Citation
Water management effects on photosynthate distribution, physiology, and nutritive value of perennial peanut

Material Information

Title:
Water management effects on photosynthate distribution, physiology, and nutritive value of perennial peanut
Creator:
Mansfield, Charles W., 1954- ( Dissertant )
Bennet, Jerry M. ( Thesis advisor )
Boote, Kenneth J. ( Reviewer )
Sollenberger, Lynn E. ( Reviewer )
Wofford, David S. ( Reviewer )
Bowes, George E. ( Reviewer )
Fry, Jack L. ( Degree grantor )
Place of Publication:
Gainesville, Fla.
Publisher:
University of Florida
Publication Date:
Copyright Date:
1990
Language:
English
Physical Description:
x, 128 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Carbon ( jstor )
Dehydration ( jstor )
Forage ( jstor )
Peanuts ( jstor )
Plants ( jstor )
Rain ( jstor )
Rhizomes ( jstor )
Soil water ( jstor )
Water management ( jstor )
Water treatment ( jstor )
Dissertations, Academic -- UF -- Agronomy
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Abstract:
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 droughty 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, 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, repetitively, as compared to the average of the WW and CS1treatments. 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 with 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 int he 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.
Thesis:
Thesis (Ph. D.)--University of Florida, 1990.
Bibliography:
Includes bibliographical references (leaves 122-127).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Charles W. Mansfield.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
023046259 ( alephbibnum )
AHJ9790 ( notis )
22913692 ( oclc )

Downloads

This item has the following downloads:

UF00103063 ( .pdf )

UF00103063 ( .zip )

watermanagemente00mans_0101.txt

watermanagemente00mans_0132.txt

watermanagemente00mans_0129.txt

watermanagemente00mans_0040.txt

watermanagemente00mans_0027.txt

watermanagemente00mans_0070.txt

watermanagemente00mans_0055.txt

watermanagemente00mans_0142.txt

watermanagemente00mans_0095.txt

watermanagemente00mans_0139.txt

watermanagemente00mans_0133.txt

watermanagemente00mans_0069.txt

watermanagemente00mans_0061.txt

watermanagemente00mans_0082.txt

watermanagemente00mans_0022.txt

watermanagemente00mans_0071.txt

watermanagemente00mans_0116.txt

watermanagemente00mans_0087.txt

watermanagemente00mans_0025.txt

watermanagemente00mans_0052.txt

watermanagemente00mans_0080.txt

watermanagemente00mans_0046.txt

watermanagemente00mans_0077.txt

watermanagemente00mans_0143.txt

watermanagemente00mans_0081.txt

watermanagemente00mans_0047.txt

watermanagemente00mans_0110.txt

watermanagemente00mans_0090.txt

watermanagemente00mans_0075.txt

watermanagemente00mans_0104.txt

watermanagemente00mans_0125.txt

watermanagemente00mans_0002.txt

watermanagemente00mans_0063.txt

watermanagemente00mans_0010.txt

watermanagemente00mans_0049.txt

watermanagemente00mans_0023.txt

watermanagemente00mans_0028.txt

watermanagemente00mans_0121.txt

watermanagemente00mans_0045.txt

watermanagemente00mans_0083.txt

watermanagemente00mans_0011.txt

watermanagemente00mans_0123.txt

watermanagemente00mans_0024.txt

watermanagemente00mans_0109.txt

watermanagemente00mans_0100.txt

watermanagemente00mans_0120.txt

watermanagemente00mans_0131.txt

watermanagemente00mans_0072.txt

watermanagemente00mans_0038.txt

watermanagemente00mans_0092.txt

watermanagemente00mans_0135.txt

watermanagemente00mans_0076.txt

watermanagemente00mans_0118.txt

watermanagemente00mans_0093.txt

watermanagemente00mans_0041.txt

watermanagemente00mans_0036.txt

watermanagemente00mans_0113.txt

watermanagemente00mans_0013.txt

watermanagemente00mans_0019.txt

watermanagemente00mans_0012.txt

watermanagemente00mans_0034.txt

watermanagemente00mans_0134.txt

watermanagemente00mans_0020.txt

watermanagemente00mans_0004.txt

watermanagemente00mans_0016.txt

watermanagemente00mans_0033.txt

watermanagemente00mans_0008.txt

watermanagemente00mans_0026.txt

watermanagemente00mans_0091.txt

watermanagemente00mans_0099.txt

watermanagemente00mans_0084.txt

watermanagemente00mans_0015.txt

watermanagemente00mans_0111.txt

watermanagemente00mans_0097.txt

watermanagemente00mans_0096.txt

watermanagemente00mans_0066.txt

watermanagemente00mans_0007.txt

watermanagemente00mans_0067.txt

watermanagemente00mans_0062.txt

watermanagemente00mans_0136.txt

watermanagemente00mans_0059.txt

watermanagemente00mans_0031.txt

watermanagemente00mans_0107.txt

watermanagemente00mans_0006.txt

watermanagemente00mans_0074.txt

watermanagemente00mans_0068.txt

watermanagemente00mans_0079.txt

watermanagemente00mans_0064.txt

watermanagemente00mans_0053.txt

watermanagemente00mans_0115.txt

watermanagemente00mans_0102.txt

watermanagemente00mans_0085.txt

watermanagemente00mans_0105.txt

watermanagemente00mans_0130.txt

watermanagemente00mans_0042.txt

watermanagemente00mans_0057.txt

watermanagemente00mans_0058.txt

watermanagemente00mans_0017.txt

watermanagemente00mans_0108.txt

watermanagemente00mans_0140.txt

watermanagemente00mans_0094.txt

watermanagemente00mans_0086.txt

watermanagemente00mans_0056.txt

watermanagemente00mans_0030.txt

watermanagemente00mans_0138.txt

watermanagemente00mans_0005.txt

watermanagemente00mans_0127.txt

watermanagemente00mans_0035.txt

watermanagemente00mans_0114.txt

watermanagemente00mans_0048.txt

watermanagemente00mans_0122.txt

watermanagemente00mans_0073.txt

watermanagemente00mans_0088.txt

watermanagemente00mans_0126.txt

watermanagemente00mans_0103.txt

watermanagemente00mans_0032.txt

watermanagemente00mans_0029.txt

watermanagemente00mans_0001.txt

watermanagemente00mans_0117.txt

watermanagemente00mans_0141.txt

watermanagemente00mans_0043.txt

watermanagemente00mans_0037.txt

watermanagemente00mans_0119.txt

watermanagemente00mans_0014.txt

watermanagemente00mans_0124.txt

watermanagemente00mans_0128.txt

watermanagemente00mans_0137.txt

watermanagemente00mans_0065.txt

watermanagemente00mans_0051.txt

watermanagemente00mans_0078.txt

watermanagemente00mans_0054.txt

watermanagemente00mans_0050.txt

watermanagemente00mans_0089.txt

watermanagemente00mans_0018.txt

watermanagemente00mans_0044.txt

watermanagemente00mans_0106.txt

watermanagemente00mans_0021.txt

watermanagemente00mans_0039.txt

watermanagemente00mans_0098.txt

watermanagemente00mans_0112.txt

watermanagemente00mans_0060.txt

UF00103063_pdf.txt


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




Full Text

PAGE 1

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

PAGE 2

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 i i

PAGE 3

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. ; i i

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGMENTS i i LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . V LIST OF FIGURES . ................ .. .. .. ......................... vii ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . i X 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 ANO 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 ANO CONCLUSIONS ......................................... 104 APPENDIX ........... ........................... .. ............. . 109 LITERATURE CITED ............................................. .. 122 BIOGRAPHICAL SKETCH ............................................. 128 iv

PAGE 5

LIST OF TABLES Table !@.gg 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 (f), and the ratio of intercellular to ambient CO 2 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 V

PAGE 6

Table 12. Single leaf carbon exchange rate (A), stomatal conductance (g), transpiration (f), and the ratio of intercellular to ambient CO 2 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 (f), and the ratio of intercellular to ambient CO 2 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 (f), and the ratio of intercellular to ambient CO 2 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 vi

PAGE 7

LI ST OF FI GU RES Figure 1. Daily rainfall distribution during the growing season for perennial peanut in 1987 at Gainesville, Florida ........... 31 2. Daily rainfall distribution during the growing season for perennial peanut in 1988 at Gainesville, Florida ........... 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. ..... ..................... ........... 39 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 .................................. 43 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 ................ ............................... 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 P~0.05 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 vii

PAGE 8

Figure 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

PAGE 9

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 g1abrata Benth.) is a relatively drought tolerant tropical legume with potential for production of high quality forage on many acres of infertile and draughty soils of Florida. However, information is lacking on the response of perennial peanut to irrigation. Dry matter (OM) 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 (CSl), 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 ix

PAGE 10

growth during periods of water deficit. Above-ground OM in the RF treatment was reduced 18 and 26% in 1987 and 1988 respectively, as compared to the average of the WW and CSl treatments Above-ground OM 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. X

PAGE 11

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

PAGE 12

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 (Arach;s 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, 1

PAGE 13

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. T his 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 i n 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 v i tro organic matter digestibility and nitrogen concentration of the plant and its

PAGE 14

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).

PAGE 15

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 4

PAGE 16

5 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 (G1ycjne max L.) more 1 4 C 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 (Arachjs hypogaea L.) was unaffected by water management on a fine sand

PAGE 17

6 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.

PAGE 18

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 Gladjolus 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.

PAGE 19

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 (lea 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

PAGE 20

9 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 C higher than for well-watered plants, indicating that stomatal closure had significantly reduced transpiration. Sharratt et al. (1983) reported a 0.2 mm h 1 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.

PAGE 21

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 ~anopy 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 Schulze (1983) found that carbon assimilation of cowpea

PAGE 22

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%. Kuppers et al. (1988) examined the relationship between stomatal conductance and intercellular CO 2 concentration (Ci) in cowpea. No unique relationship between Ci or the ratio of ambient CO 2 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 13 C in the nonirrigated treatment indicating a reduction in CO 2 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

PAGE 23

12 (Macroptilium atropurpureum) in a semi-a r id 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 (osmotic adjustment), whereas siratro was able to

PAGE 24

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

PAGE 25

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

PAGE 26

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

PAGE 27

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

PAGE 28

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).

PAGE 29

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 1 of 0-5-25 P205-l<:i0) containing 70, 8, 5, 2, and 91 g kg 1 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 1 of 0-0-61 (N-P 2 0 5 -K 2 0) 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 1, was planted on 30 and 31 January 1986, in 15 cm rows at the rate of 8.7 m 3 ha 1 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-1. Weeds were controlled with appropriate herbicide treatments and by hand18

PAGE 30

19 weeding during the 1986 season. Premerge' (dinoseb, 2-sec butyl 4,6dinitrophenol) 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-1H2 l,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

PAGE 31

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 1 of 0-0-61 P205-~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-D' (2,4-Dichlorophenoxyacetic acid) applied on 1 March, and summer annual grassy and broadleaf weeds were controlled with a preemergence application of 'Prowl' (N-(l-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 (CSl) 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 CSl 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.

PAGE 32

21 For the remainder of the season, soil water recharge for the CSl 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 gin 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 CSl 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 (CSl), 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

PAGE 33

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 CSl 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 m onitored 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 95C 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 (OM) as well as total nonstructural carbohydrate {TNC) distribution in stems, leaves, and

PAGE 34

-------------------23 rhizomes. On sampling dates, two 0.5 m 2 quadrats from each plot were clipped to ground level to determine above-ground OM, and four soil core samples (9.5 cm in diameter) were extracted to a depth of 20 cm from each quadrat to determine rhizome OM. 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 OM 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 OM (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 OM 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

PAGE 35

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 OM 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 cm 3 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 Regrowth After Clipping Regrowth from the staging height of 5 cm was monitored in 220 cm 2 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.

PAGE 36

25 Physiological Measurements Single leaf carbon exchange rate (A), stomatal conductance, and transpiration were determined with a LI-C0R 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 30C 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 (~L) of each sample. Each psychrometer had been

PAGE 37

I 26 previously calibrated using known concentrations of NaCl solutions to develop a regression equation of water potential versus psychrometer output. After ~L 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 (~rr) was determined by the method previously described. Leaf turgor (~p) was calculated by subtracting W rr from w l Leaf Carbon Exchange Rate and Leaf Nitrogen Concentration Leaflets of Florigraze perennial peanut from the WW and CSl 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 mols CO 2 m 2 s 1 2) A from 15 to 17 mols CO 2 m s 3) A from 18 to 20 mols CO 2 m 2 1 s 4) A from 21 to 23 mols CO 2 m 2 s ,, 5) A from 24 to 26 mols CO 2 2 -1 s and 6) A greater than or equal to 27 mols CO 2 m 2 s 1 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 CSl 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 CSl treatment.

PAGE 38

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 1 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-C0R 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

PAGE 39

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 OM subsamples as well as on rhizome OM 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 OM subsamples and the forage subsamples were also analyzed for in vitro digestion. In vitro organic matter digestibility (IVOMO) was determined by the two-stage technique of in vitro digestion (Tilley and Terry, 1963) as modified by Moore and Mott {1974).

PAGE 40

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 29

PAGE 41

30 Table 1 Monthly rainfall for 1987 and 30-year mean monthly rainfall for Gainesville, Florida. Month January February March April May June July August September October November December Total Rainfall 1987 30-year mean -------------mm------------106 83 138 100 261 93 11 74 108 106 75 167 99 176 137 204 93 143 7 59 109 51 30 82 1156 1338

PAGE 42

E E .. C .... C C 0:: 100 100 :::0 C ::J -+, 50 50 C 3 n 3 0 0 SEPT OCT MONTH, 1987 Figure 1. Daily rainfall distribution during the growing season for perennial peanut in 1987 at Gainesville, Florida. w

PAGE 43

32 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 CSI RF ------------------mm----------Pre-treatment 1 Jan. 24 Apr. 517 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 tAbbreviations are as follows: WW) well-watered, SP) speedling, CSI) cyclical-stress 1, RF) rainfed. 93 93

PAGE 44

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, CSl 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 CSl 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

PAGE 45

34 Table 3. Monthly rainfall for 1988 and 30-year mean monthly rainfall for Gainesville, Florida. Month January February March April May June July August September October November December Total Rainfall 1988 30-year mean -------------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 1558 1338

PAGE 46

E E C ..... C C a::: 180 180 120 120 ;o C ::::J -+, C 3 t 60 60 3 01 MAY I JUNE I JULY 11 MB IW w1rm 0 f I AUG OCT MONTH, 1988 Figure 2. Daily rainfall distribution during the growing season for perennial peanut in 1988 at Gainesville, Florida. w u,

PAGE 47

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 4.5C below the normal with temperatures dropping below 10c 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 CSI 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

PAGE 48

37 Table 4. Rainfall and irrigation specific to each harvest cycle as well as before and after the perennial peanut growing season in 1988. Irrigation treatment st Interval of time Dates Rainfall WW CSl 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, CSl) cyclical-stress 1, RF) rainfed.

PAGE 49

38 spring and summer. Total irrigation applied in 1988 was 430, 221, 120, and O mm in the WW, CSl, 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 CSl 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. Photosynthate Accumulation and Distribution 1987 Season Above-ground DM production during the 1987 growing season as well as the effect of water management on OM 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 OM accumulation during the 1987 season. Even though there were drying conditions in late May and early June,

PAGE 50

39 Well watered N 500 mm Speedhng I aa E a D Cyclical-stre ss 1 O'I CC Rainfed 400 a a::: w be :::E 300 ... g a C a a i"I ca C ' 200 ... z :::> I 0 a:: c., 100 I ' ' ' 0 0 I. MAY JUNE JULY SEP NOV 1987 HARVESTS Figure 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 P50.05 according to Duncan's New Multiple Range Test

PAGE 51

40 they had no effects on above-ground OM 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 OM accumulation during the third harvest cycle (24 July 8 September) in both the CSl 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 OM 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 OM 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 OM to occur in the rhizomes and less to occur

PAGE 52

41 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 Abovedatet treatment ground Rhizome Total Forage* LAI 2 --------------gm------------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 Ra i nfed 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 Ra i nfed 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 t 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.

PAGE 53

42 above-ground in the RF treatment as compared to the other three treatments; however, total OM accumulation was not different among any of the treatments. As expected, total OM 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 m 2 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 OM 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 OM than the other treatments (Table 5). Regrowth was also

PAGE 54

E E .. :r: I0... w 0 43 ROOT LENGTH DENSl1Y, cm cm 3 0.0 1 .0 2.0 3.0 4.0 50 150 300 450 600 750 900 1050 1200 ---Well-watered Speedling c=J Cyclical stress 1 Rainfed Figure 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

PAGE 55

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 cm 2 of ground area. Stubble units produced the most shoots during regrowth with rhizomes only giving rise to four shoots per 100 cm 2 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 OM at the end of the third harvest cycle on 8 September (Table 5). Above-ground OM was reduced 14 and 18% in the CSl 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 CSl 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 cm 2 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 CSl and RF treatments. New shoots originating from rhizomes showed a significant reduction on 5 September as a result of

PAGE 56

45 water stress, being reduced from 4 stems per 100 cm 2 in the WW treatment to 1 stem per 100 cm 2 in the CSl 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 effect(s) of previous or current plant water stress on growth of perennial peanut. Reductions in above ground OM 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 OM 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 OM was reduced 17% in the RF treatment as compared to the average of the WW and CSl treatments. However, average forage yield was negligible at 22 gm~ 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 OM 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

PAGE 57

46 cycles), the proportion of rhizome OM declined; however, it later increased to a very high value at the November harvest date. Averaged across water management treatments, rhizome OM constituted approximately two-thirds of the total OM throughout the season except in the fall (2 November) when rhizome OM was 85% of total OM. Rhizome OM increased 43% during the last harvest cycle, actually causing total OM 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 OM 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

PAGE 58

47 1400 ~Ab d ove-groun E!l3 Rhizome 1200 N I 1000 E C, 800 w i 600 400 C 200 0 MAY JUNE JULY SEP NOV 1987 HARVESTS Figure 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.

PAGE 59

C'I w a:::: Q :::r: 0 CD a:::: (3 400 300 200 w 100 0 N :::r: a:::: MAY JUNE 48 Well watered Speedling CJ Cyclical-stress 1 Rainfed JULY SEP 1987 HARVESTS a NOV Figure 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 P~0.05 according to Duncan's New Multiple Range Test.

PAGE 60

49 (ended 8 September). Above-ground OM 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 OM 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 CSl 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 1 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 treatment(s) (CSl and RF or RF alone) cause a reduction in above-ground OM or cause rhizome TNC to significantly increase (Figs. 3 and 6). Averaged across water management treatments, cumulative above ground OM production for the 1987 season was 1019 g m 4 (Table 6). Above-ground OM accumulation was 440 g m 2 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 OM for on 23 July (316 g 2 ) represents only the forage OM taken during the second harvest cycle,

PAGE 61

50 Table 6. Above-ground, rhizome, total, and forage dry matter production by perennial peanut averaged across water management treatments in 1987. Date 11 June 23 July 8 Sep. 2 Nov. Cumulative total Dry Matter Above ground Rhizome -----------------------9 440 111t 316* -97 241 154 ____u* 326 1019 494 Total Forage m~--------------------551 298 219 316 395 348 1513 241 ..11 877 t Represents average dry matter accumulation in rhizomes between the 15 May and 11 June sampling dates. Represents dry matter accumulated above the 5-cm stubble height (forage)

PAGE 62

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 OM for 8 September (241 g 2 ) and 2 November (22 g m 2 ) corresponded to forage OM. Therefore, summing the above-ground OM accumulated for each harvest cycle gives a cumulative OM production of 1019 g m 2 for the season. Average rhizome OM present increased from 580 g m 2 on 15 May to 107 4 g m 2 on 2 November, representing a net increase of 494 g m 2 Summing above-ground and rhizome OM, total OM accumulated for the season was 1513 g m 4 Assuming a 184-day growth period from May through October, the average daily rate of OM accumulation was computed to be 8.2 g m 2 Summing forage OM accumulation from each harvest cycle gives a total of 877 g m 4 for seasonal forage yield. This is similar to previous seasonal forage OM yields reported by Prine et al. (1981) and Valentim (1987). Seasonal OM accumulation and distribution for each water management treatment is shown in Table 7. Cumulative OM 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 OM in the RF treatment as compared to the WW treatment. 1988 Season Seasonal distribution of above-ground OM 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 OM accumulation on different harvest dates. Below average rainfall

PAGE 63

52 Table 7. Seasonal dry matter accumulation and distribution as affected by water management treatment in 1987. Water management treatment Well-watered Speedling Cyclical-stress Rainfed 1 Above ground Dry Matter Rhizome Total Forage -----------------------9 m~------------------1101 a* 561 a 1043 a 460 a 882 a 585 a 1662 a 1503 a 1467 a 990 a 903 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.

PAGE 64

N 600 I -~ E C'I 500 ... r:L w 400 .... ::::l: a a 300 0 0 z => 200 0 a:: (!) I 100 0 b ... .. .. ' MAY 53 a C a b C C 0 cc C ... I', ' ' ... ... ... ... JUNE JULY SEP 1988 HARVESTS Well watered D Cyclical-str mm Cycliccl-str ess 1 ess 2 lSS'SI Rainfed aaa 0 NOV Figure 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

PAGE 65

54 during the spring of 1988 caused a reduction in above-ground OM accumulation in the water stressed (CS2 and RF) treatments as compared to the WW and CSl treatments during that time. The effect of drying conditions on the reduction of OM 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 OM 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 OM accumulation causing it to be reduced in the RF treatment as compared to the WW and CSl 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 OM as illustrated in Figure 7 as well as total, rhizome, and forage OM 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 OM. As previously mentioned, below average rainfall during the spring and early summer caused a reduction in above-ground OM 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

PAGE 66

55 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 Abovedatet treatment ground Rhizome Total Forage* LAI 2 --------------gm-------------19 May Well-watered 337 a* 937 a 1274 a 3.4 a Cyclical-stress I 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 We l l-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 I 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. 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.

PAGE 67

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 OM 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 OM 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 OM being reduced 11 and 23% in the CS2 and RF treatments, respectively, as compared to the average of the WW and CSl 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 OM in the RF treatment being only 67% of that present as the average of the WW and CSI treatments. The CSl and CS2 treatments were not different on this sampling date; however, the CS2 treatment had accumulated 16% less above-ground OM 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 CSl treatments (4.4). In addition, a difference in specific leaf area (SLA) was detected at this sampling with leaves in

PAGE 68

57 the RF treatment being thicker and having lower SLA (136 cm 2 g~) as compared to 1 eaves in the WW treatment ( 148 cm 2 g 1 ). At the end of the first harvest cycle (8 June) above-ground OM was reduced 20 and 43% by water stress in the CS2 and RF treatments, respectively, as compared to the average of the WW and CSl 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 m 2 ) as compared to the average of the WW and CSl treatments ( 479 g m 2 ). Forage OM in the CS2 treatment was also reduced 36% as compared to the average of the WW and CSl 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 CSl 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 cm 2 g 1 ) as compared to the average of the WW and CSl (149 cm 2 g 1 ), and CSl and CS2 (144 cm 2 1 ) 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 OM accumulation as affected by water treatment. However, there does appear to be a trend for reduced total OM accumulation in the CS2 and RF treatments as compared to the WW and CSl 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

PAGE 69

58 observed for all water stress treatments (CSl, 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 CSl 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 CSl treatments. Water management had no effect on OM accumulation and distribution, LAI, or SLA during the rest of the season (third and fourth harvest cycles which ended 12 September and 1 November,

PAGE 70

0 150 300 450 E E 600 ::r: .,_ 750 a.. w a 900 1050 1200 2 59 SOIL WATER CONTENT, 4 6 8 10 12 Well-watered AA Cyclical stress 1 0-0 Cyclical stress 2 l:J..-l:J.. Rainfed Figure 8. Gravimetric soil water content on 8 June 1988.

PAGE 71

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 OM 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 OM present in rhizomes increased. On average, rhizome OM constituted most of the total OM 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 OM in rhizomes decreased, however it increased again during the fall. Averaged across water management treatments, rhizome OM again constituted approximately two-thirds of the total OM present during any given harvest cycle except in the fall (fourth harvest cycle ended I November) when rhizomes were 84% of total OM. Average OM 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 I November) at the expense of above-ground OM accumulation, and forage OM 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 OM at the end of 1988 (1 November) was similar to that at the beginning of

PAGE 72

61 1400 D Above-ground Rhizome 1200 N I 1000 E O"I BOO ; 0::: w "' 600 400 0 200 0 MAY JUNE JULY SEP NOV 1988 HARVESTS Figure 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.

PAGE 73

62 1988 (19 May). This is in contrast to the large increase in rhizome OM 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 OM 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 OM 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 OM 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 CSI treatments. On 8 June, average rhizome TNC in the RF and CS2 treatments (349 g kg~) was 8 and 17% higher than in the CSI and WW

PAGE 74

63 400 ...... I C, .:,/. C, 300 w 0 200 0 m 0:::: w 100 0 N I 0:: 0 a a CJ Cyclical stress 1 a a~ !88 Cyclical stress 2 b Ii, Rainfed bb I'' C a ' a0 a ' ' ' ' ' ab 0 ' be ' 0 a a ' C Q!li; ' r"'I ' ' ' ' I\ ' I\ !'I ' I\ Well-watered 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.

PAGE 75

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 1 ) as compared to the average of the WW, CSl 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 1, 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 (20C) was 2c warmer than in 1987, and average soil temperature at 100 mm was 4C higher (24C in April 1988 as compared to 20C 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

PAGE 76

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 gm~ (Table 9) Average rhizome OM remained practically the same during the 1988 season and actually decreased slightly from 994 g m 2 on 12 May to 895 g m 2 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 gm~. Again, assuming a 184day growing season from May through October, 5. 2 g m 2 d 1 was accumulated on the average. Summing forage DM from each of the harvest cycles gives an average cumulative forage yield of 1008 gm~. 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;

PAGE 77

66 Table 9. Above-ground, rhizome, total, and forage dry matter production by perennial peanut averaged across water management treatments in 1988. Date 8 June 25 July 12 Sep. 1 Nov. Cumulative total Dry Matter Above ground Rhizome -----------------------9 418 -1st 344* -294 301 39 171 -1063 -99 Total Forage 2 m ----------------------403 363 50 344 340 301 -111 964 1008 t Represents average dry matter accumulation in rhizomes between the 15 May and 11 June sampling dates. Represents dry matter accumulated above the 5-cm stubble height (forage).

PAGE 78

67 Table 10. Seasonal dry matter accumulation and distribution as affected by water management treatment in 1988. Water management treatment Well-watered Cyclical-stress Cyclical-stress Rainfed 1 2 Above ground Dry Matter Rhizome Total Forage -g m 2 1194 a* -117 a 1077 a 1167 a 1133 ab 9 a 1142 a 1124 a 1057 b -149 a 908 b 967 b 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.

PAGE 79

68 however, the CSl and CS2 treatments were not different. The CS2 and RF water management treatments caused a significant reduction in cumulative total and cumulative forage OM in 1988 as compared to the average of the WW and CSl treatments. The CS2 and RF treatments produced 18 and 35% less total DM, and 16 and 33% less forage OM, respectively, than the average of the WW and CSl 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 CSl treatment in 1988. Phvsiological 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 similar to those observed by Lopez et al. (1988) for pigeonpea, Nicolodi et al. (1988) for alfalfa, Sinclair and Horie (1989) for rice (Oryza sativa L.) and soybean, and Turner et al. (1985) for sunflower. Twenty-five millimeters of rain accumulated between 12 and 16 May and 75 mm of rainfall occurred from 19 to 21 May. Apparently those rainfall events provided sufficient water to prevent plant water deficits in the CSl, SP, and RF treatments during May. Nevertheless, gravimetric soil water content on 31 May was significantly lower in the SP, CSl, and RF treatments as compared to the WW treatment down to a depth of 750 mm in

PAGE 80

...... I en 30 N I E N 25 0 u 0 20 E :::t. 15 (I) -+-' 0 0::: (I) 10 Cl C 0 ..c 0 5 X w C 0 0 .!l I... 0 u 69 Well-watered Speedling D Cyclical-stress 1 Rainfed a aa M a i a a a a I\ a a a a b a a9~ I!; g" 9~ a ) r"I r"I b I) r"I I> I\ I'.: I> ' ) g, ' I\ M M M ) M )I ' ) ' t ) ) w ) ) " ' ) ' ' I> ' ) ' ) ' ) ' ' ) ) r"I ' 18 MAY 26 MAY 29 MAY 8 JUNE 10 JUNE 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.

PAGE 81

70 the soil profile (Fig. 12). On 8 and 10 June, the CSl 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 CSl, 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

PAGE 82

E E :I: Ia. w C 0 2 150 300 450 600 750 900 1050 1200 71 SOIL WATER CONTENT, 4 6 8 10 12 Well-watered 0-0 Speedling _... _... Cyclical stress 1 A-A Rainfed Figure 12. Gravimetric soil water content on 31 May 1987.

PAGE 83

E E LL.I C 0 2 150 300 450 600 750 900 1050 1200 72 SOIL WATER CONTENT, 4 6 8 10 12 Well-watered o-o Speedling .a.-.a. ~clical-stress 1 6-6 Rainfed Figure 13. Gravimetric soil water content on 10 June 1987.

PAGE 84

I en N I E N 0 (.) 30 25 0 20 E ::t cv 15 +-' C 0::: Q) C'I C: C .c 0 X uJ 10 5 a 0 ...___ __ 73 a a a a a Well-watered CJ Cyclical-stress 1 Rainfed a a a a ab b C 0 ..c .... 0 (.) 27 AUG 28 AUG I 7 OCT 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.

PAGE 85

74 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 CSl 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 CSl 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 i n 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 (), 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 mol CO 2 m 2 s\ 1.6 cm s 1 and 0 016 mol m 2 s\ respectively. A

PAGE 86

75 SOIL WATER CONTENT, 0 2 4 6 a 10 12 150 Well-watered 300 A-A Cyclical-stress 1 ll.-6 Rainfed 450 E E 600 750 a.. w 0 900 1050 1200 Figure 15. Gravimetric soil water content on 30 Oct. 1987.

PAGE 87

76 Table 11. Single leaf carbon exchange rate (A), stomatal conductance (g), transpiration (f), and the ratio of intercellular to ambient CO 2 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 CO 2 m2 1 cm s 1 mol m2 s1 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 I 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 Means within a column on each date followed by the same letter do not differ at Pi0.05 as determined by Duncan's New Multiple Range test.

PAGE 88

77 significant reduction in the ratio Ci/Ca was detected on 18 May in the CS! 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 L L 1 } to maximize A (at ambient CO 2 }. Leaf carbon exchange rate has been shown to respond linearly to Ci up to 230 L L 1 in various C 3 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 (CSl and RF} (Table 11). Single leaf carbon exchange rate, g and f were 28, 45 and 25% lower, respectively, under water stress (CSl 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 L L 1 in the RF treatment and 223 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 CS! treatments. Calculated Ci values for the RF treatment and the mean of the other three treatments were 208 and 227 L L 1, 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 Kuppers 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

PAGE 89

78 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 L L 1 ) 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 L, ~ or Vp Mean values for L, ~ and VP 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 CSl treatment caused vl and ff 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 ~Land ff were found, they were apparently insufficient to cause

PAGE 90

79 Table 12. Single leaf carbon exchange rate (A), stomatal conductance (g), transpiration (f), and the ratio of intercellular to ambient CO 2 concentration (Cj/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 C;/Ca mol CO 2 m 2 s 1 cm s 1 mo l m 2 s 1 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 Means within a column on each date followed by the same letter do not differ at P~0.05 as determined by Duncan s New Multiple Range test

PAGE 91

80 perturbances in A. Unfortunately, no data with respect to ~L 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 c;;ca on 7 or 22 October (Table 12). Average A, g, E, and CUCa were 21.9 mol CO 2 m 2 s 1, 1.56 cm s\ 0.017 mol m 2 s, and 0.75, respectively, on 7 October; and 20.2 mol CO 2 m 2 s 1, 3.60 cm s, 0 019 mol m 2 s 1, and 0.89, respectively, on 22 October. On 29 October, water stress (RF treatment) significantly reduced A, g, E, and Cj/Ca by 34, 53, 44, and 10%, respectively, as compared to the WW treatment. Calculated c; was 218 and 238 L L 1 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 c;;ca was reduced in the RF treatment as compared to the WW and CSl treatments, it appears that calculated Ci (218 L L 1 ) 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 2 (gc) from the RF treatment according to the method of Nicolodi et al. (1988). Calculated A for the WW treatment based on gc from the RF treatment was 13. 6 mo l CO 2 m 2 s 1 when in fact actual A was 11.6 mol CO 2 m 2 s 1 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

PAGE 92

81 (biochemical) mechanisms. Similar results were published by Nicolodi et al. (1988) for water stressed alfalfa Total leaf water potential and ~Palso 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 -I.I and 0.9 MPa on the average in the WW and CSI treatments. It is noteworthy that leaves with a ~L 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, CSl, and CS2 treatments. On 12 May, the RF treatment had received only 28 mm of r ainfall in the last 30 days. As a result of this extended dry period A was reduced by

PAGE 93

..... I (I) 25 N I E N 20 0 u 0 E 15 ::t Cl) .,J 0 0::: 10 Cl) O"I C 0 ..c 5 u X w C 0 0 .0 I... C CJ a aa ca a a r,. r,. " r,. b r,. r E " E "' r en "' r " C D e r r > r,. " 82 Well-watered D Cyclical-stress 1 00 Cyclical-stress 2 a Rainfed a a a ~b b bb be a ~b E D C a r D r a E D r D D p "' r r 0 D r, r n p D C "' r e p D D r l r r D r p D r,. D r,. D r I a b b D r r r r r r r r,. 12 MAY 16 MAY 20 MAY 31 MAY 1 JUNE l 8 JULY DATE, 1 988 Figure 16. Single l eaf 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 P50 05 accordi n g to Duncan's New Multiple Range Test

PAGE 94

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 (CSl, 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 (CSl, 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 CSI 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.

PAGE 95

E E b: w 0 0 2 150 300 450 600 750 900 1050 1200 84 SOIL WATER CONTENT, ,; 4 6 8 10 Well-watered &-A Cyclical-stress 1 0-0 Cyclical-stress 2 l:l.-l:l. Rainfed Figure 17 Gravimetric soil water content on 2 June 1988.

PAGE 96

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, CSl and CS2 treatments. Calculated Ci was 191 L 1 in the RF treatment as compared to 231 L L 1 for the average of the WW, CSl, 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 gc for the RF treatment was 12.9 mol CO 2 m 2 s 1 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, CSl, and CS2 treatments. Apparently, the previous plant water deficit prevented complete stomatal opening after soil water was replenished. Calculated Ci was 213 L L 1 in the RF treatment as compared to 229 L 1 for the average of the WW, CSl, 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 CSl, CS2 and RF treatments as compared to the WW treatment on 31 May. Average calculated Ci was 206 L l 1 for the CSl, CS2 and RF treatments and 231 L L 1 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

PAGE 97

86 Table 13. Single leaf carbon exchange rate (A), stomatal conductance (g), transpiration (f), and the ratio of intercellular to ambient CO 2 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. Date 12 May 16 May 20 May 31 May 1 June 8 July Water management treatment A We 11-watered Cyclical-stress Cyclical-stress Rainfed mo 1 CO 2 m 2 s 1 20.2 a* 1 20.5 a 2 18.3 a Well -watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well -watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed 11. 4 b 22.6 a 20.9 a 21. 2 a 19.0 a 15.7 a 17.6 a 15.1 a 11.8 a 19.8 a 14.7 b 13.7 b 11.1 b 21.1 a 16.1 ab 13. 7 be 10.0 C 25.0 a 21.3 a 12.1 b 9.5 b g 1.3 a 1.3 a 1. 1 a 0.4 b 1. 5 a 1. 5 a 1. 5 a 1.0 b 1.0 a 1.0 a 0.8 a 0.5 a 1.3 a 0.7 b 0.6 be 0.4 C 1.8 a 0.9 ab 0.5 b 0.4 b 1.4 a 1.0 a 0.3 b 0.3 b E 0.014 a 0.015 a 0.015 a 0.007 b 0.014 a 0.014 a 0.013 a 0.011 b 0.013 a 0.013 a 0.012 a 0.009 a 0.015 a 0.010 b 0.010 be 0.007 C 0.019 a 0.014 b 0.008 be 0.006 C O. 017 a 0.014 a 0.007 b 0.006 b Ci/Ca 0.73 a 0.74 a 0. 71 a 0.58 b 0.73 a 0.76 a 0.76 a 0.68 b 0.70 a 0.68 a 0.69 a 0.61 a 0.74 a 0.67 b 0.62 b 0.62 b 0.76 a 0.70 ab 0.61 be 0.58 C 0.67 a 0.64 a 0.45 b 0.44 b Means within a column on each date followed by the same letter do not differ at Pi0,05 as determined by Duncan's New Multiple Range test.

PAGE 98

87 WW treatment. Average calculated Ci was 195 LL _, in the CS2 and RF treatments and 230 L l 1 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 inf as well as causing reduced diffusion of CO 2 through stomata. Since A was affected to a lesser extent than g by plant water deficits, CO 2 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 gc 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 I June indicated that ~Land~~ were reduced from -1.44 and -1.68 MPa (average of the WW and CSl treatments), respectively, to -2.28 and -2.06 MPa under water stress (average of the CS2 and RF treatments). Calculated ~p was higher in the WW treatment (0.31 MPa) than in the CS2 and RF treatments (0 MPa) with ~Pin the CSl 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 CSl treatments

PAGE 99

88 (Table 13). Average calculated Ci was 144 L L 1 in the CS2 and RF treatments and 203 L l 1 on the average in the WW and CSl treatments. In this case lowered gc 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, CSI and CS2 treatments. Calculated ~p was not shown to be statistically different but appeared to decline with increasing water stress (0.32, 0.13, 0.05 and O MPa in the WW, CS2, CSl 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; CSI, CS2 and RF on 31 May; and CS2 and RF on 1 June and 8 July) was 12 mol CO 2 m 2 s 1, and still approximately 60% of the rate in the nonstressed treatments (21 mol CO 2 m 2 1 } (WW, CSI and CS2 on 12 May; WW on 31 May and 1 June; and WW and CSl 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 2 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 gc accounted for 66 to 94% of the observed reductions in A. Even though leaf water status indicated severe plant water deficits, at no time

PAGE 100

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, for 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 CSI 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 Nm~ (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

PAGE 101

90 Table 14. Single leaf carbon exchange rate (A), stomatal conductance (g), transpiration (), and the ratio of intercellular to ambient CO 2 concentration (C;/Ca) of perennial peanut as affected by water management treatment on 17 and 24 Oct. 1988. Water management Date treatment A g E CUCa mol CO 2 m 2 s 1 cm s 1 mol 2 s 1 17 Oct. Well-watered 22.8 a* 1. 1 a 0.012 a 0.67 a Cyclical-stress 1 21. 5 a 1.0 a 0.012 a 0.67 a Cyclical-stress 2 19.6 a 0.9 a 0.012 a 0.65 a Rainfed 22.9 a 1. 1 a 0 010 b 0.66 a 24 Oct. Well-watered 21. 7 ab 1. 1 a 0.011 a 0.68 a Cyclical-stress 1 Cyclical-stress 2 24.4 a 1. 1 a 0.011 a 0.65 a Rainfed 20.0 b 1.0 a 0. 011 a 0.67 a Means within a column on each date followed by the same letter do not differ at P~0.05 as determined by Duncan's New Multiple Range test.

PAGE 102

E E LaJ C 0 2 150 300 450 600 750 900 1050 1200 91 SOIL WATER CONTENT, !C 4 6 8 10 12 Well-watered .& A Cyclical-stress 1 0-0 Cyclical-stress 2 6-6. Rainfed Figure 18. Gravimetric soil water content on 2 Nov. 1988.

PAGE 103

..... I fl) N I E N 0 u 0 E ::t Q) C Q) C, C: C .c 0 )( w C 0 ..0 C (..) 92 40~--------------------30 20 10 Arbrook Y = 1 8. 13x 7 .34 r = 0.59 A l not included A in regression A Florigraze Y = 17.04x 10.95 r = 0.64 0 -+--+_____ _,._ ___ ----4 __ -+ ________ -----1 1.0 1.5 2.0 2.5 Specific Leaf Nitrogen, g m 2 Figure 19. Correlation between specific leaf nitrogen content and single leaf carbon exchange rate of perennial peanut.

PAGE 104

93 function for each cultivar was slightly different, a regression line was computed independently for Florigraze and Arbrook. The correlation between SLN content and A was 0.64 and 0.59 for Florigraze and Arbrook, respectively (Fig. 19). The same general trend is evident for both cultivars showing a positive relationship between SLN and A. This would be expected since increasing amounts of N would be used by the plant for additional proteins and enzymes such as light harvesting complex protein and the carboxylating enzyme, rubisco. In general, Florigraze tended to have higher SLN content than Arbrook, and exhibited lower A at a given SLN content as compared to Arbrook. However, the range observed in A was similar for the two species. Plant Nitrogen Concentration and in vitro Digestibility 1987 Season Plant N concentration was unaffected by water management treatment on any harvest date; therefore, only means (averaged across water management treatments) are presented in Table 15. Plant N concentration in the forage sample and above-ground plant components was relatively high throughout the season with average concentration being 24, 27, and 15 g kg 1 in the forage, leaves, and stems, respectively. These values represent a relatively high level of N with respect to nutritive value of forages and are similar to results reported by Romero et al. (1987) for perennial peanut. Plant fraction had a significant effect on N concentration on all harvest dates (Table 15). Nitrogen concentration was always higher in the above-ground plant components than it was in rhizomes. On the

PAGE 105

94 Table 15. Nitrogen concentration of perennial peanut forage, and leaf, stem and rhizome plant components (averaged across water management treatments) in 1987. Plant fraction Forage Leaf Stem Rhizome Date 15 May 1 June 11 June -----------------------9 28.5 aA 17.8 bA 11. 9 cB 24.2 bA* 25 2 aA 26.6 aAB 26.9 aA 15.4 cBC 13.7 bCD 11 5 dB 13.5 bA 23 July 8 Sep. 2 Nov kg 24.3 bA 23.8 aA 24.6 bA 27.4 aA 23.9 aB 28.7 aA 15.9 cB 12.9 bD 13.8 cCD 10 9 dB 9.9 cC 11.5 cB Means within a column followed by the same lowercase letter and means within a row followed by the same uppercase l etter do not differ at P~O.05 as determined by Duncan's New Multiple Range Test.

PAGE 106

95 average, stem and rhizome components contained 45 and 57% less N, respectively, than the leaf component. Mean IVOMD values, averaged across water management treatments, for each harvest are presented in Table 16. Even though statistical differences are noted across harvest dates, IVOMD was relatively high in the forage sample as well as leaf and stem components throughout the season, and there was no mid-summer decline as is often seen in other forages (Jolliff et al., 1979; Sleper and Mott, 1976). Plant fraction affected IVOMD throughout the season (except for the first harvest of relatively immature plants in mid May) with IVOMD of the stem averaging 14% less than that in the leaf. In vitro organic matter digestibility was slightly affected during the season by water management. Stems of plants from the RF treatment subjected to the prolonged dry period (41 days) in the fall were higher in digestibility (587 g kg 1 ) than those in the CSl and WW treatments ( 553 g kg 1 ) However, forage that accumulated during the fa 11 ( 1 ast harvest cycle) was negligible, and any decline in IVOMD of the plants at that time would be of little consequence in terms of animal nutrition and forage utilization. 1988 Season In general, plant N concentration in the forage sample as well as the leaf and stem components was relatively high throughout the 1988 growing season (Table 17) as had been the case in 1987. Average N concentrations for the season were 30, 31, and 18 g kg~ in the forage, leaves, and stems, respectively, being slightly higher than those reported in 1987.

PAGE 107

96 Table 16 In vjtro organic matter digestibility of perennial peanut forage, and leaf and stem components (averaged across water management treatments) in 1987. Plant fraction Forage Leaf Stem Date 15 May 1 June 11 June 23 July 8 Sep. 2 Nov. --------------------9 kg 730 aC* 735 aBC 742 aAB 736 aBC 747 aA 771 aA 725 aE 728 aDE 742 aC 733 aD 763 bB 725 aA 657 bB 641 bB 638 bB 608 bC 565 cD Means within a column followed by the same lowercase letter and means within a row followed by the same uppercase letter do not differ at P~0.05 as determined by Duncan's New Multiple Range Test.

PAGE 108

97 Table 17. Nitrogen concentration of perennial peanut forage, and leaf, stem and rhizome plant components (averaged across water management treatments) in 1988. Plant fraction Forage Leaf Stem Rhizome Date 19 May 2 June 8 June 25 July 12 Sep. 1 Nov. --------------------------9 kg_, 25.6 bC 32 6 bA 31.3 bB 27.7 28.3 aCD 29.2 aC 34.7 aA 34.8 aA 32.5 aB 17 .1 bC 16.8 bC 16.8 cc 20.4 cA 19.1 cB 15.2 bD 11. 5 co 12.6 cc 12.9 dBC 14.2 dA 13.3 dB 11.5 co Means within a column followed by the same lowercase letter and means within a row followed by the same uppercase letter do not differ at Pi0.05 as determined by Duncan's New Multiple Range Test.

PAGE 109

98 Plant fraction had a consistent effect on N concentration throughout the season (Table 17). Again, the above-ground plant components contained more N than the rhizomes and, on the average, N concentration in the stems and rhizomes was 44 and 59% lower, respectively, than that in the leaves. Water management treatment did have a slight effect on plant N concentration on two occasions in 1988. An effect of water management was observed at the 2 June harvest with leaf N concentration being reduced from 29.8 g kg 1 on the average in the WW and CSl treatments to 26.8 g kg 1 on the average in the CS2 and RF treatments, respectively. However, those differences had disappeared by 8 June at the end of the first harvest cycle. Furthermore, such small effects do not represent differences which would appreciably affect the nutritive value of the crop. Water management treatment also had an effect on plant N concentration in the leaf and stem components as well as the composite forage sample taken at the end of the second harvest cycle on 25 July (Table 18). Water stress (CS2 and RF treatments) led to relatively higher forage and leaf N concentrations as compared to the WW and CSI treatments, but only the RF treatment caused stem N to increase as compared to the other three treatments. Nitrogen concentration in the forage sample increased in the RF and CS2 treatments as compared to the WW treatment. Donovan and Meek {1983) have also reported higher protein concentrations under dry as compared to well-irrigated conditions in alfalfa. Even though the differences in N concentration are significant, perennial peanut had relatively high concentrations of N

PAGE 110

99 Table 18. Nitrogen concentration of perennial peanut forage, and leaf and stem components as affected by water management treatment on 25 July 1988 Water management treatment Well watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Plant Component Forage Leaf Stem k 1 --------------9 g ------------------29.6 c* 32 8 b 19.1 b 31. 6 be 32.7 b 36 3 a 32.9 b 35.9 a 37.0 a 19.6 b 19.7 b 22.9 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.

PAGE 111

100 under all water management treatments throughout t he entire growing season. Mean IVOMD, averaged across water management treatments, for each harvest is presented in Table 19. As in 1987, IVOMD was relatively high in the forage samples as well as the leaf and stem components throughout the season except for quite low values in the stem component on 1 November. As previously mentioned, forage accumulated during the fall was negligible and low stem digestibility at that time of year would likely be of no consequence for forage utilization. Plant component had a consistent effect on IVOMD throughout the season with stem IVOMD being 16% less than leaf IVOMD on the average. Water management treatment impacted plant IVOMD on two occasions during the 1988 growing season. Stem IVOMD w as increased by 8.5% at the end of the first harvest cycle (ended 8 June) in the RF treatment (661 g kg 1 ) as compared to the average of the WW CS 1 and CS2 treatments ( 609 g kg 1 ). Water stress (CS2 and RF treatments) al so caused an increase in IVOMD during the second harvest cycle (ended 25 July) in the leaf and stem fractions as well as the forage sample (Table 20). Even though differences occurred, the leaf component and forage IVOMD was relatively high under all water management treatments. Stem digestibility was reduced to an average of 589 g kg 1 in the WW and CSI treatments; however, samples taken represent the whole stem to ground level which includes unpalatable material that would probably never be utilized by the grazing animal or harvested by mechanical equipment. No other effects of water management treatment were detected on IVOMD during the season.

PAGE 112

101 Table 19. In vitro organic matter digestibility of perennial peanut forage, and leaf and stem components (averaged across water management treatments) in 1988. Plant fraction Forage Leaf Stem Date 19 May 2 June 8 June 25 July 12 Sep. 1 Nov. k 1 --------------------------9 g -----------------------695 bA* 700 bA 690 bA 728 aAB 713 aC 716 aC 721 aBC 701 aD 731 aA 676 bA 639 bB 622 cc 609 cD 602 co 487 bE Means within a column followed by the same lowercase letter and means within a row followed by the same uppercase letter do not differ at Pi0,05 as determined by Duncan's New Multiple Range Test.

PAGE 113

102 Table 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. Water management treatment Well -watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Plant Component Forage Leaf Stem k 1 -----------------g g --------------------683 c* 711 b 579 c 689 be 699 b 728 a 713 b 728 a 730 a 599 be 615 ab 639 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.

PAGE 114

103 Forage IVOMD and N concentration were r elatively high under the various water management treatments throughout both seasons. No decline in digestibility or N concentration was noted during the hot summer months, and perennial peanut would be considered a forage crop of high nutritive value Romero et al. (1987} in a study comparing alfalfa and perennial peanut also concluded that perennial peanut is of high nutritive value and equal to alfalfa for ruminant animal nutrition.

PAGE 115

SUMMARY AND CONCLUSIONS A field-grown perennial peanut crop, established under well irrigated conditions in 1986, was subjected in 1987 and 1988 to several water management treatments designed to create varying degrees of plant water deficits. 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 every 2 to 4 days during periods of dry weather throughout the growing season; 2) cyclical stress (CSl), received complete soil water recharge at 10 to 14 day intervals during dry weather; 3) water withheld until stomata began to close, then light frequent irrigation to replenish daily evapotranspiration (SP); and 4) rainfed (RF). The SP treatment was suspended after the first harvest cycle in 1987 (ended 11 June), and thereafter received the same irrigation as the CSl treatment during the rest of the season. In 1988, the treatments were as follows: 1) well-watered (WW), received irrigation every 2 to 4 days during periods without significant rainfall; 2) 2-week cyclical stress (CSl), received irrigation to achieve complete soil water recharge every 14 days during periods without rainfall; 3) 3-week cyclical stress (CS2), received irrigation to achieve complete soil water recharge every 21 days during periods without rainfall; and 4) rainfed (RF). Water management treatments were the same in both years except for treatment no. 3. Since the SP treatment produced similar results as the WW and CSl treatments in 1987, 104

PAGE 116

105 it was replaced with the CS2 treatment in 1988. Dry matter accumulation and distribution, rhizome TNC concentration, A, g, leaf water status, and nutritive value were determined to quantify the response of the crop to drying soil conditions. Rainfall was below the 30-year mean during the spring in both 1987 and 1988. Typical abundant summer rainfall occurred in both years and the fall was quite dry, especially in 1987. In general, rainfall patterns were typical of those observed in north Florida, and the drier than normal spring weather probably tended to amplify the responses to irrigation during those periods compared to what might be expected with more normal rainfall amounts. Gravimetric soil water data indicated that the water management regimes imposed were effective in creating real differences in available soil water at various times throughout the study. Water management treatment affected OM distribution and accumulation of perennial peanut only during periods of severe soil water deficits (fall 1987, spring and summer 1988). Increasing severity of plant water stress caused a decrease in above-ground growth and an increase in rhizome TNC, indicating that photosynthate was being partitioned underground at the expense of above-ground growth during periods of plant water deficits. Total 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 CSl treatments. Above-ground DM was also reduced 11% in the CS2 treatment in 1988 as compared to the WW treatment. Seasonal forage OM (OM above 5-cm cutting height) accumulation was unaffected by water treatment in 1987; however, an extremely dry spring followed by a summer dry period

PAGE 117

106 (30 June to 10 July) caused seasonal forage DM production to be reduced by up to one-third in the water stress treatments (CS2 and RF) as compared to the well-watered control and CSl treatments in 1988. In 1987, there was more soil water deep in the profile for the water stress treatments than occurred in 1988. A dry spring, earlier initiation of plant growth, and more fully established plants all probably contributed to depletion of soil water deep in the profile in 1988. Depletion of this soil water reserve during the spring and early summer of 1988 contributed to the greater response of plants to irrigation than was observed in 1987. Rhizomes constituted approximately two-thirds of the total OM present at the end of any given harvest cycle, except the fall (last) harvest cycle when they constituted approximately 85% of the total OM present. Rhizome DM present in the fall of 1987 was substantially higher than in the spring; however, in 1988 there was relatively little net change in rhizome OM from spring to fall. In both years, rhizome OM was depleted as rapid above-ground growth occurred in earlyand mid-summer (June-July), but accumulated rapidly at the expense of above ground growth in the fall (September-November). Single leaf carbon exchange rate and the other physiological measurements indicated that perennial peanut was relatively unresponsive to water management until severe soil water deficits occurred (fall 1987, spring and summer 1988). Even under extremely dry conditions when A was significantly reduced by plant water stress (RF, or CS2 and RF treatments), average A was found to be 66% of the WW rate in 1987 and 57% of the rate for nonstressed plants in 1988. Stomatal conductance

PAGE 118

107 was affected to a substantially greater degree than A, being reduced 61% on the average by water stress as compared to an average reduction of 39% in A. There were also two occasions when a significant reduction in g was detected, yet A was unaffected (10 June 1987 and 16 May 1988). A reduction in the Ci/Ca ratio often coincided with reduced A suggesting that stomatal control over gas diffusion may have been limiting CO 2 supply to the leaf thus causing a reduction in A due to limited substrate. Calculated c; was below 200 L 1 on several occasions and was as low as 144 L L 1 on one occasion (8 July 1988) which is certainly low enough to limit A in C 3 plants. The stomatal mechanism (restricted gas diffusion) consistently accounted for two thirds or more of the observed reductions in A under water stress. However, it is likely that both nonstomatal (biochemical) as well as stomatal mechanisms were limiting A in perennial peanut subjected to plant water deficits. Single leaf carbon exchange rate for Florigraze and Arbrook perennial peanut cultivars correlated positively to SLN content This would be expected since increasing amounts of N would be used by the plant for additional proteins and enzymes such as light harvesting complex protein and the carboxylating enzyme, rubisco. In general, Florigraze tended to have higher SLN content than Arbrook, and exhibited lower A at a given SLN content as compared to Arbrook. However, the range observed in A was similar for the two cultivars Plant component affected N concentration and IVOMD throughout both seasons. On the average, N concentration and IVOMD were 44 and 15% lower, respectively, in the stems as compared to the leaves. However,

PAGE 119

108 forage IVOMD and N concentration were relatively high under the various water management treatments throughout both seasons. No decline in digestibility or N concentration was noted during the hot summer months, and perennial peanut would be considered a forage crop of high nutritive value. Deep rooted characteristics of perennial peanut resulted in plant productivity which was 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 (summer 1988). It appeared that perennial peanut was capable of using soil water deep in the profile to avoid tissue water deficits. In addition, even when soil water was severely depleted and leaf water potential declined as low as -3.0 MPa in the RF treatment, stomata did not close completely and some carbon assimilation continued. 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.

PAGE 120

APPENDIX Table A-1. Rainfall and irrigation during the establishment season (April through October, 1986) of Florigraze perennial peanut grown at Gainesville, Florida. Date Rainfall Irrigation Date Rainfall Irrigation ---------mm------------------mm---------13 Apr. 10 11 July 8 18 13 13 8 19 21 6 21 6 8 28 13 22 5 23 3 1 May 13 24 16 5 13 26 3 8 13 27 25 16 15 28 6 18 8 19 5 2 Aug. 2 20 20 3 10 21 1 4 4 25 13 5 3 30 13 6 1 7 21 1 June 5 11 13 2 10 12 17 8 20 13 11 10 24 14 27 11 2 15 2 13 9 17 1 14 44 19 1 15 8 20 51 16 I 21 13 18 10 22 4 19 23 24 5 20 3 25 2 21 I 27 1 30 4 28 2 29 21 1 July 5 31 4 3 10 5 8 1 Sep. 1 7 36 3 1 109

PAGE 121

110 Table Al. continued. Date Rainfall Irrigation ---------mm----------5 Sep. 30 6 1 9 21 10 9 11 5 13 6 18 3 26 18 28 4 3 Oct. 19 7 10 8 6 9 16 10 36 25 2 26 1 27 13 29 6 30 8

PAGE 122

111 Table A-2. Establishment season forage dry matter production of Florigraze perennial peanut harvested on 31 Oct. 1986 Samp 1 e Site 1 2 3 4 5 6 mean Forage Dry Matter 2 -----------gm----------279 320 293 277 239 249 276

PAGE 123

112 Table A-3. Rainfall and irrigation for the four water management treatments imposed on Florigraze perennial peanut grown at Gainesville, Florida, in 1987. Irrigation WellCyclical Date Rainfall watered Speedling stress 1 Rainfed ---------------------------mm------------------------------3 Apr. 5 4 1 9 14 14 14 14 15 6 16 8 8 8 8 18 6 6 6 6 19 6 6 6 6 21 8 8 8 8 22 13 13 13 13 24 38 38 38 38 27 28 29 19 1 May 19 7 2 8 3 19 6 44 4 5 5 19 4 6 4 7 19 4 8 3 9 1 19 4 11 19 12 4 45 13 6 14 7 16 8 17 19 19 39 20 11 21 25 22 2 25 19 27 5 19 5 29 19 7 31 15 7 2 June 18 8 38 4 19 4 6 2 19 5

PAGE 124

Table A-3. continued. Date Rainfall Well watered 113 Irrigation Speedling Cyclical stress 1 Ra i nfed ---------------------------mm------------------------------8 June 19 4 10 19 4 13 19 38 38 14 5 20 19 22 14 23 7 24 1 26 6 27 35 28 1 29 5 2 July 5 3 1 6 17 7 1 8 1 9 26 10 12 14 4 15 4 18 1 19 2 8 20 1 21 20 24 14 25 1 26 1 28 5 29 19 31 1 3 Aug. 2 4 10 5 1 8 19 10 89 11 6 12 3 14 2 15 2

PAGE 125

Table A-3. continued. Date Rainfall Well watered 114 Irrigation Speedling Cyclical stress 1 Rainfed ---------------------------mm------------------------------19 Aug. 20 22 25 26 20 29 32 32 32 30 1 2 Sep. 2 3 1 5 25 6 2 8 1 9 13 11 1 12 47 13 3 18 1 19 31 25 18 29 1 30 3 1 Oct. 7 5 25 25 25 12 10 18 32 32 32

PAGE 126

115 Table A-4. Rainfall and irrigation for the four water management treatments imposed on Florigraze perennial peanut grown at Gainesville, Florida, in 1988. Date 29 Mar. 2 Apr. 7 8 11 12 13 19 20 22 27 1 May 5 8 13 14 19 23 25 30 3 June 6 7 9 12 16 19 21 24 25 26 28 29 2 July 8 11 Rainfall Well watered Irrigation Cyclical stress 1 Cyclical stress 2 Rainfed ---------------------------mm------------------------------8 13 6 5 1 3 13 39 30 10 16 1 10 6 10 12 18 2 8 19 13 13 13 13 19 19 15 18 20 19 20 25 15 18 20 18 5 25 25 38 46 46 19 25 38

PAGE 127

116 Table A-4. continued. Date 14 July 15 20 21 23 24 25 27 30 31 I Aug. 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 23 24 25 27 28 30 31 2 Sep. 3 6 7 Rainfall Wel 1watered Irrigation Cyclical stress 1 Cyclical stress 2 Rainfed ---------------------------mm------------------------------30 7 8 10 7 23 1 1 1 19 1 25 8 4 I 13 1 17 4 2 6 6 1 so 1 38 28 27 6 1 15 2 2 71 4 60 1 168 38

PAGE 128

Table A-4. continued. Date Rainfall Well watered 117 Irrigation Cyclical stress 1 Cyclical stress 2 Rainfed ---------------------------mm------------------------------8 Sep. 9 10 16 19 20 22 25 26 27 30 1 Oct. 2 3 4 11 18 22 25 29 2 34 1 2 1 7 12 1 11 5 4 4 1 24 15 18 18 15 18 15 36 36

PAGE 129

118 Table A-5. Gravimetric soil water content as affected by water treatment in 1987. Water treatment 150 300 450 Depth, mm 600 750 900 1050 1200 -----------------------% water----------------------Well-watered Speedling II.la* 10.3a 4.6b 3.8b -31 May -7.3a 6.8a 4.6b 4.5b 7.7a 4.6b 9.3a 5.2a 6.9a 10.9a 9.6a 10.Sa Cyclical-stress 1 2.7c 3.3b 4.6b 5.lb 5.4b 7 9a 5 Sa 9.7a Rainfed 2.8c 3.Sb 4.Sb 4.3b 4.Sb 7.Sa 7.4a 9.4a IO.la 9.7a 3.0b 2.7b 2.7bc 2.9b 10 June -8.4a 8.la 7.7a 3.4bc 3.Sbc 3.2c 4.2b 4.7b 5.4b 9.0a 3.3a 3.9a 6.Sa 10.3a 5.6a 9.2a 5.0a 8.3a Well-watered Speedling Cyclical-stress 1 Rainfed 2.0c 2.lb 2.4c 2.3c 2.2c 2.6a 4.6a 7.9a Well-watered 5.9a Cyclical-stress 1 2.9b Rainfed 1.6c 7.2a 4.4b 1.6c 30 October -6.5a 5.7a 4.5b 4.0b 1.8c 1.7b 5.7a 3.9b 1. 6b 7.4a 4.2a 6.0a 2.9ab 2.6a 3.2a I.Sb 2.4a I.la Means within a column for each date followed by the same letter do not differ at P~0.05 as determined by Duncan's New Multiple Range test.

PAGE 130

119 Table A-6. Gravimetric soil water content as affected by water treatment in 1988. Water treatment Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Well-watered Cyclical-stress 1 Cyclical-stress 2 Rainfed Depth, mm 150 300 450 600 750 900 1050 1200 -----------------------% water----------------------5.8a* 6.4a 2.0b 2.4b 1. 8b 2. 4b I.Sb 2.lb 6.6a 5.2b 4.6bc 3.6c 6.0a 3.Sc 4.8b 2.3d 7.2a 3.6b 4.3b I.Be 6.Sa 3.9c 5.4b 2.8d -2 June -6.la 5.5a 3.4b 3.lb 3.3b 2.7b 2.Sb 1.7b -8 June -6.6a 3.6c 4.8b 1. 9d 6.6a 3.7b 4. lb 2. le 2 November -6.4a 6.2a 4.7b 4.7b 5.6a 4.9b 3.2c 3.0c 5.8a 3. lb 6.8a 8.6a 8.3a 2.0b 2.9ab 2.6ab 3.4a I.Sb 1.8b 2.5a 1. Sb 1.3b 1.6b 3.7a 7.4a 4.4b 2.7b 2.0b 8.6a 3.7a 2.0a 1.2a 6.6a 8.9a 4.8b 4.3a 3.9bc 3.3a 3. le 2 .Sa 6.3a 3.lb 1.8b 1.6b 10.6a 3.5a 2.3a I. la 5.8a 8.9a 5.0ab 7.7a 2.8c 5.2a 3.4bc 2.9a Means within a column for each date followed by the same letter do not differ at P~0.05 as determined by Duncan's New Multiple Range test.

PAGE 131

120 Table A-7. Rhizome total nonstructural carbohydrate concentration as affected by water treatment in 1987 Water treatment Well-watered Speedling Cyclical-stress Ra i nfed 1 Date 15 May 1 June 11 June 23 July 8 Sep. 2 Nov. ------------------------g kg 1 163 a* 151 a 245 a 94 a 121 b 324 a 185 a 185 a 269 a 208 a 217 a 285 a 94 a 147 ab 323 a 234 a 186 a 296 a 137 a 167 a 359 a Means within a column followed by the same letter do not differ at Pi0.05 as determined by Duncan's New Multiple Range test.

PAGE 132

121 Table A-8. Rhizome total nonstructural carbohydrate concentration as affected by water treatment in 1988. Water treatment Well-watered Cyclical-stress Cyclical-stress Rainfed 1 2 Date 19 May 2 June 8 June 25 July 12 Sep. 1 Nov. ------------------------g kg 289 b* 284 b 298 C 97 C 109 a 268 a 305 b 298 b 323 b 123 be 117 a 283 a 337 a 332 a 342 a 156 ab 97 a 261 a 363 a 337 a 356 a 172 a 115 a 268 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.

PAGE 133

LITERATURE CITED Albrecht, K.A., J.M. Bennett, and G.M. Prine. 1989. Growth and gas exchange of rhizoma peanut (Arachis glabrata Benth.) subjected to water deficit. p. 495-496. Proc. XVI Int. Grassl. Congr., Nice, France. Albrecht, S L ., J.M. Bennett, and K.J. Boote. 1984. Relationship of nitrogenase activity to plant water stress in field-grown soybeans. Field Crops Res. 8:61-71. Bates, L.M. and A.E. Hall. 1981. Stomatal closure with soil water depletion not associated with changes in bulk leaf water status. Oecologia 50:62-65. Beardsley, A.E. and S.R. Anderson. 1960. The effect of height and frequency of cutting on various strains of birdsfoot trefoil timothy associations as compared to a Ranger alfalfa-timothy association. Am. Soc. Agron. Abstracts p. 62 Beltranena, R. 1980. Yield, growth and quality of Florigraze rhizoma peanut (Arachis glabrata Benth.) as affected by cutting height and frequency M.S. thesis. Univ. of Fla., Gainesville. Bennett, J.M., K.J. Boote, and L C. Hammond. 1981. Alterations in components of peanut leaf water potential during desiccation. J. Exp. Bot 32:1035-1043. Bennett, J.M., K.J. Boote, and L.C. Hammond 1984. Relationships among water potential components, relative water content, and stomatal resistance of field grown peanut leaves. Peanut Sci. 11:31-35 Bennett, J.M T.R. Sinclair, R.C. Muchow and S.R. Costello. 1987. Dependence of stomatal conductance on leaf water potential, turgor potential, and relative water content in field-grown soybean and maize Crop Sci. 27:984-990. Bhagsari, A.S., R.H. Brown, and J.S. Schepers. 1976. Effect of moisture stress on photosynthesis and some related physiological characteristics in peanut. Crop Sci. 16:712-715. Blackman, P.G. and W J. Davies. 1985. Root to shoot communication in maize plants of the effects of soil drying. J. Exp. Bot 36:39-48. Boyer, J.S. 1982. Plant productivity and environment Science 218:443448. 122

PAGE 134

123 Brown, P.W. and C.B. Tanner. 1983. Alfalfa stem and leaf growth during water stress. Agron. J. 75:799-805. Carter, P.R. and C.C. Sheaffer. 1983a. Alfalfa response to soil water deficits. I. Growth, forage quality, yield, water use, and water use efficiency Crop Sci. 23:669-675. Carter, P.R. and C.C. Sheaffer. 1983b. Alfalfa response to soil water deficits. II. Plant water potential, leaf conductance and canopy temperature relationships. Crop Sci. 23:676-680. Chatterton, W.J., S. Akao, G.E. Carlson, and W.E. Hungerford. 1977. Physiological components of yield and tolerance to frequent harvests in alfalfa. Crop Sci. 17:915-923. Christiansen, S. 1982. Energy reserves and agronomic characteristics of four limpograsses [Hemarthria altissima (Pair) Stapf et C.E. Hubb] for Florida's flatwoods. Ph.D. diss. Univ. of Fla., Gainesville. Cortes, P.M. and T.R. Sinclair. 1986. Water relations of field-grown soybean under drought. Crop Sci. 26:993 998. Donovan, T.J. and D.B. Meek. 1983. Alfalfa responses to irrigation treatment and environment. Agron. J 75:461-464. Farquahar, G.D. and T.D. Sharkey. 1982. Stomatal conductance and photosynthesis. Annu Rev. Plant Physiol. 33:317-345. Feltner, K.C. and M.A. Massengale. 1965. Influence of temperature and harvest management on growth, level of carbohydrates in the roots, and survival of alfalfa (Medicago sativa L.). Crop Sci. 5:585-588. Finn, G.A. and W.A. Brun. 1980. Water stress effects on CO 2 assimilation, photosynthate partitioning, stomatal resistance and nodule activity in soybean. Crop Sci. 20:431-434. Fisher, M.J. and M.M. Ludlow. 1982. Responses of siratro (Macroptilium atropurpureum) to water stress. p. 396-398. ln J.A. Smith and V.W. Hays (eds.) International Grasslands Congress Proc. XIV Int. Grassl. Congr., Lexington, KY, USA. June 15-24, 1981. Westview Press, Boulder, Colorado. Gallaher, R.N., C.O. Weldon, and J.G. Futral. 1975. An aluminum block digester for plant and soil analysis. Soil Sci. Soc. Amer. Proc. 39:803-806. Halim, R.A., D.R. Bunton, M.J. Hattendorf and R.E. Carlson. 1989. Water-stress effects on alfalfa forage quality after adjustment for maturity differences. Agron. J 81:189-194.

PAGE 135

124 Hall, M.H., C.C. Sheaffer, and G.H. Heichel. 1988. Partitioning and mobilization of photoassimilate in alfalfa subjected to water deficits. Crop Sci. 28:964-969. Hambleton, L.G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. J.A.O.A.C. 60:845-852. Jensen, E.H., W.W. Miller, C.N. Mahannah, J.J. Reed, and M.K. Kimbell. 1988. Effect of water supply on performance of alfalfa. J. Prod. Agric. 1:152-155. Jolliff, G.D., A. Garza, and J.M. Hertel. 1979. Seasonal forage nutritive value variation of Coastal and Coastcross-1 bermudagrass. Agron. J. 71:91-94. Kuppers, 8.1.L., M. Kuppers, and E.-D Schulze. 1988. Soil drying and its effect on leaf conductance and CO 2 assimilation of Vigna unguiculata (L.) Walp. I. The response to climatic factors and to the rate of soil drying in young plants. Oecologia 75:99-104. Lang, A. and M.R. Thorpe. 1986. Water potential, translocation and assimilate partitioning. J. Exp. Bot. 37:495-503. Lopez, F.B., T.L. Setter, and C.R. McDavid. 1988. Photosynthesis and water vapor exchange of pigeonpea leaves in response to water deficit and recovery. Crop Sci. 28:141-145. Ludlow, M.M., A.C.P. Chu, R.J. Clements, and R.G. Kerslake. 1983. Adaptation of species of Centrosema to water stress. Aust. J. Plant Physiol. 10:119-130. Markhart, III, A.H. 1985. Comparative water relations of Phaseolus vulgaris L. and Phaseolus acutifolius Gray. Plant Physiol. 77:113117. Mccree, K.J. and S.G. Richardson. 1987. Stomatal closure vs. osmotic adjustment: A comparison of stress responses. Crop Sci. 27:539543. Moore, J.E. and G.O. Mott. 1974. Recovery of residual organic matter from in vitro digestion of forages. J. Dairy Sci. 57:1258-1259. Muchow, R.C., T.R. Sinclair, J.M. Bennett, and L.C. Hammond. 1986. Response of leaf growth, leaf nitrogen, and stomatal conductance to water deficits during vegetative growth of field-grown soybean. Crop Sci. 26:1190-1195. Munns, R. and R.W. King. 1988. Abscisic acid is not the only stomatal inhibitor in the transpiration stream of wheat plants. Plant Physiol. 88:703-708.

PAGE 136

125 Nagarajah, S. and E.-0. Schulze. 1983. Responses of Vigna unguiculata (L.) Walp. to atmospheric and soil drought. Aust. J. Plant Physiol. 10:385-394. Nelson, C.J. and D. Smith. 1968. Growth of birdsfoot trefoil and alfalfa. III. Changes in carbohydrate reserves and growth analysis under field conditions. Crop Sci. 8:25-28. Nicolodi, C., A. Massacci, and G. Di Marco. 1988 Water status effects on net photosynthesis in field-grown alfalfa. Crop Sci 28:944948. Osonubi, 0. 1985. Responses of cowpeas [Vigna unguiculata (L.) Walp] to progressive soil drought. Oecologia 66:554-557. Pitman, W.D. and E.C. Holt. 1982. Environmental relationships with forage quality of warm season perennial grasses Crop Sci. 22:1012-1016. Pitman, W.D. E.C. Holt, B.E. Conrad, and E.C. Bashaw. 1983. Histological differences in moisture stressed and nonstressed kleingrass forage. Crop Sci. 23:793-795. Pitman, W.D., D.M. Vietor, and E.C. Holt. 1981. Digestibility of kleingrass forage grown under moisture stress. Crop Sci. 21:951953. Prine, G.M., L.S. Dunavin, J.E. Moore, and R.D. Roush. 1981. 'Florigraze' rhizoma peanut--A perennial forage legume. Circular S-275. Institute of Food and Agricultural Sciences, Univ. of Florida, Gainesville, FL. I Robertson W.K., L.C. Hammond, J.T. Johnson, and K.J. Boote. 1980. Effects of plant-water stress on root distribution of corn, soybeans, and peanuts in sandy soil. Agron. J. 72:548-550. Robinson, M., A.H. Halevy, D. Galili, and Z. Plaut. 1983. Distribution of assimilates in Gladiolus grandiflorus as affected by water deficit. Ann. Bot. 51:461-468. Romero F., H.H. Van Horn, G.M. Prine, and E.C. French. 1987. Effect of cutting interval upon yield, composition and digestibility of Florida 77 alfalfa and Florigraze rhizoma peanut. J. Anim. Sci. 65:786-796. Saldivar-F., A J. 1983. Growth analysis of 'Florigraze' rhizoma peanut. Ph.D. diss. Univ. of Fla., Gainesville. SAS Institute. 1985. SAS user's guide: statistics. SAS Institute Inc., Cary, NC.

PAGE 137

126 Schulze, E.-D. 1986a. Carbon dioxide and water vapor exchange in response to drought in the atmosphere and in the soil. Ann. Rev. Plant Physiol. 37:247-274. Schulze, E.-D. 1986b. Whole plant responses to drought. Aust J. Plant Physiol. 13:127-141. Shackel, K.A. and A.E. Hall. 1983. Comparison of water relations and osmotic adjustment in sorghum and cowpea under field conditions. Aust. J. Plant Physiol. 10:423-435. Sharratt, B.S., D.C. Reicosky, S.B. ldso, and D.G. Baker. 1983. Relationships between leaf water potential, canopy temperature, and evapotranspiration in irrigated and nonirrigated alfalfa. Agron. J. 75:891-894. Sheriff, D.W., M.J. Fisher, G. Rusitzka, and C .W Ford. 1986. Physiological reactions to an imposed drought by two twining pasture legumes: Macroptilium atropurpureum (desiccation sensitive} and Galactia striata (desiccation insensitive) Aust. J. Plant Physiol. 13:431-445. Sheriff, D.W. and M.M. Ludlow. 1984. Physiological reactions to an imposed drought by Macroptilium atropurpureum and Cenchrus ciliaris in a mixed sward. Aust. J. Plant Physiol. 11:23-34. Silvius, J.E., R.R. Johnson, and D.B. Peters 1977. Effect of water stress on carbon assimilation and distribution in soybean plants at different stages of development. Crop Sci 17:713-716. Sinclair, T.R. and T. Horie. 1989 Leaf nitrogen photosynthesis, and crop radiation use efficiency: A review. Crop Sci. 29:90-98. Sinclair, T.R and M.M. Ludlow. 1986. Influence of soil water supply on the plant water balance of four tropical grain legumes. Aust. J. Plant Physiol. 13:329-341. Sleper, D.A. and G.O. Mott. 1976. Digestibility of four digitgrass cultivars under different harvest frequencies. Agron. J. 68:993-995. Smith, D. 1962. Carbohydrate root reserves in alfalfa, red clover, and birdsfoot trefoil under several management schedules. Crop Sci. 2:75-78. Tennant, D. 1975. A test of a modified line intersect method of estimating root length. J. Ecol. 63:995-1001. Tilley, J.M.A and R.A. Terry. 1963. A two stage technique for in vitro digestion of forage crops. J. Brit Grassl. Soc. 18:104-111.

PAGE 138

127 Turner, N.C. 1986a. Adaptation to water deficits: A changing perspective. Aust. J. Plant Physiol. 13:175-190. Turner, N.C. 1986b. Crop water deficits: A decade of progress. Adv. Agron. 39:1-51. Turner, N.C., E.-D. Schulze, and T. Gollan. 1985. The responses of stomata and leaf gas exchange to vapour pressure deficits and soil water content II. In the mesophytic herbaceous species Helianthus annuus. Oecologia 65:348-355. Valentim, J.F., O.C. Ruelke, and G.M. Prine. 1987. Interplanting of alfalfa and rhizoma peanut. Proc. Soil Crop Sci. Soc. Fla. 46:52-55. Wilson, J.R. 1983a. Effects of water stress on in vitro dry matter digestibility and chemical composition of herbage of tropical pasture species. Aust. J. Agric. Res. 34:377-390. Wilson, J.R. 1983b. Effects of water stress on herbage quality. p. 470472. lrr J.A. Smith and V.W. Hays (eds.) International Grassland Congress. Proc. XIV Int. Grassl. Congr., Lexington, KY, USA. June 15-24, 1981. Westview Press, Boulder, Colorado. Wilson, J.R., M.M. Ludlow, M.J. Fisher, and E.-D. Schulze. 1980. Adaptation to water stress of the leaf water relations of four tropical forage species. Aust. J. Plant Physiol. 7:207-220. Wilson, J.R. and T.T. Ng. 1975. Influence of water stress on parameters associated with herbage quality of Panicum maximum var. trichoglume. Aust. J. Agric. Res. 26 : 127-136. Zhang, J., U. Schurr, and W.J. Davies. 1987. Control of stomatal behavior by abscisic acid which apparently originates in the roots. J. Exp. Bot. 38:1174-1181.

PAGE 139

BIOGRAPHICAL SKETCH Charles W. Mansfield was born 30 Jan. 1954 in Paducah, Kentucky, and grew up in the Glendale community of rural southern Illinois. After graduating from Pope County High School in 1972, he attended Southern Illinois University at Carbondale where he graduated with a BS in plant and soil science in 1976. He then joined the Peace Corps and went to Colombia, South America, for two and one-half years where he worked as an agriculture extension agent in rural development. While in Colombia he met and married the former Leonor Garcfa Florez. They returned to the U.S in 1979 and Chuck completed a M.S. degree at the University of Florida in 1981. Upon graduation from the University of Florida, Chuck and his family moved to Marshfield, Missouri, where he worked as an Area Agronomy Specialist for two and one-half years and then they went back to Colombia. He taught at the Fundaci6n Universitaria de Malaga for two years before returning to the U.S. in 1986 to pursue a Ph.D. program under the supervision of Dr. Jerry Bennett. Chuck is a member of the American Society of Agronomy, the Soil and Crop Science Society of Florida, and Gamma Sigma Delta. 128

PAGE 140

I certify that I conforms to acceptable adequate, in scope and Doctor of Philosophy. I certify that I conforms to acceptable adequate, in scope and Doctor of Philosophy. I certify that I conforms to acceptable adequate, in scope and Doctor of Philosophy. have read this study and that in my op1n1on it standards of scholarly presentation and is fully quality, as a dissertation for the degree of Dr~Jerfy M. Bennett, Chair Associate Professor of Agronomy have read this study and that in my op1n1on it standards of scholarly presentation and is fully quality, as a dissertation for the degree of ~ d &;t Dr. K:nneth ~sVote Professor of Agronomy have read this study and that in my op1n1on it standards of scholarly presentation and is fu}]y quality, as a dissertation for the degree of Dr f Lynn E. Sollenberger Assistant Professor of Agronom I certify that I have read this study and that in my op1n1on it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. // d~~{t/ZO // Oy. David S. Woff Alssistant Professor of Agronomy I certify that I conforms to acceptable adequate, in scope and Doctor of Philosophy. have read thL study and that in !TIY opm1 on it standards of : c holarly p r ~sentation and is fully quality, a~ & d : sscrtation for the degree of Botany

PAGE 141

This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. ( ] # May I 990 ,)!(
PAGE 142

UNIVERSITY OF FLORIDA 1111111111111111111111111111111111111111111111111111111111111111 3 1262 08553 4450