averaged across all water management
treatments) on 19 May, 8 June, 25 July, 12 Sep.,
and 1 Nov. 1988 ........................................... 61

10. Rhizome nonstructural carbohydrate concentration
of perennial peanut as affected by water management
treatment on 19 May, 8 June, 25 July, 12 Sep., and
1 Nov. 1988. Different letters above bars within a
harvest date indicate differences at P<0.05 according
to Duncan's New Multiple Range Test...........................63

11. Single leaf carbon exchange rate of perennial
peanut as affected by water management treatment
on 18, 26, and 29 May, and 8 and 10 June 1987.
Different letters above bars within a harvest date
indicate differences at P<0.05 according to Duncan's
New Multiple Range Test................................... 69

12. Gravimetric soil water content on 31 May 1987.................71

13. Gravimetric soil water content on 10 June 1987................72

14. Single leaf carbon exchange rate of perennial
peanut as affected by water management treatment
on 27 and 28 Aug., and 7, 22 and 29 Oct. 1987.
Different letters above bars within a harvest date
indicate differences at P<0.05 according to Duncan's
New Multiple Range Test..................................... ..73

15. Gravimetric soil water content on 30 Oct. 1987................75

16. Single leaf carbon exchange rate of perennial
peanut as affected by water management treatment
on 12, 16, 20 and 31 May, 1 June and 8 July 1988.
Different letters above bars within a harvest date
indicate differences at P<0.05 according to Duncan's
New Multiple Range Test................................... ..82

17. Gravimetric soil water content on 2 June 1988.................84

18. Gravimetric soil water content on 2 Nov. 1988.................91

19. Correlation between specific leaf nitrogen content
and single leaf carbon exchange rate of perennial
peanut....................................................... 92


viii














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

WATER MANAGEMENT EFFECTS ON PHOTOSYNTHATE
DISTRIBUTION, PHYSIOLOGY, AND NUTRITIVE VALUE
OF PERENNIAL PEANUT

By

CHARLES W. MANSFIELD

May 1990

Chairman: Dr. Jerry M. Bennett
Major Department: Agronomy
Perennial peanut (Arachis glabrata Benth.) is a relatively drought

tolerant tropical legume with potential for production of high quality

forage on many acres of infertile and drought soils of Florida.

However, information is lacking on the response of perennial peanut to

irrigation. Dry matter (DM) accumulation and distribution, single leaf

carbon exchange rate (A), stomatal conductance (g), leaf and soil water

status, rhizome total nonstructural carbohydrate (TNC) concentration,

and nutritive value were determined on 'Florigraze' perennial peanut

subjected to the following water management treatments in 1987 and 1988:

1) well-watered (WW), 2) 2-week cyclical stress (CS1), 3) 3-week

cyclical stress (CS2), and 4) rainfed (RF).

Increasing severity of plant water stress caused a decrease in

above-ground growth and an increase in rhizome TNC, suggesting that

photosynthate was partitioned underground at the expense of above-ground








growth during periods of water deficit. Above-ground DM in the RF

treatment was reduced 18 and 26% in 1987 and 1988, respectively, as

compared to the average of the WW and CS1 treatments. Above-ground DM

was also reduced 11% in the CS2 treatment in 1988 as compared to the WW

treatment. Single leaf carbon exchange rate, g, and leaf water

potentials were also reduced only during severe soil water deficits. On

the average, g was reduced more than A during plant water stress.

However, both stomatal and nonstomatal mechanisms appeared to limit A in

perennial peanut during periods of drought. Water management had little

effect on the nutritive value of perennial peanut which maintained

relatively high in vitro organic matter digestibility and nitrogen

concentration throughout both seasons.

Deep rooted characteristics of perennial peanut caused plant

productivity to be insensitive to water management during relatively

short periods of dry weather (2 to 3 weeks) unless soil moisture

reserves deep in the profile had been depleted. Results of this study

suggested that irrigation of perennial peanut would increase forage

production only during periods of severe soil water deficits. During

dry periods, replenishment of the soil profile with water at 2-week

intervals resulted in similar forage yields to those from treatments

irrigated more frequently.










photosynthate accumulation and distribution, regrowth after defoliation,

and forage nutritive value are not known. Opportune applications of

water may help maintain favorable plant water status for growth and

increase overall forage yields through increased numbers of cuttings,

more rapid regrowth, and improved crop persistence. It is important to

quantify the specific physiological responses as well as whole plant

responses of tropical forage legumes to drying soil conditions in order

to gain a better understanding of the mechanisms impacting

photoassimilate accumulation and distribution. This type of information

can facilitate genetic manipulation to enhance yield and plant

persistence, as well as reveal management procedures necessary to

maintain vigorous stands and high forage yields throughout the growing

season.

The research undertaken here examined some of the physiological,

morphological, and forage quality responses of Florigraze perennial

peanut to different levels of water management. The specific objectives

of the study were as follows:

1) Characterize the temporal changes in photosynthate and nitrogen

accumulation and distribution in response to four water management

treatments.

2) Determine the responses of and relationships among leaf stomatal

activity, leaf carbon exchange rate, leaf nitrogen concentration,

and leaf and soil water status during water deficits.

3) Characterize the seasonal patterns of in vitro organic matter

digestibility and nitrogen concentration of the plant and its














INTRODUCTION
Forage production by legumes in the subtropics is often reduced by

periods of drought. Extended periods with little or no rainfall are

common and often limit forage productivity at times of the year when

temperatures are otherwise favorable for plant growth. Since periodic

loss of the above-ground portion of a forage occurs each time herbage is

removed, either by livestock or by mechanical harvesting, proper

management of legumes during these periods of drought is essential to

provide a relatively constant amount of high quality feed.

Understanding the physiological impact of drying atmospheric and soil

conditions on plant growth is important to facilitate managing legumes

and other forages for optimum production and nutritive value as well as

plant persistence.

Perennial peanut (Arachis glabrata Benth.) is a deep rooted,

relatively drought tolerant tropical forage legume which is well-adapted

to the environmental conditions of Florida. Perennial peanut is

presently being propagated on farms in Florida as the cultivars

'Florigraze' and 'Arbrook' to increase the availability of rhizomes for

propagation. The cultivar Florigraze is adapted to well-drained, sandy

soils and grows vigorously alone or in association with a grass (Prine

et al., 1981). There is very little information available concerning

the physiological response of perennial peanut to different levels of

water management. Effects of drying soil on physiological processes,










2

photosynthate accumulation and distribution, regrowth after defoliation,

and forage nutritive value are not known. Opportune applications of

water may help maintain favorable plant water status for growth and

increase overall forage yields through increased numbers of cuttings,

more rapid regrowth, and improved crop persistence. It is important to

quantify the specific physiological responses as well as whole plant

responses of tropical forage legumes to drying soil conditions in order

to gain a better understanding of the mechanisms impacting

photoassimilate accumulation and distribution. This type of information

can facilitate genetic manipulation to enhance yield and plant

persistence, as well as reveal management procedures necessary to

maintain vigorous stands and high forage yields throughout the growing

season.

The research undertaken here examined some of the physiological,

morphological, and forage quality responses of Florigraze perennial

peanut to different levels of water management. The specific objectives

of the study were as follows:

1) Characterize the temporal changes in photosynthate and nitrogen

accumulation and distribution in response to four water management

treatments.

2) Determine the responses of and relationships among leaf stomatal

activity, leaf carbon exchange rate, leaf nitrogen concentration,

and leaf and soil water status during water deficits.

3) Characterize the seasonal patterns of in vitro organic matter

digestibility and nitrogen concentration of the plant and its








3

yield components (leaves and stems) in response to the water

management treatments.

Photosynthate accumulation and distribution in a field planting of

perennial peanut were monitored during two growing seasons (1987 and

1988) under four levels of water management. The effects of water

management on shoot and rhizome mass were examined along with total

nonstructural carbohydrate concentrations in different plant parts.

Also, leaf stomatal conductance, leaf carbon exchange rate, and leaf

nitrogen concentration were measured. Soil water status and leaf water

potential components were characterized throughout both seasons in order

to define the relationships among the plant responses and plant and soil

water status. Regrowth following clipping as affected by water

management was also evaluated in terms of stem numbers, rate of stem

appearance, and canopy height during one season (1987).














LITERATURE REVIEW

Very limited information is available on the physiological,

morphological and nutritive value responses of perennial peanut

subjected to water deficits; therefore, most of the information

presented in this section was compiled from research conducted on other

forage and row crops.

Photosvnthate Distribution in Plants

Nonstructural carbohydrate root and crown reserves as well as the

presence of photosynthetically active tissue are important for forage

regrowth and maintenance. Smith (1962) studied root reserves in

birdsfoot trefoil (Lotus corniculatus L.), alfalfa (Medicago sativa L.),

and red clover (Trifolium pratense L.) under several cutting treatments.

He observed that root reserves were maintained at a relatively high

level in alfalfa and red clover under three and two harvests,

respectively, per season as well as in the uncut treatment. Birdsfoot

trefoil maintained relatively low concentrations of root reserves

throughout the season across all cutting treatments until fall. Such

results suggest that trefoil depends on current photosynthesis from

existing leaf area in order to regrow and persist throughout the season.

Similar results have been obtained by Beardsley and Anderson (1960) who

compared alfalfa and trefoil. They found that birdsfoot trefoil could

withstand frequent but not complete defoliation, and that alfalfa could

withstand complete defoliation but on a relatively infrequent cutting










schedule so that nonstructural carbohydrate reserves could be restored

during the regrowth period. Nelson and Smith (1968) also observed

higher carbohydrate root reserves in alfalfa than in birdsfoot trefoil.

During most of the season alfalfa had a higher crop growth rate,

produced more dry matter, and had a higher leaf area index than trefoil.

Feltner and Massengale (1965) studied the summer reduction in growth of

alfalfa and found that day/night temperatures affected accumulation of

soluble carbohydrates in the roots and that the reserves were reduced by

frequent cutting. Chatterton et al. (1977) reported that alfalfa clones

tolerant to relatively frequent clipping were also the clones highest in

soluble carbohydrate root reserves. Maintenance of sufficient

carbohydrate root reserves and/or residual light intercepting foliage

after defoliation is important for rapid regrowth and crop persistence

of forage legumes.

Periods of drought stress may also affect the partitioning of

photosynthate. In forage plants, the above-ground portion of the plant

is of primary agronomic importance and drought may lower the amount of

foliage produced either by stimulating root growth at the expense of the

shoot, by stimulating leaf senescence, or by causing an overall

reduction in plant growth (shoots and roots). Finn and Brun (1980) and

Silvius et al. (1977) both found that in drought stressed soybean

(Glycine max L.) more 14C was partitioned to the roots at the expense of

the leaves than in the unstressed plants. However, Robertson et al.

(1980) reported that total root length and distribution in annual peanut

(Arachis hypogaea L.) was unaffected by water management on a fine sand










in Florida although pod yields increased with increasing amounts of

irrigation.

Nonstructural carbohydrate concentrations in various organs of the

plant can be used as an indicator of the temporal distribution of

photosynthate. Saldivar (1983) observed that soluble carbohydrate

concentrations in perennial peanut rhizomes were higher in a "dry" year

than in a "normal" year, although soil water status was not monitored

during the study. He also found that within a season, rhizome soluble

carbohydrate concentrations decreased during the hot, summer months but

increased in the fall and were relatively high in the spring. Saldivar

(1983) speculated that since growth was slower during the cooler parts

of the season, demands made on the reserves were not as great thus

allowing them to accumulate.

In a preliminary study of perennial peanut with and without

irrigation, it was found that forage yield was depressed 45% under the

nonirrigated treatment but canopy photosynthesis in the water-stressed

plots was 90% of that in the well-watered plots (Albrecht et al., 1989).

They also observed a 25% increase in total nonstructural carbohydrate

concentration in the rhizomes of the water-stressed plants compared to

well-watered plants indicating that carbon assimilated under water

stress was accumulated underground in rhizomes rather than utilized for

shoot growth (K.A. Albrecht, personal communication). Hall et al.

(1988) reported similar results for alfalfa. Water-stressed alfalfa had

greater nonstructural carbohydrate concentrations in the roots as

compared to well-watered plants apparently due to increased

photoassimilate partitioning to the roots during water stress.








7

Assimilate partitioning may also be related to the water status of

the plant tissue receiving the assimilate. Lang and Thorpe (1986)

showed that bathing the roots of young Phaseolus plants in solutions of

different osmolarity caused assimilate to be distributed toward areas of

more negative water potential. However, Robinson et al. (1983) studied

assimilate distribution in Gladiolus and found that even a slight water

deficit caused a decrease of assimilate deposition in the inflorescence

and increased deposition of assimilates in the corm. The corm had

consistently higher water potential than the inflorescence under water

stress. They proposed that since turgor dependent growth was slowed in

the inflorescence, more photosynthate was partitioned to the corm,

allowing it to continue growth. Schulze (1986b) reviewed whole-plant

responses to drought and stated that plant water relations interact with

carbon partitioning, but the mechanisms controlling the interaction are

obscure. He further proposed that the internal water status of a plant

probably does not affect carbon assimilation and distribution

extensively unless the plant fails to maintain water flow through the

leaf epidermis and root tip.

Physiological Responses to Water Deficits

Plants use different strategies when responding to drought and

certainly plant species differ in their ability to cope with water

stress. Plants usually respond by some mechanism which allows them to

maintain favorable water status needed for growth or survival under

drying atmospheric and soil conditions.








8

Effect of Water Deficits on Stomata and Leaf Carbon Exchange

Plants respond to water deficits by decreasing stomatal aperture.

Stomatal closure is usually related to drying soil conditions and has

been commonly related to changes in leaf water status. However, recent

evidence strongly suggests that the mechanism by which stomata respond

to water deficits is not entirely understood. Bates and Hall (1981) and

Osonubi (1985) found stomatal closure in cowpea (Vigna unguiculata L.)

to be associated with changes in soil water status rather than bulk leaf

water status. Blackman and Davies (1985) have also shown a direct

effect of soil water status (without changes in leaf water status) on

stomatal conductance in a split root experiment with maize (Zea mays

L.). Turner et al. (1985) examined the relationship between stomatal

conductance, transpiration and photosynthetic rate as related to leaf

water potential in sunflower (Helianthus annuus L.) but found no unique

relationship. Only when about two-thirds of the extractable soil water

had been depleted did stomatal conductance and net photosynthesis

decline, indicating that soil water status, not leaf water status, was

the dominant effect on stomatal behavior and photosynthesis.

Bennett et al. (1987), working with field grown soybean and maize,

examined the response of stomatal conductance to leaf water potential,

leaf turgor potential and relative water content. No significant

relationships were found when data for well-irrigated and mildly

stressed plants were analyzed; however, when data from the more severely

stressed plants were included in the analysis there were significant

relationships. Such data indicate that initial stomatal response was

not related to leaf water status. Carter and Sheaffer (1983a,b) studied










the effect of water stress on alfalfa and found stomatal conductance to

be lower at midday and higher in the morning and late afternoon in

plants under moderate water stress. However, with severe water stress,

stomatal conductance was low all day and canopy temperature was up to

8.5 oC higher than for well-watered plants, indicating that stomatal

closure had significantly reduced transpiration. Sharratt et al. (1983)

reported a 0.2 mm h" reduction in evapotranspiration, a 0.7 MPa

reduction in leaf water potential, and a 2 C canopy temperature

increase in alfalfa when extractable soil water had declined to 25% in a

water-stressed treatment.

Muchow et al. (1986) examined changes in stomatal conductance as

well as leaf area development and leaf N content of field-grown soybean

in response to soil water deficits imposed after canopy closure. Mild

water deficits were found to decrease leaf area development more than

number of leaves produced. Biomass accumulation efficiency was

decreased; however, interception of photosynthetically active radiation

was unaffected. The relative decrease in biomass accumulation was

associated with lowered stomatal conductance of the leaves in response

to drying soil. Schulze (1986a), in a review on carbon dioxide and

water vapor exchange in response to drought in the atmosphere and in the

soil, suggested that stomata may respond in three ways to drought: 1) a

direct response of stomata to humidity that is a property of the

epidermis, 2) a direct response from root "signals" that are related to

root metabolic activity in a drying soil, and 3) a response to leaf

apoplastic abscisic acid (ABA) released to the guard cells through some

unknown process that could be related to the first two.








10

Abscisic acid is now thought to be the principal hormone actually

controlling stomatal action. Zhang et al. (1987), using a split root

system with Commelina communis (L.), found that as the soil dried there

was an increased level of ABA in leaf epidermal tissue which coincided

with increased levels found in roots located in dry soil. Such results

led them to conclude that ABA, which controlled leaf stomatal response,

apparently originated in the roots. However, Munns and King (1988)

found that ABA was not the only stomatal inhibitor present in the

transpiration stream of water-stressed wheat (Triticum aestivum L.).

Even after extracting ABA from plant exudate by passing it through an

immunoaffinity column, the exudate still caused stomatal closure when

"fed" to detached leaves from nonstressed wheat plants.

Leaf carbon exchange is usually reduced as stomata close when

plants are subjected to drought. Silvius et al. (1977) examined

drought-stressed soybeans and reported that carbon exchange rate (CER)

was decreased as stomatal resistance increased, but that CER increased

to prestressed levels upon rewatering if leaf water potential did not

decline below -2.1 MPa. However, when leaf water potential declined

below -2.1 MPa, CER did not fully recover. Albrecht et al. (1984) also

reported decreased canopy CER in soybeans with increased soil drying.

Upon rewatering, stomatal conductance and leaf water potential returned

to normal, but canopy CER remained below the control rate. They

attributed such results primarily to the abscission of leaves during

stress and thus a reduction in leaf area and photosynthetic capacity, or

possibly to reduced carbon fixation capacity of the recovered leaves.

Nagarajah and Schuize (1983) found that carbon assimilation of cowpea








11

decreased less rapidly than stomatal conductance as the soil dried, and

that leaf growth continued even though stomata began to close when

available soil water was depleted by 45%. Kippers et al. (1988)

examined the relationship between stomatal conductance and intercellular

CO2 concentration (Ci) in cowpea. No unique relationship between Ci or

the ratio of ambient CO, concentration (Ca) to Ci and stomatal

conductance was found. As the soil dried, leaf carbon exchange rate was

affected less than stomatal conductance. They concluded that stomatal

conductance and leaf photosynthesis responded independently. Lopez et

al. (1988) found similar results for pigeonpea (Cajanus cajan L.) and

concluded that stomatal conductance was more sensitive to water stress

than leaf photosynthesis. Nicolodi et al. (1988) compared irrigated and

nonirrigated alfalfa and found a lower discrimination against '3C in the

nonirrigated treatment indicating a reduction in CO, diffusion into the

leaf. However, they also found that reduced photosynthesis could not

always be attributed to reduced stomatal conductance. The ratio Ci/Ca

was actually higher in water-stressed plants on several occasions

indicating that a nonstomatal mechanism (ie. reduced photosynthetic

capacity of the mesophyll) was contributing to reduced photosynthesis in

water-stressed alfalfa. They concluded that both stomatal and

nonstomatal mechanisms contributed equally to reduced photosynthesis in

droughted alfalfa.

Whole Plant Responses

Several studies have been conducted to examine the response of

tropical forage species to water deficits. Wilson et al. (1980)

compared three tropical grasses and the tropical legume siratro








12

(Macroptilium atropurpureum) in a semi-arid field environment. The

grass species were shown to adjust osmotically while siratro did not.

Fisher and Ludlow (1982) also studied the water relations and adaptive

characteristics of siratro under dry tropical conditions. They found

that siratro maintained favorable leaf water status and avoided plant

water deficits by three main mechanisms: 1) sensitive stomatal control

of water loss, 2) paraheliotropic leaf movements to minimize radiation

received, and 3) abscission of leaves to reduce transpiring tissue.

However, they concluded that although those mechanisms allow siratro to

temporarily avoid severe leaf water deficits, it cannot withstand very

negative leaf water potentials which would make it susceptible to

extended dry periods.

Ludlow et al. (1983) compared siratro to several Centrosema

species, and found that unlike siratro the Centrosema species do adjust

osmotically in order to maintain leaf turgor. Shackel and Hall (1983)

also found species differences when comparing sorghum (Sorghum bicolor

L. Moench) and cowpea under field conditions. Sorghum was shown to

tolerate leaf water deficits through osmotic adjustment, whereas cowpea

possessed an avoidance mechanism by maintaining favorable leaf water

status mainly thorough stomatal control and deep rooting but did not

osmotically adjust. Sherriff and Ludlow (1984) studied buffel grass

(Cenchrus ciliaris L.) and siratro in a mixed sward under drought stress

and found that siratro had a deeper root system than buffel grass, and

buffel grass had lower daily mean water potentials than siratro. They

concluded that buffel grass survived drought because of its inherent

desiccation tolerance osmoticc adjustment), whereas siratro was able to








13

survive because it had deeper roots and was not competing with buffel

grass for water.

Sheriff et al. (1986) studied the physiological responses of two

forage legumes of similar morphology, siratro and Galactia striata,

under drought and found that rooting depth and hydraulic conductance of

the plant-soil system were as important as leaf physiological responses

in their adaptation to drought. Siratro maintained higher leaf water

potentials than G. striata but showed no osmotic adjustment. Even

though calculated turgor was higher in siratro, leaf expansion was more

affected by drought than in G. striata. They speculated that this may

have been due to higher partitioning of assimilates to roots at the

expense of leaves in siratro. The maintenance of higher leaf

conductance and carbon exchange in siratro as drought progressed was

attributed to its relatively deep root system. Cortes and Sinclair

(1986) found similar results when comparing two soybean cultivars that

differed in their response to drought. The cultivar least affected by

drought developed a deeper root system under water stress thus

maintaining a supply of water to plant tissues. Markhart (1985) also

compared the relatively drought tolerant tepary bean (Phaseolus

acutifolius Gray) with common bean (P. vulgaris L.) and found that

tepary bean had more sensitive stomatal control of water loss and deeper

roots than common bean.

It has been speculated that annual peanut may possess unique

physiological traits that enhance its ability to grow under drought

conditions. However, Bhagsari et al. (1976) found that photosynthesis

in water-stressed annual peanut was controlled in a manner similar to








14

that in other crop species. Bennett et al. (1981, 1984) also observed

that the relationships among leaf water potential components, stomatal

resistance, and leaf water content were similar in peanut as in other

agronomic crops and the relationships observed probably conferred no

unique drought resistance mechanisms to peanut. They concluded that the

ability of peanut to maintain favorable tissue water status by

extracting water deep in the soil profile and its ability to maintain

developmental plasticity (indeterminate growth habit) may allow

sensitive growth processes to escape temporary water deficits.

Certain plant characteristics can be of direct benefit in enabling

crops to withstand water deficits; however, the direct benefit of some

other mechanisms that allow plants to tolerate or escape water deficits

is not always clear. Sinclair and Ludlow (1986) examined the water

balance of four tropical grain legumes during a soil drying cycle. They

found differences in dehydration tolerance among the four legumes based

on relative water content and ranked them as follows: pigeonpea > cowpea

> mungbean (Vigna mungo L.) > soybean. The ranking based on epidermal

conductance was the inverse; ie., soybean > mungbean > cowpea >

pigeonpea, indicating that low epidermal conductance of pigeonpea helped

the plant retard tissue desiccation thus giving it an adaptive advantage

in order to withstand temporary soil water deficits. Such mechanisms

would likely be of benefit only under severe water deficits, but might

be important for crop survival under those environments. McCree and

Richardson (1987) studied the effect of water stress on carbon gain in

cowpea and sugarbeet (Beta vulgaris L.) plants. Cowpea tended to avoid

leaf water deficits through stomatal closure and water regulation. In








15

contrast, sugarbeet allowed stomata to remain open through osmotic

adjustment, theoretically giving it a carbon gain advantage. However,

they found similar carbon gains for the two species under the same

environmental conditions and concluded that there was no advantage of

osmotic adjustment over stomatal closure in plant response to soil water

deficits.

Effect of Water Stress on Forage Nutritive Value

Water deficits frequently affect the nutritive value of forages.

Drought stress usually causes an increase in digestibility of alfalfa

due to shorter internodes which result in decreased stem to leaf ratios

in the stressed forage (Brown and Tanner, 1983; Carter and Sheaffer,

1983a; Jensen et al., 1988). Plant crude protein concentration is

generally only slightly affected by water stress; however, Donovan and

Meek (1983) found that the protein concentration of alfalfa was higher

in the spring and fall under dry as compared to well-irrigated

conditions.

Conflicting results concerning the effect of water-stress on

forage digestibility exist in the literature. Wilson (1983b) examined

three tropical grasses subjected to water deficits and found that both

stem and leaf digestibility were similar to or higher than those from

unstressed herbage. However, Pitman et al. (1981, 1983) reported a

decrease in forage quality associated with drought in tropical grasses

as did Wilson (1983a) with the perennial legume siratro. Pitman et al.

(1983) found that in water-stressed Kleingrass (Panicum coloratum L.) an

increased proportion of cell wall components and increased lignification

could account for at least part of the decreased digestibility








16

associated with water stress. Wilson and Ng (1975) also found decreased

cell wall digestibility with increasing water stress in Panicum maximum

var. trichoglume. Pitman and Holt (1982) studied the effects of

temperature, rainfall, relative humidity and daylength on three

perennial, warm season grasses in the Southwest but found only poor

relationships between these environmental factors and forage

digestibility.

It has also been speculated that physiological maturity may play a

role in accounting for differences detected in forage nutritive value

under water stress. Wilson (1983a) observed that dry weather in the

spring slowed stem development of the three grasses tested and reported

that digestibility was significantly higher than in the well-irrigated

treatment. Halim et al. (1989) studied water stress effects on alfalfa

forage nutritive value after adjusting for maturity differences, but

found that maturity differences among water treatments accounted for

only part of the differences.

Much research has focused on only one or two aspects of the many

physiological processes impacting on forage management decisions during

drought. Many experiments have examined stomatal control factors and

photosynthesis, mostly in temperate forages and other agronomic crops,

with limited information available for tropical legumes. Also,

characterization of nonstructural carbohydrate distribution in forages

(mostly temperate species) has been done in response to cutting height

and frequency but not to drought. No relationships between these

intimately related parameters have been examined on tropical species

which are void of sexually reproductive sinks. The tropical nature and








17

lack of a seed sink in species such as perennial peanut may influence

assimilate accumulation, partitioning and remobilization, as well as

plant productivity, senescence, and persistence.

In two reviews on crop water deficits and plant adaptations to

stress, Turner (1986a,b) points out the need to more fully understand

the complex relationships that exist between leaf stomatal response,

root water relations, and drying soil conditions. Plant roots seem to

mediate the relationships by chemical and/or hydraulic responses to dry

soil. Once these relationships begin to be characterized and understood

then the possibility for beneficial genetic manipulation exists (Boyer,

1982). Definitive trait(s) need to be identified which can be selected

for and incorporated into germplasm through breeding programs to make

crop plants better adapted to varying environments (Turner, 1986b).














MATERIALS AND METHODS

This study was conducted at the Irrigation Research and Education

Park located on the Agronomy Farm at the University of Florida in

Gainesville. The experimental site was on a Kendrick fine sand (a

member of the loamy, siliceous, hyperthermic family of Arenic

Paleudults), a deep, well-drained to excessively well-drained soil.

Establishment of Perennial Peanut (1986)

Soil test results at the initiation of the study indicated a pH of

5.5 to 6.0, relatively high levels of available phosphorous, and low

available potassium. A preplant application of 730 kg ha" of 0-5-25 (N-

P205-IO0) containing 70, 8, 5, 2, and 91 g kg1 of the micronutrients Mg,

Mn, Zn, B, and S, respectively, was made in November 1985. On 4

February and 19 March 1986, pelletized dolomitic limestone was applied

at the rate of 1000 kg ha'. A fall maintenance application of 220 kg

ha" of 0-0-61 (N-P20,-K20) fertilizer was also applied on 24 Sep. 1986

after establishment of the crop.

Perennial peanut cv. 'Florigraze' rhizome material, which had been

inoculated with peanut Bradyrhizobium at 25 kg ha', was planted on 30

and 31 January 1986, in 15 cm rows at the rate of 8.7 m3 ha". To

encourage rapid establishment of the crop, a total of 198 mm of water

was applied to all plots during 1986 with 114 mm being applied in April

and May. Rainfall and irrigation amounts are shown in Table A-i. Weeds

were controlled with appropriate herbicide treatments and by hand-








19

weeding during the 1986 season. 'Premerge' (dinoseb, 2-sec butyl 4,6-

dinitrophenol) and 'Lasso' (alachlor, 2-chloro-2'-6'-diethyl-N-

(methoxymethyl)acetanilide) were applied on 19 March for preemergence

control of annual grasses and broadleaf weeds. 'Fusilade' (fluazifop-

butyl, butyl(RS)-2-[4-[[5-(trifluoromethyl)-2-pyridinyl]oxy]phenoxy]

propanoate) was applied on 2 May for post-emergence grassy weed control,

and yellow nutsedge (Cyperus esculentus L.) was controlled with two

post-emergence applications of 'Basagran' (bentazon, 3-isopropyl-1H-

2,1,3-benzothiadiazin-(4)-3H-one 2,2-dioxide) on 23 June and 22 July.

By the end of July, solid rows of vigorous plants had developed, and

abundant rhizomes were detected spreading underground between the rows.

All plots were cut to a 5-cm stubble height for the first time after

planting on 30 July in anticipation that removal of shoot apical

meristems would promote sprouting from rhizomes. Subsequent sprouting

did occur from rhizomes, and plants began to cover the row middles. Six

plots were harvested to a 5-cm stubble height on 31 October to estimate

establishment season forage yields. Forage dry matter yields for each

harvested area and the mean yield are given in Table A-2.

Treatments and Data Collection (1987 and 1988)

Four water management treatments replicated three times were

imposed during the study, resulting in a total of 12 plots situated

within an existing 0.25-ha field area which was equipped with a solid

set irrigation system. The plots were arranged in a randomized complete

block design with blocking based on depth of soil to the underlying clay

layer. Depth of soil to the clay layer ranged from 950-1050 mm in block

1 to 1200-1300 mm in block 3. Each plot measured 13.7 x 6.9 m. All








20
data were analyzed using analysis of variance and Duncan's New Multiple

Range test available through Statistical Analysis System (SAS Institute,

1985).

Maintenance fertilizer applications of 220 kg ha' of 0-0-61 (N-

P205-K(0) were applied on 4 April and 22 June 1987; and on 12 February

and 14 April 1988. Weeds were controlled during the 1987 and 1988

seasons principally by hand-weeding although several herbicides were

also utilized. Fusilade was applied on 10 July 1987 for postemergence

grassy weed control, mainly common bermudagrass (Cynodon dactylon L.).

In 1988, early season broadleaf weeds were controlled postemergence with

'2,4-0' (2,4-Dichlorophenoxyacetic acid) applied on 1 March, and summer

annual grassy and broadleaf weeds were controlled with a preemergence

application of 'Prowl' (N-(1-ethylpropyl)3,4-dimethyl-2,6-dinitro

benzenamine) on 4 March.

Water Management Treatments

Four water management treatments were imposed on perennial peanut

in both 1987 and 1988. In 1987, the treatments were as follows: 1)

well-watered (WW), irrigated so that plants were never subjected to

plant water deficits; 2) cyclical stress (CS1) with periodic water

recharge of the soil profile; 3) water withheld until stomata began to

close, then light frequent irrigation to replenish daily

evapotranspiration (SP); and 4) rainfed (RF). Treatment WW received 19

mm of water every 2 to 4 days as needed during periods of dry weather

throughout the growing season. The CS1 treatment received 38 to 45 mm

of water at approximately 10-day intervals to provide complete soil

water recharge during the first harvest cycle which ended on 11 June.








21
For the remainder of the season, soil water recharge for the CS1

treatment was every 2 weeks during periods of dry weather. As part of

another study which was corollary to the study reported herein, a

Speedling irrigation system was used to apply daily replacement of

evapotranspiration on the SP treatment only during the first harvest

cycle (treatment no. 3). For purposes of the corollary study, daily

midday measurements of stomatal conductance (g) were made on the WW, SP,

and RF treatments during the first harvest cycle (25 April through 11

June). Light, frequent irrigation (often times daily) was used to

maintain g in the SP treatment at a level intermediate to the WW and RF

treatments. This treatment was suspended after the first cycle, and

thereafter received the same irrigation as the CS1 treatment during the

rest of the season. The RF treatment was only irrigated to initiate

uniform crop growth in April and was never irrigated after that time.

Rainfall and irrigation data are summarized in Table A-3 for the 1987

growing season.

Water management treatments in 1988 were as follows: 1) well-

watered (WW), received approximately 19 mm of irrigation every 2 to 4

days during periods without significant rainfall; 2) 2-week cyclical

stress (CS1), received approximately 38 to 46 mm irrigation to achieve

complete soil water recharge every 14 days during periods without

rainfall; 3) 3-week cyclical stress (CS2), received approximately 38 mm

of water to achieve complete soil water recharge every 21 days during

periods without rainfall; and 4) rainfed (RF), no irrigation. The water

management treatments were the same in both years except for treatment

no. 3. The SP treatment of 1987 was not repeated in 1988 due to data








22

which suggested this treatment did not differ from the WW treatment

during the first harvest cycle in 1987. Since the SP treatment was

assumed similar to both the WW and CS1 treatments at the end of 1987

(data presented in following sections), replacement of the SP treatment

with a 3-week cyclical stress treatment in 1988 was considered to be a

better use of resources. Table A-4 presents rainfall and irrigation

data for the 1988 growing season.

Gravimetric Soil Water Content

Gravimetric soil water content was monitored at the end of several

cyclical stress periods in order to assess the soil water deficits

created by the imposed water management treatments. Soil core samples

were taken on 31 May, 10 June and 30 Oct. 1987; and on 2 and 8 June, and

2 Nov. 1988. Soil samples were extracted at various soil depths by

utilizing a 5-cm diameter aluminum tube. Samples were collected at

150-mm depth increments to a total depth of 1200 mm or until

encountering the clay layer. The soil samples were sealed in plastic

bags until wet soil weight could be determined in the laboratory. After

weighing, the bags were opened and the samples were dried at 95"C in a

forced air oven for at least 24 h. Percent water in the soil was

calculated on a dry weight basis as follows:

% gravimetric soil moisture = ((wet wt. dry wt.)/dry wt.) 100.

Determination of Crop Dry Matter Accumulation and Distribution

Temporal changes in photosynthate distribution as influenced by

water management were monitored throughout the growing seasons in 1987

and 1988 by measuring shoot and rhizome dry matter (DM) as well as total

nonstructural carbohydrate (TNC) distribution in stems, leaves, and








23

rhizomes. On sampling dates, two 0.5 m2 quadrats from each plot were

clipped to ground level to determine above-ground DM, and four soil core

samples (9.5 cm in diameter) were extracted to a depth of 20 cm from

each quadrat to determine rhizome DM. Soil samples were placed on a

fine screen and washed with a high pressure water stream from a spray

nozzle to separate rhizomes from the soil. Total nonstructural

carbohydrate concentrations were determined on the subsamples of leaf,

stem, and rhizome tissue.

Beltranena (1980) found that the highest seasonal forage DM yields

of Florigraze perennial peanut were obtained at 6 and 8 week cutting

intervals in 1978 and 1979, respectively. However, seasonal forage

yields were reduced at 2 and 4 week clipping intervals. Based on those

data, four harvest cycles of 6 to 7 week intervals were utilized during

both years of the study. In 1987, above-ground and rhizome samples were

taken on 15 May, 1 June, 11 June (end of first harvest cycle), 23 July

(end of second harvest cycle), 8 September (end of third harvest cycle),

and 2 November (end of fourth harvest cycle). In addition to samples

for total DM (clipped to ground level), forage samples (harvested from

an area 0.86 x 3.05 m clipped to a 5-cm stubble height in order to

estimate forage available for grazing) were taken on all dates except 15

May.

Total above-ground and rhizome DM samples were taken in 1988 on 19

May, 2 June, 8 June (end of first harvest cycle), 25 July (end of second

harvest cycle), 12 September (end of third harvest cycle), and 1

November (end of fourth harvest cycle). Forage samples in 1988 were

harvested as previously described at the end of each harvest cycle








24

except for the 1 November sampling date when insufficient plant material

was available for harvesting. All plots were staged (clipped) to a 5-cm

stubble height at the end of each harvest cycle and clipped forage was

removed from the plot area.

Subsamples were taken from the above-ground DM samples to

determine leaf and stem fractions, as well as leaf area index of the

crop at the time of harvest. Fresh subsamples were also taken from the

forage samples in the field. All plant samples were dried at least 72 h

at 60C in forced-air ovens before determining dry weights. Subsamples

and rhizome samples were then ground to pass a 1-mm screen and subjected

to various laboratory analyses (see section on laboratory analyses).

Root length density expressed as cm root per cm3 soil was measured

on 19 June 1987. The modified line intersect method of estimating root

length as described by Tennant (1975) was utilized to determine root

length. Two 5-cm diameter soil cores were collected at 150-mm depth

increments to a total depth of 1200 mm. Rhizomes were separated from

the 0-150 mm increment, and their length was determined and reported

independently from roots at that depth.

Determination of Crop Reqrowth After ClippinQ

Regrowth from the staging height of 5 cm was monitored in 220 cm2

areas at two locations in each plot and described in terms of shoot

origin (rhizome or stubble) and shoot number, plant height, and light

interception during both the second (12 June 23 July) and the third

(24 July 8 September) harvest cycles in 1987.










Physiological Measurements

Single leaf carbon exchange rate (A), stomatal conductance, and

transpiration were determined with a LI-COR 6200 Portable Photosynthesis

System equipped with a 0.25-1 chamber during periods of water stress in

the spring and fall. Two pairs of leaflets from two different locations

in each plot were measured periodically at midday as plant water

deficits developed in the different treatments. Fully illuminated,

recently expanded leaves at the top of the canopy were selected for

measurement. Two of the four leaflets were removed to facilitate

insertion and positioning of the two remaining leaflets in the leaf

chamber. Measurements were taken on 18, 26 and 29 May; 8 and 10 June;

27 and 28 August; and 7, 22 and 29 October during 1987. In 1988

measurements were taken on 12, 16, 20 and 31 May; 1 June; 8 July; and 17

and 24 October. Carbon exchange data were always taken under full sun

between 1030 and 1500 h EST.

Plant water potential was also monitored during periods of plant

water stress on 8 and 10 June, 28 August, and 29 October in 1987; and on

1 June and 8 July in 1988. Two to four leaflets were collected from

each plot and quickly enclosed in thermocouple psychrometer chambers.

The psychrometers were transferred to the laboratory and placed into a

thermostatically controlled water bath at 300C and allowed to

equilibrate for 4 hours. A dewpoint microvoltmeter was utilized to pass

a 15 s cooling current through the thermocouple and to monitor output

from the psychrometers. The output was recorded with a chart recorder

and data taken from the output chart were used to calculate total leaf

water potential (0L) of each sample. Each psychrometer had been








26
previously calibrated using known concentrations of NaC1 solutions to

develop a regression equation of water potential versus psychrometer

output. After OL determination, psychrometer units containing leaf

material were placed in a freezer at -20C for at least 12 h to rupture

the cells and eliminate cell turgor. After removal from the freezer,

tissue was re-equilibrated in the water bath at 30C, and osmotic

potential (0) was determined by the method previously described. Leaf

turgor (0,) was calculated by subtracting I, from OL.

Leaf Carbon Exchange Rate and Leaf Nitrogen Concentration

Leaflets of Florigraze perennial peanut from the WW and CS1

treatments on which A had been measured were collected, dried and pooled

together according to the rate of A. Pooled leaflets were ground to

pass a 1 mm screen and then analyzed for nitrogen (N) concentration. It

was necessary to pool several leaflets to obtain adequate leaf mass for

N determination because of their small size. Leaflets from 1987 and

1988 were kept separate. Six intervals of A were chosen to group

leaflets of similar carbon exchange rates for N analysis as follows: 1)

A less than or equal to 14 pmols CO2 m2 s', 2) A from 15 to 17 pimols CO2
m-2 s1, 3) A from 18 to 20 jmols CO2 m2 s-, 4) A from 21 to 23 pimols CO2

m' s 5) A from 24 to 26 pmols CO2 m2 s', and 6) A greater than or

equal to 27 pimols CO2 m2 s Such pooling of leaflets provided enough

leaf mass in 1987 to analyze one sample for N concentration at each of

the levels 2 through 6 in the WW treatment; and 1, 3, 4, 5, and 6 for

the CS1 treatment. In 1988 sufficient plant material was collected to

analyze one sample for N concentration from each of the levels 1 through

5 for the WW treatment, and 2 through 4 for the CS1 treatment.








27
Therefore, a total of 18 composite leaflet samples were obtained for N

analysis on the cultivar Florigraze.

An establishing stand of well-watered 'Arbrook' perennial peanut

was also utilized in order to obtain leaflets of varying carbon exchange

rates corresponding to varying leaf N concentrations. Arbrook perennial

peanut rhizomes were planted in March 1988 in plots adjoining the

Florigraze cultivar, and fertilizer N treatment (with and without

ammonium nitrate) was imposed as a split-plot feature in each plot.

Nitrogen was applied three times in 1988 at the rate of 78, 67 and 61 kg

ha" on 26 May, 6 July and 3 August, respectively. Plant leaves from the

no N fertilizer plots had a pale green ("yellow") color as compared to

leaves on plants that received fertilizer N. Two fully expanded

"yellow" and "green" leaflets at the top of the canopy from the no N and

N fertilized plots, respectively, were used to measure A utilizing the

LI-COR 6200 Portable Photosynthesis System as previously described. Two

to three leaflet pairs in each sub-plot were subjected to measurement

and collected on 17, 22, 23, and 24 June; 7, 8, 13 and 15 July; and 2,

18 and 29 August. Leaflets of similar carbon exchange rates were pooled

together for N analysis as previously described for the Florigraze

cultivar. Samples from June, July and August were kept separate, and

"yellow" and "green" leaflets were pooled independently of each other.

In June, "yellow" and "green" leaflets were obtained for N analysis at

each of the levels 1 through 5 (as described above), and 3 through 6,

respectively. In July, "yellow" and "green" leaflets were grouped into

each of the levels 1 through 5, and 1 through 6, respectively; and

leaflets from August into each of the levels 4 through 6 for both








28

"yellow" and "green". Therefore, a total of 26 composite Arbrook

leaflet samples were obtained for N analysis.

Laboratory Analyses

Total nonstructural carbohydrate and N concentrations were

determined on leaf and stem components of above-ground DM subsamples as

well as on rhizome DM subsamples. Nitrogen concentrations were also

determined on the forage subsamples and composite samples of leaflets

used for measurement of A. Total nonstructural carbohydrate analyses

were performed by enzymatic digestion according to procedures described

by Christiansen (1982) in which nonstructural carbohydrate components

(ie. starch and sucrose) were digested to glucose, with subsequent

colorimetric determination of glucose. The data were compared to a

standard curve which was developed for each individual set of samples in

order to quantify the amount of TNC present. For N analysis, samples

were digested using a modification of the Kjeldahl aluminum block

digestion procedure of Gallaher et al. (1975). Ammonia in the digestate

was determined by semiautomated colorimetry (Hambleton, 1977).

Leaf and stem fractions of the above-ground DM subsamples and the

forage subsamples were also analyzed for in vitro digestion. In vitro

organic matter digestibility (IVOMD) was determined by the two-stage

technique of in vitro digestion (Tilley and Terry, 1963) as modified by

Moore and Mott (1974).














RESULTS AND DISCUSSION
Rainfall and Irrigation Amounts and Distribution
1987 Season

Rainfall in 1987 totalled 1156 mm which was similar to the 30-year

mean of 1338 mm (Table 1). The early part of the year (January through

March, before water management treatments were imposed) was wetter than

normal. However, as is typical for this region, several dry periods

occurred during the growing season from late May to late June, mid

August to mid-September and in the month of October (Fig. 1 and Table

1). Rainfall was only 56% of the 30-year average during the initial

growing period and first two harvest cycles (April through July), and

October was exceptionally dry. After mid-September to the end of the

last harvest cycle (2 November) there was no significant rainfall (only

10 mm from 20 September and 1 November). This dry period of 41 days

caused perennial peanut to exhibit severe plant water stress symptoms by

the end of October. In general, rainfall distribution during 1987 was

typical of most years with adequate rainfall during the mid summer and

dry periods during the spring and fall. However, the drier than normal

spring and fall in 1987 would be expected to amplify any responses to

irrigation observed in this study compared to those which might be

expected in normal years.

Rainfall and irrigation data specific to each of the four harvest
cycles are summarized in Table 2. Specific dates and amounts of












Table 1. Monthly rainfall for 1987 and 30-year mean
monthly rainfall for Gainesville, Florida.


Rainfall

30-year
Month 1987 mean

-------------mm-------------
January 106 83
February 138 100
March 261 93
April 11 74
May 108 106
June 75 167
July 99 176
August 137 204
September 93 143
October 7 59
November 109 51
December 30 82


Total 1156 1338


Total


1156


1338


























Rainfall, mm


e-
0






S-










3C
4-
00










0.

s-















ft
I0)

Cr







5-
r 0
o











.-"C

L.
c-



SC--


00)

r- r'



**-C


=







c=


o 0 0
0 0~


_ _ _ I


C:7


U LU 'JIDtU!DN












Table 2. Rainfall and irrigation specific to each harvest cycle as
well as before and after the perennial peanut growing season
in 1987.


Irrigation
treatmentst
Interval
of time Dates Rainfall WW SP CS1 RF

--- ----------------- mm-----------
Pre-treatment 1 Jan. 24 Apr. 517 93 93 93 93
Harvest cycle 1 25 Apr. 11 June 110 346 89 127
Harvest cycle 2 12 June 23 July 152 64 38 38
Harvest cycle 3 24 July 8 Sep. 144 160 32 32
Harvest cycle 4 9 Sep. 2 Nov. 94 98 57 57
After final harvest 3 Nov. 31 Dec. 139 -

Total 1156 761 309 347 93

tAbbreviations are as follows: WW) well-watered, SP) speedling, CS1)
cyclical-stress 1, RF) rainfed.








33

rainfall received and irrigation applied to each water management

treatment are given in Table A-3. Because of limited amount and poor

distribution of rainfall in harvest cycles 1 and 3, relatively large

amounts of irrigation were applied (346 and 160 mm, respectively) to the

WW treatment (Table 2). More rainfall and better distribution resulted

in less irrigation in mid summer (harvest cycle 2). Total irrigation

was 761, 309, 347 and 93 mm for the WW, SP, CS1 and RF treatments,

respectively, over the entire growing season. However, only 89 mm of

water was actually applied to the SP treatment utilizing the Speedling

irrigation system during the time that the treatment was active (25

April to 11 June 1987). Following termination of the early-season SP

treatment, the CS1 and SP treatments were irrigated together and

subjected to a total of four complete 2-week cyclical stress periods

during the remainder of the season. The WW treatment received more

irrigation than the other treatments during each harvest cycle since it

was irrigated rather frequently in order to avoid the development of

plant-water deficits during the season.

1988 Season

Total rainfall of 1558 mm in 1988 was again similar to the 30-year

mean rainfall of 1338 mm (Table 3). Several dry periods occurred

throughout the growing season in May, June, the first part of July, and

October (Fig. 2 and Table 3). Similar to 1987, rainfall in 1988 was 57%

below the 30-year mean during the period April through July (Table 3).

No irrigation was applied pretreatment to initiate uniform growth as had

been done in 1987. Rather, the four water management treatments were

continued beginning with early season growth. Plants in the water-












Table 3. Monthly rainfall for 1988 and 30-year mean
monthly rainfall for Gainesville, Florida.


Rainfall


Month


1988


30-year
mean


January
February
March
April
May
June
July
August
September
October
November
December


-------------mm-------------
131 83
122 100
198 93
35 74
82 106
84 167
98 176
378 204
283 143
32 59
81 51
34 82


Total 1558 1338


Total


1558


1338





























Rainfall, mm


Wu 'IJDJUID)


e-
EU
c-


0


cc
a


0


(-





E3

0
S-





m 0
L.
C '-


C I-



--0







tU CA
a o




















4-
C -
r LL)












C




5-
r- 0
> 03

*C
*-. C


1t-


S- -




























i i - i i - - . -.-..-c= .


I I I I I I I I








36

stressed treatments showed visual symptoms of plant water stress in the

first harvest cycle during periods without rainfall. Above average

rainfall occurred in August and September, while October was quite dry.

Even though there was a 28-day period in October without rainfall during

the last harvest cycle, 32 mm of rainfall were received from 1 to 4

October (Figure 3). Moreover, average night temperature in October was

4.50C below the normal with temperatures dropping below 100C on 12

occasions and to 5.5C on four occasions. Visible symptoms of plant

water stress never became apparent during the 28-day dry period at the

end of the season in 1988, and lower than average night temperatures

likely limited growth on all treatments during October more than did

water management. Again, rainfall distribution for 1988 was quite

typical with dry conditions occurring in the spring and fall, and

abundant rainfall occurring in the summer.

Table 4 summarizes the irrigation and rainfall amounts specific to

each harvest cycle in 1988. Specific dates and amounts of rainfall

received and irrigation applied to each water management treatment are

given in Table A-4. A relatively large amount of irrigation was applied

to the WW and CS1 treatments during the first harvest cycle because of

limited rainfall (Table 4). Even though rainfall was below average

during the second harvest cycle, fairly uniform distribution (except

during mid June and early July) precluded the need for large amounts of

irrigation during this time. Abundant rainfall was received during the

third harvest cycle. Even though the last four weeks of the fourth

harvest cycle were dry, cool fall temperatures precluded the need to

irrigate as frequently as was done during hot, dry periods in the late












Table 4. Rainfall and irrigation specific to each harvest cycle as
well as before and after the perennial peanut growing season
in 1988.


Irrigation
treatmentst
Interval
of time Dates Rainfall WW CS1 CS2 RF

------------------ mm-----------
Pre-treatment 1 Jan. 28 Mar. 449 -
Harvest cycle 1 29 Mar. 8 June 144 201 139 44
Harvest cycle 2 9 June 25 July 153 96 46 38
Harvest cycle 3 26 July 12 Sep. 625 34 -
Harvest cycle 4 13 Sep. 1 Nov. 72 99 36 38
After final harvest 2 Nov. 31 Dec. 115 -

Total 1558 430 221 120


tAbbreviations are as follows: WW) well-watered, SP) speedling, CS1)
cyclical-stress 1, RF) rainfed.








38
spring and summer. Total irrigation applied in 1988 was 430, 221, 120,

and 0 mm in the WW, CS1, CS2, and RF treatments, respectively. Again

the WW treatment received more water than the other treatments, and less

irrigation was applied as severity of water-stress treatment increased.

The CS1 and CS2 treatments were subjected to five and three complete

cycles of stress, respectively, during 1988. These complete cycles were

limited to the spring and fall periods. As was the case in 1987, the

below normal rainfall in spring and summer would be expected to amplify

the responses to irrigation observed in this study compared to that

which might occur with more normal rainfall patterns.

Rainfall amounts and distributions observed in 1987 and 1988 were

fairly typical of those common to North Florida and, if anything, would

result in greater than normal responses to supplemental irrigation,

especially during the spring and fall months. Results from this study

would therefore seem appropriate to provide information on the response

of perennial peanut to water management in typical or slightly drier

than normal production years.

Photosvnthate Accumulation and Distribution

1987 Season

Above-ground DM production during the 1987 growing season as well

as the effect of water management on DM accumulation is shown in Figure

3. As is typical of warm season plants, perennial peanut produced most

of its above-ground growth during the late spring and hot summer months,

with declining production in the fall. Water management had only

minimal effects on above-ground DM accumulation during the 1987 season.

Even though there were drying conditions in late May and early June,




















Well watered
Speedling
Cyclical-stress 1
Rainfed


MAY JUNE JULY SEP NOV


1987 HARVESTS


Above-ground dry matter accumulation of perennial peanut as
affected by water management treatment on 15 May, 11 June,
23 July, 8 Sep., and 2 Nov. 1987. Different letters above
bars within a harvest date indicate differences at P according to Duncan's New Multiple Range Test.


500


400


300


200


100


0.


Figure 3.










they had no effects on above-ground DM accumulation. The absence of a

response to irrigation during the spring and early summer may have been

due to above-average rainfall early in the year (January through March)

as well as pretreatment irrigations that were applied in April to

initiate uniform growth in all plots. As a result, soil water was

plentiful at the beginning of the imposition of the water management

treatments.

Another dry period from mid August to early September

significantly reduced above-ground DM accumulation during the third

harvest cycle (24 July 8 September) in both the CS1 and RF treatments

(Fig. 3, September harvest). Between mid September and the end of the

growing season there was an extended dry period of 41 days without

significant rainfall which also caused a reduction in above-ground DM

accumulation in the RF treatment during the last harvest cycle (9

September 2 November). However, above-ground DM was minimal at the

end of the fall harvest cycle and the reduction in the RF treatment

actually represented only a relatively small amount of available forage.

Table 5 shows numerical data for above-ground DM as illustrated in

Figure 3 as well as total, rhizome, and forage DM distribution and leaf

area index (LAI) for each harvest. Except for slight effects on above-

ground DM accumulation in the fall, water management treatments imposed

in 1987 had little effect on total DM or rhizome DM present on any given

sampling date (Table 5). Water management had no effect on DM

accumulation and distribution or LAI at the 15 May and 1 and 11 June

harvest dates. On 11 June, at the end of the first harvest cycle, there

was a tendency for more DM to occur in the rhizomes and less to occur










Table 5. Above-ground, rhizome, total, and forage dry matter
distribution and leaf area index (LAI) of perennial peanut as
affected by water management treatments in 1987.


Dry Matter

Harvest Water management Above-
datet treatment ground Rhizome Total Foraget LAI

--------------g m -------------
15 May Well-watered 241 a* 528 a 769 a 2.7 a
Speedling 221 a 704 a 925 a 2.4 a
Cyclical-stress 1 256 a 594 a 850 a 2.7 a
Rainfed 217 a 492 a 710 a 2.2 a

1 June Well-watered 419 a 564 a 983 a 276 a 4.3 a
Speedling 396 a 572 a 968 a 264 a 4.1 a
Cyclical-stress 1 403 a 630 a 1033 a 268 a 4.3 a
Rainfed 372 a 666 a 1037 a 244 a 3.9 a

11 June Well-watered 435 a 578 a 1013 a 323 a 4.7 a
Speedling 469 a 683 a 1152 a 292 a 4.8 a
Cyclical-stress 1 468 a 692 a 1160 a 328 a 4.8 a
Rainfed 387 a 813 a 1200 a 249 a 4.0 a

23 July Well-watered 429 a 592 a 1021 a 338 a 4.4 a
Cyclical stress 1 416 a 627 a 1043 a 315 a 4.3 a
Rainfed 379 a 564 a 943 a 295 a 4.0 a

8 September Well-watered 402 a 746 a 1148 a 303 a 4.0 a
Cyclical stress 1 344 b 790 a 1134 a 235 ab 3.5 b
Rainfed 330 c 710 a 1040 a 184 b 3.5 b

2 November Well-watered 210 a 1089 a 1299 a 25 a 1.9 a
Cyclical stress 1 202 a 1054 a 1256 a 25 a 1.8 a
Rainfed 171 b 1075 a 1249 a 15 a 1.3 a


Since plots were not staged until 11 June, the data obtained on 15
May, 1 June, and 11 June represent cumulative growth. In contrast,
above-ground data for 23 July, 8 September, and 2 November represent
only accumulation during each respective harvest cycle. Rhizome dry
matter data represent cumulative growth throughout the growing
season.

Represents dry matter above 5-cm cutting height.

* Means within a column on each harvest date followed by the same
letter do not differ at P<0.05 as determined by Duncan's New Multiple
Range test.








42

above-ground in the RF treatment as compared to the other three

treatments; however, total DM accumulation was not different among any

of the treatments. As expected, total DM present increased in each

water management treatment throughout the first harvest cycle (ended 11

June). Periods of dry weather that occurred during rapid growth in late

spring had no effect on the growth of the crop. Leaf area index

increased steadily from an average of 2.5 on 15 May (approximately

halfway through the first harvest cycle) to a relatively high value of

4.6 by 11 June. Average forage yield taken at that time was

approximately 300 g m2 which compares favorably to expected first

cutting yields of alfalfa in Florida (Romero et al. 1987) and yields

collected from first cuttings of perennial peanut by Beltranena (1980).

Root samples taken on 19 June (after the first harvest cycle)

indicated no differences in root proliferation and distribution due to

the water management treatments imposed during the dry, spring period

(Fig. 4). All treatments had roots deep in the profile and roots up to

5 mm in diameter were found in all plots penetrating into the underlying

clay layer at the 1050 to 1200 mm depth. Deep rooting characteristics

along with rainfall that occurred apparently allowed plants in the RF

treatment to avoid severe water stress during the spring. It is

unfortunate that the heavy clay layer below 1200 mm prevented accurate

sampling of roots deeper in the soil profile.

Water management also did not affect DM accumulation and

distribution or LAI at the end of the second harvest cycle on 23 July,

but there was again a tendency for the RF treatment to have less above-

ground DM than the other treatments (Table 5). Regrowth was also













ROOT LENGTH DENSITY, cm cm-3


1.0


2.0


3.0


4.0


I I I I

- --- ----I- -f 1


=


Irrrrrrrlrl


m
["---
0 CI


Well-watered
Speedling
Cyclical stress 1
Rainfed


=
mm
=Z


Root length density of perennial peanut as affected by water
management treatment on 19 June 1987. Root length density
of rhizomes is reported separately at the 50-mm depth since
they occur as a rhizome mat just beneath the soil surface.


0.0


50
150

300

450

600

750


900

1050

1200


Figure 4.


777lrll .... 2 7 1 1 1Fl








44

monitored during the second harvest cycle. Water management had no

effect on number or origin of shoots, light interception, or plant

height during regrowth. On the average, there were 15 stubble units

(stems 5 cm tall from previous staging) per 100 cm2 of ground area.

Stubble units produced the most shoots during regrowth with rhizomes

only giving rise to four shoots per 100 cm2 of ground area by the end of

the harvest cycle (23 July). The balance of new shoots originated from

basal stem buds of the stubble. Since this was a first attempt at

monitoring regrowth, a final count of all shoots originating from the

stubble units was not done. However, in the subsequent regrowth period

(third harvest cycle) each stubble unit produced two or three new

shoots. Averaged across water management treatments, light interception

increased from 62% on 1 July (20 days after staging) to 85% on 8 July

and to 97% on 22 July. Final plant height averaged 17 cm on 23 July.

As previously mentioned, water management did have an effect on

above-ground DM at the end of the third harvest cycle on 8 September

(Table 5). Above-ground DM was reduced 14 and 18% in the CS1 and RF

treatments, respectively, as compared to the WW treatment. Forage yield

was 39% lower in the RF treatment as compared to the WW treatment, and

LAI was reduced 12% by the CS1 and RF treatments as compared to the WW

treatment. Parameters monitored with respect to regrowth were only

slightly affected by water management. Stubble again averaged 15 units

per 100 cm2 throughout the harvest cycle and overall plant height was

unaffected by water management even though above-ground DM and LAI had

been reduced in the CS1 and RF treatments. New shoots originating from

rhizomes showed a significant reduction on 5 September as a result of








45
water stress, being reduced from 4 stems per 100 cm2 in the WW treatment

to I stem per 100 cm2 in the CS1 and RF treatments; however, shoots

arising from rhizomes only accounted for an average of 5% of the total

shoots present at the end of the harvest cycle, indicating that most

regrowth of perennial peanut after clipping occurred from basal stem

buds. Therefore, this procedure was not pursued as a method of

quantifying the effects) of previous or current plant water stress on

growth of perennial peanut. Reductions in above-ground DM and LAI were

more sensitive indicators of the effects of plant water stress than

shoot numbers, shoot origin, or plant height. Light interception was

inconclusive in this study because of the difficulty involved in

positioning a light sensing bar below the leaf canopy. The relatively

low staging height of 5 cm in this study led to the development of

leaves below that point, and it was often impossible to position the

light sensor at the bottom of the canopy without grossly disturbing it.

After the 8 September harvest, little above-ground DM accumulated

during the rest of the season (Table 5). Plants only attained heights

of 6 to 8 cm during this harvest cycle in all water management

treatments. At the 2 November harvest, above-ground DM was reduced 17%

in the RF treatment as compared to the average of the WW and CS1

treatments. However, average forage yield was negligible at 22 g m" as

a result of the short plant height, and LAI was also low (<2.0) for all

treatments.

On the average, rhizomes comprised most of the total DM that was

present at any given time throughout the growing season (Fig. 5 and

Table 5). During rapid, above-ground growth (June and July harvest








46

cycles), the proportion of rhizome DM declined; however, it later

increased to a very high value at the November harvest date. Averaged

across water management treatments, rhizome DM constituted approximately

two-thirds of the total DM throughout the season except in the fall (2

November) when rhizome DM was 85% of total DM. Rhizome DM increased 43%

during the last harvest cycle, actually causing total DM to be at its

maximum value for the entire season. During this time, most

photoassimilate was apparently partitioned to rhizomes (underground) at

the expense of above-ground growth; therefore, forage DM accumulated

during the last harvest cycle was very small.

Total nonstructural carbohydrate accumulation and distribution

were also monitored in rhizomes throughout the season. In general,

rhizome TNC declined during the hot, summer months and increased again

in the fall (Fig. 6). Changes in rhizome TNC would suggest that

perennial peanut relied less on rhizome carbohydrate reserves for spring

and early summer growth, but more so during the hot summer months, at

which time rapid above-ground growth of the crop occurred. In the fall,

photoassimilate partitioning was directed to the underground rhizomes.

A similar seasonal distribution pattern of rhizome TNC has been shown by

Saldivar (1983).

Water management had no effect on TNC accumulation or distribution

during the first or second harvest cycles (ended 11 June and 23 July,

respectively), although the trend for slightly higher TNC in the CSI and

RF treatments as compared to the WW treatment is noticeably consistent

(Fig. 6). Water stress (RF treatment) caused rhizome TNC to be 38%

higher as compared to the WW treatment during the third harvest cycle



















= i Above-ground
M Rhizome


JUNE


JULY


SEP


NOV


1987 HARVESTS


Above-ground, rhizome, and total dry matter of perennial
peanut (averaged across all water management treatments) on
15 May, 11 June, 23 July, 8 Sep., and 2 Nov. 1987.


1400-

1200-

1000-

800-


600-

400-


200

0


MAY


Figure 5.


LA E-E E E~


















Well watered
Speedling
Cyclical-stress 1
Rainfed


100


MAY JUNE JULY SEP NOV


1987 HARVESTS


Rhizome nonstructural carbohydrate concentration of perennial
peanut as affected by water management treatment on 15 May,
11 June, 23 July, 8 Sep., and 2 Nov. 1987. Different
letters above bars within a harvest date indicate
differences at P<0.05 according to Duncan's New Multiple
Range Test.


400


Figure 6.








49

(ended 8 September). Above-ground DM had also been reduced 18% during

the third harvest cycle in the RF as compared to the WW treatment,

suggesting that water stress caused photoassimilate to be partitioned

underground to rhizomes at the expense of above-ground growth during

soil water deficits. Even though October was relatively dry and above-

ground DM was reduced in the RF treatment during the last harvest cycle,

no significant differences were observed in rhizome TNC on 2 November.

There does, however, appear to be a trend for slightly higher TNC in the

RF treatment as compared to the WW and CS1 treatments.

Total nonstructural carbohydrate concentrations in leaves and

stems were fairly constant throughout the season. At no time did water

management treatment affect leaf or stem TNC concentrations. Seasonal

averages of TNC were 58 and 69 g kg"' for the leaves and stems,

respectively (Data not shown).

Photosynthate accumulation and distribution in perennial peanut

was relatively unresponsive to water management in 1987. Only during

periods of no appreciable rainfall for greater than two weeks did the

water stress treatments) (CS1 and RF or RF alone) cause a reduction in

above-ground DM or cause rhizome TNC to significantly increase (Figs. 3

and 6).

Averaged across water management treatments, cumulative above-

ground DM production for the 1987 season was 1019 g m2 (Table 6).

Above-ground DM accumulation was 440 g m2 through 11 June at which point

all plots were staged to a 5 cm stubble height (ie., height of cut for

forage sample). The above-ground DM for on 23 July (316 g m2)

represents only the forage DM taken during the second harvest cycle,











Table 6. Above-ground, rhizome, total, and forage dry matter production
by perennial peanut averaged across water management
treatments in 1987.


Dry Matter

Above-
Date ground Rhizome Total Forage

-------------------g m------------.......................------
11 June 440 11lt 551 298
23 July 316t -97 219 316
8 Sep. 2410 154 395 241
2 Nov. 221 326 348 22
Cumulative total 1019 494 1513 877


t Represents average dry matter accumulation
May and 11 June sampling dates.


in rhizomes between the 15


Represents dry matter accumulated above the 5-cm stubble height
(forage).








51
since the forage harvest represented all above-ground growth that had

occurred above the stubble height since the 11 June staging. Likewise,

above-ground DM for 8 September (241 g m2) and 2 November (22 g m"2)

corresponded to forage DM. Therefore, summing the above-ground DM

accumulated for each harvest cycle gives a cumulative DM production of

1019 g m" for the season. Average rhizome DM present increased from 580
g m2 on 15 May to 1074 g m2 on 2 November, representing a net increase

of 494 g m-2. Summing above-ground and rhizome DM, total DM accumulated

for the season was 1513 g m2. Assuming a 184-day growth period from May

through October, the average daily rate of DM accumulation was computed

to be 8.2 g m2. Summing forage DM accumulation from each harvest cycle

gives a total of 877 g m-2 for seasonal forage yield. This is similar to

previous seasonal forage DM yields reported by Prine et al. (1981) and

Valentim (1987).

Seasonal DM accumulation and distribution for each water
management treatment is shown in Table 7. Cumulative DM and seasonal

forage yield for the different water treatments were not different in

1987. Nevertheless, there does appear to be a trend for lower above-

ground, total, and forage DM in the RF treatment as compared to the WW

treatment.

1988 Season

Seasonal distribution of above-ground DM for 1988 was similar to

that of 1987 with most production occurring in the late spring and

summer followed by a decline in the fall (Fig. 7). Data presented in

Figure 7 also depict the effect of the four water management treatments

on DM accumulation on different harvest dates. Below average rainfall











Table 7. Seasonal dry matter accumulation and distribution as affected
by water management treatment in 1987.


Dry Matter

Water management Above-
treatment ground Rhizome Total Forage

----------------------g m-------------------
Well-watered 1101 a* 561 a 1662 a 990 a
Speedling -
Cyclical-stress 1 1043 a 460 a 1503 a 903 a
Rainfed 882 a 585 a 1467 a 744 a


* Means within a column followed by the same letter do not differ at
P<0.05 as determined by Duncan's New Multiple Range test.




















Well watered
Cyclical-stress 1
Cyclical-stress 2
Rainfed


MAY JUNE JULY SEP NOV


1988 HARVESTS


Above-ground dry matter accumulation of perennial peanut as
affected by water management treatment on 19 May, 8 June, 25
July, 12 Sep., and 1 Nov. 1988. Different letters above
bars within a harvest date indicate differences at P according to Duncan's New Multiple Range Test.


600

500

400

300

200

100

0


Figure 7.










during the spring of 1988 caused a reduction in above-ground DM

accumulation in the water stressed (CS2 and RF) treatments as compared

to the WW and CS1 treatments during that time. The effect of drying

conditions on the reduction of DM accumulation in the RF and CS2

treatments can be seen as early as 19 May (approximately halfway through

the first harvest cycle). By the end of the first harvest cycle (8

June) above-ground DM in the RF treatment had been greatly reduced as

compared to the other treatments. Another dry period in July during the

second harvest cycle also had an impact on above-ground DM accumulation

causing it to be reduced in the RF treatment as compared to the WW and

CSI treatments. Abundant rainfall during August and September precluded

any further effects of water management during the remainder of the

season. Even though there was a 28-day dry period during October, lower

than average night temperatures may have limited growth more than water

management during that time.

Table 8 shows numerical data for above-ground DM as illustrated in

Figure 7 as well as total, rhizome, and forage DM distribution and LAI

on each harvest date of the 1988 season. Water management treatment

principally affected only above-ground growth and accumulation in 1988,

as was the case in 1987, with little effect on rhizome and total DM. As

previously mentioned, below average rainfall during the spring and early

summer caused a reduction in above-ground DM accumulation in two water

management treatments (CS2 and RF treatments) during May through July as

compared to the WW and CSI treatments. It is important to realize that
pretreatment irrigations were not applied in 1988 to initiate uniform

growth as had been done in 1987. Rather, the water management









Table 8. Above-ground, rhizome, total, and forage dry matter
distribution and leaf area index (LAI) of perennial peanut as
affected by water management treatments in 1988.


Dry Matter
Harvest Water management Above-
datet treatment ground Rhizome Total Forage$ LAI

--------------g m2------------
19 May Well-watered 337 a* 937 a 1274 a 3.4 a
Cyclical-stress 1 330 a 917 a 1247 a 3.4 a
Cyclical-stress 2 296 b 1034 a 1330 a 3.1 a
Rainfed 258 c 1087 a 1345 a 2.5 b

2 June Well-watered 455 a 1059 a 1514 a 4.7 a
Cyclical-stress 1 432 ab 905 a 1337 a 4.2 ab
Cyclical-stress 2 384 b 1026 a 1410 a 3.7 b
Rainfed 296 c 1053 a 1349 a 2.8 c

8 June Well-watered 519 a 909 a 1428 a 492 a 5.2 a
Cyclical-stress 1 475 a 1007 a 1482 a 466 a 4.6 a
Cyclical-stress 2 396 b 985 a 1381 a 306 b 3.9 b
Rainfed 282 c 1016 a 1298 a 191 c 2.7 c

25 July Well-watered 486 a 599 a 1085 a 397 a 4.9 a
Cyclical stress 1 456 ab 643 a 1099 a 360 ab 4.6 a
Cyclical-stress 2 435 b 737 a 1172 a 349 b 4.6 a
Rainfed 377 c 760 a 1137 a 269 c 4.3 a

12 September Well-watered 348 a 720 a 1068 a 278 a 3.5 a
Cyclical-stress 1 347 a 800 a 1147 a 300 a 3.9 a
Cyclical-stress 2 356 a 702 a 1058 a 312 a 3.6 a
Rainfed 380 a 674 a 1054 a 307 a 3.9 a

1 November Well-watered 167 a 820 a 987 a 1.4 a
Cyclical stress 1 172 a 927 a 1099 a 1.5 a
Cyclical-stress 2 174 a 885 a 1059 a 1.5 a
Rainfed 171 a 950 a 1121 a 1.6 a


t Since all plots were not staged until 8 June, the data obtained on 19
May, 2 June, and 8 June represent cumulative growth. In contrast,
above-ground data for 25 July, 12 September, and 1 November represent
only accumulation during each respective harvest cycle. Rhizome dry
matter data represent cumulative growth throughout the growing
season.

SRepresents dry matter above 5-cm cutting height.

* Means within a column on each harvest date followed by the same
letter do not differ at P<0.05 as determined by Duncan's New Multiple
Range test.








56
treatments were continued beginning with the initiation of spring

growth. Additionally, plants in the water stress treatments had been

previously subjected to water deficits in the fall of 1987. The

previous stresses coupled with the dry spring period in 1988, and the

fact that no pretreatment irrigation was applied all probably

contributed to reduced above-ground DM accumulation in the water stress

treatments during the spring.

April precipitation was 53% below normal and there were only three

rainfall events in May. Water management significantly affected above-

ground DM accumulation and LAI during the first harvest cycle (ended 8

June) (Table 8). The effects of plant water stress were detected midway

through the first harvest cycle on 19 May with above-ground DM being

reduced 11 and 23% in the CS2 and RF treatments, respectively, as

compared to the average of the WW and CS1 treatments. A difference in

LAI was also detected on 19 May, being 2.5 in the RF treatment and

averaging 3.3 in the other three treatments. Similar differences were

detected on 2 June with above-ground DM in the RF treatment being only

67% of that present as the average of the WW and CS1 treatments. The

CS1 and CS2 treatments were not different on this sampling date;

however, the CS2 treatment had accumulated 16% less above-ground DM as

compared to the WW treatment. Leaf area continued to accumulate in the

irrigated treatments more rapidly than in the RF treatment with the RF

treatment remaining relatively low (2.8) on 2 June as compared to the

average of the WW and CS1 treatments (4.4). In addition, a difference

in specific leaf area (SLA) was detected at this sampling with leaves in








57
the RF treatment being thicker and having lower SLA (136 cm2 g 1) as

compared to leaves in the WW treatment (148 cm2 g'1).

At the end of the first harvest cycle (8 June) above-ground DM was

reduced 20 and 43% by water stress in the CS2 and RF treatments,

respectively, as compared to the average of the WW and CS1 treatments.

Forage yields (cut to 5 cm stubble height) were also reduced with

increasing water stress (Table 8). Forage yields were drastically

reduced in the RF treatment (191 g m2) as compared to the average of the

WW and CSI treatments (479 g m2). Forage DM in the CS2 treatment was

also reduced 36% as compared to the average of the WW and CS1

treatments. Final LAI of perennial peanut at the end of the first

harvest cycle was relatively high in the nonstressed plots at 4.9

(average of the WW and CS1 treatments) as compared to 3.9 and 2.7 in the

CS2 and RF treatments, respectively. Leaves in the RF treatment were

much thicker and had a lower SLA (130 cm2 g1) as compared to the average

of the WW and CS1 (149 cm2 g1), and CS1 and CS2 (144 cm2 g') treatments,

respectively. This is a typical response of leaves formed under

conditions of plant water deficits. As was the case in 1987, no

differences were detected in rhizome or total DM accumulation as

affected by water treatment. However, there does appear to be a trend

for reduced total DM accumulation in the CS2 and RF treatments as

compared to the WW and CS1 treatments at the end of the first harvest

cycle (8 June).

Gravimetric soil water content data taken at the end of the first

harvest cycle on 8 June nicely illustrates differences resulting from

the various water management treatments. Reduced soil water content was








58

observed for all water stress treatments (CS1, CS2 and RF) as compared

to the WW treatment at the 150, 300, 450, 600, 750 and 1050 mm depths

(Fig. 8). The entire soil profile was very dry in the RF treatment

(average of 1.9%) down to the 1200 mm depth. The clay layer below this

depth probably provided at least a limited amount of soil moisture to

the crop in these plots (as well as in the other treatments). Soil

water content was intermediate between the WW and RF treatments for the

CS1 and CS2 water management treatments. As previously mentioned in the

1987 section, deep rooted characteristics of perennial peanut probably

contributed to its drought tolerance during periods of up to two weeks

without rainfall or irrigation.

There was an 11-day period without rainfall midway through the

second harvest cycle (ended 25 July) when plants in the water stressed

treatments (CS2 and RF) showed signs of drought stress (paraheliotropic

leaf movements and folding of leaves). Above-ground DM was reduced in

the CS2 and RF treatments as compared to the WW treatment during the

second harvest cycle even though no differences in LAI were detected

(Table 8). The fact that the plants in the water-stressed treatments

had been under prolonged drought stress during the first harvest cycle,

and below average rainfall for June and July likely contributed to the

observed differences even though the dry period was relatively short.

Above-ground and forage DM were reduced 20 and 29%, respectively, in the

RF treatment as compared to the average of the WW and CS1 treatments.

Water management had no effect on DM accumulation and

distribution, LAI, or SLA during the rest of the season (third and

fourth harvest cycles which ended 12 September and 1 November,


















SOIL WATER CONTENT, %

0 2 4 6 8 10 12
150-A -

300- 9- Well-watered
A-A Cyclical stress 1
450- 0-0 Cyclical stress 2
E A-A Rainfed
E 600

a_ 750-

900-

1050

1200


Figure 8. Gravimetric soil water content on 8 June 1988.








60

respectively). There was above average rainfall in August and September

as well as 32 mm of rainfall during the first four days of October.

Even though there was no more precipitation during the rest of October

(remainder of the last harvest cycle), below normal night temperatures

may have limited growth in all treatments more than did water stress

during the fourth harvest cycle. Again, very little above-ground DM

accumulated during the last harvest cycle as was the case in 1987. Leaf

area index remained very low (averaging 1.5) at the end of the fourth

harvest cycle on 1 November; however, the amount of DM present in

rhizomes increased.

On average, rhizome DM constituted most of the total DM present

throughout the season in 1988 as had been the case in 1987 (Fig. 9 and

Table 8). During rapid above-ground growth (first and second harvest

cycles), the proportion of total DM in rhizomes decreased, however it

increased again during the fall. Averaged across water management

treatments, rhizome DM again constituted approximately two-thirds of the

total DM present during any given harvest cycle except in the fall

(fourth harvest cycle ended 1 November) when rhizomes were 84% of total

DM. Average DM in rhizomes increased 24% during the last harvest cycle

which was somewhat less than the 1987 increase of 43% possibly because

of cooler than average weather in October. Photoassimilate accumulated

in rhizomes during the last harvest cycle (12 September to 1 November)

at the expense of above-ground DM accumulation, and forage DM was

negligible on the last harvest date because plants had remained only 5

to 6 cm tall since the previous staging on 12 September. Rhizome DM at

the end of 1988 (1 November) was similar to that at the beginning of
























EI Above-ground
MIp Rhizome


MAY JUNE JULY SEP NOV


1988 HARVESTS


Above-ground, rhizome, and total dry matter of perennial
peanut (averaged across all water management treatments) on
19 May, 8 June, 25 July, 12 Sep., and 1 Nov. 1988.


1400


CN
E



o
*&

in


Figure 9.








62

1988 (19 May). This is in contrast to the large increase in rhizome DM

observed during the 1987 growing season (see Fig. 5). Rhizome mass

would be expected to increase rapidly in the early years of crop

establishment and then remain relatively stable thereafter. Differences

in rhizome DM accumulation during 1987 and 1988 are likely a result of

the establishment phase of perennial peanut.

Total nonstructural carbohydrate accumulation and distribution

followed a similar trend in 1988 as compared to 1987. Rhizome TNC

declined during the hot, summer months and increased during the fall

(Fig. 10), again suggesting that perennial peanut did not rely heavily

on underground reserves for spring growth, but did utilize rhizome

reserves for periods of rapid above ground growth during the hot, summer

months. In the fall, photosynthate partitioning changed with most being

directed to underground rhizomes.

Plant water stress affected rhizome TNC concentrations during

periods of dry weather resulting in higher rhizome TNC in water-stressed

treatments on 19 May, 8 June, and 25 July (Fig. 10). Above-ground DM

accumulation was also less in the water-stressed treatments on these

dates, thus indicating that photosynthate was being partitioned

underground to rhizomes (and probably the root system in general) at the

expense of above-ground growth during plant water stress. Rhizome TNC

increased and above-ground DM declined as severity of water stress

increased (Figs. 10 and 7). On the average, the RF and CS2 treatments

caused an 18% increase in TNC on 19 May as compared to the average of

the WW and CS1 treatments. On 8 June, average rhizome TNC in the RF and

CS2 treatments (349 g kg"') was 8 and 17% higher than in the CS1 and WW




















Well-watered
Cyclical stress 1
Cyclical stress 2
Rainfed


MAY JUNE JULY SEP NOV


1988 HARVESTS


Figure 10.


Rhizome nonstructural carbohydrate concentration of
perennial peanut as affected by water management treatment
on 19 May, 8 June, 25 July, 12 Sep., and 1 Nov. 1988.
Different letters above bars within a harvest date indicate
differences at P<0.05 according to Duncan's New Multiple
Range Test.








64

treatments, respectively. Plant water stress (RF and CS2 treatments)

caused rhizome TNC to be 69% higher than in the WW treatment on 25 July.

These results are similar to those reported by Hall et al. (1988) who

found increased TNC in roots of water stressed alfalfa.

No differences were detected in rhizome TNC during the third and

fourth harvest cycles (Fig. 10). Neither was above-ground DM

accumulation affected by water management during the remainder of the

season. Above average rainfall during August and September precluded

the development of plant-water deficits through the remainder of the

season.

Total nonstructural carbohydrate concentrations in leaves and

stems were fairly constant throughout the season. Leaf TNC was affected

only on 19 May, being 39% higher in the RF treatment (118 g kg") as

compared to the average of the WW, CS1 and CS2 treatments (85 g kg").

At no other time in 1988 was there a significant effect of water

management treatment on leaf or stem TNC. Average seasonal TNC for

leaves and stems were 59 and 50 g kg', respectively (Data not shown).

In 1988, perennial peanut responded more to water management than

in 1987. During 1988 initial growth began earlier (late March) and

continued with substantial growth occurring in April. This probably

occurred because of somewhat warmer weather in the spring of 1988 as

compared to 1987. Average air temperature in April 1988 (200C) was 20C

warmer than in 1987, and average soil temperature at 100 mm was 4"C

higher (24C in April 1988 as compared to 200C in 1987). Furthermore,

plants in the RF treatment had been exposed to a rather long period of

dry weather in October 1987, and initial growth in the spring of 1988








65

was also under drought conditions. Drying conditions continued

throughout the first harvest cycle with the WW treatment receiving

relatively frequent irrigations. Soil water deficits also developed

during the second harvest cycle in the first part of July. Therefore,

plants in the RF and CS2 treatments were subjected to relatively more

early season stress in 1988 as compared to 1987. The water deficits in

the RF and CS2 treatments were relieved for the remainder of the season

after frequent, abundant rainfall occurred in mid summer.

Cumulative above-ground DM for 1988 averaged across water

management treatments was calculated in a similar manner to that for

1987, giving a total of 1063 g m2 (Table 9). Average rhizome DM

remained practically the same during the 1988 season and actually

decreased slightly from 994 g m2 on 12 May to 895 g m2 on 1 November.

Even though there was no net gain in rhizome DM, it did accumulate (from

its depressed mid summer value) in the fall at the expense of above-

ground growth. Summing above-ground and net rhizome DM for the season

resulted in a total DM accumulation of 964 g m2. Again, assuming a 184-

day growing season from May through October, 5.2 g m2 d-' was accumulated

on the average. Summing forage DM from each of the harvest cycles gives

an average cumulative forage yield of 1008 g m2. Average forage yield

was higher in 1988 as compared to 1987 with the yield in each respective

harvest cycle being higher except for the last harvest cycle of 1988

when negligible forage DM accumulated.

Seasonal DM accumulation and distribution in 1988 is shown for

each water management treatment in Table 10. Above-ground DM was

reduced in the CS2 and RF treatments as compared to the WW treatment;











Table 9. Above-ground, rhizome, total, and forage dry matter production
by perennial peanut averaged across water management
treatments in 1988.


Dry Matter

Above-
Date ground Rhizome Total Forage

-----------------------g m-2-------.----.-------
8 June 418 -15t 403 363
25 July 344t -294 50 344
12 Sep. 301' 39 340 301
1 Nov. 171 171
Cumulative total 1063 -99 964 1008


t Represents average dry matter accumulation in rhizomes between the 15
May and 11 June sampling dates.

SRepresents dry matter accumulated above the 5-cm stubble height
(forage).











Table 10. Seasonal dry matter accumulation and distribution as affected
by water management treatment in 1988.


Dry Matter


Water management Above-
treatment ground Rhizome Total Forage

-----------------------g m2--------------------
Well-watered 1194 a* -117 a 1077 a 1167 a
Cyclical-stress 1 1133 ab 9 a 1142 a 1124 a
Cyclical-stress 2 1057 b -149 a 908 b 967 b
Rainfed 858 c -137 a 721 c 767 c


* Means within a column followed by the same letter do not differ at
P<0.05 as determined by Duncan's New Multiple Range test.








68

however, the CS1 and CS2 treatments were not different. The CS2 and RF

water management treatments caused a significant reduction in cumulative

total and cumulative forage DM in 1988 as compared to the average of the

WW and CS1 treatments. The CS2 and RF treatments produced 18 and 35%

less total DM, and 16 and 33% less forage DM, respectively, than the

average of the WW and CS1 treatments. These data indicate that

perennial peanut did withstand periods of dry weather for up to two

weeks in length throughout the season without adversely affecting

harvestable yield; however, extended dry periods of three weeks or more

without rainfall or irrigation did reduce available forage by up to one-

third compared to the well-watered control and CS1 treatment in 1988.

Physiological Measurements

1987 Season

Single leaf carbon exchange rates (A) on specific dates during the

first harvest cycle (ended 11 June) as affected by water management


treatment are shown in Figure 11.


Single leaf carbon exchange rates


observed for well-watered perennial peanut were s

observed by Lopez et al. (1988) for pigeonpea, Ni

for alfalfa, Sinclair and Horie (1989) for rice (

soybean, and Turner et al. (1985) for sunflower.

millimeters of rain accumulated between 12 and 16

rainfall occurred from 19 to 21 May. Apparently

provided sufficient water to prevent plant water

SP, and RF treatments during May. Nevertheless,

content on 31 May was significantly lower in the


similar to those

colodi et al. (1988)

Oryza sativa L.) and

Twenty-five

May and 75 mm of

those rainfall event;

deficits in the CS1,

gravimetric soil wati

SP, CS1, and RF


s


er


treatments as compared to the WW treatment down to a depth of 750 mm in






















Well-watered
Speedling
Cyclical-stress 1
Rainfed


(N





a-
E







C
0
-o




a
n0
(--
0
o
0


0


DATE, 1987


Figure 11.


Single leaf carbon exchange rate of perennial peanut as
affected by water management treatment on 18, 26, and 29
May, and 8 and 10 June 1987. Different letters above bars
within a harvest date indicate differences at P<0.05
according to Duncan's New Multiple Range Test.


18 MAY 26 MAY 29 MAY 8 JUNE 10 JUNE








70

the soil profile (Fig. 12). On 8 and 10 June, the CS1 treatment was 6

and 8 days, respectively, into a drying cycle and there had been no

appreciable rainfall for 17 days (since 21 May). Gravimetric soil water

content on 10 June indicated that the WW treatment had a considerably

higher soil water content than the other treatments down to a depth of

750 mm (Fig. 13). Below 750 mm there were no statistically significant

differences in soil water content. Even after 17 days without rainfall

or irrigation, only a 33% reduction in A was observed on 8 June in the

RF treatment as compared to the WW and SP treatments; however,

significant treatment differences in A were not apparent on 10 June.

There were no irrigation or rainfall events between 8 and 10 June. The

lack of a significant response of A to drought on 10 June is surprising,

since a 33% reduction had been observed two days earlier. Nevertheless,

data from both 8 and 10 June suggest that perennial peanut withstood the

19-day drought very well. Presumably, plentiful soil water in the lower

depths of the soil profile provided enough water to prevent plant water

deficits in the CS1, SP, and RF treatments. The absence of appreciable

reductions in A after extended dry periods is consistent with the

drought resistance exhibited by perennial peanut.

Figure 14 illustrates A measured during the third and fourth

harvest cycles. Measurements taken near the end of a 2-week dry period

in late August failed to reveal any water management treatment effects

on A. Only at the end of the 41-day dry period during the fourth

harvest cycle (29 October) was A significantly reduced in the RF

treatment as compared to the WW treatment. Even then, A was only

reduced by 34%. There had been only 10 mm of rainfall (30 September and


















SOIL WATER CONTENT, %

0 2 4 6 8 10 12
150

300

450- 0-0 Well-watered
E / 0-0 Speedling
E 600 A-A Cyclical stress 1
- UA-A Rainfed
. 750-

900-

1050

1200


Gravimetric soil water content on 31 May 1987.


Figure 12.




















SOIL WATER CONTENT, %

0 2 4 6 8 10 12
150

300

450
E *--* Well-watered
E 600- 0-0 Speedling
A-A Cyclical-stress 1
750 A-A Rainfed
0 750-
\


Figure 13. Gravimetric soil water content on 10 June 1987.



















a
a
aa a

a I
l






II


0
(N
E
(N
0
o
0

(.4

<->
E




0
a
c

.0

L.


a a










7 OCT


M
a


Well-watered
Cyclical-stress 1
Rainfed
a

H a


22 OCT 29 OCT


DATE, 1987


Figure 14.


Single leaf carbon exchange rate of perennial peanut as
affected by water management treatment on 27 and 28 Aug.,
and 7, 22 and 29 Oct. 1987. Different letters above bars
within a harvest date indicate differences at P<0.05
according to Duncan's New Multiple Range Test.


30-

25-

20-

15-


27 AUG 28 AUG I


m










1 October) during the previous 41 days, and gravimetric samples taken at

that time (30 October) indicated that soil water contents in the RF and

CS1 treatments were lower than in the WW treatment down to a depth of

750 mm (Fig. 15). The RF treatment was also drier than the WW treatment

at 900 mm. Below 900 mm, there were no statistically significant

differences in soil water content among treatments, but trends for lower

soil water content in the CS1 and RF treatments were clearly apparent at

900, 1050 and 1200 mm. Visible signs of drought stress in the RF

treatment were apparent at the end of October with opposite leaflets

folding together and orienting themselves in such a way as to avoid

direct light interception (paraheliotropic leaf movements) during mid-

day. This was the only time in 1987 when visible signs of severe plant

water stress were observed and gravimetric soil water data indicated

that the soil profile was extremely dry in the RF treatment (average of

1.7%) down to the underlying clay layer. Primary differences between

soil water on this date and that observed earlier in the season was the

greater depletion deep in the soil profile (below 900 mm). Depletion of

this water apparently resulted in some plant water deficits and moderate

reductions in A. From data collected, it appeared that plant water

deficits did not develop and A remained high as long as plentiful soil

water was available deep in the soil profile.

Table 11 shows the numerical data for A as illustrated in Figure

11 along with leaf conductance (g) and transpiration (E), and the ratio

Ci/Ca. Water management treatment had no effect on A, g, or E during

May. Average A, g and E for the three observations (18, 26 and 29 May)

were 20.1 lmol CO2 m2 s'1, 1.6 cm s' and 0.016 mol m2 s', respectively. A



















SOIL WATER CONTENT, %

4 6 8 10 12

0--9 Well-watered
A-A Cyclical-stress 1
A--A Rainfed


Figure 15. Gravimetric soil water content on 30 Oct. 1987.


150

300

450

600

750

900

1050

1200











Table 11.


Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular to
ambient CO2 concentration (Ci/Ca) of perennial peanut as
affected by water management treatment on 18, 26 and 29 May,
and 8 and 10 June 1987.


Water management
Date treatment A g E Ci/Ca

/mol CO2 m2 s"- cm s1 mol m 2 s"1
18 May Well-watered 17.4 a* 1.7 a 0.016 a 0.80 a
Speedling 18.8 a 1.8 a 0.018 a 0.80 a
Cyclical-stress 1 19.2 a 1.5 a 0.017 a 0.76 b
Rainfed 21.6 a 2.1 a 0.020 a 0.80 a

26 May Well-watered 19.7 a 1.6 a 0.015 a 0.79 a
Speedling 20.3 a 1.7 a 0.014 a 0.79 a
Cyclical-stress 1 18.8 a 1.3 a 0.014 a 0.74 a
Rainfed 21.3 a 1.6 a 0.016 a 0.76 a

29 May Well-watered 22.4 a 1.7 a 0.016 a 0.76 a
Speedling 21.0 a 1.5 a 0.014 a 0.74 a
Cyclical-stress 1 20.0 a 1.3 a 0.012 a 0.71 a
Rainfed 20.4 a 1.3 a 0.014 a 0.73 a

8 June Well-watered 23.9 a 1.8 a 0.018 a 0.75 a
Speedling 26.1 a 2.0 a 0.014 ab 0.76 a
Cyclical-stress 1 19.4 b 1.2 b 0.014 b 0.72 a
Rainfed 16.7 b 0.9 b 0.010 b 0.66 a

10 June Well-watered 22.6 a 1.9 a 0.016 a 0.77 a
Speedling 25.6 a 2.0 a 0.017 a 0.76 a
Cyclical-stress 1 25.2 a 2.0 a 0.017 a 0.77 a
Rainfed 20.0 a 1.1 b 0.014 a 0.68 b


the same letter do not
Multiple Range test.


* Means within a column on each date followed by
differ at P<0.05 as determined by Duncan's New










significant reduction in the ratio Ci/Ca was detected on 18 May in the

CS1 treatment (0.76) as compared to the average of the other three

treatments (0.80); however, calculated Ci was likely high enough in all

treatments (>233 iLL L') to maximize A (at ambient CO,). Leaf carbon

exchange rate has been shown to respond linearly to Ci up to 230 pL L"

in various C3 species (Farquar and Sharkey, 1982). No differences in

Ci/Ca were detected among water management treatments on 26 or 29 May.

Inconsistent differences in A, g, E, and Ci/Ca were detected on 8

and 10 June due to the water stress treatments (CS1 and RF) (Table 11).

Single leaf carbon exchange rate, g and E were 28, 45 and 25% lower,

respectively, under water stress (CS1 and RF) compared to the averages

of the WW and SP treatments on 8 June. There was a tendency for Ci/Ca

to be reduced in the RF treatment (0.66) as compared to the WW and SP

treatments (0.75). Calculated Ci was 203 AL L' in the RF treatment and

223 AL L-' for the average of the WW and SP treatments. On 10 June, only

differences in g and Ci/Ca were detected, being 44 and 11% lower,

respectively, in the RF treatment as compared to the average of the WW,

SP and CS1 treatments. Calculated Ci values for the RF treatment and

the mean of the other three treatments were 208 and 227 AL L",

respectively. The data taken on 8 and 10 June would indicate that g was

more sensitive to plant water stress than was A. Stomatal conductance

was reduced proportionately more than A on 8 June, and on 10 June no

differences in A were detected while g was reduced in the RF treatment

as compared to the other three treatments. Results obtained by Koppers

et al. (1988) for cowpea and Lopez et al. (1988) for pigeonpea have also

shown g to be more sensitive to plant water stress than A. The effect










of Ci and data calculated concerning the Ci/Ca ratio are somewhat

inconclusive. While it appears that calculated Ci may have been low

enough (203 AL L1) in the RF treatment on 8 June to have possibly

resulted in a reduction in A, the Ci/Ca ratios were not statistically

different among treatments. On 10 June, the Ci/Ca ratio was reduced in

the RF treatment, but differences in A were not detected. Deep rooted

characteristics of perennial peanut along with rainfall that occurred

may have helped plants in the RF treatment avoid severe water stress

during the spring.

Leaf water status did not differ among water management treatments

on either 8 or 10 June. There were no significant differences in 0L, 0

or 0. Mean values for 0L, 0 and 0, were -1.45, -1.75 and 0.30 MPa on 8

June, and -1.14, -1.61 and 0.47 MPa on 10 June, respectively.

Table 12 shows the numerical data for A as shown in Figure 14

along with g, E, and the ratio Ci/Ca. Measurements taken on 27 and 28

August during a dry period again revealed no significant effect of water

management on A, g, E, and Ci/Ca except for g on 28 August. As of 28

August, there had not been appreciable rainfall for two weeks since

previous precipitation of 120 mm between 11 to 15 August. Stomatal

conductance was reduced 19% in the RF treatment as compared to the WW

treatment on the 28 August sampling date. Again this would indicate

that g was more sensitive to plant water deficits than was A. Although

no differences were detected in A, the CS1 treatment caused OL and ~, to

be reduced on 28 August to -1.51 and -1.70 MPa as compared to -1.17 and

-1.45 MPa, respectively, in the WW treatment. Even though reductions in

~L and b, were found, they were apparently insufficient to cause











Table 12.


Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular to
ambient CO2 concentration (Ci/Ca) of perennial peanut as
affected by water management treatment on 27 and 28 August,
and 7, 22 and 29 Oct. 1987.


Water management
Date treatment A g E Ci/Ca

imol CO0 m2 s-1 cm s'1 mol m2 S-I
27 Aug. Well-watered 23.5 a* 2.8 a 0.024 a 0.82 a
Cyclical-stress 1 22.5 a 2.2 a 0.021 a 0.80 a
Rainfed 20.6 a 1.9 a 0.019 a 0.79 a

28 Aug. Well-watered 27.3 a 2.7 a 0.020 a 0.81 a
Cyclical-stress 1 25.5 a 2.4 ab 0.019 a 0.80 a
Rainfed 21.5 a 2.2 b 0.018 a 0.81 a

7 Oct. Well-watered 22.3 a 1.7 a 0.019 a 0.77 a
Cyclical-stress 1 21.2 a 1.5 a 0.016 a 0.75 a
Rainfed 22.1 a 1.4 a 0.016 a 0.73 a

22 Oct. Well-watered 23.2 a 4.1 a 0.020 a 0.88 a
Cyclical-stress 1 22.1 a 3.9 a 0.018 a 0.89 a
Rainfed 15.3 a 2.7 a 0.019 a 0.89 a

29 Oct. Well-watered 17.5 a 1.0 a 0.010 a 0.72 a
Cyclical-stress 1 13.9 ab 0.7 ab 0.008 ab 0.71 a
Rainfed 11.6 b 0.5 b 0.006 b 0.65 b


the same letter do not
Multiple Range test.


* Means within a column on each date followed by
differ at P<0.05 as determined by Duncan's New








80

perturbances in A. Unfortunately, no data with respect to OL were taken

in the RF treatment on 28 August.

Another series of measurements was taken during a relatively long

dry period of 41 days in October. No significant differences were

detected in A, g, E, or Ci/Ca on 7 or 22 October (Table 12). Average A,

g, E, and Ci/Ca were 21.9 limol CO2 m2 sI', 1.56 cm s", 0.017 mol m2 s",

and 0.75, respectively, on 7 October; and 20.2 pmol CO, mn2 sI', 3.60 cm

s1, 0.019 mol m2 s", and 0.89, respectively, on 22 October. On 29

October, water stress (RF treatment) significantly reduced A, g, E, and

Ci/Ca by 34, 53, 44, and 10%, respectively, as compared to the WW

treatment. Calculated Ci was 218 and 238 pL L1 in the RF and WW

treatments, respectively. Again the data show that g was reduced

proportionately more than A in perennial peanut subjected to soil water

deficits. Even though Ci/Ca was reduced in the RF treatment as compared

to the WW and CS1 treatments, it appears that calculated Ci (218 AL L")

in the RF treatment was likely not low enough to cause A to be reduced

by one-third as compared to the WW treatment. To address this question,

A for the WW treatment on 29 October was calculated utilizing the value

for stomatal conductance to CO, (g,) from the RF treatment according to

the method of Nicolodi et al. (1988). Calculated A for the WW treatment

based on g, from the RF treatment was 13.6 pmol CO2 m2 s- when in fact

actual A was 11.6 pmol CO, m2 s" in the RF treatment. Therefore, the

reduction in A (ie., 17.5 13.6 = 3.9) as a result of restricted gas

diffusion was 66% of the measured reduction (difference between WW and

RF treatment, ie., 17.5 11.6 = 5.9). The additional 34% reduction in

observed A under water stress was presumably due to nonstomatal










(biochemical) mechanisms. Similar results were published by Nicolodi et

al. (1988) for water stressed alfalfa.

Total leaf water potential and Op also differed among water

management treatments on 29 October being reduced to -3.0 and 0.0 MPa,

respectively, in the RF treatment as compared to -1.1 and 0.9 MPa on the

average in the WW and CS1 treatments. It is noteworthy that leaves with

a OL of -3.0 MPa maintained A which was 66% of that in the WW treatment.

These results suggest that perennial peanut leaves continue to function

even at quite severe water deficits.

Single leaf carbon exchange rate was never reduced by more than

one-third during the 1987 growing season, even when subjected to severe

soil water deficits during late October. Neither did the Ci/Ca ratio

ever decrease greatly nor were the decreases in calculated Ci drastic

under plant water stress. These data relating the physiological

response of perennial peanut to water stress are consistent with the

minimal effects of plant water stress on photosynthate accumulation and

distribution as discussed in the previous sections.

1988 Season

Figure 16 depicts the effect of water management treatment on A

for five dates during the first harvest cycle (ended 8 June) and on 8

July during the second harvest cycle in 1988. Water management reduced

A during the dry spring period in 1988. Approximately halfway through

the first harvest cycle on 12 May, a reduction of A was evident in the

RF treatment as compared to the WW, CS1, and CS2 treatments. On 12 May,

the RF treatment had received only 28 mm of rainfall in the last 30

days. As a result of this extended dry period A was reduced by



















Well-watered
C Cyclical-stress 1
M8 Cyclical-stress 2
M1 Rainfed


04
I

E
N
o
0

E



4-
(-

OU


0
c


I..
0
C-,


Figure 16.


Single leaf carbon exchange rate of perennial peanut as
affected by water management treatment on 12, 16, 20 and 31
May, 1 June and 8 July 1988. Different letters above bars
within a harvest date indicate differences at P<0.05
according to Duncan's New Multiple Range Test.


12 MAY 16 MAY 20 MAY 31 MAY 1JUNE 8 JULY

DATE, 1988








83

approximately 40% on 12 May. Thirty-nine millimeters of rainfall on 14

May precluded the further development of water stress and all treatments

had similar A on both 16 and 20 May. Another 30 mm of rainfall occurred

on 25 May; however, reductions in A were observed one week later on 31

May and 1 June in all water stress treatments (CS1, CS2, and RF)

compared to the WW treatment. The observed reduction of A in the water

stress treatments after only one week without rainfall was somewhat

surprising considering the absence of such responses in 1987. However,

absence of pretreatment irrigation, earlier initiation of spring growth,

and depletion of soil water by roots deep in the soil profile during the

dry spring period probably contributed to the observed responses.

Gravimetric soil water content taken on 2 June indicated a reduction in

soil water content under all water stress treatments (CS1, CS2 and RF)

down to a depth of 1200 mm (Fig. 17). Soil water content in the water

stress treatments was reduced at lower depths in the soil profile in

1988 as compared to 1987. Soil water was probably extracted from these

depths during extended periods without rainfall (ie. fall 1987 and

spring 1988), thereby depleting water reserves that the crop may have

had access to in the spring of 1987. Another dry period midway through

the second harvest cycle (30 June to 10 July) also caused A to be

drastically reduced in the CS2 and RF treatments as compared to the WW

and CS1 treatments on 8 July (Fig. 16).

Abundant rainfall between 10 July and 3 October precluded the

development of plant water stress during that time. Single leaf carbon

exchange rate was also unaffected by water management during the fall

dry period even though no rainfall occurred after 3 October.




















SOIL WATER CONTENT, %

0 2 4 6 8 10
150-

300-

450-
E /e-9 Well-watered
E 600- A-A Cyclical-stress 1
0-0 Cyclical-stress 2
,, 750 A-A Rainfed

900

1050

1200


Gravimetric soil water content on 2 June 1988.


Figure 17.








85

Table 13 shows the numerical data for A as illustrated in Figure

16 along with g, E, and Ci/Ca. On 12 May, single leaf carbon exchange

rate, g, E and Ci/Ca were reduced 42, 67, 55 and 20% in the RF treatment

as compared to the average of the WW, CS1 and CS2 treatments.

Calculated Ci was 191 AL L1 in the RF treatment as compared to 231 AL L"

for the average of the WW, CS1, and CS2 treatments, indicating that

relatively low Ci may have contributed to the reduction of A in plants

from the RF treatment. Calculated A in the WW treatment based on g, for

the RF treatment was 12.9 Amol CO2 m2 s' indicating that 82% of the

observed reduction in A under water stress (RF treatment) could be

accounted for by the stomatal component. There was 39 mm of rainfall on

14 May and even though A was not found to be significantly affected by

water treatment on 16 May, g, E and Ci/Ca were reduced 36, 19 and 9%

under the RF treatment as compared to the average of the WW, CS1, and

CS2 treatments. Apparently, the previous plant water deficit prevented

complete stomatal opening after soil water was replenished. Calculated

Ci was 213 pL L' in the RF treatment as compared to 229 AL L1 for the

average of the WW, CS1, and CS2 treatments. On 20 May, no effects of

water management treatment on A, g, E or Ci/Ca were detected.

Rainfall of 32 mm occurred on 25 May; however, visible symptoms of

plant water stress were apparent by 31 May and 1 June. On the average,

A, g, E and Ci/Ca were reduced by 34, 56, 40 and 14% in the CS1, CS2 and

RF treatments as compared to the WW treatment on 31 May. Average

calculated Ci was 206 AL L"' for the CS1, CS2 and RF treatments and 231

AL L'' in the WW treatment. On 1 June the CS2 and RF treatments caused

A, g, E and Ci/Ca to be reduced by 41, 74, 62 and 21% as compared to the











Table 13.


Single leaf carbon exchange rate (A), stomatal conductance
(g), transpiration (E), and the ratio of intercellular to
ambient CO, concentration (Ci/Ca) of perennial peanut as
affected by water management treatment on 12, 16, 20 and 31
May, 1 June and 8 July 1988.


Water management
Date treatment A g E Ci/Ca


j4mol CO2 m"2 s'


12 May Well-watered
Cyclical-stress
Cyclical-stress
Rainfed

16 May Well-watered
Cyclical-stress
Cyclical-stress
Rainfed

20 May Well-watered
Cyclical-stress
Cyclical-stress
Rainfed

31 May Well-watered
Cyclical-stress
Cyclical-stress
Rainfed

1 June Well-watered
Cyclical-stress
Cyclical-stress
Rainfed

8 July Well-watered
Cyclical-stress
Cyclical-stress
Rainfed


20.2
20.5
18.3
11.4

22.6
20.9
21.2
19.0

15.7
17.6
15.1
11.8

19.8
14.7
13.7
11.1

21.1
16.1
13.7
10.0

25.0
21.3
12.1
9.5


cm S-1
1.3 i
1.3 <
1.1
0.4 t


1.5
1.5
1.5
1.0

1.0
1.0
0.8
0.5

1.3
0.7
0.6
0.4

1.8
0.9
0.5
0.4

1.4
1.0
0.3
0.3


mol m2
0.014
0.015
0.015
0.007

0.014
0.014
0.013
0.011

0.013
0.013
0.012
0.009

0.015
0.010
0.010
0.007

0.019
0.014
0.008
0.006

0.017
0.014
0.007
0.006


0.73
0.74
0.71
0.58

0.73
0.76
0.76
0.68

0.70
0.68
0.69
0.61

0.74
0.67
0.62
0.62

0.76
0.70
0.61
0.58

0.67
0.64
0.45
0.44


the same letter do not
Multiple Range test.


* Means within a column on each date followed by
differ at P<0.05 as determined by Duncan's New








87

WW treatment. Average calculated Ci was 195 pL L1 in the CS2 and RF

treatments and 230 IL L" in the WW treatment.

Again these data indicate that in perennial peanut g was reduced

to a proportionately greater degree than was A under plant water

deficits. This in turn would have a direct effect on the observed

reduction in E as well as causing reduced diffusion of CO2 through

stomata. Since A was affected to a lesser extent than g by plant water

deficits, CO2 would continue to be assimilated at a relatively rapid

rate. However, Ci would decline because of restricted gas diffusion

into the leaf (stomatal control), thus causing the observed reductions

in Ci/Ca associated with plant water deficits. Reduced g, accounted for

73 and 78% of the observed reductions in A on 31 May and 1 June,

respectively. Therefore, it appears that the biochemical effect of

plant water stress on A contributed substantially less to the observed

reductions than did the stomatal component.

Leaf water potential measurements taken on 1 June indicated that

0, and 0, were reduced from -1.44 and -1.68 MPa (average of the WW and

CS1 treatments), respectively, to -2.28 and -2.06 MPa under water stress

(average of the CS2 and RF treatments). Calculated 0, was higher in the

WW treatment (0.31 MPa) than in the CS2 and RF treatments (0 MPa) with

b in the CSI treatment (0.16 MPa) not being significantly different

from any of the other treatments.

Measurements taken on 8 July following another dry period (30 June

to 10 July without significant rainfall) indicated that water stress

(CS2 and RF treatments) caused A, g, E and Ci/Ca to be reduced 54, 74,

58 and 32% as compared to the average of the WW and CS1 treatments








88

(Table 13). Average calculated Ci was 144 pL L' in the CS2 and RF

treatments and 203 juL L" on the average in the WW and CS1 treatments.

In this case lowered g, accounted for 94% of the observed reductions in

A for the plants under water stress (CS2 and RF treatments). As

previously mentioned, g was reduced more than A under water stress.

Leaf water potential measurements were also taken during this dry

period. Total leaf water potential in the RF treatment was reduced to

-2.15 MPa as compared to -1.44 MPa for the average of the WW, CS1 and

CS2 treatments. Calculated 0. was not shown to be statistically

different but appeared to decline with increasing water stress (0.32,

0.13, 0.05 and 0 MPa in the WW, CS2, CSI and RF treatments,

respectively).

Even though A was shown to be reduced by plant water stress on

four of the six sampling dates, average A in the water-stressed

treatments (RF on 12 May; CS1, CS2 and RF on 31 May; and CS2 and RF on 1

June and 8 July) was 12 pmol CO, m2 s', and still approximately 60% of

the rate in the nonstressed treatments (21 pmol CO, m2 s") (WW, CS1 and

CS2 on 12 May; WW on 31 May and 1 June; and WW and CS1 on 8 July). On

the average, g was only 36% of that in the nonstressed treatments, thus

indicating that g was more sensitive to plant water stress than A. The

observed reduction in Ci/Ca in the water stress treatments also supports

the more sensitive response of stomata since CO, diffusion into the leaf

is principally controlled by stomata. The Ci/Ca ratio and calculated Ci

were consistently reduced in the water stress treatments, and lowered g,

accounted for 66 to 94% of the observed reductions in A. Even though

leaf water status indicated severe plant water deficits, at no time








89

during the season did A approach zero. In contrast to many field crops

such as corn, soybean, and annual peanut, it appeared that perennial

peanut maintained reasonable rates of A even though plant water status

was considerably reduced.

Plant water deficits did not develop during the remainder of the

season because of plentiful rainfall from 10 July to 3 October.

Measurements taken on 17 and 24 October revealed no differences in A, g,

E or Ci/Ca except on 24 October when a difference in A was detected

(Table 14). Single leaf carbon exchange was reduced 18% in the RF

treatment as compared to the CS2 treatment, and A in the WW treatment

did not differ from either. No data were collected from the CS1

treatment because of a recent irrigation (18 October). Even though

gravimetric data taken on 2 November indicated that there were

differences in soil moisture content at the 150, 300, 450, 600, 750 and

1050 mm depths (Fig. 18), the cooler than average night temperatures

during October may have been limiting growth more than drought.

Single Leaf Carbon Exchange Rate and Specific Leaf N Content

Even though an attempt was made to obtain perennial peanut

leaflets with a wide range of specific leaf N (SLN) content, all samples

fell within a relatively narrow range from 1.2 to 2.2 g N m-2 (Fig. 19).

Two of the data points for Florigraze that had high SLN content but

relatively low A were excluded from the regression analysis. Therefore,

16 data points for Florigraze and 26 for Arbrook were used to compute

linear regression equations in an attempt to characterize the

relationship between SLN content and A (Fig. 19). Since the response




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.8 - mvs