Citation
Energy reserves and agronomic characteristics of four limpograsses (Hemarthria altissima (Poir) Stapf et C.E. Hubb) for Florida's flatwoods

Material Information

Title:
Energy reserves and agronomic characteristics of four limpograsses (Hemarthria altissima (Poir) Stapf et C.E. Hubb) for Florida's flatwoods
Creator:
Christiansen, Scott, 1954-
Copyright Date:
1982
Language:
English

Subjects

Subjects / Keywords:
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Scott Christiansen. 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:
821011762 ( OCLC )
ABW4323 ( LTUF )
28820886 ( ALEPH )

Downloads

This item has the following downloads:

energyreservesag00chri.pdf

UF00101080_00001.pdf

energyreservesag00chri_0098.txt

energyreservesag00chri_0068.txt

energyreservesag00chri_0173.txt

energyreservesag00chri_0189.txt

energyreservesag00chri_0096.txt

energyreservesag00chri_0035.txt

energyreservesag00chri_0084.txt

energyreservesag00chri_0028.txt

energyreservesag00chri_0194.txt

energyreservesag00chri_0063.txt

energyreservesag00chri_0027.txt

energyreservesag00chri_0147.txt

energyreservesag00chri_0148.txt

energyreservesag00chri_0158.txt

energyreservesag00chri_0059.txt

energyreservesag00chri_0058.txt

energyreservesag00chri_0025.txt

energyreservesag00chri_0047.txt

energyreservesag00chri_0054.txt

energyreservesag00chri_0196.txt

energyreservesag00chri_0162.txt

energyreservesag00chri_0153.txt

energyreservesag00chri_0193.txt

energyreservesag00chri_0015.txt

energyreservesag00chri_pdf.txt

energyreservesag00chri_0134.txt

energyreservesag00chri_0170.txt

energyreservesag00chri_0203.txt

energyreservesag00chri_0087.txt

energyreservesag00chri_0202.txt

energyreservesag00chri_0122.txt

energyreservesag00chri_0017.txt

energyreservesag00chri_0037.txt

energyreservesag00chri_0060.txt

energyreservesag00chri_0106.txt

energyreservesag00chri_0095.txt

energyreservesag00chri_0076.txt

energyreservesag00chri_0197.txt

energyreservesag00chri_0144.txt

energyreservesag00chri_0124.txt

energyreservesag00chri_0034.txt

energyreservesag00chri_0117.txt

energyreservesag00chri_0011.txt

energyreservesag00chri_0198.txt

energyreservesag00chri_0013.txt

energyreservesag00chri_0137.txt

energyreservesag00chri_0119.txt

energyreservesag00chri_0048.txt

energyreservesag00chri_0050.txt

energyreservesag00chri_0008.txt

energyreservesag00chri_0103.txt

energyreservesag00chri_0204.txt

energyreservesag00chri_0107.txt

energyreservesag00chri_0004.txt

energyreservesag00chri_0160.txt

energyreservesag00chri_0139.txt

energyreservesag00chri_0014.txt

energyreservesag00chri_0126.txt

energyreservesag00chri_0195.txt

energyreservesag00chri_0151.txt

energyreservesag00chri_0208.txt

energyreservesag00chri_0206.txt

energyreservesag00chri_0007.txt

energyreservesag00chri_0080.txt

energyreservesag00chri_0057.txt

energyreservesag00chri_0071.txt

energyreservesag00chri_0089.txt

energyreservesag00chri_0021.txt

energyreservesag00chri_0088.txt

energyreservesag00chri_0038.txt

energyreservesag00chri_0140.txt

energyreservesag00chri_0042.txt

energyreservesag00chri_0036.txt

energyreservesag00chri_0043.txt

energyreservesag00chri_0118.txt

energyreservesag00chri_0020.txt

energyreservesag00chri_0005.txt

energyreservesag00chri_0108.txt

energyreservesag00chri_0003.txt

energyreservesag00chri_0154.txt

energyreservesag00chri_0132.txt

energyreservesag00chri_0009.txt

energyreservesag00chri_0066.txt

energyreservesag00chri_0001.txt

energyreservesag00chri_0181.txt

energyreservesag00chri_0046.txt

energyreservesag00chri_0056.txt

energyreservesag00chri_0142.txt

energyreservesag00chri_0109.txt

energyreservesag00chri_0026.txt

energyreservesag00chri_0168.txt

ERTACKWW9_ID7W0Q_xml.txt

energyreservesag00chri_0041.txt

energyreservesag00chri_0157.txt

energyreservesag00chri_0199.txt

energyreservesag00chri_0175.txt

energyreservesag00chri_0190.txt

UF00101080_00001_pdf.txt

energyreservesag00chri_0064.txt

energyreservesag00chri_0167.txt

energyreservesag00chri_0086.txt

energyreservesag00chri_0172.txt

energyreservesag00chri_0052.txt

energyreservesag00chri_0081.txt

energyreservesag00chri_0183.txt

energyreservesag00chri_0078.txt

energyreservesag00chri_0016.txt

energyreservesag00chri_0141.txt

energyreservesag00chri_0101.txt

energyreservesag00chri_0129.txt

E4JJQ2A3I_2I0CHB_xml.txt

energyreservesag00chri_0163.txt

energyreservesag00chri_0044.txt

energyreservesag00chri_0179.txt

energyreservesag00chri_0123.txt

energyreservesag00chri_0164.txt

energyreservesag00chri_0174.txt

energyreservesag00chri_0178.txt

energyreservesag00chri_0200.txt

energyreservesag00chri_0055.txt

energyreservesag00chri_0115.txt

energyreservesag00chri_0191.txt

energyreservesag00chri_0138.txt

energyreservesag00chri_0131.txt

energyreservesag00chri_0171.txt

energyreservesag00chri_0155.txt

energyreservesag00chri_0150.txt

energyreservesag00chri_0049.txt

energyreservesag00chri_0075.txt

energyreservesag00chri_0188.txt

energyreservesag00chri_0061.txt

energyreservesag00chri_0152.txt

energyreservesag00chri_0184.txt

energyreservesag00chri_0201.txt

energyreservesag00chri_0146.txt

energyreservesag00chri_0031.txt

energyreservesag00chri_0127.txt

energyreservesag00chri_0069.txt

energyreservesag00chri_0136.txt

energyreservesag00chri_0077.txt

energyreservesag00chri_0143.txt

energyreservesag00chri_0192.txt

energyreservesag00chri_0120.txt

energyreservesag00chri_0177.txt

energyreservesag00chri_0012.txt

energyreservesag00chri_0187.txt

energyreservesag00chri_0074.txt

energyreservesag00chri_0073.txt

energyreservesag00chri_0099.txt

energyreservesag00chri_0186.txt

energyreservesag00chri_0104.txt

energyreservesag00chri_0082.txt

energyreservesag00chri_0207.txt

energyreservesag00chri_0121.txt

energyreservesag00chri_0169.txt

energyreservesag00chri_0085.txt

energyreservesag00chri_0070.txt

energyreservesag00chri_0018.txt

energyreservesag00chri_0205.txt

energyreservesag00chri_0161.txt

energyreservesag00chri_0033.txt

energyreservesag00chri_0112.txt

energyreservesag00chri_0176.txt

energyreservesag00chri_0128.txt

energyreservesag00chri_0079.txt

energyreservesag00chri_0010.txt

energyreservesag00chri_0051.txt

energyreservesag00chri_0091.txt

energyreservesag00chri_0102.txt

energyreservesag00chri_0113.txt

energyreservesag00chri_0159.txt

energyreservesag00chri_0062.txt

energyreservesag00chri_0133.txt

energyreservesag00chri_0032.txt

energyreservesag00chri_0145.txt

energyreservesag00chri_0093.txt

energyreservesag00chri_0029.txt

energyreservesag00chri_0097.txt

energyreservesag00chri_0022.txt

energyreservesag00chri_0024.txt

energyreservesag00chri_0209.txt

energyreservesag00chri_0165.txt

energyreservesag00chri_0130.txt

energyreservesag00chri_0105.txt

energyreservesag00chri_0125.txt

energyreservesag00chri_0006.txt

energyreservesag00chri_0185.txt

energyreservesag00chri_0023.txt

energyreservesag00chri_0114.txt

energyreservesag00chri_0030.txt

energyreservesag00chri_0067.txt

energyreservesag00chri_0111.txt

energyreservesag00chri_0040.txt

energyreservesag00chri_0110.txt

energyreservesag00chri_0039.txt

energyreservesag00chri_0156.txt

energyreservesag00chri_0149.txt

energyreservesag00chri_0002.txt

energyreservesag00chri_0094.txt

energyreservesag00chri_0100.txt

energyreservesag00chri_0090.txt

energyreservesag00chri_0116.txt

energyreservesag00chri_0166.txt

energyreservesag00chri_0000.txt

energyreservesag00chri_0045.txt

energyreservesag00chri_0182.txt

energyreservesag00chri_0083.txt

energyreservesag00chri_0180.txt

energyreservesag00chri_0135.txt

energyreservesag00chri_0053.txt

energyreservesag00chri_0065.txt

energyreservesag00chri_0072.txt

energyreservesag00chri_0092.txt

energyreservesag00chri_0019.txt


Full Text











ENERGY RESERVES AND AGRONOMIC CHARACTERISTICS OF FOUR LIMPOGRASSES (Hemarthria altissima (Poir) Stapf et C.E. Hubb) FOR FLORIDA'S FLATWOODS
















BY

SCOTT CHRISTIANSEN


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



UNIVERSITY OF FLORIDA


1982




ENERGY RESERVES AND AGRONOMIC CHARACTERISTICS OF
FOUR LIMPOGRASSES (Hemarthria altissima (Poir)
Stapf et C.E. Hubb) FOR FLORIDA'S FLATWOODS
BY
SCOTT CHRISTIANSEN
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1982


ACKNOWLEDGEMENTS
I would like to acknowledge the people most responsible for my
personal progress to date and those aiding me in the accomplishment of
my Ph.D. degree. First I thank my mother, Jean, whose compassion taught
me how to be a human being and my father, Red, who gave me the guts to
be a man. Thank you Candy and Tracy for giving me the love only a
sister and brother can give.
I am grateful for the understanding and intelligent guidance of my
friend and mentor Dr. J. C. Winters who introduced me to higher
education. I wish to acknowledge my agronomic mentors Dr. Jesse M.
Scholl, Dr. Dale W. Smith, and Dr. 0. Charles Ruelke, and thank the
balance of my Ph.D. committee, Dr. William R. Ocumpaugh, Dr. Kenneth J.
Boote, Dr. John E. Moore, and Dr. Kenneth H. Quesenberry, for their
individual help, care, and support.
And I thank my friends in Florida who participated most closely
with my "science project":
Carrie Kitts Christiansen for her love and affection;
Bernard P. Monahan for his analytical skills and altruism;
Wendy J. Carpenter for the confidence we shared in each other;
Susan E. Sladden for her trust and diligence;
Hector Urbistondo for his root scraping;
Abelardo J. Saldivar for his composure, consideration, and care;
Findlay M. Pate for his advice, guidance, and generosity;
Paul Robin Harris and Paul W. Lane for their fun loving, pragmatic,
and joyfully sarcastic attitudes;
Deborah Vinci for her overwhelming sensitivity;


Jim and Janet Dean for visitation hours when I needed them the
most; and
Janet Eldred for her intelligent and tolerant approach to the task
of typing this dissertation.
iii


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
LIST OF TABLES vii
LIST OF FIGURES ix
ABSTRACT xii
INTRODUCTION 1
LITERATURE REVIEW 2
Flistory, Potential, Limitations, and Development
of Florida's Forage-Livestock Industry 2
Development of Grazing Land in Florida 3
History and Evolution of Florida's Grazing Research ... 5
Grazing Animal Improvement 6
Florida's Role in the Forage-Livestock Industry 7
The Florida Opportunity for Finished Beef 10
Barriers to Forage Production in Florida 12
The Climate in North Central Florida 13
Flatwoods and Their Potential for Grassland
Productivity 13
The Need for Research on Limpograss 15
Limpograss Literature to Present: An Overview 16
Origin, Distribution, and Description 16
Characterization 19
Pest Screening 23
Small Plot Trials and Forage Response to
Grazing 25
Animal Response and Systems Management 29
Nonstructural Carbohydrates in Forages 30
The Characterization of Carbohydrate Reserves 32
A Progression of Nonstructural Carbohydrate Methods
Used in Agriculture 33
Organic Substances Used in Regrowth 38
Location, Seasonal, and Daily Fluctuation of Reserve
Foods 39
Management Factors Affecting TNC: An Integrated
Approach 41
IV


Page
CHAPTERS
1 REGROWTH IN DARKNESS AS INFLUENCED BY PREVIOUS
CUTTING TREATMENT OF FOUR LIMPOGRASS GENOTYPES 45
Introduction 45
Materials and Methods 46
Experiment 1 47
Experiment 2 48
Residual Effects 48
Results and Discussion 48
Experiment 1 48
Experiment 2 52
Residual Effects 54
Conclusions 54
Summary 55
2 CUTTING FREQUENCY EFFECTS ON LIMPOGRASS MORPHOLOGY
AND TOTAL NONSTRUCTURAL CARBOHYDRATE RESERVES 57
Introduction 57
Materials and Methods 58
Analysis of TNC in Plant Parts 59
TNC as Related to Clipping, Season, and
Genotype 59
Correlation of TNC versus Regrowth-in-Darkness ... 60
Results and Discussion 61
Analysis of TNC in Plant Parts 61
TNC as Related to Clipping, Season, and
Genotype 63
Correlation of TNC versus Regrowth-in-Darkness ... 65
Conclusions 68
Summary 68
3 DRY MATTER YIELD, CRUDE PROTEIN, IN VITRO ORGANIC
MATTER DIGESTIBILITY, TOTAL NONSTRUCTURAL CARBO
HYDRATE, AND PERSISTENCE IN TWO PROMISING AMD TWO
RELEASED LIMPOGRASSES: EFFECTS DUE TO NITROGEN
FERTILIZATION AMD CUTTING FREQUENCY 71
Introduction 71
Materials and Methods 75
v


Page
1979 Establishment 75
1979 Experimentation 76
1980 Experimentation 77
1981 Experimentation 82
Computing 82
Results 83
Dry Matter Yields: 1979 Experimentation 83
Dry Matter Yields: 1980 Experimentation 83
Dry Matter Yields: 1981 Experimentation 92
Crude Protein: 1979 Experiment 94
Crude Protein: 1980 Experiment 94
IVOMD: 1979 Experiment 97
IVOMD: 1980 Experiment 97
Total Nonstructural Carbohydrate as Related to
IVOMD 100
Seasonal TNC Trends 107
Total Nonstructural Carbohydrates and
Persistence 109
Discussion 116
Conclusions 122
Summary 123
4 STATISTICAL ANALYSIS SYSTEM (SAS) METHODOLOGY FOR
CONSTRUCTING RESPONSE SURFACE GRAPHICS 126
Introduction 126
Materials and Methods 128
Establishment Year: 1979 129
Experimental Year: 1980 129
Results and Discussion 132
Conclusions 146
Summary 147
SUMMARY AND CONCLUSIONS 148
APPENDICES
A TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) PROCEDURE 154
B TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) RESULTS 174
REFERENCES 177
BIOGRAPHICAL SKETCH 193
VI


LIST OF TABLES
Table Page
1 Average DM yields of etiolated "shoots'1 of four
limpograsses following 3 weeks of growth in
darkness 49
2 Comparison of etiolated "shoot" DM yields and
associated "root" and "stubble" components for
four limpograsses averaged for all cutting treat
ments following 3 weeks of growth in darkness 51
3 Average DM yields of etiolated regrowth from
Experiment 2 following 3 weeks of growth in
darkness 53
4 The main effects of limpograss genotype and clip
ping treatment on the percent total nonstructural
carbohydrates (TNC) in the bottom 2 cm of stem
base ("crown") 64
5 Total dry matter (DM) yields for four limpo
grasses clipped at three different frequencies
for 10 weeks ending on 4 October 1979 84
6 Effect of nitrogen (N) rates and cutting frequency
(F) on dry matter (DM) yield, in vitro organic
matter digestibility (IVOMD), crude protein (CP),
harvest of protein, and fertilizer N efficiency 86
7 Percent crude protein (CP) for four limpograsses
clipped at three different frequencies for 10
weeks ending on 4 October 1979 (Samples taken on
4 October were analyzed for CP.) 95
8 The main effect of limpograss and clipping treat
ment on percent in vitro organic matter digesti
bility (IVOMD) of tissue sampled 4 October 1979 98
9 Balance sheet of in vitro organic matter digesti
bility (IVOMD) and total nonstructural carbo
hydrates (TNC) in two limpograsses 104
vi i


Table Page
10 Treatment combinations in PI 364888 representing
the best compromise between yield and quality in
the 1980 experiment 120
11 A SAS program using a previously fitted regression
to obtain values for the dependent variable to
fill the holes in the treatment matrix in order to
merge with actual data and use PROC G3D 133
12 A SAS analysis of variance procedure for a model
that quantifies the maximum treatment (TRT) sum of
squares (SS) for total nonstructural carbohydrate
(TNC) in harvest 9 of the 1980 experiment 137
13 A cubic SAS general linear model that fractionates
the treatment sum of squares (SS) into SS explain
able by nitrogen (N) fertilization and frequency
(F) of defoliation for total nonstructural carbo
hydrate (TNC) in harvest 9 of the 1980 experiment .... 138
14 An F-test of models for determining a statistically
significant response lack of fit (L0F) and an
elimination method for nonsignificant terms 140
15 The SAS general linear models procedure solution
for estimates of intercept adjusted for the limpo-
grass (line) and rep parameters 142
16 The SAS general linear models procedure solutions
for estimates of nitrogen (N) and cutting frequency
(F) treatment parameters adjusted for each limpo-
grass (line) 143
A-l SAS job as submitted on cards to the computer 157
A-2 Statistical analysis system 158
A-3 Percent total nonstructural carbohydrate (TNC)
means for all treatment, dates, and limpograsses
in 1980 174
v i i i


LIST OF FIGURES
Figure Page
1 Total nonstructural carbohydrate (TNC) percent of
dry matter of four plant parts in four limpograsses
sampled at two maturities 62
2 Total nonstructural carbohydrate (TNC) percentages
in the "crown" correlated against "stubble" weights
(left) and etiolated regrowth percent of whole
plants (right) for four limpograss genotypes sub
jected to three clipping frequencies 66
3 Spatial arrangement of a three level factorial
design in two variables 78
4 Spatial arrangement of a central composite design
in three variables 78
5 Treatment matrix showing treatment numbers inside
each actual (circles) and absent (squares) treat
ment combination 79
6 Treatment combinations of nitrogen (N) and cutting
frequency plotted as grid coordinates with the
origin at 120 kg/ha/yr N and 9 weeks cutting
frequency 79
7 Yield, quality (IVOMD and CP), and total nonstruc
tural carbohydrate (TNC) sampling schedule for
1980 (open symbols indicate no sample) 81
8 Total dry matter (DM) yield for four limpograsses
subjected to five levels of nitrogen (N) fertili
zation and five frequencies (F) of defoliation
throughout the 1980 growing season at the Beef
Research Unit near Gainesville, Florida 85
9 Spring and autumn seasonal distribution of dry
matter (DM) yield for four limpograsses in 1980 90
10Dry matter (DM) yields for PI 364888 at three
levels of nitrogen (N) fertility and staged at
three different dates in the autumn of 1980 91
ix
j


Figure
Page
11 Residual clipping and fertilization effects on
dry matter (DM) yield for PI 364888 harvested
every 9 weeks in 1981 (3 June, 3 August, and
7 October (not shown)) and the total production
for the year 93
12 Comparison of crude protein (CP) percentages in
the tissue of PI 364888 and 'Redalta' for two
dates in 1980 in response to five levels of
nitrogen (N) fertilization and five frequencies
(F) of clipping 96
13 Comparison of in vitro organic matter digesti
bility (IVOMD) for four limpograsses at two
1980 harvest dates subjected to five levels of
nitrogen (N) fertilization and five frequencies
(F) of clipping 99
14 Percent total nonstructural carbohydrate (TNC)
in the "shoot" tissue of three limpograsses on
27 July 1980 subjected to five levels of
nitrogen (N) fertilization and five defoliation
frequencies (F) 101
15 A comparison of total nonstructural carbohydrate
(TNC) in the "shoots" of three limpograsses
subjected to five levels of nitrogen (N) fertili
zation and five frequencies (F) of defoliation
sampled on 30 November 1980 102
16 Percent total nonstructural carbohydrate (TNC) in
the stem base ("crowns") of two limpograsses on
two dates as affected by nitrogen (N) fertiliza
tion and defoliation frequency (F) 110
17 A comparison of total nonstructural carbohydrate
(TNC) percent in the "crowns" of two limpograsses
subjected to 3 week defoliation frequencies (F)
and fertilized at three levels of nitrogen (N)
during the 1980 growing season 112
18 Visual estimations of percent 1Bigalta and
PI 364888 for treatments having 3 week defolia
tion frequencies (F) and three levels of nitrogen
(N) in 1980 growing season 113
19 Rainfall and temperature data for the 1980 grow
ing season taken at the Beef Research Unit near
Gainesville, Florida 115
x


Figure Page
20 Visual estimates of four limpograsses subjected
to five levels of nitrogen (N) fertilization
and five frequencies (F) of clipping across
three summer harvests 117
21 An example of the response surface plotted
using SAS/GRAPH computer assistance 136
A-l Dilution one 161
A-2 Dilution two 161
A-3 The regression of optical density (0D) versus
glucose 168
xi


Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
ENERGY RESERVES AND AGRONOMIC CHARACTERISTICS OF
FOUR LIMPOGRASSES (Hemarthria altissima (Poir)
Stapf et C.E. Hubb) FOR FLORIDA'S FLATWOODS
By
Scott Christiansen
December, 1982
Chairman: 0. Charles Ruelke
Major Department: Agronomy
Higher quality forages are needed for the four million hectares of
flatwoods that are the base of the forage-livestock industry in Florida.
The improved limpograss (Hemarthria altissima (Poir) Stapf et C.E. Hubb)
cultivars 'Bigalta' and 'Redalta' were compared with PI 364888 and
PI 349753 to find better yield, quality, and persistence, as well as to
investigate morphological differences and total nonstructural carbo
hydrate (TNC) physiology as they relate to persistence.
Preliminary work in 1979 used regrowth-in-darkness studies that
quantified the morphological differences in the four 1impograsses.
Larger stubble systems were positively correlated to TNC percentages in
the stem bases. Frequent defoliation treatments in the field prior to
the darkness study caused a shift in habit from upright to prostrate
growth that permitted increases in axillary tiller formation with con
comitant increases in TNC accumulation as measured by chemical means.
XT 1


A field experiment was then initiated with nitrogen (N) fertiliza
tion added as a new variable at five levels which was combined with five
defoliation frequencies using a response surface design. Dry matter
(DM) yields, in vitro organic matter digestibility (IVOMD), crude
protein (CP), and persistence were measured.
The best yield of all the genotypes was produced by PI 364888
(29 m ton/ha/yr) with two harvests on 27 July and 30 November, represent
ing 18 week cutting frequencies. Nitrogen was applied at 240 kg/ha per
harvest interval and yields were 18 and 11 m ton/ha for the first and
second harvests, respectively.
Percentages of IVOMD and CP were very low, especially for CP, in
the mid-summer period, but the quality increased substantially in the
autumn when limpograss growth rates declined. Bigalta had superior
quality among genotypes throughout the year.
Stored TNC decreased from a high in March to a low in July. The
period of lowest percent TNC in stem bases corresponded to the duration
of greatest limpograss losses in plots that were greatly stressed by
frequent clipping and high N fertilization.
Limpograss PI 364888 represents an improvement in yield and per
sistence over Bigalta, an improvement in yield and quality over Redalta,
and is scheduled for cultivar release in 1983.
XT 1 1


INTRODUCTION
Limpograss (Hemarthria altissima (Poir) Stapf et C. E. Hubb) is a
stoloniferous grass that has adaptability as an improved forage for
flatwoods in Florida and the southeast. Of the cultivars now released,
'Bigalta' has high digestibility and low persistence while 1 Reda 1ta'
has high persistence and low digestibility. The goal of this project
was to evaluate two new limpograss genotypes, PI 364888 and PI 349753,
based on their characteristics of yield, digestibility, and persistence,
as well as to investigate the nonstructural carbohydrate physiology
that releates to persistence.
A long term objective in Florida is to replace the low quality
forages that commonly grow on flatwoods with higher quality forages
possibly including limpograss. The flatwoods are the most important
ecological site for the forage-livestock industry in the state.
Knowledge of the total nonstructural carbohydrate (TNC) content of
a plant is necessary to understand how a plant grows. Studies of TNC,
morphology, and observations of plant reactions to management can yield
practical recommendations to be passed to agricultural producers.
This study was undertaken to thoroughly explore the flexibility of
limpograss under a wide range of management practices to clearly
identify its strengths and weaknesses as a forage plant for the
flatwoods.
1


LITERATURE REVIEW
History, Potential, Limitations, and
Development of Florida's Forage-Livestock Industry
In 1513 Ponce de Leon set out to find the land with the magic
waters of Indian lore and discovered the mainland near St. Augustine.
He named the place Florida.
One of the early victims of the reputed wealth of Florida was
Hernando de Soto who had whetted his appetite for gold while accompany
ing Pizarro to Peru. He outfitted an expedition and landed in Florida
in 1539. The next four years he spent marching through the southern
states seeking gold, until in 1541 he came to a vast river, now known as
the Mississippi. As the party returned from the present states of
Arkansas and Louisiana, de Soto died. Although he dissipated his
fortune in a vain quest for riches, he accomplished a far worthier
result in opening a vast territory for those who followed (Collins,
1946).
The Indian inhabitants of Florida were hunters. Of the four major
groups, the Calusa occupied the lower west coast, the Tesesta inhabited
the lower east coast, the Timucua were located in the north, and the
Apalache--west of the Aucilla River. Fire was used to chase out game,
and hence, these people were responsible for the creation of large
expanses of grasslands (Hunter et al., 1979).
2


3
The Spaniards soon became pragmatic and recognized Florida's
natural resources after the quest for gold fizzled. Large expanses of
the landscape were grazed after Ponce de Leon landed cattle on the
western coast of Florida in 1521.
The Spanish period lasted until the early 1700's, and during that
time cattle herds utilized the natural prairies such as Paynes Prairie
in Alachua County. The Spaniards left the landscape mostly undisturbed.
Much of their effort went into establishing the port city of
St. Augustine in 1565 which is the oldest permanent European settlement
in the United States.
It was under British rule that St. Augustine became an important
port for ship timbers. It has been said that almost all merchantable
live oak within a few hauling miles of a navigable stream was cut by
1823. This harvesting of wood increased the potential area for grazing.
Furthermore, in those days, 70 percent of the woodlands burned annually
so the sight and smell of wood's fire smoke was as much a part of the
Florida scene as pine trees and thunderstorms (Hunter et al., 1979).
Development of Grazing Land in Florida
The production practices in Florida for growing range animals
remained essentially the same for over 200 years until the automobile
replaced the horse and wagon. A problem arose with cattle confronting
cars on the Florida highways, resulting in ever-increasing injury and
loss. In 1948 the state legislature passed the "no fence law" that
required livestock owners to fence in their animals.
This was an important departure from the roaming range practices of
the past. Florida has serious mineral deficiencies, and fencing helped


4
ranchers to form a mental image of the mineral problems on their land.
The cattlemen were aware of the nutrient rich areas on their open range;
the areas were referred to as "hospital farms" where a nutrient defi
cient animal could graze and convalesce before being returned to the
herd (Becker, personal communication).
In 1930, before mineral supplements were generally used, a study of
7,100 cattle on 44,516 hectares of range was conducted in central
Florida. On flatwoods the calf crop averaged 34 percent, and 68 percent
of the newborn calves died prior to 30 months of age. The market value
was $10.33 per breeding cow per year or $2.54 per hectare (Becker,
personal communication; Henderson, 1956).
Improved forages and better management have helped to alleviate
many nutritional deficiencies. Supplementation with vitamins, minerals,
protein, and energy has been an important aspect of management.
McDowell et al. (1980) surveyed mineral status of beef herds on four
soil types in Florida and concluded that mineral deficiencies were area
specific. They indicated that phosphorus, selenium, and zinc deficien
cies were present in all four of the regions studied. Protein, vitamin
A, potassium, sodium, copper, and cobalt deficiencies were found in
certain regions and were related to seasonality. Magnesium deficiency
was most often associated with cows grazing winter pastures.
In general, the most satisfactory way of providing minerals to
grazing animals is through the use of one complete mineral mixture
offered free choice. The mix should contain a minimum of 6-8 percent
phosphorus, and ideally, the calcium level should not exceed twice the
level of phosphorus. Iron, zinc, manganese, copper, and cobalt should


5
provide 50-100 percent of the daily requirements in 2 ounces of the
mixture (Ammerman, 1979).
History and Evolution of Florida's Grazing Research
The first forage crop specialist at the University of Florida began
work in 1917, and the Agronomy Department was established in 1921.
Since then research on improved pastures has expanded tremendously. The
grazing experimentation was founded upon the observations made by
centuries of cattlemen who noted that lack of sufficiently high quality
forage during the winter months limited the carrying capacity of the
range, retarded the development of immature animals, and affected the
performance of mature animals. Average weight changes of mature cows
on flatwoods rangeland from June 1933 to March 1938 were as follows:
March to June, 34 kg gain; June to September, 10 kg gain; September to
December, 11 kg loss; and December to March, 38 kg loss (Henderson,
1956).
The first grazing trials in Florida were conducted in 1929 for the
purpose of evaluating four perennial summer grasses for use on well
drained sands. The grasses used and the annual beef gains in kg/ha were
as follows: carpetgrass (Axonopus affinis Chase), 195; common bermuda-
grass (Cynodon dactyl on (L.) Pers.), 202; common bahiagrass (Paspalurn
notatum Flgge), 216; and centipedegrass (Eremochloa ophiuroides (Munro)
Hack.), 248. The popularity of carpetgrass among cattlemen in the flat-
woods area and discovery in 1937 of the requirements for successful pro
duction of clovers led to grazing trials during 1942-1945 comparing
carpetgrass alone, unfertilized; carpetgrass alone, fertilized;
carpetgrass-lespedeza (Lespedeza striata (Thunb.) H. & A.); and


6
carpetgrass-white clover (Trifolium repens L.). Average annual gains in
kg/ha were as follows: carpetgrass, 84; fertilized carpetgrass, 168;
carpetgrass-lespedeza, 246; and carpetgrass-white clover, 695,
Meanwhile, superior grasses became available through plant introduction
and breeding. In grazing trials conducted between 1943-1947, three of
the new grasses--'Pangla1 digitgrass (Digitaria decumbens Stent.),
'Pensacola' bahiagrass, and 'Coastal' bermudagrass--ferti1ized with 227
kg of 6-6-6 annually in the spring, produced yearly beef gains averaging
46 percent higher than gains produced by carpetgrass in earlier grazing
trials (Henderson, 1956).
The importance of forages in Florida is obvious. From 1950 to 1973
the forage-livestock industry grew by 70 percent. One-third of the
total range and pasture land in Florida is now planted to improved
forages. In 1973, 270 million kg of beef were produced, totaling $223
million in agricultural income. By 1985 production is expected to reach
425 million kg of beef (Agriculture in an Urban Age (AGUA) Report, 1974).
Grazing Animal Improvement
Adams (1982) describes the rancher's attitude and general percep
tion of his world "back in the old days."
In the 40's we had a lot of trouble raising cattle in Florida.
Cattle were cheap and wild. Pastures were large and the small
boned cows could outrun some of the horses. Salt sickness was
prevalent and we knew little about curing animal deficiencies.
Tick fever had been eradicated but the screw worms were eating
the cattle alive. Half the year cattle were bogged down try
ing to find a drink of water, and the other half they were
standing in water up to their sides. Our main interest was
survival; for the cattle and ourselves. There was no source
of breeding stock available that had the heat tolerance and
quality to meet the needs of a cowman. (Adams, 1982, p. 91)


7
In the past 30 years, the selection of animals has revolved around
adaptation to the Florida forages and climate. Breeders have sought to
combine the adaptability of Zebu, Brahman, or native animals with the
performance capabilities of European stock.
According to Crockett (1982), it became popular in the 1960s to
use sires representing breeds developed in Europe to increase beef pro
duction, and this led to diverse breeding combinations in the subtropi
cal zone of the Atlantic coastal plain and the Gulf coast states.
Crockett feels that breeding progress can be made in subtropical
environments using Brahman-derivative breeds such as the Beefmaster,
Braford, Brangus, or Santa Gertrudis. Fields and Hentges (1979) state
that the genetic base of Florida cattle is primarily of Brahman
extraction.
The characteristics needed in a commercial Florida herd are as
follows: (1) high fertility, (2) tolerance to the temperature and
humidity, (3) good foraging ability, (4) good maternal ability, (5)
satisfactory feedlot performance, and (6) satisfactory carcass quality
(Koger, 1982).
Florida's Role in the Forage-Livestock Industry
What must exasperate Florida producers is that there is an enormous
appetite for beef in the state filled by producers from outside the
state. Only 26 percent of the beef consumed annually by 10 million
Florida residents and their visitors is produced in Florida. The
remainder represents about 340 million kg of beef shipped in by other
means. Approximately 300 million kg of the beef introduced to Florida
is trucked from the high plains area of the United States (Baker, 1980).


8
A recent survey by Spreen and Shonkwiler (1982a) identifies Texas,
Kansas, Nebraska, and Iowa as the largest suppliers of finished beef to
Florida.
A most baffling feature of the Florida beef market is that cattle
men are selling calves to stockers 2250 to 3550 kilometers distant who
then sell to feeders who sell to packinghouses who in turn spend close
to $28 million annually to truck finished beef to Florida (Baker, 1980).
The question arises, "Why does Florida have a cow-calf industry,
and what prevents producers from breaking into the stocker and finishing
phases?" This is not an easily answered question, but one fact is that
Florida does not produce the quantity and quality of feed required to
keep calves in Florida until they reach the consumer. The nagging
thought, however, is that people simply do not persist in doing things
in a way that is less profitable than some logical alternative.
There are inherent cost advantages in keeping a geographically
dispersed cattle production system. Baker (1980) reasoned that by
virtue of escalating transportation expenses an economic incentive would
slow the exodus of calves out of Florida. Stegelin and Simpson (1980),
Spreen and Shonkwiler (1982b), and Ikerd (1981) all gave an economist's
explanation to the contrary. In short, the index of transportation
services in mid 1981 was about 300 percent of the 1974 level. But,
during this same period, the total marketing cost index has increased by
about 290 percent since 1968 and has increased 215 percent since 1974.
Therefore, costs of transportation services have increased only 3-5
percent more than other marketing costs over the past 10-15 years in
spite of rapidly rising fuel costs (Ikerd, 1981).


9
What are these locational advantages of geographical dispersion for
different phases of beef production? The cow-calf phase is land exten
sive, utilizing the less productive land suited to pastures but little
else, as in Florida. Stocker operations are located on areas with high
quality pastures such as the wheat pasture areas of Oklahoma, Texas, and
Kansas. Major cattle feeding areas are located near the major feed pro
ducing areas, and slaughter plants for cattle are located in the areas
where cattle are fed (Ikerd, 1981). Ikerd's cost analysis gives a $32
per head advantage for shipping cattle to the plains states and trucking
the meat back for consumption.
The economies of scale have allowed more cost efficiency for larger
feedlots and packinghouses making establishment difficult for indepen
dent, small competitors. Indeed, it is more likely in the next 20 years
for a continued trend toward fewer and larger specialized operations in
all phases of the system. The areas where cow-calf and stocker opera
tions are dominant are likely to remain dominant because of unique land
and pasture requirements that preclude significant changes in the struc
ture of these industries. The Florida cattle inventory reflects this
trend. Between 1955 and 1980 there was a 62 percent increase in Florida
cattle numbers compared to a 25 percent increase for the entire United
States; however, in 1960 less than 20 percent of calves were outshipped,
whereas in 1980 over 80 percent were outshipped (Spreen and Shonkwiler,
1982b).
Beef cattle production is capital intensive. An estimate of costs
for land and livestock investment is currently $5,000 per cow in the
United States (Ikerd, 1981). When tight money supply and rising inter
est rates squeeze marketing margins, packers and retailers buy less


10
cattle and carcasses. Demand is low, supply is high, and cattle prices
decrease. The cattle feeders make up their losses by offering less
money for stocker animals. The stocker operators, in turn, pass on the
higher interest costs to the cow-calf man. There will be little expan
sion in the cattle business with high interest rates and low priced
calves (Ikerd, 1981).
The Florida Opportunity for Finished Beef
The southeastern United States has the largest surplus of stocker-
feeder cattle in the country; the seven states in this area provide
about 5.5 million more calves each year than are needed for herd
replacement or fed in southeastern feedlots (Baker, 1979).
If the calf crop is to be fed until slaughter in Florida, economics
will necessitate fast, continuous, efficient growth from weaning until
slaughter. The calves must be grown on forage (pasture and/or silage)
with or without grain supplement with a finishing phase in the feedlot
on a high energy ration.
Immediately, it is obvious to wonder, "Where will the grain come
from?" The entire southeast is a grain deficient area, and corn will
continue to cost $20/ton more in Florida than in major feeding states
(Baker, 1979).
The unpredictability of the Florida climate makes corn production a
risky business. Florton et al. (1981) and Horton and Mislevy (1981) grew
corn and sorghum silages in south Florida, and their results showed that
both corn and sorghum can be grown successfully and economically using
multiple cropping. The viability of an intensive beef cattle industry


n
in Florida will depend on the ability to produce these locally grown,
high energy feeds.
Provided the necessary feeding systems were developed, more cattle
would remain on Florida ranches. The question of whether increased
cattle numbers should precede the development of feedlot and packing
industries is "putting the cart in front of the horse." Efficient feed
ing systems must evolve first, the cattle build up will follow, and
count on the feedlot and packing people to fend for themselves. These
businesses simply want a consistent, dependable supply of feed and
animals (Kaplan, 1981).
As Erwin Bryan Jr. of the Central Packing Co. put it, "It is my
feeling that we already have the know-how to feed cattle in Florida.
The weakest point in the chain, and always has been, is grazing of the
feeder calves until they are big enough to go into the feedlot" (Bryan,
1981, p. 46).
In conclusion, it would appear that 20 years from now a totally
enclosed, self-reliant beef industry in Florida could be operational.
The most likely stimuli will be (1) the development of new forage
species and management systems that permit a shift from grain-fed to
forage-fed livestock operations; (2) increased irrigation costs on the
western high plains, decreasing grain availability; (3) solution of the
agronomic problems for year round feed availability in Florida; and (4)
lower interest rates to allow economic growth in the cattle business.
These factors could shift the geographical advantage to the southeastern
United States.


12
Barriers to Forage Production in Florida
There are three phases of the forage system-production, harvest
ing, and utilization. By identifying the present weaknesses in these
phases, combinations of native range and improved pastures and/or
supplemental pastures will be developed in order to provide a distribu
tion of forage throughout the year. An even distribution of forages,
more than any other factor, will ultimately be responsible for increased
beef production in Florida (Mott, 1982b).
The biggest limitation to well distributed forage production is
probably seasonality. Most forage production occurs from April to
September, and two-thirds of the yearly growth of perennials occurs
between July 1 and October 1 (Mott, 1982b).
Uneven distribution of forage production throughout the year makes
management difficult, and supplemental feeding on pasture is often
needed to efficiently utilize the forage produced. Sloan Baker summa
rizes Florida's current forage production problems. He states that cool
season forages--rye (Secale cereale L.), ryegrass (Lolium multiflorum
Lam.), and clovers (Trifolium spp.)--are excellent, yielding good gains
of 0.5 to 0.9 kg/day for 90 to 160 days. Dry matter yields, however,
are sometimes low due to weather. Warm season forages are more produc
tive but give half the daily gains. Dry weather often limits production
in spring and autumn; while in summer, rainfall and humidity are high
enough to make hay curing difficult. Results with grass silage have
been disappointing due to insufficient quality for good gains with young
cattle (Baker, 1979).


13
The Climate in North Central Florida
The climate ranges from subtropical-oceanic in south Florida to a
typical low-altitude, continental frontal pattern in the north (Hunter
et al., 1979). The average year may be divided into two seasons: a
warm, rainy season receiving about 60 percent of the annual rain and a
cooler, dry season. The warm, rainy season runs from about the middle
of June to the end of September. The cooler, dry season dominates the
remainder of the year.
The summer rain occurs as afternoon thunderstorms, generated by
strong surface heating and fed by a double sea breeze convergence.
During the winter months, the differential cooling of land and sea, the
occasional presence of stagnated high pressure cells, and the formation
of low level inversions caused by nocturnal cooling act to maintain a
high degree of atmospheric stability, suppressing frontal activity. A
decrease in frequency of frontal movement across northern Florida is one
cause of periodic drought on the average of once every 7 years
(Dohrenwend, 1978).
The Bermuda high is common throughout the year, centered in the
Caribbean with its strongest effect during winter. If it were not for
large bodies of warm water on either side of the peninsula, Florida
would be as arid as the great subtropical deserts at the same latitudes
(Dohrenwend, 1978).
Flatwoods and Their Potential for
Grassland Productivity
There are 14,236,908 hectares of land area in Florida. At present,
there are approximately 1.21 million hectares of improved pasture, 1.62


14
million hectares of native range, and 2.02 million hectares of forest
land that provide some grazing.
Seventeen major natural vegetation types are recognized in Florida.
The most important ecotype from the standpoint of hectarage and poten
tial for animal production from forages is the flatwood site. Flatwoods
occupy 4.05 million hectares or almost 30 percent of Florida's land
area.
Most of the state's timber production also occurs on flatwoods.
Open woodland consists predominantly of one to three species of pines:
longleaf (Pinus palustris Mill.), slash (P_. elliottii Engelm.), and pond
(P_. sertina Michx.). Understory produces many grasses such as chalky
bluestenr (Andropogon capillipes Nash), broomsedge bluestem (Andropogon
virginicus L.), paspa!urns (Paspalum spp.), wiregrass (Aristida stricta
Michx.), indiangrass (Sorghastrum secundum (Ell.) Nash), and panicums
(Panicum spp.). Associated forbs include grassleaf goldaster
(Chrysopsis graminifolia (Michx.) Ell.), partridge pea (Cassia
fasciculata Michx.), and beggerweed (Desmodium sp.). Shrubs are pre
dominantly saw palmetto (Serinoa repens (Bartram) Small), wax myrtle
(Myrica cerfera L.), blackberry (Rubus sp.), and gall berry (Ilex glabra
(L.) Gray).
Amidst the flatwood areas are small hardwood forests, many cypress
ponds, prairies, marshes, and bay tree swamps. The wet areas support
very desirable forages such as maidencane (Panicum hemitomon Schult.)
and little blue maidencane (Amphicarpum muhlenbergianum (Schult.)
Hitchc.). The flatwoods are inhabited by deer and hogs, quail, gray
squirrels, and turkey; hence, this is also the most important community
type in Florida for hunting.


1 5
The soils are acid (4.0 5.5 pH), of the Spodosol Order, with a
4 cm surface horizon colored grey to grey-black by the presence of
organic matter. The 10 to 45 cm stratum is dominated by white leached
sands and from 45 to 55 cm a hard pan is found--fine particles cemented
together by sesquioxides and other compounds. From 55 to 90 cm depth a
brown, tannin-stained sand exists followed by white sand with further
increases in the profile.
Due to the hardpan formed on Spodosols in Florida, a perched water
table is created that is beneficial in reducing the rapid rate of
percolation through the sands but is a problem when rainfall is so great
as to cause standing water above the soil surface.
The acid soil condition of the flatwoods makes it unsuitable for
crop production; however, acid and wet tolerant species like limpograss
(Hemarthria altissima (Poir.) Stapf et C.E. Hubb) are well adapted to
these lands.
The Need for Research on Limpograss
The days of quantum leaps in beef gains, such as those achieved on
improved forages over native range, are over. An excellent inventory of
improved forages is now available, and possibly the old problem of
seasonal production can be partially solved by incorporating limpograss
into the yearly forage system. Limpograss is a stoloniferous, C^,
perennial, summer-growing grass that has good cool-season production.
Limpograss is adapted to the flatwoods habitat, can produce an abundance
of biomass, and can be stockpiled, ensiled, or made into hay.
Research evaluating the first four limpograss introductions from
1964 culminated in the release of cultivars 'Redalta', 'Greenalta', and


16
'Bigalta' in 1978. Since that time many limpograsses were collected and
evaluated in hopes of further exploiting the germplasm. Beneficial
characteristics of quality, yield, persistence, morphology, winter
hardiness, and adaptability to flatwoods have given researchers evidence
of the potential of limpograss as a forage-producing grass for the
future.
Limpograss Literature to Present:
An Overview
The literature on limpograss species to date will be grouped into
five phases: (1) origin, distribution, and description; (2) characteri
zation; (3) pest screening; (4) small plot trials and forage response to
grazing; and (5) animal response and systems management. Limpograss
research has just recently reached phase five.
Origin, Distribution, and Description
There are probably no more than 12 described species of limpograss
(Kretschmer and Synder, 1979), and the one of most agrnonomic importance
is Hemarthria altissima (Poir.) Stapf et C.E. Hubb. Agrostologists
guess Hemarthrias1 origin to be in tropical Africa, but it is also found
in Madras, Burma, Malaysia, Malay, Siam, Turkey, Nigeria, Italy,
Ethiopia, Tanzania, Ceylon, Northern India, Southeast Asia, and
Argentina (Bor, 1960; Bogdan, 1977; Chippindall, 1955). The species are
desirable, robust perennial fodder plants found in wet habitats such as
river banks, seasonally flooded river valleys, seasonal swamps, and so
forth.


17
Due to plant exploration and redistribution, Hemarthria introduc
tions are now found in Australia, Brazil, Bolivia, Columbia, Ecuador,
Hawaii, Malawi, Mexico, New Zealand, Paraguay, Uruquay, Venezuela, and
the Virgin Islands (Quesenberry et al., 1978; Quesenberry et al. 1981;
and Oakes, 1980).
Overwintering of H_. altissima in the United States occurs in
Alabama, Mississippi, Texas, Tennessee, and as far north as Beltsville,
Maryland (Oakes, 1973; Oakes and Foy, 1980). Later, Oakes (1980)
reported the survival of one limpograss accession from the 1964 collec
tion (PI 299039 from Rhodesia) for three winters at Pullman, Washington.
This diversity of winterhardiness may be used to extend the production
range and grazing period for limpograss in the United States.
Some common names for grasses of the genus Hemarthria are limpo
grass, teagrass (Florida), Capim gamalote (Brazil), Pasto clavel,
Gramilla canita (Argentina), Baksha, Panisharu (India), Swamp Couch, and
Rooikweek (South Africa) (Bogdan, 1977; Chippendall, 1955).
A description of the genus follows: Member of the
Andropogoneae tribe. Perennials with short rhizomes and long
spreading, decumbent, branched culms that root at the nodes;
the upper part of the stems erect or suberect reaching 150 cm
height but usually 30-80 cm. Leaves up to 20 cm long and 6
mm wide with membraneous ligules. The spike-like racemes are
compressed, 6-10 cm long with spikelets appearing opposite--
each pair composed of a bisexual sessile spikelet, 5-6 mm
long, and a smaller, pedicelled male spikelet. Glumes nearly
equal. First glume flat, 2-keeled, leathery; second glume
coriaceous, fused to the hollowed-out face of the rachis
internode (Bodgan, 1977; Hall, 1978).
Quesenberry et al. (1982) surveyed the USDA collection of
Hemarthria (containing 76 species) and found 72 percent diploid, 27 per
cent tetraploid, and 1 percent hexaploid. Thirteen of 16 tetraploids


18
were originally found north of 24 S latitude in Africa. Diploids were
more winterhardy than tetraploids.
The 1964, 1971, and 1976 USDA plant exploration missions to Africa
provided the germplasm for vigorous limpograss experimentation in the
United States for the past 15 years. Jack Oakes did most of the
collecting. Three cultivars were cooperatively released on 19 April
1978 from the Institute of Food and Agricultural Sciences, University of
Florida, and the Soil Conservation Service, USDA. The limpograsses are
Redalta, Greenalta, and Bigalta, previously Pi's 299993, 299994, and
299995, respectively (Quesenberry et al., 1979).
Redalta and Bigalta are two of the four limpograsses used in this
dissertation. Oakes obtained these genotypes from the Rietondale
Research Station near Pretoria, South Africa, in 1964 (Oakes, 1973).
Redalta originated near the Pienaars River in west central Transvaal,
and Bigalta is from an unspecified location near the Transvaal. Redalta
is a diploid (2n = 18), and Bigalta is a tetraploid (2n = 36). Also
included in this dissertation are PI 364888, collected in 1971 from a
small island in the Luvuvhu River several kilometers above its conflu
ence with the Limpopo River in Kruger National Park, and PI 349753, from
Mt. Mbya, Kenya. These two limpograsses are tetraploids (Oakes, 1973).
There are many morphological differences among limpograsses.
Leafiness, internode length, anthocyanin content, bunchiness, tillers
per unit area, and stem thickness vary. These inherent differences aid
in maintaining genotype identification.
Cytological investigations (Wilms et al., 1970) characterized the
four limpograsses collected from Africa in 1964, and reported color dif
ferences for the anthers of Redalta (purple), Greenalta (brown), and


19
Bigalta (yellow). Schank (1972) characterized chromosome numbers,
pollen stainability, and seed set for 11 more Hemarthrias in preparation
for an intrageneric breeding program. Breeding efforts to date have not
produced any limpograsses with better agronomic attributes than the
vegetatively propagated plant introductions. The logistics of low seed
set (Schank, 1972) may partially discourage plant breeders in this
regard, as well as the difficulty in getting limpograss to flower in the
greenhouse (Quesenberry, personal communication).
Schank et al. (1973) discovered that tetraploid limpograsses had
higher in vitro organic matter digestibility (IVOMD) than diploid
accessions. The mean decrease of IVOMD of the tetraploid, from 68.4
percent at 5 weeks, to 66 percent in mature plants, suggested a slower
decline of quality than for most tropical grasses. Cross-sections of
stems revealed significantly lower vascular bundle area in the tetra
ploid, as well as fewer sclerenchyma fiber cells.
Characterization
One of the tangential aspects of the incipient research on limpo
grass was that it was noted as having a tea-like odor and flavor.
Killinger (1971) after collaboration with the USDA Northern Utilization
Research Laboratory, filed for a beverage use patent on 25 January 1971.
Thus, the name "teagrass" came into usage first promulgated, in-so-far
as can be determined, by Eldridge D. Lee of the University of Florida
Agronomy Farm. Killinger and Beckham obtained United States Patent
3,709,694 for Hemarthria beverage rights in 1973.
Oakes and Foy (1980) recommended limpograss for revegetating mine
spoils due to an excellent tolerance to aluminum toxicity. Oakes (1973),


20
Oakes and Foy (1980), and Oakes (1980) also reported a wide diversity of
cold tolerance, as mentioned earlier.
A recent finding was an alleopathic effect of some limpograsses
(Ruelke and Quesenberry, 1981; Young and Bartholomew, 1981; Tang and
Young, 1982). The alleopathic compound was suspected by Ruelke and
Quesenberry (1981) in their limpograss mixtures with red clover and the
action was attributed to the clover. The red clover isoflavinoids
(Tamura et al., 1967; Chang et al., 1969) and the phenolic compounds
from limpograss roots (Young and Bartholomew, 1981; Tang and Young,
1982) may have both been active. Bigalta was more growth depressive on
'Greenleaf1desmodium (Desmodium intortum (Mill.) Urb.) than was Greenalta
limpograss. Bigalta root exudates also depressed the growth of
Greenalta.
A series of four papers characterized 10 tropical forage grasses in
Puerto Rico. Limpograss was included in these studies. The main intent
was to survey: (1) fibrous carbohydrate fractions, (2) proximate
nutrient composition, (3) mineral composition, and (4) the decline of
in vitro true digestibility with advancing maturity of 10 tropical
forage grasses. The 10 grasses were guineagrass (Panicum maximum
Jacq.), Pangla digitgrass, congograss (Brachiaria ruziziensis Germain &
Evrard.), African crabgrass (_D. swazilandensis Stent.), Venezuelan
elephantgrass (Pennisetum setosum (Swartz) L. Rich, in Pers.), giant
Pangla digitgrass (D. valida Stent.), signalgrass (13. Brizantha
(Hochst. ex A. Rich.) Stapf), buffelgrass (Cenchrus ciliaris L.),
jaragua (Hyparrhenia rufa (Nees) Stapf), and limpograss. The identity
of the Hemarthria was not reported, hence, care must be exercised in


21
extrapolating these results to other limpograsses (acknowledging varia
tions among and within ploidy levels).
Paper No. 1 (Coward-Lord et al., 1974a) on carbohydrate fiber frac
tions discussed neutral-detergent fiber (NDF), acid-detergent fiber
(ADF), acid-detergent lignin (ADL), hemicellulose, cellulose, and
silica. The NDF fraction represents the total fiber fraction, its dif
ference from 100 being the neutral-detergent solubles (NDS), or soluble
nutrients. The ADF content is a measure of the 1igno-cel1ulose
fraction. The difference between NDF and ADF is an estimate of
hemicellulose. Acid-detergent lignin (by the permanganate method) is an
acid treatment of the ADF which leaves cellulose and silica. Ashing
leaves silica.
Results showed that limpograss was one of three grasses with the
highest levels of NDF, among the four lowest in ADF content, the highest
in ADL, the highest in hemicellulose, among the lowest in cellulose, and
the lowest in silica. In other words, among the tropical grasses
studied, limpograss had a high percentage of total fiber-comprised of
high quantities of lignin and hemicellulose relative to cellulose.
In Paper No. 2, Coward-Lord et al. (1974b) studied nutrient compo
sition including crude protein (CP), dry matter, crude fiber (CF), ether
extract, ash, and nitrogen-free extract (NFE). This methodology is
slowly losing favor over the Goering and Van Soest (1970) methodology
for fractionating feedstuffs for ruminant value. The definition of CF
as a chemically uniform, non-nutritive substance cannot be reconciled
because CF represents almost all the potentially digestible cellulose
and also includes some lignin and hemicellulose. The imperfect CF
methodology has allowed most of the liginin and hemicellulose to be


22
included in the NFE, which is supposed to represent available
carbohydrate. In some cases the CF can be more digestible than the NFE,
which is clearly incongruous with the aim of the fractionation.
Nevertheless, results showed that limpograss had the highest mean NFE
value of all 10 species at 53.3 percent over 180 days. The CP values
dropped to 5.7 percent by the 90-day stage.
The third Puerto Rican Paper (Arroyo-Aqui1u and Coward-Lord, 1974a)
covered mineral composition. The ranges as percent of dry matter over
180 days for all 10 species and the averages for limpograss (in paren
theses) were calcium, 0.11-0.43 (0.15); phosphorus, 0.08-0.39 (0.13);
magnesium, 0.15-0.46 (0.20); and potassium, 0.68-7.33 (1.85).
Limpograss had the lowest phosphorus percent of all grasses at 180 days
(0.08)--clearly deficient for ruminants; magnesium and potassium were
lowest at 30 days, 0.25 and 3.0 percent, respectively.
The final Puerto Rican paper (Arroyo-Aquilu and Coward-Lord, 1974b)
discussed quality decline among the 10 tropical grasses. The mean rate
of in vitro true digestibility decline was 24,1 units from 30-180 days.
The largest decline (12.3 units) occurred between 30-60 days as compared
to declines of 4.8, 3.9, 1.3, and 1.8 units between 30 day intervals
from 60-180 days. All the grasses reacted similarly. This suggests
that the tropical grasses studied may best be utilized between 30-60
days of growth.
Hodges and Martin (1975) included three limpograsses in a study of
23 perennial sub tropical grasses and reported that Cynodons and
Di gitari as were better cool season producers than the limpograsses at
Ona, Florida. Seasonal yield distribution of numerous tropical grasses
was studied by Taylor et al. (1976b) in New Zealand. In this study, a


23
tetraploid limpograss (identity not reported) yielded poorly during the
low rainfall summer of the year. The total warm-season yield for limpo
grass nearly doubled under trickle irrigation, and the moist, cool-
season yields were twice as great as the unirrigated warm-season yields.
The digestibility data for 22 grasses were analyzed in another
Taylor et al. (1976a) publication to compare nutritive quality of the
grasses grown at Kaitaia, New Zealand (35 S latitude), with the same
grasses grown in more equatorial environments. The results for grasses
grown at Kaitaia, which has a mean warm-season temperature of 18.4 C,
were compared to the digestibility data for the same species grown at
Lawes, Australia, with a mean temperature of 24.3 C. Slower rates of
maturation in the cooler environment frequently resulted in higher
tissue digestibilities.
The non-flowering limpograss used in the New Zealand studies pro
duced tissue with moderate protein (15.7 percent in leaves, 6.1 percent
in stems), moderate fermentable carbohydrate (7.2 percent in leaves, 6.0
percent in stems), and excellent digestibility (66.8 percent in leaves,
77.2 percent in stems). The stem digestibility was the highest of all
the grasses studied.
Pest Screening
A limited number of references are available on limpograss suscep
tibility to nematodes and aphids. Boyd and Perry (1969) screened
Redalta, Greenalta, and Bigalta resistance to sting nematodes
(Belonolaimus longicaudatus Rau). This nematode is among the most
persistent, serious pests in Florida's improved pasture--Pangola digit-
grass, Pensacola bahiagrass, and Coastal bermudagrass are all


24
susceptible. The limpograsses had moderately low nematode counts but
the authors attributed this to lack of roots rather than significant
nematode resistance. Chlorosis was observed in Bigalta, and this geno
type had higher counts than the two diploid limpograsses. In 1970, Boyd
and Perry reported additional information on sting nematode damage to
17 pasture grasses. Redalta, Greenalta, and Bigalta were the three most
favorable hosts.
Boyd et al. (1972) studied the interaction of soil temperature and
sting nematodes and found that Greenalta grew best between 20 and 38 C
in uninfested soils and best between 30 and 38 C in infested soils.
Above 38 C, nematodes were reduced but the soil was too hot for good
growth of the grass. Later, Quesenberry and Dunn (1977) received 54
more limpograsses which they screened for response to the sting nematode
in the greenhouse. No available limpograss lines approached immunity,
but a few introductions had greater tolerance. Of the 10 best lines, 50
percent were tetraploid or hexaploid, while all the least tolerant were
diploid. The most tolerant introductions were collected from the
islands of Mauritius in the Indian Ocean.
Pest problems in limpograss were reported by Oakes (1978) who
studied resistance in Hemarthria to the yellow sugar-cane aphid Sipha
flava (Forbes). Variable resistance was found for 54 PL altissima
accessions. Two of the introductions included in this dissertation were
evaluated in the Oakes study--PI 364888 was most susceptible and PI
349753 was moderately resistant, while all the limpograsses had better
resistance to the sugar-cane aphid than found in Digitari a.


25
Small Plot Trials and Forage Response to Grazing
The second and third collections of limpograss were initiated in
response to previous experimental results that recommended an extended
search within Hemarthrias for favorable pasture grass attributes.
Ruelke et al. (1976) evaluated 53 limpograsses in both greenhouse and
small plot trials and this study satisfied their exploratory curiosity
with respect to the identification of superior genotypes.
In 1976-77, Quesenberry and Ocumpaugh (1977) grew Redalta,
Greenalta, and Bigaita as conserved forages. They presented their
results at the American Society of Agronomy meetings at Los Angeles in
1977, and with more detail (Quesenberry and Ocumpaugh, 1980; Ocumpaugh
and Quesenberry, 1980) following a second year of data. Stockpiling is
an inexpensive way of filling the forage-deficient months of November-
February in northern Florida. Stockpiled yields were greatest for
Redalta in 1976-77 (over 10 m tons/ha). Yields were similar for the
three cultivars in 1977-78 and lower, averaging 6 m tons/ha (Quesenberry
and Ocumpaugh, 1980). Based on these results from the Green Acres Unit
near Gainesville, Florida, stockpiling should begin by the beginning of
August to allow 6-8 weeks of growth before frost.
The above study did not include data on animal acceptance; however,
the authors observed satisfactory consumption of mature Bigalta. Other
producers (Wendy J. Carpenter, personal communication) have reported
animal rejection of similarly aged Bigalta.
Quesenberry and Ocumpaugh (1982) presented data on the tissue
sampled from the stockpiling experiments. Potassium decreased from a
high of 2.5 percent to below the National Research Council (NRC)


26
recommended level for ruminants (0.65-0.80) by early November.
Phosphorus dropped below the NRC minimum for ruminants (0.16-0.24) by
mid-October and in the second year was never above the minimum level.
Magnesium was not considered a nutritional problem, while the mean
calcium percent for both years (0.28) was adequate for mature pregnant
beef cows (NRC = 0.16) but barely sufficient for lactating beef cows
(NRC = 0.27). The authors' recommendation was to supplement with
potassium, phosphorus, calcium, and protein after mid-October.
Quesenberry et al. (1978) coalesced the pertinent limpograss pro
duction data through 1978 for six sites in Florida. At Ft. Pierce,
Bigaita had higher production in November and December than Redalta or
Greenalta, but produced less in spring. At Gainesville, the late season
production of limpograss was slightly less than that of the digitgrasses
and one bermudagrass. Bigalta, however, had the highest total season
production. Frequent clipping defoliation at Ona produced weed inva
sions and the limpograsses had intermediate yields compared to other
tropical, perennial grasses. Results from Jay demonstrated slow estab
lishment rates compared to Coastal bermudagrass, 'Transvala' digitgrass,
and one bahiagrass. From work done at Quincy, it was concluded that
limpograsses are not adapted to the dry, upland soils of the Florida
panhandle. At Belle Glade, Coleman and Pate (Quesenberry et al., 1978)
found good digestibility and acceptability of Bigalta by beef animals;
however, St. Augustinegrass (Stenetaphrum secundatum Kuntz) was better
adapted to the muckland soils at this south Florida site. Bigalta did
not persist under heavy grazing. This led to the recommendation of
rotational grazing for Bigalta.


27
Rotational grazing is too intensive for most Florida producers. It
is fortunate that some of the new limpograss introductions will persist
under continual use because this is an important criterion in a ranchers
mind (Pate, personal communication).
Ruelke (1978) studied Redalta, Greenalta, and Bigalta and found
significant yield responses to nitrogen up to 330 kg/ha/yr; however,
severe losses occurred following frequent defoliation at high nitrogen
rates, especially for Bigalta.
Bigalta's high digestibility combined with poor persistence was
disconcerting. Quesenberry et al. (1981) found that PI 364888 was a
digestible, persistent tetraploid. This accession was soon included in
the limpograss studies with Bigalta and Redalta, while Greenalta was
omitted from experimentation due to its similarity to Redalta.
Ruelke et al. (1978) included the promising accession in a limpo
grass establishment study. Denser stands, earlier production, and
higher second year dry matter yields were obtained when planting
material was sprigged, followed by disking to partially cover the stems,
and cultipacked for firm contact between sprigs and soil. In the year
after establishment, PI 364888 outyielded both Redalta and Bigalta.
Kretschmer and Synder (1979) compared growth of Redalta, Greenalta,
and Bigalta to Transvala digitgrass, Pangla digitgrass, and
1Coastcross-11 bermudagrass and found that a 2 week cutting interval
severely decreased the ability of limpograss to accumulate dry matter.
Frequent clipping also diminished the efficiency of nitrogen usage.
Delaying autumn fertilization at Ft. Pierce, Florida, from 17 September
or 1 October to 29 October resulted in greatly reduced forage production
when harvested on 17 December. The later nitrogen fertilization,


28
however, resulted in a better combination of yield and quality by rais
ing the inherently low protein content in mature limpograss. Bigalta
responded to cool-season nitrogen fertilization but gave way to weed
encroachment during the summer which led to the recommendation that it
should be rested sometime in the warm season to maintain plant
populations.
Quesenberry and Ocumpaugh (1979) studied clipping and grazing
defoliation methods for three years. The mob grazing method shortened
the time necessary to advance the new limpograss germplasm through the
early phases of agronomic evaluation. Quesenberry et al. (1981) summa
rized the events leading to the clip/graze experimentation. After
preliminary testing of 53 clones in greenhouses and small plot clipping
trials, 22 were selected for evaluation by clipping and 27 by grazing.
Eight of the best genotypes were then evaluated at four frequencies of
grazing (3, 5, 7, and 9 weeks). Ocumpaugh et al. (1981) identified
PI 364888 as superior to Bigalta under grazing due to comparable
digestibilities and higher persistence.
Meanwhile, Ruelke and Quesenberry (1982) obtained more data on
seasonal productivity for PI 364888. They found nitrogen fertilization
increased early spring growth; however, the responses they obtained were
limited due to effects of spring drought and cold temperatures. In the
autumn they studied deferred forage characteristics and suggested that
after 10 weeks of age forage quality would decline to a maintenance
level, and by 20 weeks maturity the forage would be rejected. This
statement may reflect a change in attitude with respect to stockpiling.
The differences in persistence among limpograss genotypes led
researchers to ask "why?" They reasoned that a knowledge of


29
nonstructural carbohydrate metabolism should contribute to an under
standing of limpograss behavior. Christiansen et al. (1981) studied
etiolated regrowth as an indicator of the stored energy in four
1impograsses. Some of the results indicated that morphological differ
ences were related to observations of agronomic performance. Frequent
cutting treatments caused a drain of the energy reserves of Bigalta but
not PI 364888. This suggested greater reserve energy storage for
PI 364888.
Animal Response and Systems Management
Hodges and Pitman (1981) studied Bigalta limpograss, 1 Cal lie*
bermudagrass, 'Sarasota1 stargrass (Cynodon nlemfuensis Vanderyst), and
'Ona' stargrass under year long grazing. During the cool season,
Bigalta produced average daily gains of 0.20 kg as compared to 0.23,
0.21, and 0.13 kg for Ona, Sarasota, and Callie, respectively. Warm
season average daily gains were similar for the four grasses and twice
as high as the gains produced in the cool season. The stargrasses had
much higher yearly production as reflected by animal grazing days per
hectare: Sarasota, 1089; Ona, 1072; Callie, 911; and Bigalta, 783.
Beef production over both seasons averaged 585, 558, 450, and 431 kg/ha
for Ona, Sarasota, Bigalta, and Callie, respectively.
Ocumpaugh (1982) is presently conducting the second year of a three
year animal production study comparing beef gains on Pensacola bahia-
grass and PI 364888 limpograss. Preliminary results from 1981 indicate
better average gain per yearling heifer on limpograss (78.6 vs. 57.7
kg/yr) and a 70 calendar day advantage in grazing days compared to


30
Pensacola bahiagrass. Ocumpaugh states that PI 364888 is being con
sidered for cultivar release in June 1983.
Nonstructural Carbohydrates in Forages
Plants are the primary source of carbohydrates. Cellulose is the
most prevalent organic compound on earth and is man's most important
industrial carbohydrate. The staple grains are predominantly starch
which is the chief carbohydrate in the human diet (Greenwood, 1970).
This review will be confined to a summary of the carbohydrates important
in forages. Structural carbohydrates are beta-linked molecules that are
degraded into more utilizable forms by rumen microflora. These simpler
substrates are then converted into energy and animal products useful to
man. The nonstructural plant carbohydrates can be completely utilized
in animal diets and also provide the energy necessary for bacterial
preservation of silage.
In plants, nonstructural carbohydrates are used in growth and
respiration and have been called "food reserves." The main objective of
this review will be to synopsize the evolution of non-structural carbo
hydrate research in pasture plants during the past 60 years.
Quantities of nonstructural carbohydrates in plants fluctuate
during each day as well as during the season. Variations in geography,
environment, taxonomy, anatomy, and management (or experimental condi
tions) all add to the dynamics of carbohydrate flux.
Reactions to management vary from year to year depending on
environment, species, and stress; hence, there are conflicts in the
literature due to incongruities between imminent and long term results,
between animal and pasture requirements, and between applied and


31
basic objectives. Three generations of researchers have often perpetu
ated a benign acceptance of conventional wisdoms.
The overwhelming quantity of carbohydrate studies created a need
for review articles to summarize the research findings. Many theories
were proposed to consolidate research on different species grown under
different conditions for different purposes. The most important eluci
dations were (1) the definition of what organic substances should be
considered reserves, (2) what methods were best to satisfactorily frac
tionate carbohydrate components, (3) what role carbohydrates play in
regrowth mechanisms, (4) where plant foods accumulated, (5) how non-
structural carbohydrates varied by day and season, and (6) how carbo
hydrates in storage organs were affected by management. The reader is
referred to the following publications as a chronological guide: Graber
(1931), Weinmann (1952, 1955, 1961), Troughton (1957), May and Davidson
(1958), Hunter et al. (1970), Sheard (1973), Smith (1973b), White
(1973), and Noble and Lowe (1974). For work prior to the 1930's, the
reader is referred to the literature citations of Graber et al. (1927)
and Graber (1931).
The important findings of Cugnac (1931) are alluded to by nearly
all carbohydrate reviewers. He separated the grasses into two groups--
the fructose accumulating grasses native to temperature climates and
grasses accumulating sucrose and starch that are mostly adapted to warm
regions.
Graber (1931) assessed the condition of "low" or "high" organic
reserves in pasture plants by means of dry matter yields, persistence,
and weed encroachment. He saw that plant growth behavior was related to


32
available nutrients and that quantities of organic food were likewise
correlated.
The Characterization of Carbohydrate Reserves
Leukel and Coleman (1930) in Florida measured carbohydrate frac
tions of bahiagrass and claimed that hemicellulose was transformed to
lignin and cellulose under long cutting intervals and reduced to simpler
sugars for use in tissue synthesis with a frequent defoliation regime.
McCarty (1935, 1938) studied seasonal carbohydrate fluctuations in
several range grasses and concluded that sucrose and starch were the
stored foods in California bromegrass (Bromus carinatus Hook, and Arn.).
McCarty (1938) also conjectured that hemicellulose may have been con
verted into simpler components.
Sullivan and Sprague (1943) showed no hemicellulose utilization in
the regrowth mechanism of ryegrass (Lolium perenne L.). Fructosan was
the key reserve substance and these results were reinforced by Waite and
Boyd (1953a, b) and Waite (1957, 1985). The concept that hemicellulose
participated in respiration or tissue synthesis was essentially dis
carded by the late 1940's; however, fructosan storage characterization
in important northern adapted grasses continued (Sprague and Sullivan,
1950; Waite and Gorrod, 1959; Okajima and Smith, 1964; Smith and
Grotelueschen, 1966; Smith, 1967; Grotelueschen and Smith, 1968; Smith,
1975). Smith (1968) cataloged the carbohydrate storage tendencies of
many North American grasses.
Tropical legume and grass carbohydrate characterization occurred
later and remains an active area of research. Hunter et al. (1970)
found no fructosan accumulation in the tropical plants he studied.


33
Noble and Lowe (1974) showed smaller seasonal variation of alcohol
soluble carbohydrates in tropical grasses than in temperate grasses, and
Wilson and Ford (1973) found that temperate grasses accumulated much
higher concentrations of soluble carbohydrate than the tropical grasses.
A Progression of Nonstructural Carbohydrate
Methods Used in Agriculture
A simple measure of energy reserves is obtained by the regrowth of
a defoliated plant in darkness. The technique involves the removal of
the sod, usually a 15 cm diameter plug, and allowing the defoliated
plants to regrow in darkness with adequate moisture until the energy
producing materials are exhausted. The weight of the clippings produced
during the period give an index of the regrowth potential of the plant.
Sheard (1973) reviewed etiolated regrowth studies, and his oldest refer
ence is that of Burton and Jackson (1962); however, the regrowth-in
darkness technique goes much further back in time.
According to Smith (personal communication):
The first use of growth in darkness that I know of is the
early work of L. F. Graber (1927). I have always been
intrigued with the technique, probably from being a student of
Dr. Graber's. Where Graber got the idea I do not know, but he
did his Ph.D. work with Dr. Kraus, a botanist at the Univer
sity of Chicago, who worked a great deal with carbohydrate/
nitrogen ratios in the growth of tomatoes and who may have
used darkness studies. The first person to use growth in
darkness on grasses was Vance Sprague (Sullivan and Sprague,
1943) at the USDA Pasture Lab in Pennsylvania. He picked up
the technique from Graber when he was his Ph.D. student, and
Blaser got the idea from Sprague (Ward and Blaser, 1961).
Pretreatment in darkness to vary carbohydrate concentration was
used by Davidson and Milthorpe (1966b) and others that used the
regrowth-in-darkness experimentation were Adegbola (1966), Adegbola and


34
McKell (1966b), Alberda (1966), Humphreys and Robinson (1966), Rese and
Decker (1966), Matches (1969), Watson and Ward (1970), and Christiansen
et al. (1981). Christiansen et al. (1981) found improved sensitivity in
separating treatment differences by using stem base weights as a covari
able to control variations in plant size.
Concerns over the unreliability of chemical fractions were due to
the use of acids in early applications of carbohydrate methodology.
Fructosans are water- and ethanol-soluble, as well as readily hydrolyzed
to fructose monomers by acid treatment. Consequently, Sullivan and
Sprague (1943) complained that acid analyses of starch were confounded
by the contribution of fructose from fructosan in with glucose from
starch in tests of reducing power.
In 1947, Weinmann published a method for total available carbo
hydrate (TAC) determination in plants. Total available carbohydrate was
defined as "all those carbohydrates which can be used in the plant body
as a source of energy or as a building material, either directly or
indirectly after having been broken down by enzymes" (p. 279). In
Weinmann's method, small samples of finely ground air-dry material are
digested by takadiastase in water resulting in the breakdown of starch,
dextrins, and maltose to glucose, while other sugars and fructosan are
solubilized at the same time. The latter compounds are converted to
hexose sugars by acid hydrolysis, following which the reducing power of
the cleared, neutralized hydrolysate is determined (Weinmann, 1947).
Lindahl et al. (1949) made slight modifications in the Weinmann
TAC method and hailed the efficacy of the procedure in agronomic
applications. Their major modification was to change the enzyme from
dialyzed takadiastase to "clarase"--a highly purified and concentrated


35
form of takadiastase containing invertase, maltase, and amylase. This
procedure satisfied most agricultural needs for TAC determination
throughout the 1950's; however, Smith continued to examine acid tech
niques from his lab in Wisconsin. The acid methods were faster, requir
ing 8 hours to complete analysis compared to 12 hours working time plus
44 hours incubation for the enzyme method.
Most of the hindrance of using an acid method was in finding the
proper concentrations. Smith et al. (1964) varied sulfuric acid concen
trations from 0.2 N_ 0.8 N_ and compared the results to takadiastase as
a standard. The 0.2 N_ F^SO* method most nearly duplicated the
takadiastase extraction while higher concentrations degraded
hemicellulose. The results were supported by Burris et al. (1967) who
added that the chances of obtaining erroneous data with acids were
greatest when they analyzed bermudagrass during rapid growth phases and
with tissue high in starch.
Grotelueschen and Smith (1967) qualified earlier work (Smith
et al., 1964) after examining alfalfa (Medicago sativa L.), high in
starch content. For fructosan accumulating tissue such as timothy
(Phleum pratense L.), containing little starch, dilute acid (0.005 N_
F^SO^) procedures may be used for nonstructural carbohydrate
extraction. On the other hand, with tissue high in starch such as
legumes and tropical grasses, acid strengths beginning at 0.2 N_ F^SO^
degraded hemicellulose and incompletely hydrolyzed starch as well.
Greub and Wedin (1969) supported Grotelueschen and Smith's 1967
findings and also warned that above 0.2 N F^SO^ free fructose or
fructose liberated from fructosan was being destroyed. The enzymatic


36
method was more accurate, demonstrated again at a later date, in
Portugal by Chaves and Moreira (1977).
In 1969 Smith wrote his widely accepted version of the Weinmann/
Lindahl procedure. It was and still is the most popular titrimetric
nonstructural carbohydrate method. Smith suggested the term total non-
structural carbohydrate (TNC) as a more clearly definable term than TAC
to both plant and animal investigators. The advantage of the Smith
method was its organized, complete, "cookbook" presentation which
included an information review, appendices, criteria for selecting a
method, and sample preparation. Smith (1981b) revised the TNC method
with minor modifications including an enzyme change. It was found that
Mylase 100 has rapid saccharogenic activity and will completely digest
the starch in tissues containing 30 percent or less of TNC in 20 hours.
Advances in carbohydrate methodology in the past 10 years have
been centered around the shift toward colorimetry to increase speed,
sample number, and efficiency (Haslemore and Roughan, 1976; Weier
et al., 1977; daSalveira et al., 1978; Westhafer et al., 1982).
In colorimetric methods, specific color reactions for portions of
carbohydrate molecules are due to formation of furfural or furfural
homologues in strong acids, especially following heating. These furans
or their reaction products are derived from oxidation, reduction, or
condensation processes in strong acid and can form colored products upon
reaction with sugars. There are treatmendous variabilities associated
with the reactions depending on the sugar, reagent concentration, tem
perature, and time of heating (Dische, 1962). The search for specific,
quantitative, and reproducible colorimetric assays for the determination
of carbohydrates is an important area of research.


37
Colorimetric tests have advantages over titrimetric procedures due
to their speed and equivalent precision. Two distinct steps involved
in colorimetric reactions are (1) formation of a chromocen from the
sugar, and (2) development of the color by a condensation of the
chromogen with a specific reagent (Aminoff et al., 1970). It is impor
tant to recognize that most colorimetric methods are empirical. Since
little is known about the reaction mechanism or the exact nature of the
chromogen involved, absolute stochiometry is not often obtained. As
long as the results obey Beer's Law within appropriate limits of concen
tration, the problem can be solved by using appropriate internal
standards and blanks.
In quantitative analysis, two divergent objectives must be
considered: (1) an overall analysis of all the sugars present by a very
general reaction, and (2) the selective determination of one sugar in
the presence of others. For example, Westhafer et al. (1982) found
sucrose levels in turfgrass root tissue to have the most dramatic
response to nitrogen treatment; thus, a TNC method would be less sensi
tive measure of carbohydrate concentrations changes than a test of
sucrose.
The principal chemical methods for quantitative measurement of
sugars use the action of sugar reduction on alkaline solutions of the
salts of certain metals. The most extensively used metal in sugar
analysis is copper. Most of the agricultural needs for quantitative
sugar analysis have been satisfied with the titrimetric method of
Somogyi (1945, 1952) and the following colorimetric techniques: phenol-
sulfuric (Dubois et al. 1956), anthrone, and Nelson's (1944) test for


38
reducing sugars. An excellent discussion of reducing sugar techniques
is given by Hodge and Hofreiter (1962).
Organic Substances Used in Regrowth
Perhaps the biggest controversy in the history of nonstructural
carbohydrate research was conclusively proving that food reserves were
used in the synthesis of new growth. May and Davidson (1958) stated
that a general acceptance of the importance of carbohydrate reserves in
regeneration seemed unjustifiable and argued that no causal role in
shoot regrowth was suggested by decreases of carbohydrate in storage
organs. May (1960) dismissed as conventional wisdom the Graber and
Weinmann definitions of "reserves" on the basis that the term had become
semi-technical and no longer subject to criticism. May (1960) also
cited Archbold (1945) and Bernatwoicz (1958) who discarded the idea of
stored sugars as a "purposive" reserve--"reserve" connotating provision
for the future and as such signifying teleological thinking. A better
definition was "accumulate" since it was non-committal concerning pur
pose or intent.
Carbon balance and labeling studies put an end to the controversy
(Marshall and Sagar, 1965; Alberda, 1966; Davidson and Milthorpe, 1966a;
Ehara et al., 1966; Wardlaw, 1968; Watson and Ward, 1970; Sheard, 1973).
Davidson and Milthorpe (1966a) demonstrated that nitrogenous organic
compounds were also mobilized under heavy stress to the plant; there
fore, carbohydrates were only part of the labile pool used for respira
tion and new growth.


39
Location, Seasonal, and Daily Fluctuation
of Reserve Foods
Nonstructural carbohydrates may be temporarily stored in all plant
parts. Troughton (1957) concluded that the major storage regions were
in the underground organs. Many other studies, however, have shown that
major storage parts are stem bases, including stolons, corms, and
rhizomes (Sullivan and Sprague, 1943; Baker and Garwood, 1961). Waite
and Boyd (1953a, b) and Smith (1967) found percentages of fructosans and
most sugars to be higher in the stems than in leaves in temperate
grasses. For three tropical species, Hunter et al. (1970) found a
higher concentration of reducing and total sugars in the stems than leaf
blades plus sheath. Perry and Moser (1974) reported on the TNC content
of eight range grasses and stressed the importance of locating the
specific storage organ(s) of a grass before proceeding with TNC analysis.
The products of photosynthesis may be held in the leaf blades as
sucrose or starch. Since sucrose is the main transocatable sugar and
starch is stored in chloroplasts, it explains the frequent measurement
of high amounts of these substances in leaves (Greenfield et al., 1974),
especially with cool temperatures (Garrard and West, 1972; Carter and
Garrard, 1976).
The environment governs seasonal assimilate distribution and carbo
hydrate metabolism is greatly affected by temperature. The accumulation
of carbohydrates at low temperatures indicates that growth rates are
more affected than are photosynthetic rates, and there is no conclusive
evidence that reduced translocation is ever the primary cause of limit
ing growth under low temperatures (Wardlaw, 1968). Hence, for both
temperate and tropical species, carbohydrate accumulation occurs in the
autumn or cool season. However, in temperate areas autumn turns to


40
freezing winters and TNC reserves are necessary for survival until
spring. Once perennial forages begin growth in the spring, TNC levels
generally increase through vegetative stages to anthesis and later the
carbohydrate reserves decline slowly through the summer when hot night
temperatures cause high rates of respiration. Studies to support the
above hypothesis for temperate grasses were published by Waite and Boyd
(1953b), Baker and Garwood (1961), Trlica and Cook (1972), Smith (1975),
and Mislevy et al. (1978).
For many tropical growing grasses carbohydrate accumulation occurs
during the cool or dry season and carbohydrate drain is most intense
during the summer due to high night temperatures and very active growth.
It could be speculated that plant survival is dependent upon carbohy
drate accumulations during the cool seasons. In northern Florida, for
instance, frost will kill most above ground herbage but warm day tem
peratures and adequate day lengths permit basal leaf growth in species
such as limpograss (Gaskins and Sleper, 1974). Studies showing the
accumulation of carbohydrate in tropical grass species during the cool
season were reported by Woods et al. (1959), Ferraris (1978), and Wilson
and t'Mannetje (1978).
Daily carbohydrate fluctuations occur for all species but to dif
ferent extents. Holt and Hilst (1969) showed that bromegrass (Bromus
inermus Leyss.) utilized almost one-third of the TNC in the herbage
during the night, but diurnal fluctuations were less for other grasses.
Greenfield and Smith (1974) studied switchgrass (Panicum virgatum L.)
and found that the diurnal trend was an increase of total sugars and
starch from 6 am to 6 pm and then a decrease to midnight. Basal sheaths
and internodes tended to increase in percent starch and TNC from 6 am


41
to midnight. Since these are storage parts, carbohydrates were presum
ably being translocated continuously from upper parts to these lower
sinks for storage, especially after 6 pm.
Management Factors Affecting TNC:
An Integrated Approach
Applied management of forage plants must optimize yield, quality,
and plant persistence. Plant behavior can be modified by cultural
practices but these procedures must integrate the plant physiology
involved in maintaining vigor. Too often experiments disect plant
response in order to study one variable at a time. A holistic approach
that takes into account environment and plant growth stage will be
necessary for management systems to be of any practical use.
While it is generally known that carbohydrate accumulation varies
inversely with the growth rate of the plant (McCarty, 1935; Brown and
Blaser, 1965; Colby et al., 1965; Blaser et al., 1966), few management
systems are based upon these findings. Carbohydrate analyses alone
cannot unambigously identify a superior management regime because of
confounding variations in residual leaf area, crown structure, axillary
bud number, leaf age, and altered root characteristics (Humphreys,
1966).
Much effort was expended in the understanding of residual leaf
area, light interception, and carbohydrate reserves in explaining
regrowth following defoliation (Ward and Blaser, 1961; Pearce et al.,
1965; Davidson and Milthorpe, 1966b; Humphreys and Robinson, 1966).
When moisture and nutrients are in adequate supply, residual leaf area
is generally more important than food reserves; however, this concept


42
of explaining regrowth is dependent upon the intensity of defoliation,
the presence or absence of buds, the age of leaf tissue, and species
differences. Further, there is no clear relationship between rate of
growth and leaf area index. Hence, this is an oversimplified model on
which to base a management system (Milthorpe and Davidson, 1966).
The role of hormones in releasing apical dominance during frequent
defoliation is often neglected. Certainly some plants have a better
ability to adopt a prostrate growth habit with many small leaf blades
that maintain assimilate supply to the plant. The classical study of
bermudagrass by Weinmann and Goldsmith (1948) as reviewed by Weinmann
(1961) comes to mind. Close cutting of a well fertilized green of
Cynodon dactyl on 91 times in a season did not result in TAC depletion
due to high residual leaf area; however, complete defoliation by means
of scissors, repeated at weekly intervals, nearly exhausted TAC
reserves. Graber (1931) and Leukel and Coleman (1930) recognized the
ability of plants to defend against frequent defoliation by altered
morphology.
Nitrogen fertilization will stimulate herbage growth and, in
general, will cause a reduction of carbohydrate reserves as they are
used as carbon skeletons for protein synthesis (Waite, 1958; Alexander
and McCloud, 1962; Colby et al., 1965; Adegbola and McKell, 1966a;
Alberda, 1966; Auda et al., 1966; Gallaher and Brown, 1977; Wilson and
t'Mannetje, 1978).
Nitrogen fertilization can cause increases or decreases in carbo
hydrate storage depending on the amount applied and time of sampling
(Sprague and Sullivan, 1950). Carbohydrate reserves are only utilized
for a short time following defoliation (2 days--Davidson and Milthorpe,


43
1965, 1966a, b; 6 daysEhara et al., 1966; 7 days--Sul1ivan and
Sprague, 1943). Moderate nitrogen fertilization promotes growth, photo
synthesis, and TNC storage; therefore, sampling too late in non-stressed
plants will not show a carbohydrate decline.
In frequently cut and highly fertilized swards, carbohydrate drain
from storage organs can continue to the detriment of plant persistence.
Alberda (1966) reported the death of tillers following severe defolia
tion stress and stated, "It may be supposed that a considerable part of
. . these tillers is broken down and translocated to the remaining
tillers to be used for new leaf formation, but this has not been proven"
(p. 147).
The rate of degradation of a stressed sward is accelerated once a
state of tiller degradation occurs. As nitrogen fertilization forces
herbage growth and frequent cutting disallows full leaf expansion,
little assimilate is mobilized to roots and buds (Wardlaw, 1968).
Energy reaching the roots is inadequate to meet the needs of respira
tion, root growth, and absorption have slowed or stopped, and root degra
dation occurs as other substances are scavenged as a last defense
(Davidson and Milthorpe, 1966a).
Further complications in this theoretical example of high plant
stress are imposed by aggressive weeds competing for nutrients, mois
ture, and light. Due to the combined effects of all the above factors,
the botanical composition of the desired species declines and the sward
degenerates. In conclusion, when TNC studies are used as a single tool
complemented with other information, they greatly aid in understanding
the total dynamics of plant behavior. However, emphasis must be placed
on a balanced understanding of all factors involved in plant growth if


44
it is the objective of research to lead to sound management
recommendations.


CHAPTER 1
REGROWTH IN DARKNESS AS INFLUENCED BY PREVIOUS CUTTING
TREATMENT OF FOUR LIMPOGRASS GENOTYPES
Introduction
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) is
native to the humid subtropics of Africa. A brief resume of its intro
duction into American grassland agriculture has been reported by Oakes
(1973). Agronomic evaluations (Kretschmer and Snyder, 1979; Quesenberry
et al., 1978; Quesenberry and Ocumpaugh, 1980; Ruelke et al., 1978) have
generated information supporting the view that limpograss has forage
potential in subtropical regions and on soils that are intermittently
flooded. This study of etiolated regrowth was performed as a means of
characterizing the energy reserves and morphologies of four 1impograsses.
The regrowth-in-darkness technique has been used by many
investigators (Adegbola, 1966; Adegbola and McKell, 1966b; Burton and
Jackson, 1962; Dovrat and Cohen, 1970; Matches, 1969; Rese and Decker,
1966; Ward and Blaser, 1961; Watson and Ward, 1970). The objective of
this study was to determine whether frequent clipping in the field could
reduce energy reserves below the critical amounts necessary to maintain
stands through the dormant season. Although chemical tests were not
conducted, much evidence exists substantiating a strong correlation
between etiolated regrowth and nonstructural carbohydrates in storage
organs (Adegbola, 1966; Adegbola and McKell, 1966b; Dovrat and Cohen,
1970; Rese and Decker, 1966).
45


46
No carbohydrate studies of limpograss could be found; hence, it was
desired to learn the location of carbohydrate storage in these
stoloniferous grasses before full scale chemical analyses were con
ducted in later phases of research.
Materials and Methods
A clipping study was conducted on four well-established plantings
of limpograss on a Wachula sand, a poorly drained siliceous hyperthermic
ultic haplaquod soil at the Beef Research Unit of the University of
Florida near Gainesville. The grasses were PI 349753, PI 299995,
PI 299993, and PI 364888. Grasses PI 299995 and PI 299993 have been
released as 1Bigalta1 and 'Redalta1, respectively. Redalta is a diploid
(2n = 18) and the other genotypes used here are tetraploids (2n = 36).
The four blocks of grasses representing genotypes were unreplicated;
hence, the design was a split plot without replication of main plots.
The genotype*treatment (rep) term was used to test genotype, treatment,'
and interaction effects. Within main plots, clipping treatments were
randomly assigned with three replications. Clipping treatments were
initiated by mowing all plots to 5 cm on 27 July 1979 and harvesting at
2.5, 5, and 10 week intervals until 4 October 1979. Additional areas
without replication were reserved to allow limpograss top growth to
reach 15 and 25 weeks of age by 14 November 1979, when the regrowth-in
darkness Experiment 1 began.
The limpograsses were fertilized 28 May and 30 July 1979 with 280
kg/ha 17-5-10 (N-P^Og-K^O) containing 1 percent of a microelement mix.
Soil test results from samples taken on 3 December 1979 showed a pH of
6.6, 27 kg/ha phosphorus, and 22 kg/ha of potassium.


47
Experiment 1
On 14 November 1979, 15 cm diameter cores were removed from the
2.5, 5, 10, 15, and 25 week treatments in each of the four limpograss
plantings. The cores were placed in black plastic pots with the plant
material trimmed to 2.5 cm height. Two subsamples were taken from each
of the 2.5, 5, and 10 week field plots, and six samples were taken from
the areas assigned a 15 and 25 week cutting interval. Hence, there were
six pots containing limpograss for each treatment, and they were
arranged on shelves in a dark room using a randomized, complete block
design. A small electric heater was used to maintain the air tempera
ture at approximately 30 C. Plants were watered and sprayed with fungi
cide when necessary. The study was terminated after 3 weeks when growth
had ceased. "Shoot" regrowth was removed before the remainder of the
plant was washed free of soil and separated into "roots" and "stubble."
All three components were then dried at 60 C and weighed. The dry
matter (DM) yields were subjected to analysis of variance and regression
analysis. The mean yields for treatments were subjected to the Waller-
Duncan Multiple Comparison Test.
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Northeast Regional Data Center of the
State University System of Florida, located on the campus of the
University of Florida in Gainesville.


48
Experiment 2
A second etiolated regrowth experiment was begun on 6 February 1980
using new cores of PI 364888 from each of the five treatments described
in Experiment 1. In this study plants were trimmed to ground level
leaving short "stem bases" below ground. A completely randomized design
was used with nine replications in the dark room. The regrowth period
was 3 weeks in length, and plants were processed as in Experiment 1. In
Experiment 1, various statistical models were employed to explain the
etiolated regrowth as a function of the weight of "stubble" and/or
"roots," and in Experiment 2, the reduced "stubble," i.e., "stem base"
weight, was selected as a covariable in order to increase precision by
accounting for variations in plant size.
Residual Effects
The limpograsses were clipped at a height of 5 cm on 30 April 1980.
The forage was collected and dried at 60 C, weighed, and analyzed for
yield differences created by the 1979 clipping treatments.
Results and Discussion
Experiment 1
The results obtained in Experiment 1 are summarized in Table 1.
Clipping treatments caused etiolated regrowth yields to be different for
PI 349753 and Bigalta (P = 0.026 and P < 0.01, respectively), whereas
Redalta and PI 364888 were not affected by previous cutting treatment in
the field (P = 0.138 and P = 0.157, respectively). For both PI 349753
and Bigalta, a maximum regrowth was observed at the 10 week cutting


49
Table 1. Average DM yields of etiolated "shoots" for four limpograsses
following 3 weeks of growth in darkness
Clipping
frequency
(weeks)
Limpograss
number or name
PI 349753
Bigalta
Redalta
PI 364888
y/^ "
*
2.5
0.74 ab
0.95 ab
0.43 a
0.36 a
5
0.76 ab
0.81 b
0.36 a
0.45 a
10
0.87 a
1.08 a
0.62 a
0.42 a
15
0.52 b
0.34 c
0.65 a
0.45 a
25
0.51 b
0.41 c
0.55 a
0.63 a

Values within each column followed by the same letter are not
significantly different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.


50
frequency with lower yields of regrowth for the 15 and 25 week
treatments. Within the field plots representing the two longest cutting
frequencies, new growth was noticed emerging from plant bases below a
dense canopy. It is believed that this light-starved growth in
PI 349753 and Bigaita contributed to a respiratory drain of the energy
reserves. The same trend for lower DM yields in the 25 week treatment
was observed for Redalta but not for PI 364888 which had more resistance
to lodging allowing more light to reach new tillers beneath the canopy.
Weights of "shoot," "root," and "stubble" components were averaged
for all cutting frequencies in order to compare plant form among
1impograsses. As shown in Table 2, Bigalta and PI 349753 had signifi
cantly heavier "shoot" weights compared to Redalta and PI 364888.
Bigalta and PI 349753 also had the lowest "stubble" weights. These data
agree with field notes characterizing Bigalta and PI 349753 as having
larger but fewer "shoots" per unit area than Redalta and PI 364888. Of
the four grasses, PI 349753 had the lowest "root" weight. Redalta has
a bunch-type growth habit and a large "root" mass, both of which may
contribute to its excellent persistence.
These results suggest that the large amount of "stubble" in Redalta
and PI 364888 might have provided a reservoir of energy reserves which
buffered the cutting pressure on these two lines. Matches (1969) showed
higher regrowth yields with increasing height of cutting, indicating
an energy reserve sink in the stem bases of tall fescue (Festuca
arundinacea Schreb.).
In this study of limpograss there was a possibility that a lower
cutting height in darkness would reduce sink size and separate treatment
effects in a significant way.


51
Table 2. Comparison of etiolated "shoot" DM yields and associated
"root" and "stubble" components for four limpograsses averaged
for all cutting treatments following 3 weeks of growth in
darkness
Limpograss
Component
"Shoot"
"Root"
"Stubble"
PI 349753
0.68 a*
g/pot
9.66 c
2.46 c
Bigalta
0.72 a
14.47 b
2.33 c
Redalta
0.52 b
17.46 a
3.14 b
PI 364888
0.46 b
12.36 b
4.71 a
k
Values within each column followed by the same letter are not
significantly different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.


52
Experiment 2
The etiolated "shoot" weights from Experiment 1 were modeled as a
function of the cutting treatments and "roots"; cutting treatments and
"stubble"; and cutting treatments, "roots," and "stubble." The reduced
models, which explained regrowth yields as a function of treatment and
"stubble" (or treatment and "roots"), described etiolated regrowth as
well as the full model for all four 1impograsses. "Root" weights were
affected (P = 0.006) by treatments in Experiment 1, whereas "stubble"
weights were not (P = 0.712). Covariables are not supposed to be
affected by treatments, and since the "stem bases" from Experiment 2
were actually that portion of the "stubble" below ground, the "stem
base" weight was used as a covariable to explain some residual error in
the analysis of Experiment 2 regrowth.
Analysis of variance of etiolated regrowth in Experiment 2 did not
detect treatment differences (P = 0.180) but analysis of covariance did
(P = 0.026). Table 3 shows the Haller-Duncan Multiple Comparison Test
of average DM yields of "shoots" obtained for each clipping frequency in
PI 364888. "Shoot" yields were lower and the range narrower than found
in Experiment 1 for PI 364888, but the data followed the same trend.
Lower yields can be attributed to the loss of some energy reserves that
could be located in the above ground stem tissue removed prior to the
start of Experiment 2.
Results agree with those of Matches (1969), who suggested that with
shorter heights of cut the ranking order of treatments would remain
nearly the same but the magnitude of difference of energy reserves might
be less. Matches also stated that in etiolated regrowth experiments


53
Table 3. Average DM yields of etiolated regrowth from
Experiment 2 following 3 weeks of growth in darkness
Clipping frequency
(weeks)
PI 364888 "shoots"
(g/pot)
2.5
0.28 c
5
0.32 be
10
0.34 abc
15
0.38 ab
25
0.42 a
k
Values followed by the same letter are not significantly
different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.


54
higher cutting heights permit more regrowth and allow greater differ
entiation of treatment effects. That statement is further substantiated
by Watson and Ward (1970) who demonstrated a 25 percent reduction in
food reserves when cutting height was reduced from 7.5 to 2.5 cm in
dallisgrass (Paspa!urn dilatatum Poir.).
In this study the results of Experiment 2 show that trimming the
plants to ground level required a covariable ("stem base" weight) to
detect treatment differences. However, reducing rather than increasing
residual plant height together with the small differences in regrowth
yields between Experiment 1 and Experiment 2 inferred that the majority
of food storage occurs in the very basal portions of the culm ("stem
base").
Residual Effects
Yield data obtained from 30 April 1980 harvest of the limpograsses
showed no evidence of a residual treatment effect on the sward. No
sward damage was observed even in the limpograss plots harvested for
four cycles of 2.5 weeks. Hence, the limpograsses were not sufficiently
stressed to cause stand deterioration in 1980.
Conclusions
1. Experiment 1 showed that etiolated regrowth weights were maxi
mized for the 10 week clipping treatment in Bigalta (P < 0.01) and
PI 349753 (P = 0.026), but no differences were found due to previous
cutting for Redalta (P = 0.138) or PI 364888 (P = 0.157).


55
2. Following the complete removal of above ground stem tissue at
the start of Experiment 2, the production of considerable etiolated
growth suggested energy reserve storage lower on the culm.
3. "Stem base" weights were used as a covariable in Experiment 2,
and differences (P = 0.026) were found in yield of etiolated regrowth
due to previous cutting pressures on limpograss PI 364888 with longer
cutting intervals allowing greater yields of regrowth.
Summary
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb.)
research has reached a stage where management recommendations are needed
to fully implement limpograss' usefulness for the large hectarage of
improved pastures on Spodosols like Florida's flatwood soils. A pre
liminary study of energy reserves in Redalta, Bigaita, and two other
promising introductions was conducted using the regrowth-in-darkness
technique.
Cutting frequencies of 2.5, 5, 10, 15, and 25 weeks on limpograss
swards in the field were imposed to establish various levels of reserves.
The reserve energy pool was then measured by regrowth yields of plant
cores placed in a dark room.
Two regrowth-in-darkness experiments were conducted. Experiment 1
used a randomized, complete block design with six replications harvested
at 2.5 cm "stubble" height. "Stubble" was classified as all the stem
between the cutting height and "roots." The first experiment showed
that etiolated regrowth weights were maximized for the 10 week clipping
treatment in Bigalta and PI 349753 (P < 0.01 and P = 0.026,
respectively), but Redalta and PI 364888, which had higher "stubble"


56
weights, showed no treatment effects on weight of etiolated "shoots."
In Experiment 2, PI 364888, the line with the highest "stubble" weight
from Experiment 1, was evaluated using a completely randomized design
and nine replications with all plant material removed to ground level.
Hence, only the "stem bases" remaining below the soil surface were
responsible for regrowth in Experiment 2. No treatment effects were
found (P = 0.180) until the data were analyzed with covariance tech
niques using "stem base" weight as a covariable. Statistical sensi
tivity improved, and increases (P = 0.026) in regrowth potential were
detected with clipping treatments of a longer cutting interval. The
covariable analysis represented an improvement in the regrowth-in
darkness technique.
Removing all plant material to the soil level prior to the begin
ning of the second experiment left 2-3 cm of "stem base" below ground.
The close similarity of regrowth yields in Experiment 1, where plants
were clipped to 2.5 cm above the soil surface, and those of Experiment 2
having no above ground tissue, suggested that the energy reserves were
predominantly located in the bottom 2-3 cm of the stem.


CHAPTER 2
CUTTING FREQUENCY EFFECTS ON LIMPOGRASS MORPHOLOGY
AND TOTAL NONSTRUCTURAL CARBOHYDRATE RESERVES
Introduction
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) has
promise as an adapted, warm-season grass for Florida's vast hectarage of
acid flatwood soils. Unfortunately, 'Bigalta1, most favored by ranchers
because of its high forage quality, is also the least persistent of the
cultivars which have been released (Quesenberry et al., 1978; Ruelke,
1978; Kretschmer and Snyder, 1979; Quesenberry et al., 1981). Other
studies of limpograss have identified a few promising plant introduc
tions that have comparable quality and better persistence than Bigalta
(Ruelke, 1978; Quesenberry and Ocumpaugh, 1979; Ocumpaugh and
Quesenberry, 1980; Ocumpaugh et al., 1981; Ocumpaugh, 1982).
In Chapter 1, Christiansen conducted regrowth-in-darkness experi
ments with limpograss that indicated the area of compressed, lower nodes
on the stem base as the major site of energy reserve in these robust,
stoloniferous, perennial fodder plants. The regrowth-in-darkness tech
nique is a simple way to measure energy reserves by analyzing regrowth
from defoliated plants placed in darkness (Burton and Jackson, 1962;
Sheard, 1973). Perry and Moser (1974) stressed the importance of locat
ing the specific carbohydrate storage organ of a grass before proceeding
with total nonstructural carbohydrate (TNC) analyses. Studies of TNC
content were deemed necessary to verify the regrowth-in-darkness results.
57


58
Many researchers have indicated a positive correlation between etiolated
regrowth and TNC in the storage organs (Adegbola, 1966; Adegbola and
McKell, 1966b; Dovrat and Cohen, 1970; Rese and Decker, 1966).
Therefore, it was of interest to see if this positive correlation held
for limpograss TNC versus etiolated regrowth.
Leukel and Coleman (1930) and Graber (1931) discussed the ability
of various forage plants to transform an upright habit to a prostrate
habit following frequent defoliation. Some limpograsses may be sus
pected to have greater flexibility than others in altering their growth
habit. The objective was to study morphological changes induced by
frequent cutting and elucidate subsequent differences in TNC accumula
tion for two promising and two released cultivars of limpograss.
Materials and Methods
The experiments to follow were conducted on four well established
plantings of limpograss on a Wachula sand; a poorly drained siliceous
hyperthermic ultic haplaquod soil, at the Beef Research Unit of the
University of Florida. The grasses were PI 349753, PI 299995,
PI 299993, and PI 364888. Both PI 299995 and PI 299993 have been
released as Bigalta and Redalta', respectively. Redalta is a diploid
(2n = 18) and the other genotypes used here are tetraploids (2n = 36).
The limpograsses were fertilized 28 May and 30 July 1979, with
280 kg/ha of 17-5-10 (N-P20g-!<20) containing 1 percent of a microelement
mix. Soil test results from samples taken on 3 December 1979 showed a
pH of 6.6, 27 kg/ha phosphorus, and 22 kg/ha of potassium.


59
Analysis of TNC in Plant Parts
Prior to the layout of the clipping experiment, whole plant samples
of limpograss were taken on 3 July and 26 July 1979, representing 6 and
9 week old plant maturities, respectively. Five large plants of each
genotype were randomly selected, dug, washed free of soil, and arranged
in cotton sample bags. The cotton bags were then packed in a plastic
bag and put on ice. Upon reaching the lab, the samples were dried in a
Thelco forced-air oven at 100 C for 30-45 minutes after which the tem
perature was lowered to 70 C until the samples were removed 36-48 hours
later (Smith, 1973a).
The plants were carefully fractionated into "shoots," "stubble,"
"crown," and "roots." In the plant part analysis, "crown" was desig
nated as the bottom 2 cm of the stem base, and "stubble" was classified
as the immediate 2 cm above the "crown." "Shoots" included all herbage
above the "stubble," and "roots" were severed from the "crown" by knife.
Each component was ground through a 1 mm screen in a small Wiley mill
and then reground through a UDY Cyclone sample mill fitted with a 0.5 mm
screen. The samples were stored in plastic, 20 ml Dilu-vials and then
analyzed for TNC using enzymatic hydrolysis and spectrophotometric
measurement of reducing sugars as described in Appendix A.
TNC as Related to Clipping, Season, and Genotype
A 10 week long clipping study was initiated by mowing all plots to
5 cm on 27 July 1979 and harvesting at 2.5 cm height with a Jari sickle
bar mower at 2.5, 5, and 10 week intervals until 4 October 1979. The
four blocks of grasses representing genotypes were unreplicated; hence,


60
the design was a split plot without replication of the main plots. The
genotype treatment (rep) term was used to test genotypes, clipping
treatments, and interaction effects. Within main plots the clipping
treatments were randomly assigned and replicated three times.
Samples for TNC analysis were dug prior to clipping on 13 August,
30 August, 17 September, 4 October, and 11 November and processed as
before except that only "crown" (bottom 2 cm of stem base) samples were
saved for TNC analysis. The "crowns" were scraped free of roots, leaf
sheaths, and sand by using a wire buffing wheel attached to the extended
shaft of a small electric motor (20 W, 1525 RPM). The "crowns" were
then ground through a 0.5 mm screen in the UDY mill and stored until
tested for TNC.
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Northeast Regional Data Center of the
State University System of Florida, located on the campus of the Univ
ersity of Florida in Gainesville. The plant part analyses and the per
cent TNC data for all subsequent samplings were subjected to analysis of
variance and means were compared using the Waller-Duncan Multiple
Comparison Test.
Correlation of TNC versus Regrowth-in-Darkness
Prior to discussing the methods used in the correlation of TNC and
etiolated regrowth, terminology will be reviewed for clarity. The
regrowth-in-darkness experiments were completed before any chemical
analysis took place. Hence, the first use of the term "stubble" was in
Experiment 1 of Chapter 1 and represented the 2.5 cm of stem tissue


61
above the soil level as well as the stem base below ground. Recall that
the stem material above ground level was removed at the start of the
second regrowth-in-darkness study (Experment 2, Chapter 1), and the term
"stem bases" was used to define the 2-3 cm long stem segments from below
ground. In the plant part experimentation an effort was made to more
specifically focus on the location of TNC accumulation. The "stem
bases" from Experiment 2 in Chapter 1 were believed to contain most of
the stored carbohydrate so in the plant part experiment the bottom 2 cm
of the stem was called the "crown." The 2 cm of stem immediately above
the "crown" was called "stubble" corresponding to common agronomic
usage, i.e., "residual above ground tissue following defoliation."
Limpograss samples taken on 11 November 1979 were analyzed for per
cent TNC to quantitatively characterize the concentration of reserves
present in the "crown" at the beginning of the regrowth-in-darkness
study (Experiment 1, Chapter 1). The darkness study began on 14
November 1979 and lasted 3 weeks. The dry weights of etiolated
"shoots," as well as weights of "stubble" (2.5 cm of stem above ground
plus stem bases below ground) and "roots" per pot were recorded. The
"crown" TNC data from 11 November 1979 were correlated against "stubble"
and against "shoot" percentages of the "stubble" plus "roots" present
in the pots (("Shoot"/("stubble" + "roots')) 10).
Results and Discussion
Analysis of TNC in Plant Parts
Results for TNC analyses of plant parts are shown in Figure 1. The
limpograsses were different (P < 0.01) in their percent TNC; however,
significant two and three way interactions among parts, age, and


_ PI 364888
12i
o
T 8
N
6 b
c
REDALTA
a
BIGALTA PI 349753
d
4
O i i-
6 9
AGE OF LIMPOGRASS TISSUE (weeks)
crowns roots o stubble Dshoots
Figure 1. Total nonstructural carbohydrate (TNC) percent of dry matter of four plant parts in
four 1impograsses sampled at two maturities
(Different letters within an age of limpograss tissue indicate a significant differ
ence (P < 0.05) of percent TNC in the plant parts.)


63
genotypes prohibited a comparison of overall means for the four
1impograsses. When each limpograss was analyzed separately, all except
PI 364888 showed different (P < 0.01) accumulation of carbohydrate to
plant parts at 6 weeks as compared to 9 weeks. Between 6 and 9 weeks
Redalta increased TNC to the "roots" at the expense of other parts,
while Bigalta did just the opposite. In PI 349753 "roots" and "shoots"
increased in TNC while "crown" and "stubble" declined between the two
maturities. These results were interesting for Bigalta and Redalta
because of the high or low contribution, respectively, of readily
digested carbohydrates which would be measured in tests of in vitro
organic matter digestibilities (IVOMD). Many studies have shown large
IVOMD differences between these two grasses (Schank et al., 1973;
Quesenberry et al., 1978).
Criteria necessary in selecting a plant part for TNC analysis were
(1) disqualification of "shoots" because they were not perenniating
parts, and (2) rejection of "roots" because of sand and dead tissue
contamination. Figure 1 shows that the "crown" was statistically as
high or higher in percent TNC than the "stubble" at all times. The
bottom 2 cm of the stem base ("crown") was selected for further TNC
analysis by chemical means.
TNC as Related to Clipping, Season, and Genotype
Table 4 shows the TNC results for four 1impograsses subjected to
three cutting frequencies in the 1979 growing season. In a combined
analysis of sample dates, 1impograsses, and cutting treatments, each
factor influenced (P < 0.01) the TNC measured in "crowns." The two
way interactions (P < 0.01) of limpograss and clipping frequency with


64
Table 4. The main effects of limpograss genotype and clipping
treatment on the percent total nonstructural carbo
hydrates (TNC) in the bottom 2 cm of stem base
("crown")
Limpograss
1979 sampling date
8/13
8/30 9/17 10/4
11/11
TNC %
PI 364888
10.6b*
13.9a
12.4a
10.1a
13.6a
Redalta
6.6d
6.9d
6.4d
6.0d
10.8b
Bigalta
11.0a
11.8b
8.4b
9.9b
8.3c
PI 349753
9.1c
8.7c
8.0c
8.2c
8.Id
Clipping
frequency
(weeks)
2.5
9.8a
10.5a
10.2a
8.7a
12.6a
5.0
9.9a
10.2a
8.8b
9.2a
10.6b
10.0
8.2a
10.2a
7.5c
7.7a
7.5c
Different
letters
within a
sampling date
represent sig-
nificant (P < 0.
05) differences '
in percent TNC.


65
sampling date suggested that TNC concentration was independently altered
by both cutting treatment and seasonal carbohydrate flux. There was no
three way interaction (P = 0.47) or interaction between limpograss geno
types and clipping (P = 0.76); therefore, TNC concentrations were chang
ing according to cutting frequency, but the reaction was similar among
limpograsses.
Analysis of percentage TNC by sampling date revealed the rankings
shown in Table 4. Bigalta was higher in TNC than other limpograsses on
13 August. From 30 August to 4 October PI 364888 was highest in TNC.
In the last sampling on 11 November, the percent TNC in Redalta surged
above Bigalta and PI 349753 but not above that of PI 364888. Perhaps
the late autumn surge of TNC seen in PI 364888 and Redalta contributed
to their higher persistence as opposed to the plateau or decline of TNC
in PI 349753 and Bigalta.
Frequent cutting promoted lateral growth of stolons and increased
the number of leaf blades per unit area. This change toward a turf-like
morphology took some time to effect. In the bottom portion of Table 4
no differences in stored TNC were found for any frequency of cut in the
limpograsses until 17 September. On 17 September and 11 November
shorter cutting intervals enhanced TNC in the "crowns," suggesting an
enhanced ability of the plants with a turf-like habit to accumulate TNC.
Correlation of TNC versus Regrowth-in-Darkness
In Figure 2 data from the regrowth study (Experiment 1, Chapter 1)
was used to correlate against TNC in "crowns" of samples taken at the
start of the regrowth experiment. The best correlations were found for


%
TNC
16
14
12
10
8
6
4
9
ri
9
5
5 a


9
a

B
3 B*
O O
cutting

freq.
9
r value
(wks.)
9*
0.74**
0 2.5
0.65*
5
m
n.s.
OIO
s ^ ^
H
9
9
BSB
O
B
B

a
cutting
9
freq.

r value
(wks.)
a %
-0.65*
o 2.5
-0.67*
5
O

-0.65*
OIO
> ,
S3 ^
S3
PI 364888 a
REDALTA B
BIG ALTA S3
PI 349753
2 3 4 5
Wt. of Stubble.g
2 4 6 8 10 12
Etiolated Shoot % of
Stubble and Roots
14
cn
Figure 2. Total nonstructural carbohydrate (TNC) percentages in the "crown" correlated against
"stubble" weights (left) and etiolated regrowth percent of whole plants (right) for
four limpograss genotypes subjected to three clipping frequencies
P < 0.01


67
TNC versus "stubble" and TNC versus percent plant regrowth in darkness
(("shoots"/("stubble" + "roots")) 100).
In Figure 2 the symbols represent the clipping frequencies used
within each limpograss (genotypes are identified within a symbol by the
coloration pattern). The important relationships are found within each
cutting frequency.
Genotypes having greater "stubble" weight per pot had higher con
tent of TNC in the "crowns." Frequent clipping induced a prostrate
habit in all genotypes, but PI 364888 and Redalta provided more sites
for leaf emergence--as reflected by the accumulation of TNC in "crowns"
for the frequently cut treatments.
A negative correlation was observed in Figure 2 when the TNC in
"crowns" were correlated against the "shoot" percentage of "stubble"
plus "roots." Plants with more "stubble" (PI 364888) and "roots"
(Redalta) (see Table 2, Chapter 1) produced small "shoots" that were
greater in number and lower in weight; hence, were a lower percentage
of the "stubble" plus "roots" (1 to 4 percent in Figure 2). Bigalta and
PI 349753 had fewer but larger "shoots," as well as lower weights of
"roots" and "stubble" (see Table 2, Chapter 1); hence, "shoots" were a
larger percentage of the "stubble" plus "roots" in pots (4 to 14 percent
in Figure 2).
When TNC was correlated directly against the dry weight of etio
lated "shoots," a negative correlation was also found (-0.63) but not
presented. Other experimenters (Adegbola, 1966; Adegbola and McKell,
1966b; Dovrat and Cohen, 1970; Rese and Decker, 1966) have found posi
tive correlations of TNC versus etiolated regrowth, i.e., higher TNC in
plants having longer rest intervals. These studies, however, used


68
upright instead of stoloniferous grasses, and used a single genotype of
a species instead of four genotypes having variable morphologies.
Conclusions
1. Analysis of limpograss plant parts indicated that the bottom
2 cm of stem base ("crown") was a sight of TNC accumulation.
2. Cutting treatments and harvest date contributed independently
to the significant (P > 0.01) variations seen in limpograss carbohydrate
flux.
3. The more frequently clipped limpograsses effected a prostrate
morphology which allowed more axillary tiller formation and TNC accumu
lation for all limpograsses.
4. Redalta and PI 364888 showed an autumn surge of TNC that might
contribute to their better persistence as opposed to a decline or
plateau of TNC for Bigalta and PI 349753.
5. Plants having greater weights of "stubble" also had higher
concentrations of TNC in the bottom 2 cm of the stem base ("crown").
Summary
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) could
significantly contribute to the forage-livestock economy in the south
eastern United States; however, cultivars must have good quality and
high persistence before an impact will be made. Of the cultivars now
available, 'Redalta' and 'Greenalta' have low digestibility and adequate
persistence, while 'Bigalta' has excellent digestibility and poor
persistence.


69
Current research efforts have identified PI 364888 limpograss as
both persistent and digestible. The purpose of this study was to com
pare PI 364888 and another promising limpograss with Bigalta and Redalta
to try to understand persistence relative to the total nonstructural
carbohydrate (TNC) status of the plants.
The objective of this study was to locate the TNC storage site and
characterize seasonal, genotypic, and clipping frequency effects on the
TNC concentration in two released and two promising 1impograsses.
Secondly, it was desired to compare the chemical data with the regrowth-
in-darkness results obtained in a previous study.
A sampling of 6 and 9 week old limpograss on 3 July and 26 July
1979, fractionated into four plant parts ("root," "shoot," "crown," and
"stubble"), was chemically analyzed for percent TNC using an enzymatic
sugar hydrolysis and spectrophotometric measurement of reducing sugars.
The bottom 2 cm of stem base ("crown") was identified as the primary TNC
storage site and, subsequently, served to compare the carbohydrate
status in the four 1impograsses.
A clipping study was conducted to variously deplete the reserves by
cutting the limpograss field plots every 2.5, 5, or 10 weeks from 27
July to 4 October 1979. Results showed that frequent clipping induced
a prostrate growth behavior and the turf-like condition allowed greater
carbohydrate accumulation in the storage sites. Generally, PI 364888
was highest in stored TNC throughout the study and along with Redalta
showed a surge of TNC in samples taken on 11 November 1979. Bigalta and
PI 349753 did not react in this manner in late autumn, and they were
also observed to be less persistent in previous studies. Limpograss


70
PI 364888 had the largest stolon system, stored the most TNC, and was
very persistent.
The TNC results for 11 November 1979 were correlated against
"shoot" and other plant weight data obtained from a 3 week regrowth-in
darkness study initiated on 14 November 1979. The results revealed a
positive relationship (r = 0.56) across limpograsses for stubble weight
versus TNC and a negative correlation (r = -0.56) for etiolated shoots
expressed as a percent of the roots plus stubble. The significant (P <
0.01) regressions were caused by inherent differences in limpograss
morphologies rather than cutting frequency effects within each
1impograss.


CHAPTER 3
DRY MATTER YIELD, CRUDE PROTEIN, IN VITRO ORGANIC
MATTER DIGESTIBILITY, TOTAL NONSTRUCTURAL
CARBOHYDRATE, AND PERSISTENCE IN TWO PROMISING
AND TWO RELEASED LIMPOGRASSES: EFFECTS DUE TO
NITROGEN FERTILIZATION AND CUTTING FREQUENCY
Introduction
The southeastern United States is a subtropical zone that permits
the growth of a wide variety of grasses; only a handful of which are
agronomically important. Bahiagrass (Paspalum notatum Fl'gge) is the
most widespread improved grass in Florida (Mott and Moore, 1977) because
of its excellent persistence and broad adaptability. Bermudagrass
(Cynodon dactyl on (L.) Pers.) cultivars are economically important as
hay and grazing crops due to successful plant breeding programs; but
these cultivars generally grow better on upland sites. Digitgrasses
(Digitaria decumbens Stent.) and stargrasses (Cynodon nlemfuensis
Vanderyst) are grown more in south and central Florida due to their
lower frost tolerance and winter hardiness. St. Augustinegrass
(Stenotaphrum secundatum (Walt.) Kuntze and paragrass (Brachiaria mutica
(Forsk) Stapf) have special adaptabilities for organic soils but are not
widely used as pasture grasses elswhere in Florida.
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) is a
viable alternative to bahiagrass for flatwood sites. Limpograss can be
equally persistent, but higher in quality than bahiagrass (Moore et al.,
1981). Immature bahiagrass has potentially good quality; however,
71


72
quality rapidly declines in tissue greater than 6 weeks of age (Moore
et al., 1970) and requires heavy utilization for optimum yield and
quality (Beaty et al., 1980).
Ocumpaugh (1982) reported better yearly beef production and similar
average daily gains from PI 364888 limpograss than from 'Pensacola'
bahiagrass because of a 70 day grazing advantage. Gaskins and Sleper
(1974) showed that daylength sensitivity was not a limiting factor for
cool-season growth of limpograss but was for digitgrass and bermuda-
grass. Perhaps bahiagrass also falls into the latter category.
Bahiagrass is primarily used for grazing, not hay, and the major
forage related problems in the southeastern United States are seasonal
forage distribution and lack of preservation practices (Mott, 1982b).
Limpograss is more seasonably flexible and can produce copious amounts
of biomass. Killinger (1971) produced 15 m tons/ha of dry matter (DM)
by 23 May and 23 m tons/ha by 14 August in north central Florida. This
surpassed any of the yearly yields for five hybrid bermudagrasses,
'Pangla' digitgrass, and Pensacola bahiagrass reported in 1971 by
Ruelke and Prine. Hodges and Martin (1975) studied the warm- and cool-
season (1 November 15 May) production of 23 subtropical grasses at
Ona, Florida, and found the digitgrasses and Cynodons outyielded limpo-
grasses; however, limpograsses were superior to Pensacola bahiagrass at
all three fertility levels during the cool season. Kretschmer and
Snyder (1979) obtained 21.8 m tons of dry matter (DM) with two harvests
of 'Bigalta' limpograss using a 12 week cutting frequency and 168 kg/ha
N per interval. In a management study, Ruelke obtained yields of 13.5-
23.8 m tons for 'Redalta', 'Greenalta', and Bigalta limpograsses
(Quesenberry et al., 1978).


73
The quality characteristics of tetraploid limpograsses are well
known. Schank et al. (1973) identified high in vitro organic matter
digestibilities (IVOMD) and a slower rate of IVOMD decline in the
tetraploids. Taylor et al. (1976a), in New Zealand, also reported high
in vitro digestibilities (leaves: 66.8 percent; stems: 77.2 percent).
The tetraploid limpograss stem digestibility was the highest of all 25
summer grasses studied.
The above findings encouraged Quesenberry and Ocumpaugh (1977) to
initiate stockpiling experimentation. They reported the beginning of
August as the proper staging period to initiate regrowth of adequate
yields of standing forage. Redalta produced 10.8-11.9 m tons DM but
declined rapidly in quality, whereas Bigalta produced 6.2 m tons and
maintained a 45 percent IVOMD when measured the following March. The
rate of IVOMD decline was similar, but the intercept started 13 units
higher in Bigalta than for Redalta and Greenalta (Quesenberry and
Ocumpaugh, 1980).
Ruelke's management studies using limpograsses (Ruelke, 1978;
Quesenberry et al., 1978) showed that close, frequent defoliation and
high nitrogen fertilization led to the loss of Bigalta stands. Mob
grazing and clipping trials by Quesenberry and Ocumpaugh (1979) also
showed Bigalta's poor persistence. Quesenberry et al. (1981) identified
PI 364888 as having superior persistence and comparable quality to
Bigalta. Ocumpaugh et al. (1981) identified another promising limpo
grass (PI 349753) that persisted under mob grazing but that was of a
slightly lower acceptability by animals than PI 364888 or Bigalta.
Total nonstructural carbohydrate (TNC) analysis has for decades
been a useful technique in quantifying the fluctuation of organic


74
reserves in response to season (McCarty, 1935; Waite and Boyd, 1953a),
defoliation (Sullivan and Sprague, 1943; May, 1960; Baker and Garwood,
1961; Marshall and Sagar, 1965), nitrogen fertilization (Adegbola and
McKell, 1966a; Rese and Decker, 1966; Ford and Williams, 1973), and
persistence (Graber, 1931; Alberda, 1966). The efforts of Dale Smith
(1981a) and others in understanding the carbohydrate metabolism in
alfalfa (Medicago sativa L.) is a prime example of the TNC studies
yielding practical management recommendations. According to Graber
(1931) "under field conditions, the limitations of root growth and the
modifications of the internal environment resulting from low reserves
may reduce the absorptive capacity of a plant so greatly and may so
increase its susceptibility to drought, winter injury, weed encroach
ments, insect injury, and other hazards as to jeopardize its permanence"
(p. 47).
Various reports have shown that high rates of nitrogen fertilizer
have led to reduced persistence in digitgrass (Creel, 1957; Ruelke,
1960; Kien et al., 1975) and in orchardgrass (Dactylis glomerata L.)
(Alexander and McCloud, 1962), but no carbohydrate analyses were
presented. No chemical data relating TNC to persistence in limpograss
could be located. Christiansen (Chapter 1) used regrowth-in-darkness
studies as preliminary tests of reserve energy in limpograss. In
Chapter 2 of this report the site of nonstructural carbohydrate accumu
lation was determined, as well as clipping effects on TNC percent of
limpograss in 1979.
Redalta has poor quality but excellent persistence, while Bigalta
has the reverse situation; neither combination being advantageous to the
producer. The objective of this study was to thoroughly evaluate two


75
promising introductions against Bigaita and Redalta. The goal was to
obtain a persistent, high quality, high yielding limpograss for flatwood
sites in Florida and the southeast.
Materials and Methods
Two experimental areas were used in this study. The 1979 experi
mentation was conducted on four well established plantings of limpograss
on a Wachula sand, a poorly drained siliceous hyperthermic ultic
haploquod soil at the Beef Research Unit of the University of Florida.
Another field was established in 1979 by vegetative propagation from
material taken from the 1979 experimental area. The new planting was in
an Adamsville sand. The limpograsses used were PI 349753, PI 364888,
Redalta, and Bigalta. Redalta is diploid (2n = 18) and the other
grasses used here are tetraploids (2n = 36).
1979 Establishment
The experimental area was 30 x 118 m and supported a mature stand
of rye (Secale cereale L.) prior to cultivating with a Ground Hawg
rototiller on 22 May 1979. The field was blocked into three sections
and within each section four main plots (genotypes) were marked to
measure 7 x 30 m with 3 m alleys between each main plot. The alleys
were planted to 'Argentine' bahiagrass.
Waist high stands of each limpograss growing in the 1979 experi
mental area were mowed with a sickle bar set at 7.5 cm height, raked,
and carried to an appropriately assigned, random location within each
block. The herbage was evenly distributed over the soil, lightly disked
to 15 cm, and the entire field was rolled using a cultipacker seeder to


76
assure good soil to stem contact. The field was irrigated when neces
sary to insure establishment and fertilized on 29 June and 23 August
1979 with 336 kg/ha of 17-5-10 (N^O^-FOjO) containing 3.4 kg/ha of a
micronutrient mix. Soil test results from 7 June 1979 showed a pH of
6.2, 20.3 kg/ha phosphorus (P) and 40 kg/ha potassium (K). The field
was flail chopped on 6 December 1979 to 7.5 cm height to remove frosted
forage.
1979 Experimentation
Within each established limpograss sward in the 1979 experimental
area plots were marked out for a clipping study. The blocks were
unreplicated; hence, the design was a split plot design without replica
tion of main plots. The clipping treatments within a limpograss block
were randomly assigned and replicated three times. The genotype *
treatment (rep) term was used to test genotype, clipping treatment, and
interaction effects for the responses that were studied.
The clipping experiment was initiated by mowing all plots to 5 cm
on 27 July 1979 and harvesting at 2.5, 5 and 10 week intervals until
4 October. The limpograsses were fertilized on 28 May and 30 July 1979
with 280 kg/ha of 17-5-10 (N^O^-I^O) containing 1 percent of a micro
nutrient mix. Soil test results from samples taken on 3 December 1979
showed a pH of 6.6, 28.2 kg/ha P, and 22.5 kg/ha K.
The 2.5 week interval treatments were clipped at 5 cm height on
13 August, 30 August, 17 September, and 4 October 1979 using a Jari
sickle bar mower. Yields were determined and "shoot" samples (tissue
above cutting height) were taken for IV0MD and crude protein (CP).
The 5 week interval treatments were harvested on 30 August and


77
4 October 1979. The 10 week treatment was harvested and sampled for
IVOMD, yield, and CP only on 4 October 1979.
1980 Experimentation
Soil test results from the newly established field of limpograss
revealed a pH of 6.1, 16.4 kg/ha P, and 53.1 kg/ha K. On 23 March 1980
the field was uniformly mowed to 7.5 cm height and treatments consisting
of five levels of nitrogen (N) fertilization and five cutting frequen
cies (F) were imposed on the main plots. The two factors (N and F) were
t h s
combined to represent 13/25 of a 5 x 5 complete factorial. Figure 3
shows a 3 x 3 factoral and the spatial arrangement of two factors, and
Figure 4 shows a central composite design in three factors. The 13
treatment combinations used in this study are shown graphically in
Figure 5 and the design points are plotted as grid coordinates in
Figure 6. The arrangement of treatment combinations in Figure 6 shows
the underlying difference between the classical central composite design
(Figure 3) and the modified central composite design described by
Littell and Mott (1975). Note in Figure 6 how information is concen
trated in what was considered the "management realm" (0-120 kg/ha N) and
3 to 9 week defoliation frequencies. Also note the geometric scaling of
N and the arithmetric progression for cutting frequency (with a skip at
the 15 week level). As the dotted lines suggest, the design could be
2 2
considered a superimposition of two complete factorials (2 and 3 ).
Consequently, the grasses were replicated as main plots in a split
plot design, and the 13 treatments provided a response surface with all
points replicated three times. Each main plot contained 13 plots


78
Figure 3. Spatial arrangement of a three level factorial design
in two variables
Figure 4. Spatial arrangement of a central composite desian in
three variables


79
TREATMENT MATRIX
480

0

|~23¡

N 240
0

0
25
FERTILIZATION
(kg/ha/yr) *20

0

|]

60
0

0

0

0

0

3
6
9
12
18
CUTTING FREQUENCY
(wks)
Figure 5. Treatment matrix showing treatment numbers inside
each actual (circles) and absent (squares) treat
ment combination
N
FERTILIZATION
(kg/ha/yr)
ACTUAL ORIENTATION
OF FACTOR SPACE
(wks)
Figure 6. Treatment combinations of nitrogen (N) and cutting
frequency plotted as grid coordinates with the origin
at 120 kg/ha/yr N and 9 weeks cutting frequency


80
(2.3 x 7 m) for treatments. The entire field had 156 total plots (3
replications x 4 grasses x 13 treatments).
Figure 7 shows the treatment levels of N and F, the treatment
number, and the harvest (FI) number, date, and chronological stage of the
1980 experiment. Sampling for carbohydrate analyses followed visual
estimations of percent limpograss but preceded sampling for herbage
quality and clipping for yield. Fertilization followed clipping, and
the N rates were applied by hand as ammonium nitrate dispersed in sand
to ensure uniform application. The N application was split according to
the number of defoliations; an equal rate applied at the start of the
experiment and after every defoliation except the last. The entire
field was fertilized with 300 kg/ha 0-10-20 (N^O^-I^O) containing
microelements prior to the start of the experiment and at the halfway
point late in July.
The sampling schedule is shown by symbols in Figure 7 for IV0MD,
CP (CP = 6.25 x N), DM yield determination, and TNC measurement. Forage
cut for yield and quality was oven dried at 60 C. The dried forage was
ground in a Wiley mill to pass a 1 mm screen and analyzed for percent N
by the Kjeldahl procedure. The IV0MD determination used a modification
of the Tilley and Terry technique (Moore et al., 1972). Total nonstruc-
tural carbohydrate (TNC) samples were randomly selected, dug, washed
free of soil, and arranged in cotton bags. The cotton bags were then
packed in a plastic bag and put on ice. Upon reaching the lab, the
samples were dried in a Thelco forced-air oven at 100 C for 30-45
minutes after which the temperature was lowered to 70 C until the dried
samples were removed 36-48 hours later (Smith, 1973a). The stem bases
were scraped free of roots, leaf sheaths, and sand by using a wire


YIELD AND QUALITY A AND TNC k SAMPLING IN 1980
Trt.
No.
N F
(kg/ha/yr) (wks)
Reps
1
0
3
3
2
120
3
3
3
240
3
3
4
60
6
3
5
240
6
3
6
0
9
3
7
120
9
3
8
480
9
3
9
60
12
3
10
240
12
3
II
0
18
3
12
120
18
3
13
480
18
3
HARVEST,
DATE,
AND WEEK
HO HI H2 H3 H4 H5 H6 H7 H8 H9 HIO Hll HI2
23 16 3 25 14 7 27 16 6 28 18 8 30
Mar Apr May May Jun Jul Jul Aug Sep Sep Oct Nov Nov
0 3 6 9 12 15 18 21 24 27 30 33 36
AAAAAAUAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
AAAAAAAAAAAAA
Figure 7. Yield, quality (IVOMD and CP), and total nonstructural carbohydrate (TNC)
sampling schedule for 1980 (open symbols indicate no sample)


82
buffing wheel attached to the extended shaft of a small electric motor
(20 W, 1525 RPM). The bottom 2 cm of the stem ("crowns") were severed
and ground through a 0.5 mm screen in a UDY Cyclone mill and stored
until tested for TNC. "Shoot" tissue was analyzed for TNC on the 27
July and 30 November dates. "Shoot" material was also ground through
the UDY mill, and all samples were stored in plastic 20 ml Dilu-vials
and then analyzed for TNC using an enzymatic hydrolysis and spectro-
photometric measurement of reducing sugars as described in Appendix A.
1981 Experimentation
The 1980 experimental area was studied in 1981 for residual effects
of cutting frequency in 1980 on 1981 DM yields. The limpograsses were
harvested for three 9 week cutting intervals on 3 June, 3 August, and
7 October 1981. No fertilizer was applied. Soil test results from
cores taken on 26 February 1981 indicated a pH of 6.2, 14.2 kg/ha P, and
46.5 kg/ha K.
Computing
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Department of Agricultural Engineering
and the Northeast Regional Data Center of the State University System of
Florida, located on the campus of the University of Florida in
Gainesville.
Means were analyzed using the Waller/Duncan Multiple Comparison
Test. Analysis of variance and regression were used to test response


83
surface models for adequate fit. A complete explanation of the SAS
methodology for constructing surface plots is given in Chapter 4.
Results
Dry Matter Yields: 1979 Experimentation
Table 5 shows the DM yields obtained for limpograsses cut and
weighed four times, twice, and once for a 10 week period beginning 27
July 1979 and ending 4 October 1979. Within limpograss genotypes each
treatment had a significantly different DM production. Among grasses,
no differences were found for combined yields from plots cut every 2.5
weeks, but if grasses were allowed to grow for two cycles of 5 weeks,
Bigalta and PI 364888 had higher DM accumulation, as was seen again for
the 10 week treatment yield.
Dry Matter Yields: 1980 Experimentation
The total limpograss DM yields in the 1980 experimentation were
summed and modelled to construct response surface plots in Figure 8.
For total DM yields in 1980 at an actual treatment combination, con
sult Table 6. Figure 8 and Table 6 both show DM yields for PI 364888 >
PI 349753 > Bigalta > Redalta for the 480*18 (N*F) treatment. The sur
face plots show that Redalta and Bigalta plateaued in the surface region
focused at the 480*18 treatment combination. These two limpograsses
lodged at this combination of N and F, and it is believed that their
decumbent habit caused some loss of DM.
Varying DM yield advantages were shown for limpograsses under dif
ferent treatment regimes than 480*18. Bigalta, for instance, had a


84
Table 5. Total dry matter (DM) yields for four limpograsses
clipped at three different frequencies for 10 weeks
ending on 4 October 1979
Limpograss
Clipping frequency
(wks)
2.5 5 10
DM (kg/ha)
PI 364888
1756 a
3261 ab
8327 a
'Redalta'
1718 a
2328 c
4592 d
'Bigaita1
1903 a
3988 a
6616 b
PI 349753
1723 a
3133 b
5425 c

Clipping treatments within a limpograss genotype are
compared using the underlining technique and letters within a
column compare limpograsses for a treatment. Any values
sharing a commun underline or letter are not different (P <
0.05) using the Wal1er-Duncan Multiple Comparison Test.


TOTAL DM PRODUCTION FOR FOUR LIMPOGRASSES IN 1980
PI 364888 'Redalta' 'Bigalta' PI 349753
Figure 8. Total dry matter (DM) yield for four limpograsses subjected to five levels of nitrogen
(N) fertilization and five frequencies (F) of defoliation throughout the 1980 growing
season at the Beef Research Unit near Gainesville, Florida


Table 6. Effect of nitrogen (N) rates and cutting frequency (F) on dry matter (DM) yield, in vitro
organic matter digestibility (IVOMD), crude protein (CP), harvest of protein, andTertTTYzer
N efficiency
Fert. N
(kg/ha)
Defoliation
Llmpograss
Total
DM yield
(m ton/ha/yr)
DM yield
added by N
(m ton/ha)
Harvest
of protein
(kg/ha/yr)
N
harvested
applied N
(X)
27 July
1980
30 November 1980
Stand
W
Freq
F
(wks)
Total
no.
ca Total
DM yield
(m ton/ha)
IVOMD
(*)
CP
(%)
DM yield
(m ton/ha)
IVOMD
(%)
CP
(X)
0 0
3
12
PI 364888
2.7
171
0.5
49.9
7.5
0
51 .9
6.6
92
Redalta
3.3

221

0.7
41.5
7.4
0
40.0
5.9
92
Bigalta
2.7

219

0.5
58.3
8.8
0
65.6
10.6
62
PI 349753
3.1
---
215
---
0.3
47.0
9.1
0
52.6
7.5
92
10 120
3
12
PI 364888
4.4
1.7
361
26
0.6
50.7
9.0
0
56.8
9.8
82
Redalta
4.7
1.4
357
18
1.0
39.8
7.6
0
42.1
8.2
90
Bigalta
3.1
0.4
281
8
0 4
58.2
9.2
0
69.5
12.2
50
PI 349753
2.9
0
225
2
0.4
45.0
7.3
0
57.5
10.3
83
40 480
3
12
PI 364888
6.8
4.1
708
18
0.5
51.8
9.7
0
61.8
12.7
78
Redalta
7.6
4.3
744
18
0.8
44.3
9.8
0
47.9
10.4
88
Bigalta
5.6
2.9
666
15
0.1
63.7
13.2
0
72.0
14.6
23
PI 349753
6.1
3.1
643
14
0.4
45.2
9.7
0
61.5
12.5
68
10 60
6
6
PI 364888
7.1
0
384
47
3.5
49.8
4.4
0.2
56.0
7.6
95
Redalta
7.4
1.8
413
37
2.6
40.8
5.1
0.2
44.1
8.3
97
Bigalta
7.4
2.2
448
32
3.7
51.1
4.8
0.1
64.2
9.3
68
PI 349753
6.4
0
371
27
2.6
43.3
5.3
0.2
54.6
8.3
97
60 240
6
6
PI 364888
11.7
4.5
731
35
4.6
48.3
4.7
0.4
60.0
10.3
100
Redalta
8.2
2.6
557
19
3.4
39.4
6.0
0.1
69.4
10.3
82
Bigalta
12.9
7.7
921
39
4.8
55.1
5.3
0.1
69.9
13.4
70
PI 349753
10.7
4.3
704
29
4.1
49.1
5.4
0.2
60.7
10.6
95
0 0
9
4
PI 364888
5.9

262
3.8
45.8
2.9
0.1
53.0
6.6
98
Redalta
6.5

329

2.5
41.6
4.8
0.2
45.9
7.7
98
Bigalta
9.1

448

5.0
52.2
3.0
0.2
64.3
8.6
92
PI 349753
6.8

324

4.0
45.8
3.7
0.1
49.9
6.3
98


Full Text

PAGE 1

(1(5*< 5(6(59(6 $1' $*52120,& &+$5$&7(5,67,&6 2) )285 /,032*5$66(6 +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf )25 )/25,'$n6 )/$7:22'6 %< 6&277 &+5,67,$16(1 $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( &281&,/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

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

PAGE 3

-LP DQG -DQHW 'HDQ IRU YLVLWDWLRQ KRXUV ZKHQ QHHGHG WKHP WKH PRVW DQG -DQHW (OGUHG IRU KHU LQWHOOLJHQW DQG WROHUDQW DSSURDFK WR WKH WDVN RI W\SLQJ WKLV GLVVHUWDWLRQ LLL

PAGE 4

7$%/( 2) &217(176 3DJH $&.12:/('*(0(176 LL /,67 2) 7$%/(6 YLL /,67 2) ),*85(6 L[ $%675$&7 [LL ,1752'8&7,21 /,7(5$785( 5(9,(: )OLVWRU\ 3RWHQWLDO /LPLWDWLRQV DQG 'HYHORSPHQW RI )ORULGDnV )RUDJH/LYHVWRFN ,QGXVWU\ 'HYHORSPHQW RI *UD]LQJ /DQG LQ )ORULGD +LVWRU\ DQG (YROXWLRQ RI )ORULGDnV *UD]LQJ 5HVHDUFK *UD]LQJ $QLPDO ,PSURYHPHQW )ORULGDn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

PAGE 5

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n +<'5$7( $1' 3(56,67(1&( ,1 7:2 3520,6,1* $0' 7:2 5(/($6(' /,032*5$66(6 ())(&76 '8( 72 1,752*(1 )(57,/,=$7,21 $0' &877,1* )5(48(1&< ,QWURGXFWLRQ 0DWHULDOV DQG 0HWKRGV Y

PAGE 6

3DJH (VWDEOLVKPHQW ([SHULPHQWDWLRQ ([SHULPHQWDWLRQ ([SHULPHQWDWLRQ &RPSXWLQJ 5HVXOWV 'U\ 0DWWHU
PAGE 7

/,67 2) 7$%/(6 7DEOH 3DJH $YHUDJH '0 \LHOGV RI HWLRODWHG VKRRWVn RI IRXU OLPSRJUDVVHV IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV &RPSDULVRQ RI HWLRODWHG VKRRW '0 \LHOGV DQG DVVRFLDWHG URRW DQG VWXEEOH FRPSRQHQWV IRU IRXU OLPSRJUDVVHV DYHUDJHG IRU DOO FXWWLQJ WUHDWn PHQWV IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV $YHUDJH '0 \LHOGV RI HWLRODWHG UHJURZWK IURP ([SHULPHQW IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV 7KH PDLQ HIIHFWV RI OLPSRJUDVV JHQRW\SH DQG FOLSn SLQJ WUHDWPHQW RQ WKH SHUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWHV 71&f LQ WKH ERWWRP FP RI VWHP EDVH FURZQf 7RWDO GU\ PDWWHU '0f \LHOGV IRU IRXU OLPSRn JUDVVHV FOLSSHG DW WKUHH GLIIHUHQW IUHTXHQFLHV IRU ZHHNV HQGLQJ RQ 2FWREHU (IIHFW RI QLWURJHQ 1f UDWHV DQG FXWWLQJ IUHTXHQF\ )f RQ GU\ PDWWHU '0f \LHOG LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f FUXGH SURWHLQ &3f KDUYHVW RI SURWHLQ DQG IHUWLOL]HU 1 HIILFLHQF\ 3HUFHQW FUXGH SURWHLQ &3f IRU IRXU OLPSRJUDVVHV FOLSSHG DW WKUHH GLIIHUHQW IUHTXHQFLHV IRU ZHHNV HQGLQJ RQ 2FWREHU 6DPSOHV WDNHQ RQ 2FWREHU ZHUH DQDO\]HG IRU &3f 7KH PDLQ HIIHFW RI OLPSRJUDVV DQG FOLSSLQJ WUHDWn PHQW RQ SHUFHQW LQ YLWUR RUJDQLF PDWWHU GLJHVWLn ELOLW\ ,920'f RI WLVVXH VDPSOHG 2FWREHU %DODQFH VKHHW RI LQ YLWUR RUJDQLF PDWWHU GLJHVWLn ELOLW\ ,920'f DQG WRWDO QRQVWUXFWXUDO FDUERn K\GUDWHV 71&f LQ WZR OLPSRJUDVVHV YL L

PAGE 8

7DEOH 3DJH 7UHDWPHQW FRPELQDWLRQV LQ 3, UHSUHVHQWLQJ WKH EHVW FRPSURPLVH EHWZHHQ \LHOG DQG TXDOLW\ LQ WKH H[SHULPHQW $ 6$6 SURJUDP XVLQJ D SUHYLRXVO\ ILWWHG UHJUHVVLRQ WR REWDLQ YDOXHV IRU WKH GHSHQGHQW YDULDEOH WR ILOO WKH KROHV LQ WKH WUHDWPHQW PDWUL[ LQ RUGHU WR PHUJH ZLWK DFWXDO GDWD DQG XVH 352& *' $ 6$6 DQDO\VLV RI YDULDQFH SURFHGXUH IRU D PRGHO WKDW TXDQWLILHV WKH PD[LPXP WUHDWPHQW 757f VXP RI VTXDUHV 66f IRU WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ KDUYHVW RI WKH H[SHULPHQW $ FXELF 6$6 JHQHUDO OLQHDU PRGHO WKDW IUDFWLRQDWHV WKH WUHDWPHQW VXP RI VTXDUHV 66f LQWR 66 H[SODLQn DEOH E\ QLWURJHQ 1f IHUWLOL]DWLRQ DQG IUHTXHQF\ )f RI GHIROLDWLRQ IRU WRWDO QRQVWUXFWXUDO FDUERn K\GUDWH 71&f LQ KDUYHVW RI WKH H[SHULPHQW $Q )WHVW RI PRGHOV IRU GHWHUPLQLQJ D VWDWLVWLFDOO\ VLJQLILFDQW UHVSRQVH ODFN RI ILW /)f DQG DQ HOLPLQDWLRQ PHWKRG IRU QRQVLJQLILFDQW WHUPV 7KH 6$6 JHQHUDO OLQHDU PRGHOV SURFHGXUH VROXWLRQ IRU HVWLPDWHV RI LQWHUFHSW DGMXVWHG IRU WKH OLPSR JUDVV OLQHf DQG UHS SDUDPHWHUV 7KH 6$6 JHQHUDO OLQHDU PRGHOV SURFHGXUH VROXWLRQV IRU HVWLPDWHV RI QLWURJHQ 1f DQG FXWWLQJ IUHTXHQF\ )f WUHDWPHQW SDUDPHWHUV DGMXVWHG IRU HDFK OLPSR JUDVV OLQHf $O 6$6 MRE DV VXEPLWWHG RQ FDUGV WR WKH FRPSXWHU $ 6WDWLVWLFDO DQDO\VLV V\VWHP $ 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f PHDQV IRU DOO WUHDWPHQW GDWHV DQG OLPSRJUDVVHV LQ Y L L L

PAGE 9

/,67 2) ),*85(6 )LJXUH 3DJH 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQW RI GU\ PDWWHU RI IRXU SODQW SDUWV LQ IRXU OLPSRJUDVVHV VDPSOHG DW WZR PDWXULWLHV 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQWDJHV LQ WKH FURZQ FRUUHODWHG DJDLQVW VWXEEOH ZHLJKWV OHIWf DQG HWLRODWHG UHJURZWK SHUFHQW RI ZKROH SODQWV ULJKWf IRU IRXU OLPSRJUDVV JHQRW\SHV VXEn MHFWHG WR WKUHH FOLSSLQJ IUHTXHQFLHV 6SDWLDO DUUDQJHPHQW RI D WKUHH OHYHO IDFWRULDO GHVLJQ LQ WZR YDULDEOHV 6SDWLDO DUUDQJHPHQW RI D FHQWUDO FRPSRVLWH GHVLJQ LQ WKUHH YDULDEOHV 7UHDWPHQW PDWUL[ VKRZLQJ WUHDWPHQW QXPEHUV LQVLGH HDFK DFWXDO FLUFOHVf DQG DEVHQW VTXDUHVf WUHDWn PHQW FRPELQDWLRQ 7UHDWPHQW FRPELQDWLRQV RI QLWURJHQ 1f DQG FXWWLQJ IUHTXHQF\ SORWWHG DV JULG FRRUGLQDWHV ZLWK WKH RULJLQ DW NJKD\U 1 DQG ZHHNV FXWWLQJ IUHTXHQF\
PAGE 10

)LJXUH 3DJH 5HVLGXDO FOLSSLQJ DQG IHUWLOL]DWLRQ HIIHFWV RQ GU\ PDWWHU '0f \LHOG IRU 3, KDUYHVWHG HYHU\ ZHHNV LQ -XQH $XJXVW DQG 2FWREHU QRW VKRZQff DQG WKH WRWDO SURGXFWLRQ IRU WKH \HDU &RPSDULVRQ RI FUXGH SURWHLQ &3f SHUFHQWDJHV LQ WKH WLVVXH RI 3, DQG n5HGDOWDn IRU WZR GDWHV LQ LQ UHVSRQVH WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI FOLSSLQJ &RPSDULVRQ RI LQ YLWUR RUJDQLF PDWWHU GLJHVWLn ELOLW\ ,920'f IRU IRXU OLPSRJUDVVHV DW WZR KDUYHVW GDWHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI FOLSSLQJ 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VKRRW WLVVXH RI WKUHH OLPSRJUDVVHV RQ -XO\ VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH GHIROLDWLRQ IUHTXHQFLHV )f $ FRPSDULVRQ RI WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VKRRWV RI WKUHH OLPSRJUDVVHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOLn ]DWLRQ DQG ILYH IUHTXHQFLHV )f RI GHIROLDWLRQ VDPSOHG RQ 1RYHPEHU 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VWHP EDVH FURZQVf RI WZR OLPSRJUDVVHV RQ WZR GDWHV DV DIIHFWHG E\ QLWURJHQ 1f IHUWLOL]Dn WLRQ DQG GHIROLDWLRQ IUHTXHQF\ )f $ FRPSDULVRQ RI WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQW LQ WKH FURZQV RI WZR OLPSRJUDVVHV VXEMHFWHG WR ZHHN GHIROLDWLRQ IUHTXHQFLHV )f DQG IHUWLOL]HG DW WKUHH OHYHOV RI QLWURJHQ 1f GXULQJ WKH JURZLQJ VHDVRQ 9LVXDO HVWLPDWLRQV RI SHUFHQW %LJDOWDf DQG 3, IRU WUHDWPHQWV KDYLQJ ZHHN GHIROLDn WLRQ IUHTXHQFLHV )f DQG WKUHH OHYHOV RI QLWURJHQ 1f LQ JURZLQJ VHDVRQ 5DLQIDOO DQG WHPSHUDWXUH GDWD IRU WKH JURZn LQJ VHDVRQ WDNHQ DW WKH %HHI 5HVHDUFK 8QLW QHDU *DLQHVYLOOH )ORULGD [

PAGE 11

)LJXUH 3DJH 9LVXDO HVWLPDWHV RI IRXU OLPSRJUDVVHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI FOLSSLQJ DFURVV WKUHH VXPPHU KDUYHVWV $Q H[DPSOH RI WKH UHVSRQVH VXUIDFH SORWWHG XVLQJ 6$6*5$3+ FRPSXWHU DVVLVWDQFH $O 'LOXWLRQ RQH $ 'LOXWLRQ WZR $ 7KH UHJUHVVLRQ RI RSWLFDO GHQVLW\ 'f YHUVXV JOXFRVH [L

PAGE 12

$EVWUDFW RI 'LVVHUWDWLRQ 3UHVHQWHG WR WKH *UDGXDWH &RXQFLO RI WKH 8QLYHUVLW\ RI )ORULGD LQ 3DUWLDO )XOILOOPHQW RI WKH 5HTXLUHPHQWV IRU WKH 'HJUHH RI 'RFWRU RI 3KLORVRSK\ (1(5*< 5(6(59(6 $1' $*52120,& &+$5$&7(5,67,&6 2) )285 /,032*5$66(6 +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf )25 )/25,'$n6 )/$7:22'6 %\ 6FRWW &KULVWLDQVHQ 'HFHPEHU &KDLUPDQ &KDUOHV 5XHONH 0DMRU 'HSDUWPHQW $JURQRP\ +LJKHU TXDOLW\ IRUDJHV DUH QHHGHG IRU WKH IRXU PLOOLRQ KHFWDUHV RI IODWZRRGV WKDW DUH WKH EDVH RI WKH IRUDJHOLYHVWRFN LQGXVWU\ LQ )ORULGD 7KH LPSURYHG OLPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf FXOWLYDUV n%LJDOWDn DQG n5HGDOWDn ZHUH FRPSDUHG ZLWK 3, DQG 3, WR ILQG EHWWHU \LHOG TXDOLW\ DQG SHUVLVWHQFH DV ZHOO DV WR LQYHVWLJDWH PRUSKRORJLFDO GLIIHUHQFHV DQG WRWDO QRQVWUXFWXUDO FDUERn K\GUDWH 71&f SK\VLRORJ\ DV WKH\ UHODWH WR SHUVLVWHQFH 3UHOLPLQDU\ ZRUN LQ XVHG UHJURZWKLQGDUNQHVV VWXGLHV WKDW TXDQWLILHG WKH PRUSKRORJLFDO GLIIHUHQFHV LQ WKH IRXU LPSRJUDVVHV /DUJHU VWXEEOH V\VWHPV ZHUH SRVLWLYHO\ FRUUHODWHG WR 71& SHUFHQWDJHV LQ WKH VWHP EDVHV )UHTXHQW GHIROLDWLRQ WUHDWPHQWV LQ WKH ILHOG SULRU WR WKH GDUNQHVV VWXG\ FDXVHG D VKLIW LQ KDELW IURP XSULJKW WR SURVWUDWH JURZWK WKDW SHUPLWWHG LQFUHDVHV LQ D[LOODU\ WLOOHU IRUPDWLRQ ZLWK FRQn FRPLWDQW LQFUHDVHV LQ 71& DFFXPXODWLRQ DV PHDVXUHG E\ FKHPLFDO PHDQV ;7

PAGE 13

$ ILHOG H[SHULPHQW ZDV WKHQ LQLWLDWHG ZLWK QLWURJHQ 1f IHUWLOL]Dn WLRQ DGGHG DV D QHZ YDULDEOH DW ILYH OHYHOV ZKLFK ZDV FRPELQHG ZLWK ILYH GHIROLDWLRQ IUHTXHQFLHV XVLQJ D UHVSRQVH VXUIDFH GHVLJQ 'U\ PDWWHU '0f \LHOGV LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f FUXGH SURWHLQ &3f DQG SHUVLVWHQFH ZHUH PHDVXUHG 7KH EHVW \LHOG RI DOO WKH JHQRW\SHV ZDV SURGXFHG E\ 3, P WRQKD\Uf ZLWK WZR KDUYHVWV RQ -XO\ DQG 1RYHPEHU UHSUHVHQWn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n VLVWHQFH RYHU %LJDOWD DQ LPSURYHPHQW LQ \LHOG DQG TXDOLW\ RYHU 5HGDOWD DQG LV VFKHGXOHG IRU FXOWLYDU UHOHDVH LQ ;7

PAGE 14

,1752'8&7,21 /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW & ( +XEEf LV D VWRORQLIHURXV JUDVV WKDW KDV DGDSWDELOLW\ DV DQ LPSURYHG IRUDJH IRU IODWZRRGV LQ )ORULGD DQG WKH VRXWKHDVW 2I WKH FXOWLYDUV QRZ UHOHDVHG n%LJDOWDn KDV KLJK GLJHVWLELOLW\ DQG ORZ SHUVLVWHQFH ZKLOH 5HGD WDn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f FRQWHQW RI D SODQW LV QHFHVVDU\ WR XQGHUVWDQG KRZ D SODQW JURZV 6WXGLHV RI 71& PRUSKRORJ\ DQG REVHUYDWLRQV RI SODQW UHDFWLRQV WR PDQDJHPHQW FDQ \LHOG SUDFWLFDO UHFRPPHQGDWLRQV WR EH SDVVHG WR DJULFXOWXUDO SURGXFHUV 7KLV VWXG\ ZDV XQGHUWDNHQ WR WKRURXJKO\ H[SORUH WKH IOH[LELOLW\ RI OLPSRJUDVV XQGHU D ZLGH UDQJH RI PDQDJHPHQW SUDFWLFHV WR FOHDUO\ LGHQWLI\ LWV VWUHQJWKV DQG ZHDNQHVVHV DV D IRUDJH SODQW IRU WKH IODWZRRGV

PAGE 15

/,7(5$785( 5(9,(: +LVWRU\ 3RWHQWLDO /LPLWDWLRQV DQG 'HYHORSPHQW RI )ORULGDnV )RUDJH/LYHVWRFN ,QGXVWU\ ,Q 3RQFH GH /HRQ VHW RXW WR ILQG WKH ODQG ZLWK WKH PDJLF ZDWHUV RI ,QGLDQ ORUH DQG GLVFRYHUHG WKH PDLQODQG QHDU 6W $XJXVWLQH +H QDPHG WKH SODFH )ORULGD 2QH RI WKH HDUO\ YLFWLPV RI WKH UHSXWHG ZHDOWK RI )ORULGD ZDV +HUQDQGR GH 6RWR ZKR KDG ZKHWWHG KLV DSSHWLWH IRU JROG ZKLOH DFFRPSDQ\n LQJ 3L]DUUR WR 3HUX +H RXWILWWHG DQ H[SHGLWLRQ DQG ODQGHG LQ )ORULGD LQ 7KH QH[W IRXU \HDUV KH VSHQW PDUFKLQJ WKURXJK WKH VRXWKHUQ VWDWHV VHHNLQJ JROG XQWLO LQ KH FDPH WR D YDVW ULYHU QRZ NQRZQ DV WKH 0LVVLVVLSSL $V WKH SDUW\ UHWXUQHG IURP WKH SUHVHQW VWDWHV RI $UNDQVDV DQG /RXLVLDQD GH 6RWR GLHG $OWKRXJK KH GLVVLSDWHG KLV IRUWXQH LQ D YDLQ TXHVW IRU ULFKHV KH DFFRPSOLVKHG D IDU ZRUWKLHU UHVXOW LQ RSHQLQJ D YDVW WHUULWRU\ IRU WKRVH ZKR IROORZHG &ROOLQV f 7KH ,QGLDQ LQKDELWDQWV RI )ORULGD ZHUH KXQWHUV 2I WKH IRXU PDMRU JURXSV WKH &DOXVD RFFXSLHG WKH ORZHU ZHVW FRDVW WKH 7HVHVWD LQKDELWHG WKH ORZHU HDVW FRDVW WKH 7LPXFXD ZHUH ORFDWHG LQ WKH QRUWK DQG WKH $SDODFKHZHVW RI WKH $XFLOOD 5LYHU )LUH ZDV XVHG WR FKDVH RXW JDPH DQG KHQFH WKHVH SHRSOH ZHUH UHVSRQVLEOH IRU WKH FUHDWLRQ RI ODUJH H[SDQVHV RI JUDVVODQGV +XQWHU HW DO f

PAGE 16

7KH 6SDQLDUGV VRRQ EHFDPH SUDJPDWLF DQG UHFRJQL]HG )ORULGDnV QDWXUDO UHVRXUFHV DIWHU WKH TXHVW IRU JROG IL]]OHG /DUJH H[SDQVHV RI WKH ODQGVFDSH ZHUH JUD]HG DIWHU 3RQFH GH /HRQ ODQGHG FDWWOH RQ WKH ZHVWHUQ FRDVW RI )ORULGD LQ 7KH 6SDQLVK SHULRG ODVWHG XQWLO WKH HDUO\ n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nV ILUH VPRNH ZDV DV PXFK D SDUW RI WKH )ORULGD VFHQH DV SLQH WUHHV DQG WKXQGHUVWRUPV +XQWHU HW DO f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

PAGE 17

UDQFKHUV WR IRUP D PHQWDO LPDJH RI WKH PLQHUDO SUREOHPV RQ WKHLU ODQG 7KH FDWWOHPHQ ZHUH DZDUH RI WKH QXWULHQW ULFK DUHDV RQ WKHLU RSHQ UDQJH WKH DUHDV ZHUH UHIHUUHG WR DV KRVSLWDO IDUPV ZKHUH D QXWULHQW GHILn FLHQW DQLPDO FRXOG JUD]H DQG FRQYDOHVFH EHIRUH EHLQJ UHWXUQHG WR WKH KHUG %HFNHU SHUVRQDO FRPPXQLFDWLRQf ,Q EHIRUH PLQHUDO VXSSOHPHQWV ZHUH JHQHUDOO\ XVHG D VWXG\ RI FDWWOH RQ KHFWDUHV RI UDQJH ZDV FRQGXFWHG LQ FHQWUDO )ORULGD 2Q IODWZRRGV WKH FDOI FURS DYHUDJHG SHUFHQW DQG SHUFHQW RI WKH QHZERUQ FDOYHV GLHG SULRU WR PRQWKV RI DJH 7KH PDUNHW YDOXH ZDV SHU EUHHGLQJ FRZ SHU \HDU RU SHU KHFWDUH %HFNHU SHUVRQDO FRPPXQLFDWLRQ +HQGHUVRQ f ,PSURYHG IRUDJHV DQG EHWWHU PDQDJHPHQW KDYH KHOSHG WR DOOHYLDWH PDQ\ QXWULWLRQDO GHILFLHQFLHV 6XSSOHPHQWDWLRQ ZLWK YLWDPLQV PLQHUDOV SURWHLQ DQG HQHUJ\ KDV EHHQ DQ LPSRUWDQW DVSHFW RI PDQDJHPHQW 0F'RZHOO HW DO f VXUYH\HG PLQHUDO VWDWXV RI EHHI KHUGV RQ IRXU VRLO W\SHV LQ )ORULGD DQG FRQFOXGHG WKDW PLQHUDO GHILFLHQFLHV ZHUH DUHD VSHFLILF 7KH\ LQGLFDWHG WKDW SKRVSKRUXV VHOHQLXP DQG ]LQF GHILFLHQn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

PAGE 18

SURYLGH SHUFHQW RI WKH GDLO\ UHTXLUHPHQWV LQ RXQFHV RI WKH PL[WXUH $PPHUPDQ f +LVWRU\ DQG (YROXWLRQ RI )ORULGDn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f 7KH ILUVW JUD]LQJ WULDOV LQ )ORULGD ZHUH FRQGXFWHG LQ IRU WKH SXUSRVH RI HYDOXDWLQJ IRXU SHUHQQLDO VXPPHU JUDVVHV IRU XVH RQ ZHOO GUDLQHG VDQGV 7KH JUDVVHV XVHG DQG WKH DQQXDO EHHI JDLQV LQ NJKD ZHUH DV IROORZV FDUSHWJUDVV $[RQRSXV DIILQLV &KDVHf FRPPRQ EHUPXGD JUDVV &\QRGRQ GDFW\O RQ /f 3HUVf FRPPRQ EDKLDJUDVV 3DVSDOXUQ QRWDWXP )OJJHf DQG FHQWLSHGHJUDVV (UHPRFKORD RSKLXURLGHV 0XQURf +DFNf 7KH SRSXODULW\ RI FDUSHWJUDVV DPRQJ FDWWOHPHQ LQ WKH IODW ZRRGV DUHD DQG GLVFRYHU\ LQ RI WKH UHTXLUHPHQWV IRU VXFFHVVIXO SURn GXFWLRQ RI FORYHUV OHG WR JUD]LQJ WULDOV GXULQJ FRPSDULQJ FDUSHWJUDVV DORQH XQIHUWLOL]HG FDUSHWJUDVV DORQH IHUWLOL]HG FDUSHWJUDVVOHVSHGH]D /HVSHGH]D VWULDWD 7KXQEf + t $f DQG

PAGE 19

FDUSHWJUDVVZKLWH FORYHU 7ULIROLXP UHSHQV /f $YHUDJH DQQXDO JDLQV LQ NJKD ZHUH DV IROORZV FDUSHWJUDVV IHUWLOL]HG FDUSHWJUDVV FDUSHWJUDVVOHVSHGH]D DQG FDUSHWJUDVVZKLWH FORYHU 0HDQZKLOH VXSHULRU JUDVVHV EHFDPH DYDLODEOH WKURXJK SODQW LQWURGXFWLRQ DQG EUHHGLQJ ,Q JUD]LQJ WULDOV FRQGXFWHG EHWZHHQ WKUHH RI WKH QHZ JUDVVHVn3DQJOD GLJLWJUDVV 'LJLWDULD GHFXPEHQV 6WHQWf n3HQVDFRODn EDKLDJUDVV DQG n&RDVWDOn EHUPXGDJUDVVIHUWLL]HG ZLWK NJ RI DQQXDOO\ LQ WKH VSULQJ SURGXFHG \HDUO\ EHHI JDLQV DYHUDJLQJ SHUFHQW KLJKHU WKDQ JDLQV SURGXFHG E\ FDUSHWJUDVV LQ HDUOLHU JUD]LQJ WULDOV +HQGHUVRQ f 7KH LPSRUWDQFH RI IRUDJHV LQ )ORULGD LV REYLRXV )URP WR WKH IRUDJHOLYHVWRFN LQGXVWU\ JUHZ E\ SHUFHQW 2QHWKLUG RI WKH WRWDO UDQJH DQG SDVWXUH ODQG LQ )ORULGD LV QRZ SODQWHG WR LPSURYHG IRUDJHV ,Q PLOOLRQ NJ RI EHHI ZHUH SURGXFHG WRWDOLQJ PLOOLRQ LQ DJULFXOWXUDO LQFRPH %\ SURGXFWLRQ LV H[SHFWHG WR UHDFK PLOOLRQ NJ RI EHHI $JULFXOWXUH LQ DQ 8UEDQ $JH $*8$f 5HSRUW f *UD]LQJ $QLPDO ,PSURYHPHQW $GDPV f GHVFULEHV WKH UDQFKHUnV DWWLWXGH DQG JHQHUDO SHUFHSn WLRQ RI KLV ZRUOG EDFN LQ WKH ROG GD\V ,Q WKH nV ZH KDG D ORW RI WURXEOH UDLVLQJ FDWWOH LQ )ORULGD &DWWOH ZHUH FKHDS DQG ZLOG 3DVWXUHV ZHUH ODUJH DQG WKH VPDOO ERQHG FRZV FRXOG RXWUXQ VRPH RI WKH KRUVHV 6DOW VLFNQHVV ZDV SUHYDOHQW DQG ZH NQHZ OLWWOH DERXW FXULQJ DQLPDO GHILFLHQFLHV 7LFN IHYHU KDG EHHQ HUDGLFDWHG EXW WKH VFUHZ ZRUPV ZHUH HDWLQJ WKH FDWWOH DOLYH +DOI WKH \HDU FDWWOH ZHUH ERJJHG GRZQ WU\n LQJ WR ILQG D GULQN RI ZDWHU DQG WKH RWKHU KDOI WKH\ ZHUH VWDQGLQJ LQ ZDWHU XS WR WKHLU VLGHV 2XU PDLQ LQWHUHVW ZDV VXUYLYDO IRU WKH FDWWOH DQG RXUVHOYHV 7KHUH ZDV QR VRXUFH RI EUHHGLQJ VWRFN DYDLODEOH WKDW KDG WKH KHDW WROHUDQFH DQG TXDOLW\ WR PHHW WKH QHHGV RI D FRZPDQ $GDPV S f

PAGE 20

,Q WKH SDVW \HDUV WKH VHOHFWLRQ RI DQLPDOV KDV UHYROYHG DURXQG DGDSWDWLRQ WR WKH )ORULGD IRUDJHV DQG FOLPDWH %UHHGHUV KDYH VRXJKW WR FRPELQH WKH DGDSWDELOLW\ RI =HEX %UDKPDQ RU QDWLYH DQLPDOV ZLWK WKH SHUIRUPDQFH FDSDELOLWLHV RI (XURSHDQ VWRFN $FFRUGLQJ WR &URFNHWW f LW EHFDPH SRSXODU LQ WKH fV WR XVH VLUHV UHSUHVHQWLQJ EUHHGV GHYHORSHG LQ (XURSH WR LQFUHDVH EHHI SURn GXFWLRQ DQG WKLV OHG WR GLYHUVH EUHHGLQJ FRPELQDWLRQV LQ WKH VXEWURSLn FDO ]RQH RI WKH $WODQWLF FRDVWDO SODLQ DQG WKH *XOI FRDVW VWDWHV &URFNHWW IHHOV WKDW EUHHGLQJ SURJUHVV FDQ EH PDGH LQ VXEWURSLFDO HQYLURQPHQWV XVLQJ %UDKPDQGHULYDWLYH EUHHGV VXFK DV WKH %HHIPDVWHU %UDIRUG %UDQJXV RU 6DQWD *HUWUXGLV )LHOGV DQG +HQWJHV f VWDWH WKDW WKH JHQHWLF EDVH RI )ORULGD FDWWOH LV SULPDULO\ RI %UDKPDQ H[WUDFWLRQ 7KH FKDUDFWHULVWLFV QHHGHG LQ D FRPPHUFLDO )ORULGD KHUG DUH DV IROORZV f KLJK IHUWLOLW\ f WROHUDQFH WR WKH WHPSHUDWXUH DQG KXPLGLW\ f JRRG IRUDJLQJ DELOLW\ f JRRG PDWHUQDO DELOLW\ f VDWLVIDFWRU\ IHHGORW SHUIRUPDQFH DQG f VDWLVIDFWRU\ FDUFDVV TXDOLW\ .RJHU f )ORULGDnV 5ROH LQ WKH )RUDJH/LYHVWRFN ,QGXVWU\ :KDW PXVW H[DVSHUDWH )ORULGD SURGXFHUV LV WKDW WKHUH LV DQ HQRUPRXV DSSHWLWH IRU EHHI LQ WKH VWDWH ILOOHG E\ SURGXFHUV IURP RXWVLGH WKH VWDWH 2QO\ SHUFHQW RI WKH EHHI FRQVXPHG DQQXDOO\ E\ PLOOLRQ )ORULGD UHVLGHQWV DQG WKHLU YLVLWRUV LV SURGXFHG LQ )ORULGD 7KH UHPDLQGHU UHSUHVHQWV DERXW PLOOLRQ NJ RI EHHI VKLSSHG LQ E\ RWKHU PHDQV $SSUR[LPDWHO\ PLOOLRQ NJ RI WKH EHHI LQWURGXFHG WR )ORULGD LV WUXFNHG IURP WKH KLJK SODLQV DUHD RI WKH 8QLWHG 6WDWHV %DNHU f

PAGE 21

$ UHFHQW VXUYH\ E\ 6SUHHQ DQG 6KRQNZLOHU Df LGHQWLILHV 7H[DV .DQVDV 1HEUDVND DQG ,RZD DV WKH ODUJHVW VXSSOLHUV RI ILQLVKHG EHHI WR )ORULGD $ PRVW EDIIOLQJ IHDWXUH RI WKH )ORULGD EHHI PDUNHW LV WKDW FDWWOHn PHQ DUH VHOOLQJ FDOYHV WR VWRFNHUV WR NLORPHWHUV GLVWDQW ZKR WKHQ VHOO WR IHHGHUV ZKR VHOO WR SDFNLQJKRXVHV ZKR LQ WXUQ VSHQG FORVH WR PLOOLRQ DQQXDOO\ WR WUXFN ILQLVKHG EHHI WR )ORULGD %DNHU f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f UHDVRQHG WKDW E\ YLUWXH RI HVFDODWLQJ WUDQVSRUWDWLRQ H[SHQVHV DQ HFRQRPLF LQFHQWLYH ZRXOG VORZ WKH H[RGXV RI FDOYHV RXW RI )ORULGD 6WHJHOLQ DQG 6LPSVRQ f 6SUHHQ DQG 6KRQNZLOHU Ef DQG ,NHUG f DOO JDYH DQ HFRQRPLVWnV H[SODQDWLRQ WR WKH FRQWUDU\ ,Q VKRUW WKH LQGH[ RI WUDQVSRUWDWLRQ VHUYLFHV LQ PLG ZDV DERXW SHUFHQW RI WKH OHYHO %XW GXULQJ WKLV VDPH SHULRG WKH WRWDO PDUNHWLQJ FRVW LQGH[ KDV LQFUHDVHG E\ DERXW SHUFHQW VLQFH DQG KDV LQFUHDVHG SHUFHQW VLQFH 7KHUHIRUH FRVWV RI WUDQVSRUWDWLRQ VHUYLFHV KDYH LQFUHDVHG RQO\ SHUFHQW PRUH WKDQ RWKHU PDUNHWLQJ FRVWV RYHU WKH SDVW \HDUV LQ VSLWH RI UDSLGO\ ULVLQJ IXHO FRVWV ,NHUG f

PAGE 22

:KDW DUH WKHVH ORFDWLRQDO DGYDQWDJHV RI JHRJUDSKLFDO GLVSHUVLRQ IRU GLIIHUHQW SKDVHV RI EHHI SURGXFWLRQ" 7KH FRZFDOI SKDVH LV ODQG H[WHQn VLYH XWLOL]LQJ WKH OHVV SURGXFWLYH ODQG VXLWHG WR SDVWXUHV EXW OLWWOH HOVH DV LQ )ORULGD 6WRFNHU RSHUDWLRQV DUH ORFDWHG RQ DUHDV ZLWK KLJK TXDOLW\ SDVWXUHV VXFK DV WKH ZKHDW SDVWXUH DUHDV RI 2NODKRPD 7H[DV DQG .DQVDV 0DMRU FDWWOH IHHGLQJ DUHDV DUH ORFDWHG QHDU WKH PDMRU IHHG SURn GXFLQJ DUHDV DQG VODXJKWHU SODQWV IRU FDWWOH DUH ORFDWHG LQ WKH DUHDV ZKHUH FDWWOH DUH IHG ,NHUG f ,NHUGnV FRVW DQDO\VLV JLYHV D SHU KHDG DGYDQWDJH IRU VKLSSLQJ FDWWOH WR WKH SODLQV VWDWHV DQG WUXFNLQJ WKH PHDW EDFN IRU FRQVXPSWLRQ 7KH HFRQRPLHV RI VFDOH KDYH DOORZHG PRUH FRVW HIILFLHQF\ IRU ODUJHU IHHGORWV DQG SDFNLQJKRXVHV PDNLQJ HVWDEOLVKPHQW GLIILFXOW IRU LQGHSHQn GHQW VPDOO FRPSHWLWRUV ,QGHHG LW LV PRUH OLNHO\ LQ WKH QH[W \HDUV IRU D FRQWLQXHG WUHQG WRZDUG IHZHU DQG ODUJHU VSHFLDOL]HG RSHUDWLRQV LQ DOO SKDVHV RI WKH V\VWHP 7KH DUHDV ZKHUH FRZFDOI DQG VWRFNHU RSHUDn WLRQV DUH GRPLQDQW DUH OLNHO\ WR UHPDLQ GRPLQDQW EHFDXVH RI XQLTXH ODQG DQG SDVWXUH UHTXLUHPHQWV WKDW SUHFOXGH VLJQLILFDQW FKDQJHV LQ WKH VWUXFn WXUH RI WKHVH LQGXVWULHV 7KH )ORULGD FDWWOH LQYHQWRU\ UHIOHFWV WKLV WUHQG %HWZHHQ DQG WKHUH ZDV D SHUFHQW LQFUHDVH LQ )ORULGD FDWWOH QXPEHUV FRPSDUHG WR D SHUFHQW LQFUHDVH IRU WKH HQWLUH 8QLWHG 6WDWHV KRZHYHU LQ OHVV WKDQ SHUFHQW RI FDOYHV ZHUH RXWVKLSSHG ZKHUHDV LQ RYHU SHUFHQW ZHUH RXWVKLSSHG 6SUHHQ DQG 6KRQNZLOHU Ef %HHI FDWWOH SURGXFWLRQ LV FDSLWDO LQWHQVLYH $Q HVWLPDWH RI FRVWV IRU ODQG DQG OLYHVWRFN LQYHVWPHQW LV FXUUHQWO\ SHU FRZ LQ WKH 8QLWHG 6WDWHV ,NHUG f :KHQ WLJKW PRQH\ VXSSO\ DQG ULVLQJ LQWHUn HVW UDWHV VTXHH]H PDUNHWLQJ PDUJLQV SDFNHUV DQG UHWDLOHUV EX\ OHVV

PAGE 23

FDWWOH DQG FDUFDVVHV 'HPDQG LV ORZ VXSSO\ LV KLJK DQG FDWWOH SULFHV GHFUHDVH 7KH FDWWOH IHHGHUV PDNH XS WKHLU ORVVHV E\ RIIHULQJ OHVV PRQH\ IRU VWRFNHU DQLPDOV 7KH VWRFNHU RSHUDWRUV LQ WXUQ SDVV RQ WKH KLJKHU LQWHUHVW FRVWV WR WKH FRZFDOI PDQ 7KHUH ZLOO EH OLWWOH H[SDQn VLRQ LQ WKH FDWWOH EXVLQHVV ZLWK KLJK LQWHUHVW UDWHV DQG ORZ SULFHG FDOYHV ,NHUG f 7KH )ORULGD 2SSRUWXQLW\ IRU )LQLVKHG %HHI 7KH VRXWKHDVWHUQ 8QLWHG 6WDWHV KDV WKH ODUJHVW VXUSOXV RI VWRFNHU IHHGHU FDWWOH LQ WKH FRXQWU\ WKH VHYHQ VWDWHV LQ WKLV DUHD SURYLGH DERXW PLOOLRQ PRUH FDOYHV HDFK \HDU WKDQ DUH QHHGHG IRU KHUG UHSODFHPHQW RU IHG LQ VRXWKHDVWHUQ IHHGORWV %DNHU f ,I WKH FDOI FURS LV WR EH IHG XQWLO VODXJKWHU LQ )ORULGD HFRQRPLFV ZLOO QHFHVVLWDWH IDVW FRQWLQXRXV HIILFLHQW JURZWK IURP ZHDQLQJ XQWLO VODXJKWHU 7KH FDOYHV PXVW EH JURZQ RQ IRUDJH SDVWXUH DQGRU VLODJHf ZLWK RU ZLWKRXW JUDLQ VXSSOHPHQW ZLWK D ILQLVKLQJ SKDVH LQ WKH IHHGORW RQ D KLJK HQHUJ\ UDWLRQ ,PPHGLDWHO\ LW LV REYLRXV WR ZRQGHU :KHUH ZLOO WKH JUDLQ FRPH IURP" 7KH HQWLUH VRXWKHDVW LV D JUDLQ GHILFLHQW DUHD DQG FRUQ ZLOO FRQWLQXH WR FRVW WRQ PRUH LQ )ORULGD WKDQ LQ PDMRU IHHGLQJ VWDWHV %DNHU f 7KH XQSUHGLFWDELOLW\ RI WKH )ORULGD FOLPDWH PDNHV FRUQ SURGXFWLRQ D ULVN\ EXVLQHVV )ORUWRQ HW DO f DQG +RUWRQ DQG 0LVOHY\ f JUHZ FRUQ DQG VRUJKXP VLODJHV LQ VRXWK )ORULGD DQG WKHLU UHVXOWV VKRZHG WKDW ERWK FRUQ DQG VRUJKXP FDQ EH JURZQ VXFFHVVIXOO\ DQG HFRQRPLFDOO\ XVLQJ PXOWLSOH FURSSLQJ 7KH YLDELOLW\ RI DQ LQWHQVLYH EHHI FDWWOH LQGXVWU\

PAGE 24

Q LQ )ORULGD ZLOO GHSHQG RQ WKH DELOLW\ WR SURGXFH WKHVH ORFDOO\ JURZQ KLJK HQHUJ\ IHHGV 3URYLGHG WKH QHFHVVDU\ IHHGLQJ V\VWHPV ZHUH GHYHORSHG PRUH FDWWOH ZRXOG UHPDLQ RQ )ORULGD UDQFKHV 7KH TXHVWLRQ RI ZKHWKHU LQFUHDVHG FDWWOH QXPEHUV VKRXOG SUHFHGH WKH GHYHORSPHQW RI IHHGORW DQG SDFNLQJ LQGXVWULHV LV SXWWLQJ WKH FDUW LQ IURQW RI WKH KRUVH (IILFLHQW IHHGn LQJ V\VWHPV PXVW HYROYH ILUVW WKH FDWWOH EXLOG XS ZLOO IROORZ DQG FRXQW RQ WKH IHHGORW DQG SDFNLQJ SHRSOH WR IHQG IRU WKHPVHOYHV 7KHVH EXVLQHVVHV VLPSO\ ZDQW D FRQVLVWHQW GHSHQGDEOH VXSSO\ RI IHHG DQG DQLPDOV .DSODQ f $V (UZLQ %U\DQ -U RI WKH &HQWUDO 3DFNLQJ &R SXW LW ,W LV P\ IHHOLQJ WKDW ZH DOUHDG\ KDYH WKH NQRZKRZ WR IHHG FDWWOH LQ )ORULGD 7KH ZHDNHVW SRLQW LQ WKH FKDLQ DQG DOZD\V KDV EHHQ LV JUD]LQJ RI WKH IHHGHU FDOYHV XQWLO WKH\ DUH ELJ HQRXJK WR JR LQWR WKH IHHGORW %U\DQ S f ,Q FRQFOXVLRQ LW ZRXOG DSSHDU WKDW \HDUV IURP QRZ D WRWDOO\ HQFORVHG VHOIUHOLDQW EHHI LQGXVWU\ LQ )ORULGD FRXOG EH RSHUDWLRQDO 7KH PRVW OLNHO\ VWLPXOL ZLOO EH f WKH GHYHORSPHQW RI QHZ IRUDJH VSHFLHV DQG PDQDJHPHQW V\VWHPV WKDW SHUPLW D VKLIW IURP JUDLQIHG WR IRUDJHIHG OLYHVWRFN RSHUDWLRQV f LQFUHDVHG LUULJDWLRQ FRVWV RQ WKH ZHVWHUQ KLJK SODLQV GHFUHDVLQJ JUDLQ DYDLODELOLW\ f VROXWLRQ RI WKH DJURQRPLF SUREOHPV IRU \HDU URXQG IHHG DYDLODELOLW\ LQ )ORULGD DQG f ORZHU LQWHUHVW UDWHV WR DOORZ HFRQRPLF JURZWK LQ WKH FDWWOH EXVLQHVV 7KHVH IDFWRUV FRXOG VKLIW WKH JHRJUDSKLFDO DGYDQWDJH WR WKH VRXWKHDVWHUQ 8QLWHG 6WDWHV

PAGE 25

%DUULHUV WR )RUDJH 3URGXFWLRQ LQ )ORULGD 7KHUH DUH WKUHH SKDVHV RI WKH IRUDJH V\VWHPSURGXFWLRQ KDUYHVWn LQJ DQG XWLOL]DWLRQ %\ LGHQWLI\LQJ WKH SUHVHQW ZHDNQHVVHV LQ WKHVH SKDVHV FRPELQDWLRQV RI QDWLYH UDQJH DQG LPSURYHG SDVWXUHV DQGRU VXSSOHPHQWDO SDVWXUHV ZLOO EH GHYHORSHG LQ RUGHU WR SURYLGH D GLVWULEXn WLRQ RI IRUDJH WKURXJKRXW WKH \HDU $Q HYHQ GLVWULEXWLRQ RI IRUDJHV PRUH WKDQ DQ\ RWKHU IDFWRU ZLOO XOWLPDWHO\ EH UHVSRQVLEOH IRU LQFUHDVHG EHHI SURGXFWLRQ LQ )ORULGD 0RWW Ef 7KH ELJJHVW OLPLWDWLRQ WR ZHOO GLVWULEXWHG IRUDJH SURGXFWLRQ LV SUREDEO\ VHDVRQDOLW\ 0RVW IRUDJH SURGXFWLRQ RFFXUV IURP $SULO WR 6HSWHPEHU DQG WZRWKLUGV RI WKH \HDUO\ JURZWK RI SHUHQQLDOV RFFXUV EHWZHHQ -XO\ DQG 2FWREHU 0RWW Ef 8QHYHQ GLVWULEXWLRQ RI IRUDJH SURGXFWLRQ WKURXJKRXW WKH \HDU PDNHV PDQDJHPHQW GLIILFXOW DQG VXSSOHPHQWDO IHHGLQJ RQ SDVWXUH LV RIWHQ QHHGHG WR HIILFLHQWO\ XWLOL]H WKH IRUDJH SURGXFHG 6ORDQ %DNHU VXPPDn UL]HV )ORULGDnV FXUUHQW IRUDJH SURGXFWLRQ SUREOHPV +H VWDWHV WKDW FRRO VHDVRQ IRUDJHVU\H 6HFDOH FHUHDOH /f U\HJUDVV /ROLXP PXOWLIORUXP /DPf DQG FORYHUV 7ULIROLXP VSSfDUH H[FHOOHQW \LHOGLQJ JRRG JDLQV RI WR NJGD\ IRU WR GD\V 'U\ PDWWHU \LHOGV KRZHYHU DUH VRPHWLPHV ORZ GXH WR ZHDWKHU :DUP VHDVRQ IRUDJHV DUH PRUH SURGXFn WLYH EXW JLYH KDOI WKH GDLO\ JDLQV 'U\ ZHDWKHU RIWHQ OLPLWV SURGXFWLRQ LQ VSULQJ DQG DXWXPQ ZKLOH LQ VXPPHU UDLQIDOO DQG KXPLGLW\ DUH KLJK HQRXJK WR PDNH KD\ FXULQJ GLIILFXOW 5HVXOWV ZLWK JUDVV VLODJH KDYH EHHQ GLVDSSRLQWLQJ GXH WR LQVXIILFLHQW TXDOLW\ IRU JRRG JDLQV ZLWK \RXQJ FDWWOH %DNHU f

PAGE 26

7KH &OLPDWH LQ 1RUWK &HQWUDO )ORULGD 7KH FOLPDWH UDQJHV IURP VXEWURSLFDORFHDQLF LQ VRXWK )ORULGD WR D W\SLFDO ORZDOWLWXGH FRQWLQHQWDO IURQWDO SDWWHUQ LQ WKH QRUWK +XQWHU HW DO f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f 7KH %HUPXGD KLJK LV FRPPRQ WKURXJKRXW WKH \HDU FHQWHUHG LQ WKH &DULEEHDQ ZLWK LWV VWURQJHVW HIIHFW GXULQJ ZLQWHU ,I LW ZHUH QRW IRU ODUJH ERGLHV RI ZDUP ZDWHU RQ HLWKHU VLGH RI WKH SHQLQVXOD )ORULGD ZRXOG EH DV DULG DV WKH JUHDW VXEWURSLFDO GHVHUWV DW WKH VDPH ODWLWXGHV 'RKUHQZHQG f )ODWZRRGV DQG 7KHLU 3RWHQWLDO IRU *UDVVODQG 3URGXFWLYLW\ 7KHUH DUH KHFWDUHV RI ODQG DUHD LQ )ORULGD $W SUHVHQW WKHUH DUH DSSUR[LPDWHO\ PLOOLRQ KHFWDUHV RI LPSURYHG SDVWXUH

PAGE 27

PLOOLRQ KHFWDUHV RI QDWLYH UDQJH DQG PLOOLRQ KHFWDUHV RI IRUHVWn ODQG WKDW SURYLGH VRPH JUD]LQJ 6HYHQWHHQ PDMRU QDWXUDO YHJHWDWLRQ W\SHV DUH UHFRJQL]HG LQ )ORULGD 7KH PRVW LPSRUWDQW HFRW\SH IURP WKH VWDQGSRLQW RI KHFWDUDJH DQG SRWHQn WLDO IRU DQLPDO SURGXFWLRQ IURP IRUDJHV LV WKH IODWZRRG VLWH )ODWZRRGV RFFXS\ PLOOLRQ KHFWDUHV RU DOPRVW SHUFHQW RI )ORULGDnV ODQG DUHD 0RVW RI WKH VWDWHnV WLPEHU SURGXFWLRQ DOVR RFFXUV RQ IODWZRRGV 2SHQ ZRRGODQG FRQVLVWV SUHGRPLQDQWO\ RI RQH WR WKUHH VSHFLHV RI SLQHV ORQJOHDI 3LQXV SDOXVWULV 0LOOf VODVK 3B HOOLRWWLL (QJHOPf DQG SRQG 3B VHUWLQD 0LFK[f 8QGHUVWRU\ SURGXFHV PDQ\ JUDVVHV VXFK DV FKDON\ EOXHVWHQU $QGURSRJRQ FDSLOOLSHV 1DVKf EURRPVHGJH EOXHVWHP $QGURSRJRQ YLUJLQLFXV /f SDVSDXUQV 3DVSDOXP VSSf ZLUHJUDVV $ULVWLGD VWULFWD 0LFK[f LQGLDQJUDVV 6RUJKDVWUXP VHFXQGXP (OOf 1DVKf DQG SDQLFXPV 3DQLFXP VSSf $VVRFLDWHG IRUEV LQFOXGH JUDVVOHDI JROGDVWHU &KU\VRSVLV JUDPLQLIROLD 0LFK[f (OOf SDUWULGJH SHD &DVVLD IDVFLFXODWD 0LFK[f DQG EHJJHUZHHG 'HVPRGLXP VSf 6KUXEV DUH SUHn GRPLQDQWO\ VDZ SDOPHWWR 6HULQRD UHSHQV %DUWUDPf 6PDOOf ZD[ P\UWOH 0\ULFD FHUIHUD /f EODFNEHUU\ 5XEXV VSf DQG JDOO EHUU\ ,OH[ JODEUD /f *UD\f $PLGVW WKH IODWZRRG DUHDV DUH VPDOO KDUGZRRG IRUHVWV PDQ\ F\SUHVV SRQGV SUDLULHV PDUVKHV DQG ED\ WUHH VZDPSV 7KH ZHW DUHDV VXSSRUW YHU\ GHVLUDEOH IRUDJHV VXFK DV PDLGHQFDQH 3DQLFXP KHPLWRPRQ 6FKXOWf DQG OLWWOH EOXH PDLGHQFDQH $PSKLFDUSXP PXKOHQEHUJLDQXP 6FKXOWf +LWFKFf 7KH IODWZRRGV DUH LQKDELWHG E\ GHHU DQG KRJV TXDLO JUD\ VTXLUUHOV DQG WXUNH\ KHQFH WKLV LV DOVR WKH PRVW LPSRUWDQW FRPPXQLW\ W\SH LQ )ORULGD IRU KXQWLQJ

PAGE 28

7KH VRLOV DUH DFLG S+f RI WKH 6SRGRVRO 2UGHU ZLWK D FP VXUIDFH KRUL]RQ FRORUHG JUH\ WR JUH\EODFN E\ WKH SUHVHQFH RI RUJDQLF PDWWHU 7KH WR FP VWUDWXP LV GRPLQDWHG E\ ZKLWH OHDFKHG VDQGV DQG IURP WR FP D KDUG SDQ LV IRXQGILQH SDUWLFOHV FHPHQWHG WRJHWKHU E\ VHVTXLR[LGHV DQG RWKHU FRPSRXQGV )URP WR FP GHSWK D EURZQ WDQQLQVWDLQHG VDQG H[LVWV IROORZHG E\ ZKLWH VDQG ZLWK IXUWKHU LQFUHDVHV LQ WKH SURILOH 'XH WR WKH KDUGSDQ IRUPHG RQ 6SRGRVROV LQ )ORULGD D SHUFKHG ZDWHU WDEOH LV FUHDWHG WKDW LV EHQHILFLDO LQ UHGXFLQJ WKH UDSLG UDWH RI SHUFRODWLRQ WKURXJK WKH VDQGV EXW LV D SUREOHP ZKHQ UDLQIDOO LV VR JUHDW DV WR FDXVH VWDQGLQJ ZDWHU DERYH WKH VRLO VXUIDFH 7KH DFLG VRLO FRQGLWLRQ RI WKH IODWZRRGV PDNHV LW XQVXLWDEOH IRU FURS SURGXFWLRQ KRZHYHU DFLG DQG ZHW WROHUDQW VSHFLHV OLNH OLPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf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n5HGDOWDn n*UHHQDOWDn DQG

PAGE 29

n%LJDOWDn LQ 6LQFH WKDW WLPH PDQ\ OLPSRJUDVVHV ZHUH FROOHFWHG DQG HYDOXDWHG LQ KRSHV RI IXUWKHU H[SORLWLQJ WKH JHUPSODVP %HQHILFLDO FKDUDFWHULVWLFV RI TXDOLW\ \LHOG SHUVLVWHQFH PRUSKRORJ\ ZLQWHUn KDUGLQHVV DQG DGDSWDELOLW\ WR IODWZRRGV KDYH JLYHQ UHVHDUFKHUV HYLGHQFH RI WKH SRWHQWLDO RI OLPSRJUDVV DV D IRUDJHSURGXFLQJ JUDVV IRU WKH IXWXUH /LPSRJUDVV /LWHUDWXUH WR 3UHVHQW $Q 2YHUYLHZ 7KH OLWHUDWXUH RQ OLPSRJUDVV VSHFLHV WR GDWH ZLOO EH JURXSHG LQWR ILYH SKDVHV f RULJLQ GLVWULEXWLRQ DQG GHVFULSWLRQ f FKDUDFWHULn ]DWLRQ f SHVW VFUHHQLQJ f VPDOO SORW WULDOV DQG IRUDJH UHVSRQVH WR JUD]LQJ DQG f DQLPDO UHVSRQVH DQG V\VWHPV PDQDJHPHQW /LPSRJUDVV UHVHDUFK KDV MXVW UHFHQWO\ UHDFKHG SKDVH ILYH 2ULJLQ 'LVWULEXWLRQ DQG 'HVFULSWLRQ 7KHUH DUH SUREDEO\ QR PRUH WKDQ GHVFULEHG VSHFLHV RI OLPSRJUDVV .UHWVFKPHU DQG 6\QGHU f DQG WKH RQH RI PRVW DJUQRQRPLF LPSRUWDQFH LV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEE $JURVWRORJLVWV JXHVV +HPDUWKULDV RULJLQ WR EH LQ WURSLFDO $IULFD EXW LW LV DOVR IRXQG LQ 0DGUDV %XUPD 0DOD\VLD 0DOD\ 6LDP 7XUNH\ 1LJHULD ,WDO\ (WKLRSLD 7DQ]DQLD &H\ORQ 1RUWKHUQ ,QGLD 6RXWKHDVW $VLD DQG $UJHQWLQD %RU %RJGDQ &KLSSLQGDOO f 7KH VSHFLHV DUH GHVLUDEOH UREXVW SHUHQQLDO IRGGHU SODQWV IRXQG LQ ZHW KDELWDWV VXFK DV ULYHU EDQNV VHDVRQDOO\ IORRGHG ULYHU YDOOH\V VHDVRQDO VZDPSV DQG VR IRUWK

PAGE 30

'XH WR SODQW H[SORUDWLRQ DQG UHGLVWULEXWLRQ +HPDUWKULD LQWURGXFn WLRQV DUH QRZ IRXQG LQ $XVWUDOLD %UD]LO %ROLYLD &ROXPELD (FXDGRU +DZDLL 0DODZL 0H[LFR 1HZ =HDODQG 3DUDJXD\ 8UXTXD\ 9HQH]XHOD DQG WKH 9LUJLQ ,VODQGV 4XHVHQEHUU\ HW DO 4XHVHQEHUU\ HW DO DQG 2DNHV f 2YHUZLQWHULQJ RI +B DOWLVVLPD LQ WKH 8QLWHG 6WDWHV RFFXUV LQ $ODEDPD 0LVVLVVLSSL 7H[DV 7HQQHVVHH DQG DV IDU QRUWK DV %HOWVYLOOH 0DU\ODQG 2DNHV 2DNHV DQG )R\ f /DWHU 2DNHV f UHSRUWHG WKH VXUYLYDO RI RQH OLPSRJUDVV DFFHVVLRQ IURP WKH FROOHFn WLRQ 3, IURP 5KRGHVLDf IRU WKUHH ZLQWHUV DW 3XOOPDQ :DVKLQJWRQ 7KLV GLYHUVLW\ RI ZLQWHUKDUGLQHVV PD\ EH XVHG WR H[WHQG WKH SURGXFWLRQ UDQJH DQG JUD]LQJ SHULRG IRU OLPSRJUDVV LQ WKH 8QLWHG 6WDWHV 6RPH FRPPRQ QDPHV IRU JUDVVHV RI WKH JHQXV +HPDUWKULD DUH OLPSRn JUDVV WHDJUDVV )ORULGDf &DSLP JDPDORWH %UD]LOf 3DVWR FODYHO *UDPLOOD FDQLWD $UJHQWLQDf %DNVKD 3DQLVKDUX ,QGLDf 6ZDPS &RXFK DQG 5RRLNZHHN 6RXWK $IULFDf %RJGDQ &KLSSHQGDOO f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f 4XHVHQEHUU\ HW DO f VXUYH\HG WKH 86'$ FROOHFWLRQ RI +HPDUWKULD FRQWDLQLQJ VSHFLHVf DQG IRXQG SHUFHQW GLSORLG SHUn FHQW WHWUDSORLG DQG SHUFHQW KH[DSORLG 7KLUWHHQ RI WHWUDSORLGV

PAGE 31

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nV DQG UHVSHFWLYHO\ 4XHVHQEHUU\ HW DO f 5HGDOWD DQG %LJDOWD DUH WZR RI WKH IRXU OLPSRJUDVVHV XVHG LQ WKLV GLVVHUWDWLRQ 2DNHV REWDLQHG WKHVH JHQRW\SHV IURP WKH 5LHWRQGDOH 5HVHDUFK 6WDWLRQ QHDU 3UHWRULD 6RXWK $IULFD LQ 2DNHV f 5HGDOWD RULJLQDWHG QHDU WKH 3LHQDDUV 5LYHU LQ ZHVW FHQWUDO 7UDQVYDDO DQG %LJDOWD LV IURP DQ XQVSHFLILHG ORFDWLRQ QHDU WKH 7UDQVYDDO 5HGDOWD LV D GLSORLG Q f DQG %LJDOWD LV D WHWUDSORLG Q f $OVR LQFOXGHG LQ WKLV GLVVHUWDWLRQ DUH 3, FROOHFWHG LQ IURP D VPDOO LVODQG LQ WKH /XYXYKX 5LYHU VHYHUDO NLORPHWHUV DERYH LWV FRQIOXn HQFH ZLWK WKH /LPSRSR 5LYHU LQ .UXJHU 1DWLRQDO 3DUN DQG 3, IURP 0W 0E\D .HQ\D 7KHVH WZR OLPSRJUDVVHV DUH WHWUDSORLGV 2DNHV f 7KHUH DUH PDQ\ PRUSKRORJLFDO GLIIHUHQFHV DPRQJ OLPSRJUDVVHV /HDILQHVV LQWHUQRGH OHQJWK DQWKRF\DQLQ FRQWHQW EXQFKLQHVV WLOOHUV SHU XQLW DUHD DQG VWHP WKLFNQHVV YDU\ 7KHVH LQKHUHQW GLIIHUHQFHV DLG LQ PDLQWDLQLQJ JHQRW\SH LGHQWLILFDWLRQ &\WRORJLFDO LQYHVWLJDWLRQV :LOPV HW DO f FKDUDFWHUL]HG WKH IRXU OLPSRJUDVVHV FROOHFWHG IURP $IULFD LQ DQG UHSRUWHG FRORU GLIn IHUHQFHV IRU WKH DQWKHUV RI 5HGDOWD SXUSOHf *UHHQDOWD EURZQf DQG

PAGE 32

%LJDOWD \HOORZf 6FKDQN f FKDUDFWHUL]HG FKURPRVRPH QXPEHUV SROOHQ VWDLQDELOLW\ DQG VHHG VHW IRU PRUH +HPDUWKULDV LQ SUHSDUDWLRQ IRU DQ LQWUDJHQHULF EUHHGLQJ SURJUDP %UHHGLQJ HIIRUWV WR GDWH KDYH QRW SURGXFHG DQ\ OLPSRJUDVVHV ZLWK EHWWHU DJURQRPLF DWWULEXWHV WKDQ WKH YHJHWDWLYHO\ SURSDJDWHG SODQW LQWURGXFWLRQV 7KH ORJLVWLFV RI ORZ VHHG VHW 6FKDQN f PD\ SDUWLDOO\ GLVFRXUDJH SODQW EUHHGHUV LQ WKLV UHJDUG DV ZHOO DV WKH GLIILFXOW\ LQ JHWWLQJ OLPSRJUDVV WR IORZHU LQ WKH JUHHQKRXVH 4XHVHQEHUU\ SHUVRQDO FRPPXQLFDWLRQf 6FKDQN HW DO f GLVFRYHUHG WKDW WHWUDSORLG OLPSRJUDVVHV KDG KLJKHU LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f WKDQ GLSORLG DFFHVVLRQV 7KH PHDQ GHFUHDVH RI ,920' RI WKH WHWUDSORLG IURP SHUFHQW DW ZHHNV WR SHUFHQW LQ PDWXUH SODQWV VXJJHVWHG D VORZHU GHFOLQH RI TXDOLW\ WKDQ IRU PRVW WURSLFDO JUDVVHV &URVVVHFWLRQV RI VWHPV UHYHDOHG VLJQLILFDQWO\ ORZHU YDVFXODU EXQGOH DUHD LQ WKH WHWUDn SORLG DV ZHOO DV IHZHU VFOHUHQFK\PD ILEHU FHOOV &KDUDFWHUL]DWLRQ 2QH RI WKH WDQJHQWLDO DVSHFWV RI WKH LQFLSLHQW UHVHDUFK RQ OLPSRn JUDVV ZDV WKDW LW ZDV QRWHG DV KDYLQJ D WHDOLNH RGRU DQG IODYRU .LOOLQJHU f DIWHU FROODERUDWLRQ ZLWK WKH 86'$ 1RUWKHUQ 8WLOL]DWLRQ 5HVHDUFK /DERUDWRU\ ILOHG IRU D EHYHUDJH XVH SDWHQW RQ -DQXDU\ 7KXV WKH QDPH WHDJUDVV FDPH LQWR XVDJH ILUVW SURPXOJDWHG LQVRIDU DV FDQ EH GHWHUPLQHG E\ (OGULGJH /HH RI WKH 8QLYHUVLW\ RI )ORULGD $JURQRP\ )DUP .LOOLQJHU DQG %HFNKDP REWDLQHG 8QLWHG 6WDWHV 3DWHQW IRU +HPDUWKULD EHYHUDJH ULJKWV LQ 2DNHV DQG )R\ f UHFRPPHQGHG OLPSRJUDVV IRU UHYHJHWDWLQJ PLQH VSRLOV GXH WR DQ H[FHOOHQW WROHUDQFH WR DOXPLQXP WR[LFLW\ 2DNHV f

PAGE 33

2DNHV DQG )R\ f DQG 2DNHV f DOVR UHSRUWHG D ZLGH GLYHUVLW\ RI FROG WROHUDQFH DV PHQWLRQHG HDUOLHU $ UHFHQW ILQGLQJ ZDV DQ DOOHRSDWKLF HIIHFW RI VRPH OLPSRJUDVVHV 5XHONH DQG 4XHVHQEHUU\
PAGE 34

H[WUDSRODWLQJ WKHVH UHVXOWV WR RWKHU OLPSRJUDVVHV DFNQRZOHGJLQJ YDULDn WLRQV DPRQJ DQG ZLWKLQ SORLG\ OHYHOVf 3DSHU 1R &RZDUG/RUG HW DO Df RQ FDUERK\GUDWH ILEHU IUDFn WLRQV GLVFXVVHG QHXWUDOGHWHUJHQW ILEHU 1')f DFLGGHWHUJHQW ILEHU $')f DFLGGHWHUJHQW OLJQLQ $'/f KHPLFHOOXORVH FHOOXORVH DQG VLOLFD 7KH 1') IUDFWLRQ UHSUHVHQWV WKH WRWDO ILEHU IUDFWLRQ LWV GLIn IHUHQFH IURP EHLQJ WKH QHXWUDOGHWHUJHQW VROXEOHV 1'6f RU VROXEOH QXWULHQWV 7KH $') FRQWHQW LV D PHDVXUH RI WKH LJQRFHOXORVH IUDFWLRQ 7KH GLIIHUHQFH EHWZHHQ 1') DQG $') LV DQ HVWLPDWH RI KHPLFHOOXORVH $FLGGHWHUJHQW OLJQLQ E\ WKH SHUPDQJDQDWH PHWKRGf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f VWXGLHG QXWULHQW FRPSRn VLWLRQ LQFOXGLQJ FUXGH SURWHLQ &3f GU\ PDWWHU FUXGH ILEHU &)f HWKHU H[WUDFW DVK DQG QLWURJHQIUHH H[WUDFW 1)(f 7KLV PHWKRGRORJ\ LV VORZO\ ORVLQJ IDYRU RYHU WKH *RHULQJ DQG 9DQ 6RHVW f PHWKRGRORJ\ IRU IUDFWLRQDWLQJ IHHGVWXIIV IRU UXPLQDQW YDOXH 7KH GHILQLWLRQ RI &) DV D FKHPLFDOO\ XQLIRUP QRQQXWULWLYH VXEVWDQFH FDQQRW EH UHFRQFLOHG EHFDXVH &) UHSUHVHQWV DOPRVW DOO WKH SRWHQWLDOO\ GLJHVWLEOH FHOOXORVH DQG DOVR LQFOXGHV VRPH OLJQLQ DQG KHPLFHOOXORVH 7KH LPSHUIHFW &) PHWKRGRORJ\ KDV DOORZHG PRVW RI WKH OLJLQLQ DQG KHPLFHOOXORVH WR EH

PAGE 35

LQFOXGHG LQ WKH 1)( ZKLFK LV VXSSRVHG WR UHSUHVHQW DYDLODEOH FDUERK\GUDWH ,Q VRPH FDVHV WKH &) FDQ EH PRUH GLJHVWLEOH WKDQ WKH 1)( ZKLFK LV FOHDUO\ LQFRQJUXRXV ZLWK WKH DLP RI WKH IUDFWLRQDWLRQ 1HYHUWKHOHVV UHVXOWV VKRZHG WKDW OLPSRJUDVV KDG WKH KLJKHVW PHDQ 1)( YDOXH RI DOO VSHFLHV DW SHUFHQW RYHU GD\V 7KH &3 YDOXHV GURSSHG WR SHUFHQW E\ WKH GD\ VWDJH 7KH WKLUG 3XHUWR 5LFDQ 3DSHU $UUR\R$TXLX DQG &RZDUG/RUG Df FRYHUHG PLQHUDO FRPSRVLWLRQ 7KH UDQJHV DV SHUFHQW RI GU\ PDWWHU RYHU GD\V IRU DOO VSHFLHV DQG WKH DYHUDJHV IRU OLPSRJUDVV LQ SDUHQn WKHVHVf ZHUH FDOFLXP f SKRVSKRUXV f PDJQHVLXP f DQG SRWDVVLXP f /LPSRJUDVV KDG WKH ORZHVW SKRVSKRUXV SHUFHQW RI DOO JUDVVHV DW GD\V fFOHDUO\ GHILFLHQW IRU UXPLQDQWV PDJQHVLXP DQG SRWDVVLXP ZHUH ORZHVW DW GD\V DQG SHUFHQW UHVSHFWLYHO\ 7KH ILQDO 3XHUWR 5LFDQ SDSHU $UUR\R$TXLOX DQG &RZDUG/RUG Ef GLVFXVVHG TXDOLW\ GHFOLQH DPRQJ WKH WURSLFDO JUDVVHV 7KH PHDQ UDWH RI LQ YLWUR WUXH GLJHVWLELOLW\ GHFOLQH ZDV XQLWV IURP GD\V 7KH ODUJHVW GHFOLQH XQLWVf RFFXUUHG EHWZHHQ GD\V DV FRPSDUHG WR GHFOLQHV RI DQG XQLWV EHWZHHQ GD\ LQWHUYDOV IURP GD\V $OO WKH JUDVVHV UHDFWHG VLPLODUO\ 7KLV VXJJHVWV WKDW WKH WURSLFDO JUDVVHV VWXGLHG PD\ EHVW EH XWLOL]HG EHWZHHQ GD\V RI JURZWK +RGJHV DQG 0DUWLQ f LQFOXGHG WKUHH OLPSRJUDVVHV LQ D VWXG\ RI SHUHQQLDO VXE WURSLFDO JUDVVHV DQG UHSRUWHG WKDW &\QRGRQV DQG 'L JLWDUL DV ZHUH EHWWHU FRRO VHDVRQ SURGXFHUV WKDQ WKH OLPSRJUDVVHV DW 2QD )ORULGD 6HDVRQDO \LHOG GLVWULEXWLRQ RI QXPHURXV WURSLFDO JUDVVHV ZDV VWXGLHG E\ 7D\ORU HW DO Ef LQ 1HZ =HDODQG ,Q WKLV VWXG\ D

PAGE 36

WHWUDSORLG OLPSRJUDVV LGHQWLW\ QRW UHSRUWHGf \LHOGHG SRRUO\ GXULQJ WKH ORZ UDLQIDOO VXPPHU RI WKH \HDU 7KH WRWDO ZDUPVHDVRQ \LHOG IRU OLPSRn JUDVV QHDUO\ GRXEOHG XQGHU WULFNOH LUULJDWLRQ DQG WKH PRLVW FRRO VHDVRQ \LHOGV ZHUH WZLFH DV JUHDW DV WKH XQLUULJDWHG ZDUPVHDVRQ \LHOGV 7KH GLJHVWLELOLW\ GDWD IRU JUDVVHV ZHUH DQDO\]HG LQ DQRWKHU 7D\ORU HW DO Df SXEOLFDWLRQ WR FRPSDUH QXWULWLYH TXDOLW\ RI WKH JUDVVHV JURZQ DW .DLWDLD 1HZ =HDODQG 6 ODWLWXGHf ZLWK WKH VDPH JUDVVHV JURZQ LQ PRUH HTXDWRULDO HQYLURQPHQWV 7KH UHVXOWV IRU JUDVVHV JURZQ DW .DLWDLD ZKLFK KDV D PHDQ ZDUPVHDVRQ WHPSHUDWXUH RI & ZHUH FRPSDUHG WR WKH GLJHVWLELOLW\ GDWD IRU WKH VDPH VSHFLHV JURZQ DW /DZHV $XVWUDOLD ZLWK D PHDQ WHPSHUDWXUH RI & 6ORZHU UDWHV RI PDWXUDWLRQ LQ WKH FRROHU HQYLURQPHQW IUHTXHQWO\ UHVXOWHG LQ KLJKHU WLVVXH GLJHVWLELOLWLHV 7KH QRQIORZHULQJ OLPSRJUDVV XVHG LQ WKH 1HZ =HDODQG VWXGLHV SURn GXFHG WLVVXH ZLWK PRGHUDWH SURWHLQ SHUFHQW LQ OHDYHV SHUFHQW LQ VWHPVf PRGHUDWH IHUPHQWDEOH FDUERK\GUDWH SHUFHQW LQ OHDYHV SHUFHQW LQ VWHPVf DQG H[FHOOHQW GLJHVWLELOLW\ SHUFHQW LQ OHDYHV SHUFHQW LQ VWHPVf 7KH VWHP GLJHVWLELOLW\ ZDV WKH KLJKHVW RI DOO WKH JUDVVHV VWXGLHG 3HVW 6FUHHQLQJ $ OLPLWHG QXPEHU RI UHIHUHQFHV DUH DYDLODEOH RQ OLPSRJUDVV VXVFHSn WLELOLW\ WR QHPDWRGHV DQG DSKLGV %R\G DQG 3HUU\ f VFUHHQHG 5HGDOWD *UHHQDOWD DQG %LJDOWD UHVLVWDQFH WR VWLQJ QHPDWRGHV %HORQRODLPXV ORQJLFDXGDWXV 5DXf 7KLV QHPDWRGH LV DPRQJ WKH PRVW SHUVLVWHQW VHULRXV SHVWV LQ )ORULGDnV LPSURYHG SDVWXUH3DQJROD GLJLW JUDVV 3HQVDFROD EDKLDJUDVV DQG &RDVWDO EHUPXGDJUDVV DUH DOO

PAGE 37

VXVFHSWLEOH 7KH OLPSRJUDVVHV KDG PRGHUDWHO\ ORZ QHPDWRGH FRXQWV EXW WKH DXWKRUV DWWULEXWHG WKLV WR ODFN RI URRWV UDWKHU WKDQ VLJQLILFDQW QHPDWRGH UHVLVWDQFH &KORURVLV ZDV REVHUYHG LQ %LJDOWD DQG WKLV JHQRn W\SH KDG KLJKHU FRXQWV WKDQ WKH WZR GLSORLG OLPSRJUDVVHV ,Q %R\G DQG 3HUU\ UHSRUWHG DGGLWLRQDO LQIRUPDWLRQ RQ VWLQJ QHPDWRGH GDPDJH WR SDVWXUH JUDVVHV 5HGDOWD *UHHQDOWD DQG %LJDOWD ZHUH WKH WKUHH PRVW IDYRUDEOH KRVWV %R\G HW DO f VWXGLHG WKH LQWHUDFWLRQ RI VRLO WHPSHUDWXUH DQG VWLQJ QHPDWRGHV DQG IRXQG WKDW *UHHQDOWD JUHZ EHVW EHWZHHQ DQG & LQ XQLQIHVWHG VRLOV DQG EHVW EHWZHHQ DQG & LQ LQIHVWHG VRLOV $ERYH & QHPDWRGHV ZHUH UHGXFHG EXW WKH VRLO ZDV WRR KRW IRU JRRG JURZWK RI WKH JUDVV /DWHU 4XHVHQEHUU\ DQG 'XQQ f UHFHLYHG PRUH OLPSRJUDVVHV ZKLFK WKH\ VFUHHQHG IRU UHVSRQVH WR WKH VWLQJ QHPDWRGH LQ WKH JUHHQKRXVH 1R DYDLODEOH OLPSRJUDVV OLQHV DSSURDFKHG LPPXQLW\ EXW D IHZ LQWURGXFWLRQV KDG JUHDWHU WROHUDQFH 2I WKH EHVW OLQHV SHUFHQW ZHUH WHWUDSORLG RU KH[DSORLG ZKLOH DOO WKH OHDVW WROHUDQW ZHUH GLSORLG 7KH PRVW WROHUDQW LQWURGXFWLRQV ZHUH FROOHFWHG IURP WKH LVODQGV RI 0DXULWLXV LQ WKH ,QGLDQ 2FHDQ 3HVW SUREOHPV LQ OLPSRJUDVV ZHUH UHSRUWHG E\ 2DNHV f ZKR VWXGLHG UHVLVWDQFH LQ +HPDUWKULD WR WKH \HOORZ VXJDUFDQH DSKLG 6LSKD IODYD )RUEHVf 9DULDEOH UHVLVWDQFH ZDV IRXQG IRU 3/ DOWLVVLPD DFFHVVLRQV 7ZR RI WKH LQWURGXFWLRQV LQFOXGHG LQ WKLV GLVVHUWDWLRQ ZHUH HYDOXDWHG LQ WKH 2DNHV VWXG\3, ZDV PRVW VXVFHSWLEOH DQG 3, ZDV PRGHUDWHO\ UHVLVWDQW ZKLOH DOO WKH OLPSRJUDVVHV KDG EHWWHU UHVLVWDQFH WR WKH VXJDUFDQH DSKLG WKDQ IRXQG LQ 'LJLWDUL D

PAGE 38

6PDOO 3ORW 7ULDOV DQG )RUDJH 5HVSRQVH WR *UD]LQJ 7KH VHFRQG DQG WKLUG FROOHFWLRQV RI OLPSRJUDVV ZHUH LQLWLDWHG LQ UHVSRQVH WR SUHYLRXV H[SHULPHQWDO UHVXOWV WKDW UHFRPPHQGHG DQ H[WHQGHG VHDUFK ZLWKLQ +HPDUWKULDV IRU IDYRUDEOH SDVWXUH JUDVV DWWULEXWHV 5XHONH HW DO f HYDOXDWHG OLPSRJUDVVHV LQ ERWK JUHHQKRXVH DQG VPDOO SORW WULDOV DQG WKLV VWXG\ VDWLVILHG WKHLU H[SORUDWRU\ FXULRVLW\ ZLWK UHVSHFW WR WKH LGHQWLILFDWLRQ RI VXSHULRU JHQRW\SHV ,Q 4XHVHQEHUU\ DQG 2FXPSDXJK f JUHZ 5HGDOWD *UHHQDOWD DQG %LJDLWD DV FRQVHUYHG IRUDJHV 7KH\ SUHVHQWHG WKHLU UHVXOWV DW WKH $PHULFDQ 6RFLHW\ RI $JURQRP\ PHHWLQJV DW /RV $QJHOHV LQ DQG ZLWK PRUH GHWDLO 4XHVHQEHUU\ DQG 2FXPSDXJK 2FXPSDXJK DQG 4XHVHQEHUU\ f IROORZLQJ D VHFRQG \HDU RI GDWD 6WRFNSLOLQJ LV DQ LQH[SHQVLYH ZD\ RI ILOOLQJ WKH IRUDJHGHILFLHQW PRQWKV RI 1RYHPEHU )HEUXDU\ LQ QRUWKHUQ )ORULGD 6WRFNSLOHG \LHOGV ZHUH JUHDWHVW IRU 5HGDOWD LQ RYHU P WRQVKDf
PAGE 39

UHFRPPHQGHG OHYHO IRU UXPLQDQWV f E\ HDUO\ 1RYHPEHU 3KRVSKRUXV GURSSHG EHORZ WKH 15& PLQLPXP IRU UXPLQDQWV f E\ PLG2FWREHU DQG LQ WKH VHFRQG \HDU ZDV QHYHU DERYH WKH PLQLPXP OHYHO 0DJQHVLXP ZDV QRW FRQVLGHUHG D QXWULWLRQDO SUREOHP ZKLOH WKH PHDQ FDOFLXP SHUFHQW IRU ERWK \HDUV f ZDV DGHTXDWH IRU PDWXUH SUHJQDQW EHHI FRZV 15& f EXW EDUHO\ VXIILFLHQW IRU ODFWDWLQJ EHHI FRZV 15& f 7KH DXWKRUVn UHFRPPHQGDWLRQ ZDV WR VXSSOHPHQW ZLWK SRWDVVLXP SKRVSKRUXV FDOFLXP DQG SURWHLQ DIWHU PLG2FWREHU 4XHVHQEHUU\ HW DO f FRDOHVFHG WKH SHUWLQHQW OLPSRJUDVV SURn GXFWLRQ GDWD WKURXJK IRU VL[ VLWHV LQ )ORULGD $W )W 3LHUFH %LJDLWD KDG KLJKHU SURGXFWLRQ LQ 1RYHPEHU DQG 'HFHPEHU WKDQ 5HGDOWD RU *UHHQDOWD EXW SURGXFHG OHVV LQ VSULQJ $W *DLQHVYLOOH WKH ODWH VHDVRQ SURGXFWLRQ RI OLPSRJUDVV ZDV VOLJKWO\ OHVV WKDQ WKDW RI WKH GLJLWJUDVVHV DQG RQH EHUPXGDJUDVV %LJDOWD KRZHYHU KDG WKH KLJKHVW WRWDO VHDVRQ SURGXFWLRQ )UHTXHQW FOLSSLQJ GHIROLDWLRQ DW 2QD SURGXFHG ZHHG LQYDn VLRQV DQG WKH OLPSRJUDVVHV KDG LQWHUPHGLDWH \LHOGV FRPSDUHG WR RWKHU WURSLFDO SHUHQQLDO JUDVVHV 5HVXOWV IURP -D\ GHPRQVWUDWHG VORZ HVWDEn OLVKPHQW UDWHV FRPSDUHG WR &RDVWDO EHUPXGDJUDVV n7UDQVYDODn GLJLWJUDVV DQG RQH EDKLDJUDVV )URP ZRUN GRQH DW 4XLQF\ LW ZDV FRQFOXGHG WKDW OLPSRJUDVVHV DUH QRW DGDSWHG WR WKH GU\ XSODQG VRLOV RI WKH )ORULGD SDQKDQGOH $W %HOOH *ODGH &ROHPDQ DQG 3DWH 4XHVHQEHUU\ HW DO f IRXQG JRRG GLJHVWLELOLW\ DQG DFFHSWDELOLW\ RI %LJDOWD E\ EHHI DQLPDOV KRZHYHU 6W $XJXVWLQHJUDVV 6WHQHWDSKUXP VHFXQGDWXP .XQW]f ZDV EHWWHU DGDSWHG WR WKH PXFNODQG VRLOV DW WKLV VRXWK )ORULGD VLWH %LJDOWD GLG QRW SHUVLVW XQGHU KHDY\ JUD]LQJ 7KLV OHG WR WKH UHFRPPHQGDWLRQ RI URWDWLRQDO JUD]LQJ IRU %LJDOWD

PAGE 40

5RWDWLRQDO JUD]LQJ LV WRR LQWHQVLYH IRU PRVW )ORULGD SURGXFHUV ,W LV IRUWXQDWH WKDW VRPH RI WKH QHZ OLPSRJUDVV LQWURGXFWLRQV ZLOO SHUVLVW XQGHU FRQWLQXDO XVH EHFDXVH WKLV LV DQ LPSRUWDQW FULWHULRQ LQ D UDQFKHUV PLQG 3DWH SHUVRQDO FRPPXQLFDWLRQf 5XHONH f VWXGLHG 5HGDOWD *UHHQDOWD DQG %LJDOWD DQG IRXQG VLJQLILFDQW \LHOG UHVSRQVHV WR QLWURJHQ XS WR NJKD\U KRZHYHU VHYHUH ORVVHV RFFXUUHG IROORZLQJ IUHTXHQW GHIROLDWLRQ DW KLJK QLWURJHQ UDWHV HVSHFLDOO\ IRU %LJDOWD %LJDOWDnV KLJK GLJHVWLELOLW\ FRPELQHG ZLWK SRRU SHUVLVWHQFH ZDV GLVFRQFHUWLQJ 4XHVHQEHUU\ HW DO f IRXQG WKDW 3, ZDV D GLJHVWLEOH SHUVLVWHQW WHWUDSORLG 7KLV DFFHVVLRQ ZDV VRRQ LQFOXGHG LQ WKH OLPSRJUDVV VWXGLHV ZLWK %LJDOWD DQG 5HGDOWD ZKLOH *UHHQDOWD ZDV RPLWWHG IURP H[SHULPHQWDWLRQ GXH WR LWV VLPLODULW\ WR 5HGDOWD 5XHONH HW DO f LQFOXGHG WKH SURPLVLQJ DFFHVVLRQ LQ D OLPSRn JUDVV HVWDEOLVKPHQW VWXG\ 'HQVHU VWDQGV HDUOLHU SURGXFWLRQ DQG KLJKHU VHFRQG \HDU GU\ PDWWHU \LHOGV ZHUH REWDLQHG ZKHQ SODQWLQJ PDWHULDO ZDV VSULJJHG IROORZHG E\ GLVNLQJ WR SDUWLDOO\ FRYHU WKH VWHPV DQG FXOWLSDFNHG IRU ILUP FRQWDFW EHWZHHQ VSULJV DQG VRLO ,Q WKH \HDU DIWHU HVWDEOLVKPHQW 3, RXW\LHOGHG ERWK 5HGDOWD DQG %LJDOWD .UHWVFKPHU DQG 6\QGHU f FRPSDUHG JURZWK RI 5HGDOWD *UHHQDOWD DQG %LJDOWD WR 7UDQVYDOD GLJLWJUDVV 3DQJOD GLJLWJUDVV DQG &RDVWFURVV EHUPXGDJUDVV DQG IRXQG WKDW D ZHHN FXWWLQJ LQWHUYDO VHYHUHO\ GHFUHDVHG WKH DELOLW\ RI OLPSRJUDVV WR DFFXPXODWH GU\ PDWWHU )UHTXHQW FOLSSLQJ DOVR GLPLQLVKHG WKH HIILFLHQF\ RI QLWURJHQ XVDJH 'HOD\LQJ DXWXPQ IHUWLOL]DWLRQ DW )W 3LHUFH )ORULGD IURP 6HSWHPEHU RU 2FWREHU WR 2FWREHU UHVXOWHG LQ JUHDWO\ UHGXFHG IRUDJH SURGXFWLRQ ZKHQ KDUYHVWHG RQ 'HFHPEHU 7KH ODWHU QLWURJHQ IHUWLOL]DWLRQ

PAGE 41

KRZHYHU UHVXOWHG LQ D EHWWHU FRPELQDWLRQ RI \LHOG DQG TXDOLW\ E\ UDLVn LQJ WKH LQKHUHQWO\ ORZ SURWHLQ FRQWHQW LQ PDWXUH OLPSRJUDVV %LJDOWD UHVSRQGHG WR FRROVHDVRQ QLWURJHQ IHUWLOL]DWLRQ EXW JDYH ZD\ WR ZHHG HQFURDFKPHQW GXULQJ WKH VXPPHU ZKLFK OHG WR WKH UHFRPPHQGDWLRQ WKDW LW VKRXOG EH UHVWHG VRPHWLPH LQ WKH ZDUP VHDVRQ WR PDLQWDLQ SODQW SRSXODWLRQV 4XHVHQEHUU\ DQG 2FXPSDXJK f VWXGLHG FOLSSLQJ DQG JUD]LQJ GHIROLDWLRQ PHWKRGV IRU WKUHH \HDUV 7KH PRE JUD]LQJ PHWKRG VKRUWHQHG WKH WLPH QHFHVVDU\ WR DGYDQFH WKH QHZ OLPSRJUDVV JHUPSODVP WKURXJK WKH HDUO\ SKDVHV RI DJURQRPLF HYDOXDWLRQ 4XHVHQEHUU\ HW DO f VXPPDn UL]HG WKH HYHQWV OHDGLQJ WR WKH FOLSJUD]H H[SHULPHQWDWLRQ $IWHU SUHOLPLQDU\ WHVWLQJ RI FORQHV LQ JUHHQKRXVHV DQG VPDOO SORW FOLSSLQJ WULDOV ZHUH VHOHFWHG IRU HYDOXDWLRQ E\ FOLSSLQJ DQG E\ JUD]LQJ (LJKW RI WKH EHVW JHQRW\SHV ZHUH WKHQ HYDOXDWHG DW IRXU IUHTXHQFLHV RI JUD]LQJ DQG ZHHNVf 2FXPSDXJK HW DO f LGHQWLILHG 3, DV VXSHULRU WR %LJDOWD XQGHU JUD]LQJ GXH WR FRPSDUDEOH GLJHVWLELOLWLHV DQG KLJKHU SHUVLVWHQFH 0HDQZKLOH 5XHONH DQG 4XHVHQEHUU\ f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

PAGE 42

QRQVWUXFWXUDO FDUERK\GUDWH PHWDEROLVP VKRXOG FRQWULEXWH WR DQ XQGHUn VWDQGLQJ RI OLPSRJUDVV EHKDYLRU &KULVWLDQVHQ HW DO f VWXGLHG HWLRODWHG UHJURZWK DV DQ LQGLFDWRU RI WKH VWRUHG HQHUJ\ LQ IRXU LPSRJUDVVHV 6RPH RI WKH UHVXOWV LQGLFDWHG WKDW PRUSKRORJLFDO GLIIHUn HQFHV ZHUH UHODWHG WR REVHUYDWLRQV RI DJURQRPLF SHUIRUPDQFH )UHTXHQW FXWWLQJ WUHDWPHQWV FDXVHG D GUDLQ RI WKH HQHUJ\ UHVHUYHV RI %LJDOWD EXW QRW 3, 7KLV VXJJHVWHG JUHDWHU UHVHUYH HQHUJ\ VWRUDJH IRU 3, $QLPDO 5HVSRQVH DQG 6\VWHPV 0DQDJHPHQW +RGJHV DQG 3LWPDQ f VWXGLHG %LJDOWD OLPSRJUDVV &DO OLHr EHUPXGDJUDVV n6DUDVRWD VWDUJUDVV &\QRGRQ QOHPIXHQVLV 9DQGHU\VWf DQG n2QDn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f LV SUHVHQWO\ FRQGXFWLQJ WKH VHFRQG \HDU RI D WKUHH \HDU DQLPDO SURGXFWLRQ VWXG\ FRPSDULQJ EHHI JDLQV RQ 3HQVDFROD EDKLD JUDVV DQG 3, OLPSRJUDVV 3UHOLPLQDU\ UHVXOWV IURP LQGLFDWH EHWWHU DYHUDJH JDLQ SHU \HDUOLQJ KHLIHU RQ OLPSRJUDVV YV NJ\Uf DQG D FDOHQGDU GD\ DGYDQWDJH LQ JUD]LQJ GD\V FRPSDUHG WR

PAGE 43

3HQVDFROD EDKLDJUDVV 2FXPSDXJK VWDWHV WKDW 3, LV EHLQJ FRQn VLGHUHG IRU FXOWLYDU UHOHDVH LQ -XQH 1RQVWUXFWXUDO &DUERK\GUDWHV LQ )RUDJHV 3ODQWV DUH WKH SULPDU\ VRXUFH RI FDUERK\GUDWHV &HOOXORVH LV WKH PRVW SUHYDOHQW RUJDQLF FRPSRXQG RQ HDUWK DQG LV PDQnV PRVW LPSRUWDQW LQGXVWULDO FDUERK\GUDWH 7KH VWDSOH JUDLQV DUH SUHGRPLQDQWO\ VWDUFK ZKLFK LV WKH FKLHI FDUERK\GUDWH LQ WKH KXPDQ GLHW *UHHQZRRG f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n K\GUDWH UHVHDUFK LQ SDVWXUH SODQWV GXULQJ WKH SDVW \HDUV 4XDQWLWLHV RI QRQVWUXFWXUDO FDUERK\GUDWHV LQ SODQWV IOXFWXDWH GXULQJ HDFK GD\ DV ZHOO DV GXULQJ WKH VHDVRQ 9DULDWLRQV LQ JHRJUDSK\ HQYLURQPHQW WD[RQRP\ DQDWRP\ DQG PDQDJHPHQW RU H[SHULPHQWDO FRQGLn WLRQVf DOO DGG WR WKH G\QDPLFV RI FDUERK\GUDWH IOX[ 5HDFWLRQV WR PDQDJHPHQW YDU\ IURP \HDU WR \HDU GHSHQGLQJ RQ HQYLURQPHQW VSHFLHV DQG VWUHVV KHQFH WKHUH DUH FRQIOLFWV LQ WKH OLWHUDWXUH GXH WR LQFRQJUXLWLHV EHWZHHQ LPPLQHQW DQG ORQJ WHUP UHVXOWV EHWZHHQ DQLPDO DQG SDVWXUH UHTXLUHPHQWV DQG EHWZHHQ DSSOLHG DQG

PAGE 44

EDVLF REMHFWLYHV 7KUHH JHQHUDWLRQV RI UHVHDUFKHUV KDYH RIWHQ SHUSHWXn DWHG D EHQLJQ DFFHSWDQFH RI FRQYHQWLRQDO ZLVGRPV 7KH RYHUZKHOPLQJ TXDQWLW\ RI FDUERK\GUDWH VWXGLHV FUHDWHG D QHHG IRU UHYLHZ DUWLFOHV WR VXPPDUL]H WKH UHVHDUFK ILQGLQJV 0DQ\ WKHRULHV ZHUH SURSRVHG WR FRQVROLGDWH UHVHDUFK RQ GLIIHUHQW VSHFLHV JURZQ XQGHU GLIIHUHQW FRQGLWLRQV IRU GLIIHUHQW SXUSRVHV 7KH PRVW LPSRUWDQW HOXFLn GDWLRQV ZHUH f WKH GHILQLWLRQ RI ZKDW RUJDQLF VXEVWDQFHV VKRXOG EH FRQVLGHUHG UHVHUYHV f ZKDW PHWKRGV ZHUH EHVW WR VDWLVIDFWRULO\ IUDFn WLRQDWH FDUERK\GUDWH FRPSRQHQWV f ZKDW UROH FDUERK\GUDWHV SOD\ LQ UHJURZWK PHFKDQLVPV f ZKHUH SODQW IRRGV DFFXPXODWHG f KRZ QRQ VWUXFWXUDO FDUERK\GUDWHV YDULHG E\ GD\ DQG VHDVRQ DQG f KRZ FDUERn K\GUDWHV LQ VWRUDJH RUJDQV ZHUH DIIHFWHG E\ PDQDJHPHQW 7KH UHDGHU LV UHIHUUHG WR WKH IROORZLQJ SXEOLFDWLRQV DV D FKURQRORJLFDO JXLGH *UDEHU f :HLQPDQQ f 7URXJKWRQ f 0D\ DQG 'DYLGVRQ f +XQWHU HW DO f 6KHDUG f 6PLWK Ef :KLWH f DQG 1REOH DQG /RZH f )RU ZRUN SULRU WR WKH nV WKH UHDGHU LV UHIHUUHG WR WKH OLWHUDWXUH FLWDWLRQV RI *UDEHU HW DO f DQG *UDEHU f 7KH LPSRUWDQW ILQGLQJV RI &XJQDF f DUH DOOXGHG WR E\ QHDUO\ DOO FDUERK\GUDWH UHYLHZHUV +H VHSDUDWHG WKH JUDVVHV LQWR WZR JURXSV WKH IUXFWRVH DFFXPXODWLQJ JUDVVHV QDWLYH WR WHPSHUDWXUH FOLPDWHV DQG JUDVVHV DFFXPXODWLQJ VXFURVH DQG VWDUFK WKDW DUH PRVWO\ DGDSWHG WR ZDUP UHJLRQV *UDEHU f DVVHVVHG WKH FRQGLWLRQ RI ORZ RU KLJK RUJDQLF UHVHUYHV LQ SDVWXUH SODQWV E\ PHDQV RI GU\ PDWWHU \LHOGV SHUVLVWHQFH DQG ZHHG HQFURDFKPHQW +H VDZ WKDW SODQW JURZWK EHKDYLRU ZDV UHODWHG WR

PAGE 45

DYDLODEOH QXWULHQWV DQG WKDW TXDQWLWLHV RI RUJDQLF IRRG ZHUH OLNHZLVH FRUUHODWHG 7KH &KDUDFWHUL]DWLRQ RI &DUERK\GUDWH 5HVHUYHV /HXNHO DQG &ROHPDQ f LQ )ORULGD PHDVXUHG FDUERK\GUDWH IUDFn WLRQV RI EDKLDJUDVV DQG FODLPHG WKDW KHPLFHOOXORVH ZDV WUDQVIRUPHG WR OLJQLQ DQG FHOOXORVH XQGHU ORQJ FXWWLQJ LQWHUYDOV DQG UHGXFHG WR VLPSOHU VXJDUV IRU XVH LQ WLVVXH V\QWKHVLV ZLWK D IUHTXHQW GHIROLDWLRQ UHJLPH 0F&DUW\ f VWXGLHG VHDVRQDO FDUERK\GUDWH IOXFWXDWLRQV LQ VHYHUDO UDQJH JUDVVHV DQG FRQFOXGHG WKDW VXFURVH DQG VWDUFK ZHUH WKH VWRUHG IRRGV LQ &DOLIRUQLD EURPHJUDVV %URPXV FDULQDWXV +RRN DQG $UQf 0F&DUW\ f DOVR FRQMHFWXUHG WKDW KHPLFHOOXORVH PD\ KDYH EHHQ FRQn YHUWHG LQWR VLPSOHU FRPSRQHQWV 6XOOLYDQ DQG 6SUDJXH f VKRZHG QR KHPLFHOOXORVH XWLOL]DWLRQ LQ WKH UHJURZWK PHFKDQLVP RI U\HJUDVV /ROLXP SHUHQQH /f )UXFWRVDQ ZDV WKH NH\ UHVHUYH VXEVWDQFH DQG WKHVH UHVXOWV ZHUH UHLQIRUFHG E\ :DLWH DQG %R\G D Ef DQG :DLWH f 7KH FRQFHSW WKDW KHPLFHOOXORVH SDUWLFLSDWHG LQ UHVSLUDWLRQ RU WLVVXH V\QWKHVLV ZDV HVVHQWLDOO\ GLVn FDUGHG E\ WKH ODWH nV KRZHYHU IUXFWRVDQ VWRUDJH FKDUDFWHUL]DWLRQ LQ LPSRUWDQW QRUWKHUQ DGDSWHG JUDVVHV FRQWLQXHG 6SUDJXH DQG 6XOOLYDQ :DLWH DQG *RUURG 2NDMLPD DQG 6PLWK 6PLWK DQG *URWHOXHVFKHQ 6PLWK *URWHOXHVFKHQ DQG 6PLWK 6PLWK f 6PLWK f FDWDORJHG WKH FDUERK\GUDWH VWRUDJH WHQGHQFLHV RI PDQ\ 1RUWK $PHULFDQ JUDVVHV 7URSLFDO OHJXPH DQG JUDVV FDUERK\GUDWH FKDUDFWHUL]DWLRQ RFFXUUHG ODWHU DQG UHPDLQV DQ DFWLYH DUHD RI UHVHDUFK +XQWHU HW DO f IRXQG QR IUXFWRVDQ DFFXPXODWLRQ LQ WKH WURSLFDO SODQWV KH VWXGLHG

PAGE 46

1REOH DQG /RZH f VKRZHG VPDOOHU VHDVRQDO YDULDWLRQ RI DOFRKRO VROXEOH FDUERK\GUDWHV LQ WURSLFDO JUDVVHV WKDQ LQ WHPSHUDWH JUDVVHV DQG :LOVRQ DQG )RUG f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f UHYLHZHG HWLRODWHG UHJURZWK VWXGLHV DQG KLV ROGHVW UHIHUn HQFH LV WKDW RI %XUWRQ DQG -DFNVRQ f KRZHYHU WKH UHJURZWKLQn GDUNQHVV WHFKQLTXH JRHV PXFK IXUWKHU EDFN LQ WLPH $FFRUGLQJ WR 6PLWK SHUVRQDO FRPPXQLFDWLRQf 7KH ILUVW XVH RI JURZWK LQ GDUNQHVV WKDW NQRZ RI LV WKH HDUO\ ZRUN RI / ) *UDEHU f KDYH DOZD\V EHHQ LQWULJXHG ZLWK WKH WHFKQLTXH SUREDEO\ IURP EHLQJ D VWXGHQW RI 'U *UDEHUnV :KHUH *UDEHU JRW WKH LGHD GR QRW NQRZ EXW KH GLG KLV 3K' ZRUN ZLWK 'U .UDXV D ERWDQLVW DW WKH 8QLYHUn VLW\ RI &KLFDJR ZKR ZRUNHG D JUHDW GHDO ZLWK FDUERK\GUDWH QLWURJHQ UDWLRV LQ WKH JURZWK RI WRPDWRHV DQG ZKR PD\ KDYH XVHG GDUNQHVV VWXGLHV 7KH ILUVW SHUVRQ WR XVH JURZWK LQ GDUNQHVV RQ JUDVVHV ZDV 9DQFH 6SUDJXH 6XOOLYDQ DQG 6SUDJXH f DW WKH 86'$ 3DVWXUH /DE LQ 3HQQV\OYDQLD +H SLFNHG XS WKH WHFKQLTXH IURP *UDEHU ZKHQ KH ZDV KLV 3K' VWXGHQW DQG %ODVHU JRW WKH LGHD IURP 6SUDJXH :DUG DQG %ODVHU f 3UHWUHDWPHQW LQ GDUNQHVV WR YDU\ FDUERK\GUDWH FRQFHQWUDWLRQ ZDV XVHG E\ 'DYLGVRQ DQG 0LOWKRUSH Ef DQG RWKHUV WKDW XVHG WKH UHJURZWKLQGDUNQHVV H[SHULPHQWDWLRQ ZHUH $GHJEROD f $GHJEROD DQG

PAGE 47

0F.HOO Ef $OEHUGD f +XPSKUH\V DQG 5RELQVRQ f 5£HVH DQG 'HFNHU f 0DWFKHV f :DWVRQ DQG :DUG f DQG &KULVWLDQVHQ HW DO f &KULVWLDQVHQ HW DO f IRXQG LPSURYHG VHQVLWLYLW\ LQ VHSDUDWLQJ WUHDWPHQW GLIIHUHQFHV E\ XVLQJ VWHP EDVH ZHLJKWV DV D FRYDULn DEOH WR FRQWURO YDULDWLRQV LQ SODQW VL]H &RQFHUQV RYHU WKH XQUHOLDELOLW\ RI FKHPLFDO IUDFWLRQV ZHUH GXH WR WKH XVH RI DFLGV LQ HDUO\ DSSOLFDWLRQV RI FDUERK\GUDWH PHWKRGRORJ\ )UXFWRVDQV DUH ZDWHU DQG HWKDQROVROXEOH DV ZHOO DV UHDGLO\ K\GURO\]HG WR IUXFWRVH PRQRPHUV E\ DFLG WUHDWPHQW &RQVHTXHQWO\ 6XOOLYDQ DQG 6SUDJXH f FRPSODLQHG WKDW DFLG DQDO\VHV RI VWDUFK ZHUH FRQIRXQGHG E\ WKH FRQWULEXWLRQ RI IUXFWRVH IURP IUXFWRVDQ LQ ZLWK JOXFRVH IURP VWDUFK LQ WHVWV RI UHGXFLQJ SRZHU ,Q :HLQPDQQ SXEOLVKHG D PHWKRG IRU WRWDO DYDLODEOH FDUERn K\GUDWH 7$&f GHWHUPLQDWLRQ LQ SODQWV 7RWDO DYDLODEOH FDUERK\GUDWH ZDV GHILQHG DV DOO WKRVH FDUERK\GUDWHV ZKLFK FDQ EH XVHG LQ WKH SODQW ERG\ DV D VRXUFH RI HQHUJ\ RU DV D EXLOGLQJ PDWHULDO HLWKHU GLUHFWO\ RU LQGLUHFWO\ DIWHU KDYLQJ EHHQ EURNHQ GRZQ E\ HQ]\PHV S f ,Q :HLQPDQQnV PHWKRG VPDOO VDPSOHV RI ILQHO\ JURXQG DLUGU\ PDWHULDO DUH GLJHVWHG E\ WDNDGLDVWDVH LQ ZDWHU UHVXOWLQJ LQ WKH EUHDNGRZQ RI VWDUFK GH[WULQV DQG PDOWRVH WR JOXFRVH ZKLOH RWKHU VXJDUV DQG IUXFWRVDQ DUH VROXELOL]HG DW WKH VDPH WLPH 7KH ODWWHU FRPSRXQGV DUH FRQYHUWHG WR KH[RVH VXJDUV E\ DFLG K\GURO\VLV IROORZLQJ ZKLFK WKH UHGXFLQJ SRZHU RI WKH FOHDUHG QHXWUDOL]HG K\GURO\VDWH LV GHWHUPLQHG :HLQPDQQ f /LQGDKO HW DO f PDGH VOLJKW PRGLILFDWLRQV LQ WKH :HLQPDQQ 7$& PHWKRG DQG KDLOHG WKH HIILFDF\ RI WKH SURFHGXUH LQ DJURQRPLF DSSOLFDWLRQV 7KHLU PDMRU PRGLILFDWLRQ ZDV WR FKDQJH WKH HQ]\PH IURP GLDO\]HG WDNDGLDVWDVH WR FODUDVHD KLJKO\ SXULILHG DQG FRQFHQWUDWHG

PAGE 48

IRUP RI WDNDGLDVWDVH FRQWDLQLQJ LQYHUWDVH PDOWDVH DQG DP\ODVH 7KLV SURFHGXUH VDWLVILHG PRVW DJULFXOWXUDO QHHGV IRU 7$& GHWHUPLQDWLRQ WKURXJKRXW WKH nV KRZHYHU 6PLWK FRQWLQXHG WR H[DPLQH DFLG WHFKn QLTXHV IURP KLV ODE LQ :LVFRQVLQ 7KH DFLG PHWKRGV ZHUH IDVWHU UHTXLU LQJ KRXUV WR FRPSOHWH DQDO\VLV FRPSDUHG WR KRXUV ZRUNLQJ WLPH SOXV KRXUV LQFXEDWLRQ IRU WKH HQ]\PH PHWKRG 0RVW RI WKH KLQGUDQFH RI XVLQJ DQ DFLG PHWKRG ZDV LQ ILQGLQJ WKH SURSHU FRQFHQWUDWLRQV 6PLWK HW DO f YDULHG VXOIXULF DFLG FRQFHQ WUDWLRQV IURP 1B 1B DQG FRPSDUHG WKH UHVXOWV WR WDNDGLDVWDVH DV D VWDQGDUG 7KH 1B )A62r PHWKRG PRVW QHDUO\ GXSOLFDWHG WKH WDNDGLDVWDVH H[WUDFWLRQ ZKLOH KLJKHU FRQFHQWUDWLRQV GHJUDGHG KHPLFHOOXORVH 7KH UHVXOWV ZHUH VXSSRUWHG E\ %XUULV HW DO f ZKR DGGHG WKDW WKH FKDQFHV RI REWDLQLQJ HUURQHRXV GDWD ZLWK DFLGV ZHUH JUHDWHVW ZKHQ WKH\ DQDO\]HG EHUPXGDJUDVV GXULQJ UDSLG JURZWK SKDVHV DQG ZLWK WLVVXH KLJK LQ VWDUFK *URWHOXHVFKHQ DQG 6PLWK f TXDOLILHG HDUOLHU ZRUN 6PLWK HW DO f DIWHU H[DPLQLQJ DOIDOID 0HGLFDJR VDWLYD /f KLJK LQ VWDUFK FRQWHQW )RU IUXFWRVDQ DFFXPXODWLQJ WLVVXH VXFK DV WLPRWK\ 3KOHXP SUDWHQVH /f FRQWDLQLQJ OLWWOH VWDUFK GLOXWH DFLG 1B )A62Af SURFHGXUHV PD\ EH XVHG IRU QRQVWUXFWXUDO FDUERK\GUDWH H[WUDFWLRQ 2Q WKH RWKHU KDQG ZLWK WLVVXH KLJK LQ VWDUFK VXFK DV OHJXPHV DQG WURSLFDO JUDVVHV DFLG VWUHQJWKV EHJLQQLQJ DW 1B )A62A GHJUDGHG KHPLFHOOXORVH DQG LQFRPSOHWHO\ K\GURO\]HG VWDUFK DV ZHOO *UHXE DQG :HGLQ f VXSSRUWHG *URWHOXHVFKHQ DQG 6PLWKnV ILQGLQJV DQG DOVR ZDUQHG WKDW DERYH 1 )A62A IUHH IUXFWRVH RU IUXFWRVH OLEHUDWHG IURP IUXFWRVDQ ZDV EHLQJ GHVWUR\HG 7KH HQ]\PDWLF

PAGE 49

PHWKRG ZDV PRUH DFFXUDWH GHPRQVWUDWHG DJDLQ DW D ODWHU GDWH LQ 3RUWXJDO E\ &KDYHV DQG 0RUHLUD f ,Q 6PLWK ZURWH KLV ZLGHO\ DFFHSWHG YHUVLRQ RI WKH :HLQPDQQ /LQGDKO SURFHGXUH ,W ZDV DQG VWLOO LV WKH PRVW SRSXODU WLWULPHWULF QRQVWUXFWXUDO FDUERK\GUDWH PHWKRG 6PLWK VXJJHVWHG WKH WHUP WRWDO QRQ VWUXFWXUDO FDUERK\GUDWH 71&f DV D PRUH FOHDUO\ GHILQDEOH WHUP WKDQ 7$& WR ERWK SODQW DQG DQLPDO LQYHVWLJDWRUV 7KH DGYDQWDJH RI WKH 6PLWK PHWKRG ZDV LWV RUJDQL]HG FRPSOHWH FRRNERRN SUHVHQWDWLRQ ZKLFK LQFOXGHG DQ LQIRUPDWLRQ UHYLHZ DSSHQGLFHV FULWHULD IRU VHOHFWLQJ D PHWKRG DQG VDPSOH SUHSDUDWLRQ 6PLWK Ef UHYLVHG WKH 71& PHWKRG ZLWK PLQRU PRGLILFDWLRQV LQFOXGLQJ DQ HQ]\PH FKDQJH ,W ZDV IRXQG WKDW 0\ODVH KDV UDSLG VDFFKDURJHQLF DFWLYLW\ DQG ZLOO FRPSOHWHO\ GLJHVW WKH VWDUFK LQ WLVVXHV FRQWDLQLQJ SHUFHQW RU OHVV RI 71& LQ KRXUV $GYDQFHV LQ FDUERK\GUDWH PHWKRGRORJ\ LQ WKH SDVW \HDUV KDYH EHHQ FHQWHUHG DURXQG WKH VKLIW WRZDUG FRORULPHWU\ WR LQFUHDVH VSHHG VDPSOH QXPEHU DQG HIILFLHQF\ +DVOHPRUH DQG 5RXJKDQ :HLHU HW DO GD6DOYHLUD HW DO :HVWKDIHU HW DO f ,Q FRORULPHWULF PHWKRGV VSHFLILF FRORU UHDFWLRQV IRU SRUWLRQV RI FDUERK\GUDWH PROHFXOHV DUH GXH WR IRUPDWLRQ RI IXUIXUDO RU IXUIXUDO KRPRORJXHV LQ VWURQJ DFLGV HVSHFLDOO\ IROORZLQJ KHDWLQJ 7KHVH IXUDQV RU WKHLU UHDFWLRQ SURGXFWV DUH GHULYHG IURP R[LGDWLRQ UHGXFWLRQ RU FRQGHQVDWLRQ SURFHVVHV LQ VWURQJ DFLG DQG FDQ IRUP FRORUHG SURGXFWV XSRQ UHDFWLRQ ZLWK VXJDUV 7KHUH DUH WUHDWPHQGRXV YDULDELOLWLHV DVVRFLDWHG ZLWK WKH UHDFWLRQV GHSHQGLQJ RQ WKH VXJDU UHDJHQW FRQFHQWUDWLRQ WHPn SHUDWXUH DQG WLPH RI KHDWLQJ 'LVFKH f 7KH VHDUFK IRU VSHFLILF TXDQWLWDWLYH DQG UHSURGXFLEOH FRORULPHWULF DVVD\V IRU WKH GHWHUPLQDWLRQ RI FDUERK\GUDWHV LV DQ LPSRUWDQW DUHD RI UHVHDUFK

PAGE 50

&RORULPHWULF WHVWV KDYH DGYDQWDJHV RYHU WLWULPHWULF SURFHGXUHV GXH WR WKHLU VSHHG DQG HTXLYDOHQW SUHFLVLRQ 7ZR GLVWLQFW VWHSV LQYROYHG LQ FRORULPHWULF UHDFWLRQV DUH f IRUPDWLRQ RI D FKURPRFHQ IURP WKH VXJDU DQG f GHYHORSPHQW RI WKH FRORU E\ D FRQGHQVDWLRQ RI WKH FKURPRJHQ ZLWK D VSHFLILF UHDJHQW $PLQRII HW DO f ,W LV LPSRUn WDQW WR UHFRJQL]H WKDW PRVW FRORULPHWULF PHWKRGV DUH HPSLULFDO 6LQFH OLWWOH LV NQRZQ DERXW WKH UHDFWLRQ PHFKDQLVP RU WKH H[DFW QDWXUH RI WKH FKURPRJHQ LQYROYHG DEVROXWH VWRFKLRPHWU\ LV QRW RIWHQ REWDLQHG $V ORQJ DV WKH UHVXOWV REH\ %HHUnV /DZ ZLWKLQ DSSURSULDWH OLPLWV RI FRQFHQn WUDWLRQ WKH SUREOHP FDQ EH VROYHG E\ XVLQJ DSSURSULDWH LQWHUQDO VWDQGDUGV DQG EODQNV ,Q TXDQWLWDWLYH DQDO\VLV WZR GLYHUJHQW REMHFWLYHV PXVW EH FRQVLGHUHG f DQ RYHUDOO DQDO\VLV RI DOO WKH VXJDUV SUHVHQW E\ D YHU\ JHQHUDO UHDFWLRQ DQG f WKH VHOHFWLYH GHWHUPLQDWLRQ RI RQH VXJDU LQ WKH SUHVHQFH RI RWKHUV )RU H[DPSOH :HVWKDIHU HW DO f IRXQG VXFURVH OHYHOV LQ WXUIJUDVV URRW WLVVXH WR KDYH WKH PRVW GUDPDWLF UHVSRQVH WR QLWURJHQ WUHDWPHQW WKXV D 71& PHWKRG ZRXOG EH OHVV VHQVLn WLYH PHDVXUH RI FDUERK\GUDWH FRQFHQWUDWLRQV FKDQJHV WKDQ D WHVW RI VXFURVH 7KH SULQFLSDO FKHPLFDO PHWKRGV IRU TXDQWLWDWLYH PHDVXUHPHQW RI VXJDUV XVH WKH DFWLRQ RI VXJDU UHGXFWLRQ RQ DONDOLQH VROXWLRQV RI WKH VDOWV RI FHUWDLQ PHWDOV 7KH PRVW H[WHQVLYHO\ XVHG PHWDO LQ VXJDU DQDO\VLV LV FRSSHU 0RVW RI WKH DJULFXOWXUDO QHHGV IRU TXDQWLWDWLYH VXJDU DQDO\VLV KDYH EHHQ VDWLVILHG ZLWK WKH WLWULPHWULF PHWKRG RI 6RPRJ\L f DQG WKH IROORZLQJ FRORULPHWULF WHFKQLTXHV SKHQRO VXOIXULF 'XERLV HW DO f DQWKURQH DQG 1HOVRQnV f WHVW IRU

PAGE 51

UHGXFLQJ VXJDUV $Q H[FHOOHQW GLVFXVVLRQ RI UHGXFLQJ VXJDU WHFKQLTXHV LV JLYHQ E\ +RGJH DQG +RIUHLWHU f 2UJDQLF 6XEVWDQFHV 8VHG LQ 5HJURZWK 3HUKDSV WKH ELJJHVW FRQWURYHUV\ LQ WKH KLVWRU\ RI QRQVWUXFWXUDO FDUERK\GUDWH UHVHDUFK ZDV FRQFOXVLYHO\ SURYLQJ WKDW IRRG UHVHUYHV ZHUH XVHG LQ WKH V\QWKHVLV RI QHZ JURZWK 0D\ DQG 'DYLGVRQ f VWDWHG WKDW D JHQHUDO DFFHSWDQFH RI WKH LPSRUWDQFH RI FDUERK\GUDWH UHVHUYHV LQ UHJHQHUDWLRQ VHHPHG XQMXVWLILDEOH DQG DUJXHG WKDW QR FDXVDO UROH LQ VKRRW UHJURZWK ZDV VXJJHVWHG E\ GHFUHDVHV RI FDUERK\GUDWH LQ VWRUDJH RUJDQV 0D\ f GLVPLVVHG DV FRQYHQWLRQDO ZLVGRP WKH *UDEHU DQG :HLQPDQQ GHILQLWLRQV RI UHVHUYHV RQ WKH EDVLV WKDW WKH WHUP KDG EHFRPH VHPLWHFKQLFDO DQG QR ORQJHU VXEMHFW WR FULWLFLVP 0D\ f DOVR FLWHG $UFKEROG f DQG %HUQDWZRLF] f ZKR GLVFDUGHG WKH LGHD RI VWRUHG VXJDUV DV D SXUSRVLYH UHVHUYHUHVHUYH FRQQRWDWLQJ SURYLVLRQ IRU WKH IXWXUH DQG DV VXFK VLJQLI\LQJ WHOHRORJLFDO WKLQNLQJ $ EHWWHU GHILQLWLRQ ZDV DFFXPXODWH VLQFH LW ZDV QRQFRPPLWWDO FRQFHUQLQJ SXUn SRVH RU LQWHQW &DUERQ EDODQFH DQG ODEHOLQJ VWXGLHV SXW DQ HQG WR WKH FRQWURYHUV\ 0DUVKDOO DQG 6DJDU $OEHUGD 'DYLGVRQ DQG 0LOWKRUSH D (KDUD HW DO :DUGODZ :DWVRQ DQG :DUG 6KHDUG f 'DYLGVRQ DQG 0LOWKRUSH Df GHPRQVWUDWHG WKDW QLWURJHQRXV RUJDQLF FRPSRXQGV ZHUH DOVR PRELOL]HG XQGHU KHDY\ VWUHVV WR WKH SODQW WKHUHn IRUH FDUERK\GUDWHV ZHUH RQO\ SDUW RI WKH ODELOH SRRO XVHG IRU UHVSLUDn WLRQ DQG QHZ JURZWK

PAGE 52

/RFDWLRQ 6HDVRQDO DQG 'DLO\ )OXFWXDWLRQ RI 5HVHUYH )RRGV 1RQVWUXFWXUDO FDUERK\GUDWHV PD\ EH WHPSRUDULO\ VWRUHG LQ DOO SODQW SDUWV 7URXJKWRQ f FRQFOXGHG WKDW WKH PDMRU VWRUDJH UHJLRQV ZHUH LQ WKH XQGHUJURXQG RUJDQV 0DQ\ RWKHU VWXGLHV KRZHYHU KDYH VKRZQ WKDW PDMRU VWRUDJH SDUWV DUH VWHP EDVHV LQFOXGLQJ VWRORQV FRUPV DQG UKL]RPHV 6XOOLYDQ DQG 6SUDJXH %DNHU DQG *DUZRRG f :DLWH DQG %R\G D Ef DQG 6PLWK f IRXQG SHUFHQWDJHV RI IUXFWRVDQV DQG PRVW VXJDUV WR EH KLJKHU LQ WKH VWHPV WKDQ LQ OHDYHV LQ WHPSHUDWH JUDVVHV )RU WKUHH WURSLFDO VSHFLHV +XQWHU HW DO f IRXQG D KLJKHU FRQFHQWUDWLRQ RI UHGXFLQJ DQG WRWDO VXJDUV LQ WKH VWHPV WKDQ OHDI EODGHV SOXV VKHDWK 3HUU\ DQG 0RVHU f UHSRUWHG RQ WKH 71& FRQWHQW RI HLJKW UDQJH JUDVVHV DQG VWUHVVHG WKH LPSRUWDQFH RI ORFDWLQJ WKH VSHFLILF VWRUDJH RUJDQVf RI D JUDVV EHIRUH SURFHHGLQJ ZLWK 71& DQDO\VLV 7KH SURGXFWV RI SKRWRV\QWKHVLV PD\ EH KHOG LQ WKH OHDI EODGHV DV VXFURVH RU VWDUFK 6LQFH VXFURVH LV WKH PDLQ WUDQVRFDWDEOH VXJDU DQG VWDUFK LV VWRUHG LQ FKORURSODVWV LW H[SODLQV WKH IUHTXHQW PHDVXUHPHQW RI KLJK DPRXQWV RI WKHVH VXEVWDQFHV LQ OHDYHV *UHHQILHOG HW DO f HVSHFLDOO\ ZLWK FRRO WHPSHUDWXUHV *DUUDUG DQG :HVW &DUWHU DQG *DUUDUG f 7KH HQYLURQPHQW JRYHUQV VHDVRQDO DVVLPLODWH GLVWULEXWLRQ DQG FDUERn K\GUDWH PHWDEROLVP LV JUHDWO\ DIIHFWHG E\ WHPSHUDWXUH 7KH DFFXPXODWLRQ RI FDUERK\GUDWHV DW ORZ WHPSHUDWXUHV LQGLFDWHV WKDW JURZWK UDWHV DUH PRUH DIIHFWHG WKDQ DUH SKRWRV\QWKHWLF UDWHV DQG WKHUH LV QR FRQFOXVLYH HYLGHQFH WKDW UHGXFHG WUDQVORFDWLRQ LV HYHU WKH SULPDU\ FDXVH RI OLPLWn LQJ JURZWK XQGHU ORZ WHPSHUDWXUHV :DUGODZ f +HQFH IRU ERWK WHPSHUDWH DQG WURSLFDO VSHFLHV FDUERK\GUDWH DFFXPXODWLRQ RFFXUV LQ WKH DXWXPQ RU FRRO VHDVRQ +RZHYHU LQ WHPSHUDWH DUHDV DXWXPQ WXUQV WR

PAGE 53

IUHH]LQJ ZLQWHUV DQG 71& UHVHUYHV DUH QHFHVVDU\ IRU VXUYLYDO XQWLO VSULQJ 2QFH SHUHQQLDO IRUDJHV EHJLQ JURZWK LQ WKH VSULQJ 71& OHYHOV JHQHUDOO\ LQFUHDVH WKURXJK YHJHWDWLYH VWDJHV WR DQWKHVLV DQG ODWHU WKH FDUERK\GUDWH UHVHUYHV GHFOLQH VORZO\ WKURXJK WKH VXPPHU ZKHQ KRW QLJKW WHPSHUDWXUHV FDXVH KLJK UDWHV RI UHVSLUDWLRQ 6WXGLHV WR VXSSRUW WKH DERYH K\SRWKHVLV IRU WHPSHUDWH JUDVVHV ZHUH SXEOLVKHG E\ :DLWH DQG %R\G Ef %DNHU DQG *DUZRRG f 7UOLFD DQG &RRN f 6PLWK f DQG 0LVOHY\ HW DO f )RU PDQ\ WURSLFDO JURZLQJ JUDVVHV FDUERK\GUDWH DFFXPXODWLRQ RFFXUV GXULQJ WKH FRRO RU GU\ VHDVRQ DQG FDUERK\GUDWH GUDLQ LV PRVW LQWHQVH GXULQJ WKH VXPPHU GXH WR KLJK QLJKW WHPSHUDWXUHV DQG YHU\ DFWLYH JURZWK ,W FRXOG EH VSHFXODWHG WKDW SODQW VXUYLYDO LV GHSHQGHQW XSRQ FDUERK\n GUDWH DFFXPXODWLRQV GXULQJ WKH FRRO VHDVRQV ,Q QRUWKHUQ )ORULGD IRU LQVWDQFH IURVW ZLOO NLOO PRVW DERYH JURXQG KHUEDJH EXW ZDUP GD\ WHPn SHUDWXUHV DQG DGHTXDWH GD\ OHQJWKV SHUPLW EDVDO OHDI JURZWK LQ VSHFLHV VXFK DV OLPSRJUDVV *DVNLQV DQG 6OHSHU f 6WXGLHV VKRZLQJ WKH DFFXPXODWLRQ RI FDUERK\GUDWH LQ WURSLFDO JUDVV VSHFLHV GXULQJ WKH FRRO VHDVRQ ZHUH UHSRUWHG E\ :RRGV HW DO f )HUUDULV f DQG :LOVRQ DQG Wn0DQQHWMH f 'DLO\ FDUERK\GUDWH IOXFWXDWLRQV RFFXU IRU DOO VSHFLHV EXW WR GLIn IHUHQW H[WHQWV +ROW DQG +LOVW f VKRZHG WKDW EURPHJUDVV %URPXV LQHUPXV /H\VVf XWLOL]HG DOPRVW RQHWKLUG RI WKH 71& LQ WKH KHUEDJH GXULQJ WKH QLJKW EXW GLXUQDO IOXFWXDWLRQV ZHUH OHVV IRU RWKHU JUDVVHV *UHHQILHOG DQG 6PLWK f VWXGLHG VZLWFKJUDVV 3DQLFXP YLUJDWXP /f DQG IRXQG WKDW WKH GLXUQDO WUHQG ZDV DQ LQFUHDVH RI WRWDO VXJDUV DQG VWDUFK IURP DP WR SP DQG WKHQ D GHFUHDVH WR PLGQLJKW %DVDO VKHDWKV DQG LQWHUQRGHV WHQGHG WR LQFUHDVH LQ SHUFHQW VWDUFK DQG 71& IURP DP

PAGE 54

WR PLGQLJKW 6LQFH WKHVH DUH VWRUDJH SDUWV FDUERK\GUDWHV ZHUH SUHVXPn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f IHZ PDQDJHPHQW V\VWHPV DUH EDVHG XSRQ WKHVH ILQGLQJV &DUERK\GUDWH DQDO\VHV DORQH FDQQRW XQDPELJRXVO\ LGHQWLI\ D VXSHULRU PDQDJHPHQW UHJLPH EHFDXVH RI FRQIRXQGLQJ YDULDWLRQV LQ UHVLGXDO OHDI DUHD FURZQ VWUXFWXUH D[LOODU\ EXG QXPEHU OHDI DJH DQG DOWHUHG URRW FKDUDFWHULVWLFV +XPSKUH\V f 0XFK HIIRUW ZDV H[SHQGHG LQ WKH XQGHUVWDQGLQJ RI UHVLGXDO OHDI DUHD OLJKW LQWHUFHSWLRQ DQG FDUERK\GUDWH UHVHUYHV LQ H[SODLQLQJ UHJURZWK IROORZLQJ GHIROLDWLRQ :DUG DQG %ODVHU 3HDUFH HW DO 'DYLGVRQ DQG 0LOWKRUSH E +XPSKUH\V DQG 5RELQVRQ f :KHQ PRLVWXUH DQG QXWULHQWV DUH LQ DGHTXDWH VXSSO\ UHVLGXDO OHDI DUHD LV JHQHUDOO\ PRUH LPSRUWDQW WKDQ IRRG UHVHUYHV KRZHYHU WKLV FRQFHSW

PAGE 55

RI H[SODLQLQJ UHJURZWK LV GHSHQGHQW XSRQ WKH LQWHQVLW\ RI GHIROLDWLRQ WKH SUHVHQFH RU DEVHQFH RI EXGV WKH DJH RI OHDI WLVVXH DQG VSHFLHV GLIIHUHQFHV )XUWKHU WKHUH LV QR FOHDU UHODWLRQVKLS EHWZHHQ UDWH RI JURZWK DQG OHDI DUHD LQGH[ +HQFH WKLV LV DQ RYHUVLPSOLILHG PRGHO RQ ZKLFK WR EDVH D PDQDJHPHQW V\VWHP 0LOWKRUSH DQG 'DYLGVRQ f 7KH UROH RI KRUPRQHV LQ UHOHDVLQJ DSLFDO GRPLQDQFH GXULQJ IUHTXHQW GHIROLDWLRQ LV RIWHQ QHJOHFWHG &HUWDLQO\ VRPH SODQWV KDYH D EHWWHU DELOLW\ WR DGRSW D SURVWUDWH JURZWK KDELW ZLWK PDQ\ VPDOO OHDI EODGHV WKDW PDLQWDLQ DVVLPLODWH VXSSO\ WR WKH SODQW 7KH FODVVLFDO VWXG\ RI EHUPXGDJUDVV E\ :HLQPDQQ DQG *ROGVPLWK f DV UHYLHZHG E\ :HLQPDQQ f FRPHV WR PLQG &ORVH FXWWLQJ RI D ZHOO IHUWLOL]HG JUHHQ RI &\QRGRQ GDFW\O RQ WLPHV LQ D VHDVRQ GLG QRW UHVXOW LQ 7$& GHSOHWLRQ GXH WR KLJK UHVLGXDO OHDI DUHD KRZHYHU FRPSOHWH GHIROLDWLRQ E\ PHDQV RI VFLVVRUV UHSHDWHG DW ZHHNO\ LQWHUYDOV QHDUO\ H[KDXVWHG 7$& UHVHUYHV *UDEHU f DQG /HXNHO DQG &ROHPDQ f UHFRJQL]HG WKH DELOLW\ RI SODQWV WR GHIHQG DJDLQVW IUHTXHQW GHIROLDWLRQ E\ DOWHUHG PRUSKRORJ\ 1LWURJHQ IHUWLOL]DWLRQ ZLOO VWLPXODWH KHUEDJH JURZWK DQG LQ JHQHUDO ZLOO FDXVH D UHGXFWLRQ RI FDUERK\GUDWH UHVHUYHV DV WKH\ DUH XVHG DV FDUERQ VNHOHWRQV IRU SURWHLQ V\QWKHVLV :DLWH $OH[DQGHU DQG 0F&ORXG &ROE\ HW DO $GHJEROD DQG 0F.HOO D $OEHUGD $XGD HW DO *DOODKHU DQG %URZQ :LOVRQ DQG Wn0DQQHWMH f 1LWURJHQ IHUWLOL]DWLRQ FDQ FDXVH LQFUHDVHV RU GHFUHDVHV LQ FDUERn K\GUDWH VWRUDJH GHSHQGLQJ RQ WKH DPRXQW DSSOLHG DQG WLPH RI VDPSOLQJ 6SUDJXH DQG 6XOOLYDQ f &DUERK\GUDWH UHVHUYHV DUH RQO\ XWLOL]HG IRU D VKRUW WLPH IROORZLQJ GHIROLDWLRQ GD\V'DYLGVRQ DQG 0LOWKRUSH

PAGE 56

D E GD\Vf§(KDUD HW DO GD\V6XOLYDQ DQG 6SUDJXH f 0RGHUDWH QLWURJHQ IHUWLOL]DWLRQ SURPRWHV JURZWK SKRWRn V\QWKHVLV DQG 71& VWRUDJH WKHUHIRUH VDPSOLQJ WRR ODWH LQ QRQVWUHVVHG SODQWV ZLOO QRW VKRZ D FDUERK\GUDWH GHFOLQH ,Q IUHTXHQWO\ FXW DQG KLJKO\ IHUWLOL]HG VZDUGV FDUERK\GUDWH GUDLQ IURP VWRUDJH RUJDQV FDQ FRQWLQXH WR WKH GHWULPHQW RI SODQW SHUVLVWHQFH $OEHUGD f UHSRUWHG WKH GHDWK RI WLOOHUV IROORZLQJ VHYHUH GHIROLDn WLRQ VWUHVV DQG VWDWHG ,W PD\ EH VXSSRVHG WKDW D FRQVLGHUDEOH SDUW RI WKHVH WLOOHUV LV EURNHQ GRZQ DQG WUDQVORFDWHG WR WKH UHPDLQLQJ WLOOHUV WR EH XVHG IRU QHZ OHDI IRUPDWLRQ EXW WKLV KDV QRW EHHQ SURYHQ S f 7KH UDWH RI GHJUDGDWLRQ RI D VWUHVVHG VZDUG LV DFFHOHUDWHG RQFH D VWDWH RI WLOOHU GHJUDGDWLRQ RFFXUV $V QLWURJHQ IHUWLOL]DWLRQ IRUFHV KHUEDJH JURZWK DQG IUHTXHQW FXWWLQJ GLVDOORZV IXOO OHDI H[SDQVLRQ OLWWOH DVVLPLODWH LV PRELOL]HG WR URRWV DQG EXGV :DUGODZ f (QHUJ\ UHDFKLQJ WKH URRWV LV LQDGHTXDWH WR PHHW WKH QHHGV RI UHVSLUDn WLRQ URRW JURZWK DQG DEVRUSWLRQ KDYH VORZHG RU VWRSSHG DQG URRW GHJUDn GDWLRQ RFFXUV DV RWKHU VXEVWDQFHV DUH VFDYHQJHG DV D ODVW GHIHQVH 'DYLGVRQ DQG 0LOWKRUSH Df )XUWKHU FRPSOLFDWLRQV LQ WKLV WKHRUHWLFDO H[DPSOH RI KLJK SODQW VWUHVV DUH LPSRVHG E\ DJJUHVVLYH ZHHGV FRPSHWLQJ IRU QXWULHQWV PRLVn WXUH DQG OLJKW 'XH WR WKH FRPELQHG HIIHFWV RI DOO WKH DERYH IDFWRUV WKH ERWDQLFDO FRPSRVLWLRQ RI WKH GHVLUHG VSHFLHV GHFOLQHV DQG WKH VZDUG GHJHQHUDWHV ,Q FRQFOXVLRQ ZKHQ 71& VWXGLHV DUH XVHG DV D VLQJOH WRRO FRPSOHPHQWHG ZLWK RWKHU LQIRUPDWLRQ WKH\ JUHDWO\ DLG LQ XQGHUVWDQGLQJ WKH WRWDO G\QDPLFV RI SODQW EHKDYLRU +RZHYHU HPSKDVLV PXVW EH SODFHG RQ D EDODQFHG XQGHUVWDQGLQJ RI DOO IDFWRUV LQYROYHG LQ SODQW JURZWK LI

PAGE 57

LW LV WKH REMHFWLYH RI UHVHDUFK WR OHDG WR VRXQG PDQDJHPHQW UHFRPPHQGDWLRQV

PAGE 58

&+$37(5 5(*52:7+ ,1 '$5.1(66 $6 ,1)/8(1&(' %< 35(9,286 &877,1* 75($70(17 2) )285 /,032*5$66 *(127<3(6 ,QWURGXFWLRQ /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf LV QDWLYH WR WKH KXPLG VXEWURSLFV RI $IULFD $ EULHI UHVXPH RI LWV LQWURn GXFWLRQ LQWR $PHULFDQ JUDVVODQG DJULFXOWXUH KDV EHHQ UHSRUWHG E\ 2DNHV f $JURQRPLF HYDOXDWLRQV .UHWVFKPHU DQG 6Q\GHU 4XHVHQEHUU\ HW DO 4XHVHQEHUU\ DQG 2FXPSDXJK 5XHONH HW DO f KDYH JHQHUDWHG LQIRUPDWLRQ VXSSRUWLQJ WKH YLHZ WKDW OLPSRJUDVV KDV IRUDJH SRWHQWLDO LQ VXEWURSLFDO UHJLRQV DQG RQ VRLOV WKDW DUH LQWHUPLWWHQWO\ IORRGHG 7KLV VWXG\ RI HWLRODWHG UHJURZWK ZDV SHUIRUPHG DV D PHDQV RI FKDUDFWHUL]LQJ WKH HQHUJ\ UHVHUYHV DQG PRUSKRORJLHV RI IRXU LPSRJUDVVHV 7KH UHJURZWKLQGDUNQHVV WHFKQLTXH KDV EHHQ XVHG E\ PDQ\ LQYHVWLJDWRUV $GHJEROD $GHJEROD DQG 0F.HOO E %XUWRQ DQG -DFNVRQ 'RYUDW DQG &RKHQ 0DWFKHV 5£HVH DQG 'HFNHU :DUG DQG %ODVHU :DWVRQ DQG :DUG f 7KH REMHFWLYH RI WKLV VWXG\ ZDV WR GHWHUPLQH ZKHWKHU IUHTXHQW FOLSSLQJ LQ WKH ILHOG FRXOG UHGXFH HQHUJ\ UHVHUYHV EHORZ WKH FULWLFDO DPRXQWV QHFHVVDU\ WR PDLQWDLQ VWDQGV WKURXJK WKH GRUPDQW VHDVRQ $OWKRXJK FKHPLFDO WHVWV ZHUH QRW FRQGXFWHG PXFK HYLGHQFH H[LVWV VXEVWDQWLDWLQJ D VWURQJ FRUUHODWLRQ EHWZHHQ HWLRODWHG UHJURZWK DQG QRQVWUXFWXUDO FDUERK\GUDWHV LQ VWRUDJH RUJDQV $GHJEROD $GHJEROD DQG 0F.HOO E 'RYUDW DQG &RKHQ 5£HVH DQG 'HFNHU f

PAGE 59

1R FDUERK\GUDWH VWXGLHV RI OLPSRJUDVV FRXOG EH IRXQG KHQFH LW ZDV GHVLUHG WR OHDUQ WKH ORFDWLRQ RI FDUERK\GUDWH VWRUDJH LQ WKHVH VWRORQLIHURXV JUDVVHV EHIRUH IXOO VFDOH FKHPLFDO DQDO\VHV ZHUH FRQn GXFWHG LQ ODWHU SKDVHV RI UHVHDUFK 0DWHULDOV DQG 0HWKRGV $ FOLSSLQJ VWXG\ ZDV FRQGXFWHG RQ IRXU ZHOOHVWDEOLVKHG SODQWLQJV RI OLPSRJUDVV RQ D :DFKXOD VDQG D SRRUO\ GUDLQHG VLOLFHRXV K\SHUWKHUPLF XOWLF KDSODTXRG VRLO DW WKH %HHI 5HVHDUFK 8QLW RI WKH 8QLYHUVLW\ RI )ORULGD QHDU *DLQHVYLOOH 7KH JUDVVHV ZHUH 3, 3, 3, DQG 3, *UDVVHV 3, DQG 3, KDYH EHHQ UHOHDVHG DV %LJDOWD DQG n5HGDOWD UHVSHFWLYHO\ 5HGDOWD LV D GLSORLG Q f DQG WKH RWKHU JHQRW\SHV XVHG KHUH DUH WHWUDSORLGV Q f 7KH IRXU EORFNV RI JUDVVHV UHSUHVHQWLQJ JHQRW\SHV ZHUH XQUHSOLFDWHG KHQFH WKH GHVLJQ ZDV D VSOLW SORW ZLWKRXW UHSOLFDWLRQ RI PDLQ SORWV 7KH JHQRW\SHrWUHDWPHQW UHSf WHUP ZDV XVHG WR WHVW JHQRW\SH WUHDWPHQWn DQG LQWHUDFWLRQ HIIHFWV :LWKLQ PDLQ SORWV FOLSSLQJ WUHDWPHQWV ZHUH UDQGRPO\ DVVLJQHG ZLWK WKUHH UHSOLFDWLRQV &OLSSLQJ WUHDWPHQWV ZHUH LQLWLDWHG E\ PRZLQJ DOO SORWV WR FP RQ -XO\ DQG KDUYHVWLQJ DW DQG ZHHN LQWHUYDOV XQWLO 2FWREHU $GGLWLRQDO DUHDV ZLWKRXW UHSOLFDWLRQ ZHUH UHVHUYHG WR DOORZ OLPSRJUDVV WRS JURZWK WR UHDFK DQG ZHHNV RI DJH E\ 1RYHPEHU ZKHQ WKH UHJURZWKLQn GDUNQHVV ([SHULPHQW EHJDQ 7KH OLPSRJUDVVHV ZHUH IHUWLOL]HG 0D\ DQG -XO\ ZLWK NJKD 13A2J.A2f FRQWDLQLQJ SHUFHQW RI D PLFURHOHPHQW PL[ 6RLO WHVW UHVXOWV IURP VDPSOHV WDNHQ RQ 'HFHPEHU VKRZHG D S+ RI NJKD SKRVSKRUXV DQG NJKD RI SRWDVVLXP

PAGE 60

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n WXUH DW DSSUR[LPDWHO\ & 3ODQWV ZHUH ZDWHUHG DQG VSUD\HG ZLWK IXQJLn FLGH ZKHQ QHFHVVDU\ 7KH VWXG\ ZDV WHUPLQDWHG DIWHU ZHHNV ZKHQ JURZWK KDG FHDVHG 6KRRW UHJURZWK ZDV UHPRYHG EHIRUH WKH UHPDLQGHU RI WKH SODQW ZDV ZDVKHG IUHH RI VRLO DQG VHSDUDWHG LQWR URRWV DQG VWXEEOH $OO WKUHH FRPSRQHQWV ZHUH WKHQ GULHG DW & DQG ZHLJKHG 7KH GU\ PDWWHU '0f \LHOGV ZHUH VXEMHFWHG WR DQDO\VLV RI YDULDQFH DQG UHJUHVVLRQ DQDO\VLV 7KH PHDQ \LHOGV IRU WUHDWPHQWV ZHUH VXEMHFWHG WR WKH :DOOHU 'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW 'DWD ZHUH DQDO\]HG XVLQJ WKH 6WDWLVWLFDO $QDO\VLV 6\VWHP 6$6f RQ DQ $PGDKO 9 ZLWK 26096 5HOHDVH DQG -(61-( 5HOHDVH &RPSXWLQJ ZDV SHUIRUPHG DW WKH 1RUWKHDVW 5HJLRQDO 'DWD &HQWHU RI WKH 6WDWH 8QLYHUVLW\ 6\VWHP RI )ORULGD ORFDWHG RQ WKH FDPSXV RI WKH 8QLYHUVLW\ RI )ORULGD LQ *DLQHVYLOOH

PAGE 61

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f ZKHUHDV 5HGDOWD DQG 3, ZHUH QRW DIIHFWHG E\ SUHYLRXV FXWWLQJ WUHDWPHQW LQ WKH ILHOG 3 DQG 3 UHVSHFWLYHO\f )RU ERWK 3, DQG %LJDOWD D PD[LPXP UHJURZWK ZDV REVHUYHG DW WKH ZHHN FXWWLQJ

PAGE 62

7DEOH $YHUDJH '0 \LHOGV RI HWLRODWHG VKRRWV IRU IRXU OLPSRJUDVVHV IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV &OLSSLQJ IUHTXHQF\ ZHHNVf /LPSRJUDVV QXPEHU RU QDPH 3, %LJDOWD 5HGDOWD 3, \A r DE DE D D DE E D D D D D D E F D D E F D D 9DOXHV ZLWKLQ HDFK FROXPQ IROORZHG E\ WKH VDPH OHWWHU DUH QRW VLJQLILFDQWO\ GLIIHUHQW 3 f EDVHG RQ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW

PAGE 63

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n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f VKRZHG KLJKHU UHJURZWK \LHOGV ZLWK LQFUHDVLQJ KHLJKW RI FXWWLQJ LQGLFDWLQJ DQ HQHUJ\ UHVHUYH VLQN LQ WKH VWHP EDVHV RI WDOO IHVFXH )HVWXFD DUXQGLQDFHD 6FKUHEf ,Q WKLV VWXG\ RI OLPSRJUDVV WKHUH ZDV D SRVVLELOLW\ WKDW D ORZHU FXWWLQJ KHLJKW LQ GDUNQHVV ZRXOG UHGXFH VLQN VL]H DQG VHSDUDWH WUHDWPHQW HIIHFWV LQ D VLJQLILFDQW ZD\

PAGE 64

7DEOH &RPSDULVRQ RI HWLRODWHG VKRRW '0 \LHOGV DQG DVVRFLDWHG URRW DQG VWXEEOH FRPSRQHQWV IRU IRXU OLPSRJUDVVHV DYHUDJHG IRU DOO FXWWLQJ WUHDWPHQWV IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV /LPSRJUDVV &RPSRQHQW 6KRRW 5RRW 6WXEEOH 3, Dr JSRW F F %LJDOWD D E F 5HGDOWD E D E 3, E E D ‘N 9DOXHV ZLWKLQ HDFK FROXPQ IROORZHG E\ WKH VDPH OHWWHU DUH QRW VLJQLILFDQWO\ GLIIHUHQW 3 f EDVHG RQ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW

PAGE 65

([SHULPHQW 7KH HWLRODWHG VKRRW ZHLJKWV IURP ([SHULPHQW ZHUH PRGHOHG DV D IXQFWLRQ RI WKH FXWWLQJ WUHDWPHQWV DQG URRWV FXWWLQJ WUHDWPHQWV DQG VWXEEOH DQG FXWWLQJ WUHDWPHQWV URRWV DQG VWXEEOH 7KH UHGXFHG PRGHOV ZKLFK H[SODLQHG UHJURZWK \LHOGV DV D IXQFWLRQ RI WUHDWPHQW DQG VWXEEOH RU WUHDWPHQW DQG URRWVf GHVFULEHG HWLRODWHG UHJURZWK DV ZHOO DV WKH IXOO PRGHO IRU DOO IRXU LPSRJUDVVHV 5RRW ZHLJKWV ZHUH DIIHFWHG 3 f E\ WUHDWPHQWV LQ ([SHULPHQW ZKHUHDV VWXEEOH ZHLJKWV ZHUH QRW 3 f &RYDULDEOHV DUH QRW VXSSRVHG WR EH DIIHFWHG E\ WUHDWPHQWV DQG VLQFH WKH VWHP EDVHV IURP ([SHULPHQW ZHUH DFWXDOO\ WKDW SRUWLRQ RI WKH VWXEEOH EHORZ JURXQG WKH VWHP EDVH ZHLJKW ZDV XVHG DV D FRYDULDEOH WR H[SODLQ VRPH UHVLGXDO HUURU LQ WKH DQDO\VLV RI ([SHULPHQW UHJURZWK $QDO\VLV RI YDULDQFH RI HWLRODWHG UHJURZWK LQ ([SHULPHQW GLG QRW GHWHFW WUHDWPHQW GLIIHUHQFHV 3 f EXW DQDO\VLV RI FRYDULDQFH GLG 3 f 7DEOH VKRZV WKH +DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW RI DYHUDJH '0 \LHOGV RI VKRRWV REWDLQHG IRU HDFK FOLSSLQJ IUHTXHQF\ LQ 3, 6KRRW \LHOGV ZHUH ORZHU DQG WKH UDQJH QDUURZHU WKDQ IRXQG LQ ([SHULPHQW IRU 3, EXW WKH GDWD IROORZHG WKH VDPH WUHQG /RZHU \LHOGV FDQ EH DWWULEXWHG WR WKH ORVV RI VRPH HQHUJ\ UHVHUYHV WKDW FRXOG EH ORFDWHG LQ WKH DERYH JURXQG VWHP WLVVXH UHPRYHG SULRU WR WKH VWDUW RI ([SHULPHQW 5HVXOWV DJUHH ZLWK WKRVH RI 0DWFKHV f ZKR VXJJHVWHG WKDW ZLWK VKRUWHU KHLJKWV RI FXW WKH UDQNLQJ RUGHU RI WUHDWPHQWV ZRXOG UHPDLQ QHDUO\ WKH VDPH EXW WKH PDJQLWXGH RI GLIIHUHQFH RI HQHUJ\ UHVHUYHV PLJKW EH OHVV 0DWFKHV DOVR VWDWHG WKDW LQ HWLRODWHG UHJURZWK H[SHULPHQWV

PAGE 66

7DEOH $YHUDJH '0 \LHOGV RI HWLRODWHG UHJURZWK IURP ([SHULPHQW IROORZLQJ ZHHNV RI JURZWK LQ GDUNQHVV &OLSSLQJ IUHTXHQF\ ZHHNVf 3, VKRRWV JSRWf F EH DEF DE D ‘N 9DOXHV IROORZHG E\ WKH VDPH OHWWHU DUH QRW VLJQLILFDQWO\ GLIIHUHQW 3 f EDVHG RQ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW

PAGE 67

KLJKHU FXWWLQJ KHLJKWV SHUPLW PRUH UHJURZWK DQG DOORZ JUHDWHU GLIIHUn HQWLDWLRQ RI WUHDWPHQW HIIHFWV 7KDW VWDWHPHQW LV IXUWKHU VXEVWDQWLDWHG E\ :DWVRQ DQG :DUG f ZKR GHPRQVWUDWHG D SHUFHQW UHGXFWLRQ LQ IRRG UHVHUYHV ZKHQ FXWWLQJ KHLJKW ZDV UHGXFHG IURP WR FP LQ GDOOLVJUDVV 3DVSDXUQ GLODWDWXP 3RLUf ,Q WKLV VWXG\ WKH UHVXOWV RI ([SHULPHQW VKRZ WKDW WULPPLQJ WKH SODQWV WR JURXQG OHYHO UHTXLUHG D FRYDULDEOH VWHP EDVH ZHLJKWf WR GHWHFW WUHDWPHQW GLIIHUHQFHV +RZHYHU UHGXFLQJ UDWKHU WKDQ LQFUHDVLQJ UHVLGXDO SODQW KHLJKW WRJHWKHU ZLWK WKH VPDOO GLIIHUHQFHV LQ UHJURZWK \LHOGV EHWZHHQ ([SHULPHQW DQG ([SHULPHQW LQIHUUHG WKDW WKH PDMRULW\ RI IRRG VWRUDJH RFFXUV LQ WKH YHU\ EDVDO SRUWLRQV RI WKH FXOP VWHP EDVHf 5HVLGXDO (IIHFWV
PAGE 68

)ROORZLQJ WKH FRPSOHWH UHPRYDO RI DERYH JURXQG VWHP WLVVXH DW WKH VWDUW RI ([SHULPHQW WKH SURGXFWLRQ RI FRQVLGHUDEOH HWLRODWHG JURZWK VXJJHVWHG HQHUJ\ UHVHUYH VWRUDJH ORZHU RQ WKH FXOP 6WHP EDVH ZHLJKWV ZHUH XVHG DV D FRYDULDEOH LQ ([SHULPHQW DQG GLIIHUHQFHV 3 f ZHUH IRXQG LQ \LHOG RI HWLRODWHG UHJURZWK GXH WR SUHYLRXV FXWWLQJ SUHVVXUHV RQ OLPSRJUDVV 3, ZLWK ORQJHU FXWWLQJ LQWHUYDOV DOORZLQJ JUHDWHU \LHOGV RI UHJURZWK 6XPPDU\ /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf UHVHDUFK KDV UHDFKHG D VWDJH ZKHUH PDQDJHPHQW UHFRPPHQGDWLRQV DUH QHHGHG WR IXOO\ LPSOHPHQW OLPSRJUDVVn XVHIXOQHVV IRU WKH ODUJH KHFWDUDJH RI LPSURYHG SDVWXUHV RQ 6SRGRVROV OLNH )ORULGDnV IODWZRRG VRLOV $ SUHn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f EXW 5HGDOWD DQG 3, ZKLFK KDG KLJKHU VWXEEOH

PAGE 69

ZHLJKWV VKRZHG QR WUHDWPHQW HIIHFWV RQ ZHLJKW RI HWLRODWHG VKRRWV ,Q ([SHULPHQW 3, WKH OLQH ZLWK WKH KLJKHVW VWXEEOH ZHLJKW IURP ([SHULPHQW ZDV HYDOXDWHG XVLQJ D FRPSOHWHO\ UDQGRPL]HG GHVLJQ DQG QLQH UHSOLFDWLRQV ZLWK DOO SODQW PDWHULDO UHPRYHG WR JURXQG OHYHO +HQFH RQO\ WKH VWHP EDVHV UHPDLQLQJ EHORZ WKH VRLO VXUIDFH ZHUH UHVSRQVLEOH IRU UHJURZWK LQ ([SHULPHQW 1R WUHDWPHQW HIIHFWV ZHUH IRXQG 3 f XQWLO WKH GDWD ZHUH DQDO\]HG ZLWK FRYDULDQFH WHFKn QLTXHV XVLQJ VWHP EDVH ZHLJKW DV D FRYDULDEOH 6WDWLVWLFDO VHQVLn WLYLW\ LPSURYHG DQG LQFUHDVHV 3 f LQ UHJURZWK SRWHQWLDO ZHUH GHWHFWHG ZLWK FOLSSLQJ WUHDWPHQWV RI D ORQJHU FXWWLQJ LQWHUYDO 7KH FRYDULDEOH DQDO\VLV UHSUHVHQWHG DQ LPSURYHPHQW LQ WKH UHJURZWKLQn GDUNQHVV WHFKQLTXH 5HPRYLQJ DOO SODQW PDWHULDO WR WKH VRLO OHYHO SULRU WR WKH EHJLQn QLQJ RI WKH VHFRQG H[SHULPHQW OHIW FP RI VWHP EDVH EHORZ JURXQG 7KH FORVH VLPLODULW\ RI UHJURZWK \LHOGV LQ ([SHULPHQW ZKHUH SODQWV ZHUH FOLSSHG WR FP DERYH WKH VRLO VXUIDFH DQG WKRVH RI ([SHULPHQW KDYLQJ QR DERYH JURXQG WLVVXH VXJJHVWHG WKDW WKH HQHUJ\ UHVHUYHV ZHUH SUHGRPLQDQWO\ ORFDWHG LQ WKH ERWWRP FP RI WKH VWHP

PAGE 70

&+$37(5 &877,1* )5(48(1&< ())(&76 21 /,032*5$66 0253+2/2*< $1' 727$/ 1216758&785$/ &$5%2+<'5$7( 5(6(59(6 ,QWURGXFWLRQ /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf KDV SURPLVH DV DQ DGDSWHG ZDUPVHDVRQ JUDVV IRU )ORULGDnV YDVW KHFWDUDJH RI DFLG IODWZRRG VRLOV 8QIRUWXQDWHO\ n%LJDOWD PRVW IDYRUHG E\ UDQFKHUV EHFDXVH RI LWV KLJK IRUDJH TXDOLW\ LV DOVR WKH OHDVW SHUVLVWHQW RI WKH FXOWLYDUV ZKLFK KDYH EHHQ UHOHDVHG 4XHVHQEHUU\ HW DO 5XHONH .UHWVFKPHU DQG 6Q\GHU 4XHVHQEHUU\ HW DO f 2WKHU VWXGLHV RI OLPSRJUDVV KDYH LGHQWLILHG D IHZ SURPLVLQJ SODQW LQWURGXFn WLRQV WKDW KDYH FRPSDUDEOH TXDOLW\ DQG EHWWHU SHUVLVWHQFH WKDQ %LJDOWD 5XHONH 4XHVHQEHUU\ DQG 2FXPSDXJK 2FXPSDXJK DQG 4XHVHQEHUU\ 2FXPSDXJK HW DO 2FXPSDXJK f ,Q &KDSWHU &KULVWLDQVHQ FRQGXFWHG UHJURZWKLQGDUNQHVV H[SHULn PHQWV ZLWK OLPSRJUDVV WKDW LQGLFDWHG WKH DUHD RI FRPSUHVVHG ORZHU QRGHV RQ WKH VWHP EDVH DV WKH PDMRU VLWH RI HQHUJ\ UHVHUYH LQ WKHVH UREXVW VWRORQLIHURXV SHUHQQLDO IRGGHU SODQWV 7KH UHJURZWKLQGDUNQHVV WHFKn QLTXH LV D VLPSOH ZD\ WR PHDVXUH HQHUJ\ UHVHUYHV E\ DQDO\]LQJ UHJURZWK IURP GHIROLDWHG SODQWV SODFHG LQ GDUNQHVV %XUWRQ DQG -DFNVRQ 6KHDUG f 3HUU\ DQG 0RVHU f VWUHVVHG WKH LPSRUWDQFH RI ORFDWn LQJ WKH VSHFLILF FDUERK\GUDWH VWRUDJH RUJDQ RI D JUDVV EHIRUH SURFHHGLQJ ZLWK WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f DQDO\VHV 6WXGLHV RI 71& FRQWHQW ZHUH GHHPHG QHFHVVDU\ WR YHULI\ WKH UHJURZWKLQGDUNQHVV UHVXOWV

PAGE 71

0DQ\ UHVHDUFKHUV KDYH LQGLFDWHG D SRVLWLYH FRUUHODWLRQ EHWZHHQ HWLRODWHG UHJURZWK DQG 71& LQ WKH VWRUDJH RUJDQV $GHJEROD $GHJEROD DQG 0F.HOO E 'RYUDW DQG &RKHQ 5£HVH DQG 'HFNHU f 7KHUHIRUH LW ZDV RI LQWHUHVW WR VHH LI WKLV SRVLWLYH FRUUHODWLRQ KHOG IRU OLPSRJUDVV 71& YHUVXV HWLRODWHG UHJURZWK /HXNHO DQG &ROHPDQ f DQG *UDEHU f GLVFXVVHG WKH DELOLW\ RI YDULRXV IRUDJH SODQWV WR WUDQVIRUP DQ XSULJKW KDELW WR D SURVWUDWH KDELW IROORZLQJ IUHTXHQW GHIROLDWLRQ 6RPH OLPSRJUDVVHV PD\ EH VXVn SHFWHG WR KDYH JUHDWHU IOH[LELOLW\ WKDQ RWKHUV LQ DOWHULQJ WKHLU JURZWK KDELW 7KH REMHFWLYH ZDV WR VWXG\ PRUSKRORJLFDO FKDQJHV LQGXFHG E\ IUHTXHQW FXWWLQJ DQG HOXFLGDWH VXEVHTXHQW GLIIHUHQFHV LQ 71& DFFXPXODn WLRQ IRU WZR SURPLVLQJ DQG WZR UHOHDVHG FXOWLYDUV RI OLPSRJUDVV 0DWHULDOV DQG 0HWKRGV 7KH H[SHULPHQWV WR IROORZ ZHUH FRQGXFWHG RQ IRXU ZHOO HVWDEOLVKHG SODQWLQJV RI OLPSRJUDVV RQ D :DFKXOD VDQG D SRRUO\ GUDLQHG VLOLFHRXV K\SHUWKHUPLF XOWLF KDSODTXRG VRLO DW WKH %HHI 5HVHDUFK 8QLW RI WKH 8QLYHUVLW\ RI )ORULGD 7KH JUDVVHV ZHUH 3, 3, 3, DQG 3, %RWK 3, DQG 3, KDYH EHHQ UHOHDVHG DV %LJDOWD DQG f5HGDOWDn UHVSHFWLYHO\ 5HGDOWD LV D GLSORLG Q f DQG WKH RWKHU JHQRW\SHV XVHG KHUH DUH WHWUDSORLGV Q f 7KH OLPSRJUDVVHV ZHUH IHUWLOL]HG 0D\ DQG -XO\ ZLWK NJKD RI 13Jf FRQWDLQLQJ SHUFHQW RI D PLFURHOHPHQW PL[ 6RLO WHVW UHVXOWV IURP VDPSOHV WDNHQ RQ 'HFHPEHU VKRZHG D S+ RI NJKD SKRVSKRUXV DQG NJKD RI SRWDVVLXP

PAGE 72

$QDO\VLV RI 71& LQ 3ODQW 3DUWV 3ULRU WR WKH OD\RXW RI WKH FOLSSLQJ H[SHULPHQW ZKROH SODQW VDPSOHV RI OLPSRJUDVV ZHUH WDNHQ RQ -XO\ DQG -XO\ UHSUHVHQWLQJ DQG ZHHN ROG SODQW PDWXULWLHV UHVSHFWLYHO\ )LYH ODUJH SODQWV RI HDFK JHQRW\SH ZHUH UDQGRPO\ VHOHFWHG GXJ ZDVKHG IUHH RI VRLO DQG DUUDQJHG LQ FRWWRQ VDPSOH EDJV 7KH FRWWRQ EDJV ZHUH WKHQ SDFNHG LQ D SODVWLF EDJ DQG SXW RQ LFH 8SRQ UHDFKLQJ WKH ODE WKH VDPSOHV ZHUH GULHG LQ D 7KHOFR IRUFHGDLU RYHQ DW & IRU PLQXWHV DIWHU ZKLFK WKH WHPn SHUDWXUH ZDV ORZHUHG WR & XQWLO WKH VDPSOHV ZHUH UHPRYHG KRXUV ODWHU 6PLWK Df 7KH SODQWV ZHUH FDUHIXOO\ IUDFWLRQDWHG LQWR VKRRWV VWXEEOH FURZQ DQG URRWV ,Q WKH SODQW SDUW DQDO\VLV FURZQ ZDV GHVLJn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

PAGE 73

WKH GHVLJQ ZDV D VSOLW SORW ZLWKRXW UHSOLFDWLRQ RI WKH PDLQ SORWV 7KH JHQRW\SH r WUHDWPHQW UHSf WHUP ZDV XVHG WR WHVW JHQRW\SHV FOLSSLQJ WUHDWPHQWV DQG LQWHUDFWLRQ HIIHFWV :LWKLQ PDLQ SORWV WKH FOLSSLQJ WUHDWPHQWV ZHUH UDQGRPO\ DVVLJQHG DQG UHSOLFDWHG WKUHH WLPHV 6DPSOHV IRU 71& DQDO\VLV ZHUH GXJ SULRU WR FOLSSLQJ RQ $XJXVW $XJXVW 6HSWHPEHU 2FWREHU DQG 1RYHPEHU DQG SURFHVVHG DV EHIRUH H[FHSW WKDW RQO\ FURZQ ERWWRP FP RI VWHP EDVHf VDPSOHV ZHUH VDYHG IRU 71& DQDO\VLV 7KH FURZQV ZHUH VFUDSHG IUHH RI URRWV OHDI VKHDWKV DQG VDQG E\ XVLQJ D ZLUH EXIILQJ ZKHHO DWWDFKHG WR WKH H[WHQGHG VKDIW RI D VPDOO HOHFWULF PRWRU : 530f 7KH FURZQV ZHUH WKHQ JURXQG WKURXJK D PP VFUHHQ LQ WKH 8'< PLOO DQG VWRUHG XQWLO WHVWHG IRU 71& 'DWD ZHUH DQDO\]HG XVLQJ WKH 6WDWLVWLFDO $QDO\VLV 6\VWHP 6$6f RQ DQ $PGDKO 9 ZLWK 26096 5HOHDVH DQG -(61-( 5HOHDVH &RPSXWLQJ ZDV SHUIRUPHG DW WKH 1RUWKHDVW 5HJLRQDO 'DWD &HQWHU RI WKH 6WDWH 8QLYHUVLW\ 6\VWHP RI )ORULGD ORFDWHG RQ WKH FDPSXV RI WKH 8QLYn HUVLW\ RI )ORULGD LQ *DLQHVYLOOH 7KH SODQW SDUW DQDO\VHV DQG WKH SHUn FHQW 71& GDWD IRU DOO VXEVHTXHQW VDPSOLQJV ZHUH VXEMHFWHG WR DQDO\VLV RI YDULDQFH DQG PHDQV ZHUH FRPSDUHG XVLQJ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW &RUUHODWLRQ RI 71& YHUVXV 5HJURZWKLQ'DUNQHVV 3ULRU WR GLVFXVVLQJ WKH PHWKRGV XVHG LQ WKH FRUUHODWLRQ RI 71& DQG HWLRODWHG UHJURZWK WHUPLQRORJ\ ZLOO EH UHYLHZHG IRU FODULW\ 7KH UHJURZWKLQGDUNQHVV H[SHULPHQWV ZHUH FRPSOHWHG EHIRUH DQ\ FKHPLFDO DQDO\VLV WRRN SODFH +HQFH WKH ILUVW XVH RI WKH WHUP VWXEEOH ZDV LQ ([SHULPHQW RI &KDSWHU DQG UHSUHVHQWHG WKH FP RI VWHP WLVVXH

PAGE 74

DERYH WKH VRLO OHYHO DV ZHOO DV WKH VWHP EDVH EHORZ JURXQG 5HFDOO WKDW WKH VWHP PDWHULDO DERYH JURXQG OHYHO ZDV UHPRYHG DW WKH VWDUW RI WKH VHFRQG UHJURZWKLQGDUNQHVV VWXG\ ([SHUPHQW &KDSWHU f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n FHQW 71& WR TXDQWLWDWLYHO\ FKDUDFWHUL]H WKH FRQFHQWUDWLRQ RI UHVHUYHV SUHVHQW LQ WKH FURZQ DW WKH EHJLQQLQJ RI WKH UHJURZWKLQGDUNQHVV VWXG\ ([SHULPHQW &KDSWHU f 7KH GDUNQHVV VWXG\ EHJDQ RQ 1RYHPEHU DQG ODVWHG ZHHNV 7KH GU\ ZHLJKWV RI HWLRODWHG VKRRWV DV ZHOO DV ZHLJKWV RI VWXEEOH FP RI VWHP DERYH JURXQG SOXV VWHP EDVHV EHORZ JURXQGf DQG URRWV SHU SRW ZHUH UHFRUGHG 7KH FURZQ 71& GDWD IURP 1RYHPEHU ZHUH FRUUHODWHG DJDLQVW VWXEEOH DQG DJDLQVW VKRRW SHUFHQWDJHV RI WKH VWXEEOH SOXV URRWV SUHVHQW LQ WKH SRWV 6KRRWVWXEEOH URRWVnff r •f 5HVXOWV DQG 'LVFXVVLRQ $QDO\VLV RI 71& LQ 3ODQW 3DUWV 5HVXOWV IRU 71& DQDO\VHV RI SODQW SDUWV DUH VKRZQ LQ )LJXUH 7KH OLPSRJUDVVHV ZHUH GLIIHUHQW 3 f LQ WKHLU SHUFHQW 71& KRZHYHU VLJQLILFDQW WZR DQG WKUHH ZD\ LQWHUDFWLRQV DPRQJ SDUWV DJH DQG

PAGE 75

B 3, L R 7 1 ‘ E F 5('$/7$ D %,*$/7$ 3, G 2 f§L L $*( 2) /,032*5$66 7,668( ZHHNVf f FURZQV URRWV R VWXEEOH 'VKRRWV )LJXUH 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQW RI GU\ PDWWHU RI IRXU SODQW SDUWV LQ IRXU LPSRJUDVVHV VDPSOHG DW WZR PDWXULWLHV 'LIIHUHQW OHWWHUV ZLWKLQ DQ DJH RI OLPSRJUDVV WLVVXH LQGLFDWH D VLJQLILFDQW GLIIHUn HQFH 3 f RI SHUFHQW 71& LQ WKH SODQW SDUWVf

PAGE 76

JHQRW\SHV SURKLELWHG D FRPSDULVRQ RI RYHUDOO PHDQV IRU WKH IRXU LPSRJUDVVHV :KHQ HDFK OLPSRJUDVV ZDV DQDO\]HG VHSDUDWHO\ DOO H[FHSW 3, VKRZHG GLIIHUHQW 3 f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f 0DQ\ VWXGLHV KDYH VKRZQ ODUJH ,920' GLIIHUHQFHV EHWZHHQ WKHVH WZR JUDVVHV 6FKDQN HW DO 4XHVHQEHUU\ HW DO f &ULWHULD QHFHVVDU\ LQ VHOHFWLQJ D SODQW SDUW IRU 71& DQDO\VLV ZHUH f GLVTXDOLILFDWLRQ RI VKRRWV EHFDXVH WKH\ ZHUH QRW SHUHQQLDWLQJ SDUWV DQG f UHMHFWLRQ RI URRWV EHFDXVH RI VDQG DQG GHDG WLVVXH FRQWDPLQDWLRQ )LJXUH VKRZV WKDW WKH FURZQ ZDV VWDWLVWLFDOO\ DV KLJK RU KLJKHU LQ SHUFHQW 71& WKDQ WKH VWXEEOH DW DOO WLPHV 7KH ERWWRP FP RI WKH VWHP EDVH FURZQf ZDV VHOHFWHG IRU IXUWKHU 71& DQDO\VLV E\ FKHPLFDO PHDQV 71& DV 5HODWHG WR &OLSSLQJ 6HDVRQ DQG *HQRW\SH 7DEOH VKRZV WKH 71& UHVXOWV IRU IRXU LPSRJUDVVHV VXEMHFWHG WR WKUHH FXWWLQJ IUHTXHQFLHV LQ WKH JURZLQJ VHDVRQ ,Q D FRPELQHG DQDO\VLV RI VDPSOH GDWHV LPSRJUDVVHV DQG FXWWLQJ WUHDWPHQWV HDFK IDFWRU LQIOXHQFHG 3 f WKH 71& PHDVXUHG LQ FURZQV 7KH WZR ZD\ LQWHUDFWLRQV 3 f RI OLPSRJUDVV DQG FOLSSLQJ IUHTXHQF\ ZLWK

PAGE 77

7DEOH 7KH PDLQ HIIHFWV RI OLPSRJUDVV JHQRW\SH DQG FOLSSLQJ WUHDWPHQW RQ WKH SHUFHQW WRWDO QRQVWUXFWXUDO FDUERn K\GUDWHV 71&f LQ WKH ERWWRP FP RI VWHP EDVH FURZQf /LPSRJUDVV VDPSOLQJ GDWH 71& b 3, Er D D D D 5HGDOWD G G G G E %LJDOWD D E E E F 3, F F F F ,G &OLSSLQJ IUHTXHQF\ ZHHNVf D D D D D D D E D E D D F D F 'LIIHUHQW OHWWHUV ZLWKLQ D VDPSOLQJ GDWH UHSUHVHQW VLJ QLILFDQW 3 f GLIIHUHQFHV n LQ SHUFHQW 71&

PAGE 78

VDPSOLQJ GDWH VXJJHVWHG WKDW 71& FRQFHQWUDWLRQ ZDV LQGHSHQGHQWO\ DOWHUHG E\ ERWK FXWWLQJ WUHDWPHQW DQG VHDVRQDO FDUERK\GUDWH IOX[ 7KHUH ZDV QR WKUHH ZD\ LQWHUDFWLRQ 3 f RU LQWHUDFWLRQ EHWZHHQ OLPSRJUDVV JHQRn W\SHV DQG FOLSSLQJ 3 f WKHUHIRUH 71& FRQFHQWUDWLRQV ZHUH FKDQJn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f ZDV XVHG WR FRUUHODWH DJDLQVW 71& LQ FURZQV RI VDPSOHV WDNHQ DW WKH VWDUW RI WKH UHJURZWK H[SHULPHQW 7KH EHVW FRUUHODWLRQV ZHUH IRXQG IRU

PAGE 79

b 71& UL k D k f D f % %r 2 2 FXWWLQJ k IUHT U YDOXH ZNVf r rr r ’ P QV 2,2 V A A + %6% 2 % % k D FXWWLQJ IUHT k U YDOXH ZNVf D b r R r ’ 2 f r 2,2 A 6 A 6 3, D 5('$/7$ % %,* $/7$ 6 3, ‘ :W RI 6WXEEOHJ (WLRODWHG 6KRRW b RI 6WXEEOH DQG 5RRWV FQ )LJXUH 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQWDJHV LQ WKH FURZQ FRUUHODWHG DJDLQVW VWXEEOH ZHLJKWV OHIWf DQG HWLRODWHG UHJURZWK SHUFHQW RI ZKROH SODQWV ULJKWf IRU IRXU OLPSRJUDVV JHQRW\SHV VXEMHFWHG WR WKUHH FOLSSLQJ IUHTXHQFLHV 3

PAGE 80

71& YHUVXV VWXEEOH DQG 71& YHUVXV SHUFHQW SODQW UHJURZWK LQ GDUNQHVV VKRRWVVWXEEOH URRWVff r f ,Q )LJXUH WKH V\PEROV UHSUHVHQW WKH FOLSSLQJ IUHTXHQFLHV XVHG ZLWKLQ HDFK OLPSRJUDVV JHQRW\SHV DUH LGHQWLILHG ZLWKLQ D V\PERO E\ WKH FRORUDWLRQ SDWWHUQf 7KH LPSRUWDQW UHODWLRQVKLSV DUH IRXQG ZLWKLQ HDFK FXWWLQJ IUHTXHQF\ *HQRW\SHV KDYLQJ JUHDWHU VWXEEOH ZHLJKW SHU SRW KDG KLJKHU FRQn WHQW RI 71& LQ WKH FURZQV )UHTXHQW FOLSSLQJ LQGXFHG D SURVWUDWH KDELW LQ DOO JHQRW\SHV EXW 3, DQG 5HGDOWD SURYLGHG PRUH VLWHV IRU OHDI HPHUJHQFHDV UHIOHFWHG E\ WKH DFFXPXODWLRQ RI 71& LQ FURZQV IRU WKH IUHTXHQWO\ FXW WUHDWPHQWV $ QHJDWLYH FRUUHODWLRQ ZDV REVHUYHG LQ )LJXUH ZKHQ WKH 71& LQ FURZQV ZHUH FRUUHODWHG DJDLQVW WKH VKRRW SHUFHQWDJH RI VWXEEOH SOXV URRWV 3ODQWV ZLWK PRUH VWXEEOH 3, f DQG URRWV 5HGDOWDf VHH 7DEOH &KDSWHU f SURGXFHG VPDOO VKRRWV WKDW ZHUH JUHDWHU LQ QXPEHU DQG ORZHU LQ ZHLJKW KHQFH ZHUH D ORZHU SHUFHQWDJH RI WKH VWXEEOH SOXV URRWV WR SHUFHQW LQ )LJXUH f %LJDOWD DQG 3, KDG IHZHU EXW ODUJHU VKRRWV DV ZHOO DV ORZHU ZHLJKWV RI URRWV DQG VWXEEOH VHH 7DEOH &KDSWHU f KHQFH VKRRWV ZHUH D ODUJHU SHUFHQWDJH RI WKH VWXEEOH SOXV URRWV LQ SRWV WR SHUFHQW LQ )LJXUH f :KHQ 71& ZDV FRUUHODWHG GLUHFWO\ DJDLQVW WKH GU\ ZHLJKW RI HWLRn ODWHG VKRRWV D QHJDWLYH FRUUHODWLRQ ZDV DOVR IRXQG f EXW QRW SUHVHQWHG 2WKHU H[SHULPHQWHUV $GHJEROD $GHJEROD DQG 0F.HOO E 'RYUDW DQG &RKHQ 5£HVH DQG 'HFNHU f KDYH IRXQG SRVLn WLYH FRUUHODWLRQV RI 71& YHUVXV HWLRODWHG UHJURZWK LH KLJKHU 71& LQ SODQWV KDYLQJ ORQJHU UHVW LQWHUYDOV 7KHVH VWXGLHV KRZHYHU XVHG

PAGE 81

XSULJKW LQVWHDG RI VWRORQLIHURXV JUDVVHV DQG XVHG D VLQJOH JHQRW\SH RI D VSHFLHV LQVWHDG RI IRXU JHQRW\SHV KDYLQJ YDULDEOH PRUSKRORJLHV &RQFOXVLRQV $QDO\VLV RI OLPSRJUDVV SODQW SDUWV LQGLFDWHG WKDW WKH ERWWRP FP RI VWHP EDVH FURZQf ZDV D VLJKW RI 71& DFFXPXODWLRQ &XWWLQJ WUHDWPHQWV DQG KDUYHVW GDWH FRQWULEXWHG LQGHSHQGHQWO\ WR WKH VLJQLILFDQW 3 f YDULDWLRQV VHHQ LQ OLPSRJUDVV FDUERK\GUDWH IOX[ 7KH PRUH IUHTXHQWO\ FOLSSHG OLPSRJUDVVHV HIIHFWHG D SURVWUDWH PRUSKRORJ\ ZKLFK DOORZHG PRUH D[LOODU\ WLOOHU IRUPDWLRQ DQG 71& DFFXPXn ODWLRQ IRU DOO OLPSRJUDVVHV 5HGDOWD DQG 3, VKRZHG DQ DXWXPQ VXUJH RI 71& WKDW PLJKW FRQWULEXWH WR WKHLU EHWWHU SHUVLVWHQFH DV RSSRVHG WR D GHFOLQH RU SODWHDX RI 71& IRU %LJDOWD DQG 3, 3ODQWV KDYLQJ JUHDWHU ZHLJKWV RI VWXEEOH DOVR KDG KLJKHU FRQFHQWUDWLRQV RI 71& LQ WKH ERWWRP FP RI WKH VWHP EDVH FURZQf 6XPPDU\ /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf FRXOG VLJQLILFDQWO\ FRQWULEXWH WR WKH IRUDJHOLYHVWRFN HFRQRP\ LQ WKH VRXWKn HDVWHUQ 8QLWHG 6WDWHV KRZHYHU FXOWLYDUV PXVW KDYH JRRG TXDOLW\ DQG KLJK SHUVLVWHQFH EHIRUH DQ LPSDFW ZLOO EH PDGH 2I WKH FXOWLYDUV QRZ DYDLODEOH n5HGDOWDn DQG n*UHHQDOWDn KDYH ORZ GLJHVWLELOLW\ DQG DGHTXDWH SHUVLVWHQFH ZKLOH n%LJDOWDn KDV H[FHOOHQW GLJHVWLELOLW\ DQG SRRU SHUVLVWHQFH

PAGE 82

&XUUHQW UHVHDUFK HIIRUWV KDYH LGHQWLILHG 3, OLPSRJUDVV DV ERWK SHUVLVWHQW DQG GLJHVWLEOH 7KH SXUSRVH RI WKLV VWXG\ ZDV WR FRPn SDUH 3, DQG DQRWKHU SURPLVLQJ OLPSRJUDVV ZLWK %LJDOWD DQG 5HGDOWD WR WU\ WR XQGHUVWDQG SHUVLVWHQFH UHODWLYH WR WKH WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f VWDWXV RI WKH SODQWV 7KH REMHFWLYH RI WKLV VWXG\ ZDV WR ORFDWH WKH 71& VWRUDJH VLWH DQG FKDUDFWHUL]H VHDVRQDO JHQRW\SLF DQG FOLSSLQJ IUHTXHQF\ HIIHFWV RQ WKH 71& FRQFHQWUDWLRQ LQ WZR UHOHDVHG DQG WZR SURPLVLQJ LPSRJUDVVHV 6HFRQGO\ LW ZDV GHVLUHG WR FRPSDUH WKH FKHPLFDO GDWD ZLWK WKH UHJURZWK LQGDUNQHVV UHVXOWV REWDLQHG LQ D SUHYLRXV VWXG\ $ VDPSOLQJ RI DQG ZHHN ROG OLPSRJUDVV RQ -XO\ DQG -XO\ IUDFWLRQDWHG LQWR IRXU SODQW SDUWV URRW VKRRW FURZQ DQG VWXEEOHf ZDV FKHPLFDOO\ DQDO\]HG IRU SHUFHQW 71& XVLQJ DQ HQ]\PDWLF VXJDU K\GURO\VLV DQG VSHFWURSKRWRPHWULF PHDVXUHPHQW RI UHGXFLQJ VXJDUV 7KH ERWWRP FP RI VWHP EDVH FURZQf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

PAGE 83

3, KDG WKH ODUJHVW VWRORQ V\VWHP VWRUHG WKH PRVW 71& DQG ZDV YHU\ SHUVLVWHQW 7KH 71& UHVXOWV IRU 1RYHPEHU ZHUH FRUUHODWHG DJDLQVW VKRRW DQG RWKHU SODQW ZHLJKW GDWD REWDLQHG IURP D ZHHN UHJURZWKLQn GDUNQHVV VWXG\ LQLWLDWHG RQ 1RYHPEHU 7KH UHVXOWV UHYHDOHG D SRVLWLYH UHODWLRQVKLS U f DFURVV OLPSRJUDVVHV IRU VWXEEOH ZHLJKW YHUVXV 71& DQG D QHJDWLYH FRUUHODWLRQ U f IRU HWLRODWHG VKRRWV H[SUHVVHG DV D SHUFHQW RI WKH URRWV SOXV VWXEEOH 7KH VLJQLILFDQW 3 f UHJUHVVLRQV ZHUH FDXVHG E\ LQKHUHQW GLIIHUHQFHV LQ OLPSRJUDVV PRUSKRORJLHV UDWKHU WKDQ FXWWLQJ IUHTXHQF\ HIIHFWV ZLWKLQ HDFK LPSRJUDVV

PAGE 84

&+$37(5 '5< 0$77(5 <,(/' &58'( 3527(,1 ,1 9,752 25*$1,& 0$77(5 ',*(67,%,/,7< 727$/ 1216758&785$/ &$5%2+<'5$7( $1' 3(56,67(1&( ,1 7:2 3520,6,1* $1' 7:2 5(/($6(' /,032*5$66(6 ())(&76 '8( 72 1,752*(1 )(57,/,=$7,21 $1' &877,1* )5(48(1&< ,QWURGXFWLRQ 7KH VRXWKHDVWHUQ 8QLWHG 6WDWHV LV D VXEWURSLFDO ]RQH WKDW SHUPLWV WKH JURZWK RI D ZLGH YDULHW\ RI JUDVVHV RQO\ D KDQGIXO RI ZKLFK DUH DJURQRPLFDOO\ LPSRUWDQW %DKLDJUDVV 3DVSDOXP QRWDWXP )OnJJHf LV WKH PRVW ZLGHVSUHDG LPSURYHG JUDVV LQ )ORULGD 0RWW DQG 0RRUH f EHFDXVH RI LWV H[FHOOHQW SHUVLVWHQFH DQG EURDG DGDSWDELOLW\ %HUPXGDJUDVV &\QRGRQ GDFW\O RQ /f 3HUVf FXOWLYDUV DUH HFRQRPLFDOO\ LPSRUWDQW DV KD\ DQG JUD]LQJ FURSV GXH WR VXFFHVVIXO SODQW EUHHGLQJ SURJUDPV EXW WKHVH FXOWLYDUV JHQHUDOO\ JURZ EHWWHU RQ XSODQG VLWHV 'LJLWJUDVVHV 'LJLWDULD GHFXPEHQV 6WHQWf DQG VWDUJUDVVHV &\QRGRQ QOHPIXHQVLV 9DQGHU\VWf DUH JURZQ PRUH LQ VRXWK DQG FHQWUDO )ORULGD GXH WR WKHLU ORZHU IURVW WROHUDQFH DQG ZLQWHU KDUGLQHVV 6W $XJXVWLQHJUDVV 6WHQRWDSKUXP VHFXQGDWXP :DOWf .XQW]H DQG SDUDJUDVV %UDFKLDULD PXWLFD )RUVNf 6WDSIf KDYH VSHFLDO DGDSWDELOLWLHV IRU RUJDQLF VRLOV EXW DUH QRW ZLGHO\ XVHG DV SDVWXUH JUDVVHV HOVZKHUH LQ )ORULGD /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf LV D YLDEOH DOWHUQDWLYH WR EDKLDJUDVV IRU IODWZRRG VLWHV /LPSRJUDVV FDQ EH HTXDOO\ SHUVLVWHQW EXW KLJKHU LQ TXDOLW\ WKDQ EDKLDJUDVV 0RRUH HW DO f ,PPDWXUH EDKLDJUDVV KDV SRWHQWLDOO\ JRRG TXDOLW\ KRZHYHU

PAGE 85

TXDOLW\ UDSLGO\ GHFOLQHV LQ WLVVXH JUHDWHU WKDQ ZHHNV RI DJH 0RRUH HW DO f DQG UHTXLUHV KHDY\ XWLOL]DWLRQ IRU RSWLPXP \LHOG DQG TXDOLW\ %HDW\ HW DO f 2FXPSDXJK f UHSRUWHG EHWWHU \HDUO\ EHHI SURGXFWLRQ DQG VLPLODU DYHUDJH GDLO\ JDLQV IURP 3, OLPSRJUDVV WKDQ IURP n3HQVDFRODn EDKLDJUDVV EHFDXVH RI D GD\ JUD]LQJ DGYDQWDJH *DVNLQV DQG 6OHSHU f VKRZHG WKDW GD\OHQJWK VHQVLWLYLW\ ZDV QRW D OLPLWLQJ IDFWRU IRU FRROVHDVRQ JURZWK RI OLPSRJUDVV EXW ZDV IRU GLJLWJUDVV DQG EHUPXGD JUDVV 3HUKDSV EDKLDJUDVV DOVR IDOOV LQWR WKH ODWWHU FDWHJRU\ %DKLDJUDVV LV SULPDULO\ XVHG IRU JUD]LQJ QRW KD\ DQG WKH PDMRU IRUDJH UHODWHG SUREOHPV LQ WKH VRXWKHDVWHUQ 8QLWHG 6WDWHV DUH VHDVRQDO IRUDJH GLVWULEXWLRQ DQG ODFN RI SUHVHUYDWLRQ SUDFWLFHV 0RWW Ef /LPSRJUDVV LV PRUH VHDVRQDEO\ IOH[LEOH DQG FDQ SURGXFH FRSLRXV DPRXQWV RI ELRPDVV .LOOLQJHU f SURGXFHG P WRQVKD RI GU\ PDWWHU '0f E\ 0D\ DQG P WRQVKD E\ $XJXVW LQ QRUWK FHQWUDO )ORULGD 7KLV VXUSDVVHG DQ\ RI WKH \HDUO\ \LHOGV IRU ILYH K\EULG EHUPXGDJUDVVHV n3DQJODn GLJLWJUDVV DQG 3HQVDFROD EDKLDJUDVV UHSRUWHG LQ E\ 5XHONH DQG 3ULQH +RGJHV DQG 0DUWLQ f VWXGLHG WKH ZDUP DQG FRRO VHDVRQ 1RYHPEHU 0D\f SURGXFWLRQ RI VXEWURSLFDO JUDVVHV DW 2QD )ORULGD DQG IRXQG WKH GLJLWJUDVVHV DQG &\QRGRQV RXW\LHOGHG OLPSR JUDVVHV KRZHYHU OLPSRJUDVVHV ZHUH VXSHULRU WR 3HQVDFROD EDKLDJUDVV DW DOO WKUHH IHUWLOLW\ OHYHOV GXULQJ WKH FRRO VHDVRQ .UHWVFKPHU DQG 6Q\GHU f REWDLQHG P WRQV RI GU\ PDWWHU '0f ZLWK WZR KDUYHVWV RI n%LJDOWDn OLPSRJUDVV XVLQJ D ZHHN FXWWLQJ IUHTXHQF\ DQG NJKD 1 SHU LQWHUYDO ,Q D PDQDJHPHQW VWXG\ 5XHONH REWDLQHG \LHOGV RI P WRQV IRU n5HGDOWDn n*UHHQDOWDn DQG %LJDOWD OLPSRJUDVVHV 4XHVHQEHUU\ HW DO f

PAGE 86

7KH TXDOLW\ FKDUDFWHULVWLFV RI WHWUDSORLG OLPSRJUDVVHV DUH ZHOO NQRZQ 6FKDQN HW DO f LGHQWLILHG KLJK LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLWLHV ,920'f DQG D VORZHU UDWH RI ,920' GHFOLQH LQ WKH WHWUDSORLGV 7D\ORU HW DO Df LQ 1HZ =HDODQG DOVR UHSRUWHG KLJK LQ YLWUR GLJHVWLELOLWLHV OHDYHV SHUFHQW VWHPV SHUFHQWf 7KH WHWUDSORLG OLPSRJUDVV VWHP GLJHVWLELOLW\ ZDV WKH KLJKHVW RI DOO VXPPHU JUDVVHV VWXGLHG 7KH DERYH ILQGLQJV HQFRXUDJHG 4XHVHQEHUU\ DQG 2FXPSDXJK f WR LQLWLDWH VWRFNSLOLQJ H[SHULPHQWDWLRQ 7KH\ UHSRUWHG WKH EHJLQQLQJ RI $XJXVW DV WKH SURSHU VWDJLQJ SHULRG WR LQLWLDWH UHJURZWK RI DGHTXDWH \LHOGV RI VWDQGLQJ IRUDJH 5HGDOWD SURGXFHG P WRQV '0 EXW GHFOLQHG UDSLGO\ LQ TXDOLW\ ZKHUHDV %LJDOWD SURGXFHG P WRQV DQG PDLQWDLQHG D SHUFHQW ,920' ZKHQ PHDVXUHG WKH IROORZLQJ 0DUFK 7KH UDWH RI ,920' GHFOLQH ZDV VLPLODU EXW WKH LQWHUFHSW VWDUWHG XQLWV KLJKHU LQ %LJDOWD WKDQ IRU 5HGDOWD DQG *UHHQDOWD 4XHVHQEHUU\ DQG 2FXPSDXJK f 5XHONHnV PDQDJHPHQW VWXGLHV XVLQJ OLPSRJUDVVHV 5XHONH 4XHVHQEHUU\ HW DO f VKRZHG WKDW FORVH IUHTXHQW GHIROLDWLRQ DQG KLJK QLWURJHQ IHUWLOL]DWLRQ OHG WR WKH ORVV RI %LJDOWD VWDQGV 0RE JUD]LQJ DQG FOLSSLQJ WULDOV E\ 4XHVHQEHUU\ DQG 2FXPSDXJK f DOVR VKRZHG %LJDOWDnV SRRU SHUVLVWHQFH 4XHVHQEHUU\ HW DO f LGHQWLILHG 3, DV KDYLQJ VXSHULRU SHUVLVWHQFH DQG FRPSDUDEOH TXDOLW\ WR %LJDOWD 2FXPSDXJK HW DO f LGHQWLILHG DQRWKHU SURPLVLQJ OLPSRn JUDVV 3, f WKDW SHUVLVWHG XQGHU PRE JUD]LQJ EXW WKDW ZDV RI D VOLJKWO\ ORZHU DFFHSWDELOLW\ E\ DQLPDOV WKDQ 3, RU %LJDOWD 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f DQDO\VLV KDV IRU GHFDGHV EHHQ D XVHIXO WHFKQLTXH LQ TXDQWLI\LQJ WKH IOXFWXDWLRQ RI RUJDQLF

PAGE 87

UHVHUYHV LQ UHVSRQVH WR VHDVRQ 0F&DUW\ :DLWH DQG %R\G Df GHIROLDWLRQ 6XOOLYDQ DQG 6SUDJXH 0D\ %DNHU DQG *DUZRRG 0DUVKDOO DQG 6DJDU f QLWURJHQ IHUWLOL]DWLRQ $GHJEROD DQG 0F.HOO D 5£HVH DQG 'HFNHU )RUG DQG :LOOLDPV f DQG SHUVLVWHQFH *UDEHU $OEHUGD f 7KH HIIRUWV RI 'DOH 6PLWK Df DQG RWKHUV LQ XQGHUVWDQGLQJ WKH FDUERK\GUDWH PHWDEROLVP LQ DOIDOID 0HGLFDJR VDWLYD /f LV D SULPH H[DPSOH RI WKH 71& VWXGLHV \LHOGLQJ SUDFWLFDO PDQDJHPHQW UHFRPPHQGDWLRQV $FFRUGLQJ WR *UDEHU f XQGHU ILHOG FRQGLWLRQV WKH OLPLWDWLRQV RI URRW JURZWK DQG WKH PRGLILFDWLRQV RI WKH LQWHUQDO HQYLURQPHQW UHVXOWLQJ IURP ORZ UHVHUYHV PD\ UHGXFH WKH DEVRUSWLYH FDSDFLW\ RI D SODQW VR JUHDWO\ DQG PD\ VR LQFUHDVH LWV VXVFHSWLELOLW\ WR GURXJKW ZLQWHU LQMXU\ ZHHG HQFURDFKn PHQWV LQVHFW LQMXU\ DQG RWKHU KD]DUGV DV WR MHRSDUGL]H LWV SHUPDQHQFH S f 9DULRXV UHSRUWV KDYH VKRZQ WKDW KLJK UDWHV RI QLWURJHQ IHUWLOL]HU KDYH OHG WR UHGXFHG SHUVLVWHQFH LQ GLJLWJUDVV &UHHO 5XHONH .LHQ HW DO f DQG LQ RUFKDUGJUDVV 'DFW\OLV JORPHUDWD /f $OH[DQGHU DQG 0F&ORXG f EXW QR FDUERK\GUDWH DQDO\VHV ZHUH SUHVHQWHG 1R FKHPLFDO GDWD UHODWLQJ 71& WR SHUVLVWHQFH LQ OLPSRJUDVV FRXOG EH ORFDWHG &KULVWLDQVHQ &KDSWHU f XVHG UHJURZWKLQGDUNQHVV VWXGLHV DV SUHOLPLQDU\ WHVWV RI UHVHUYH HQHUJ\ LQ OLPSRJUDVV ,Q &KDSWHU RI WKLV UHSRUW WKH VLWH RI QRQVWUXFWXUDO FDUERK\GUDWH DFFXPXn ODWLRQ ZDV GHWHUPLQHG DV ZHOO DV FOLSSLQJ HIIHFWV RQ 71& SHUFHQW RI OLPSRJUDVV LQ 5HGDOWD KDV SRRU TXDOLW\ EXW H[FHOOHQW SHUVLVWHQFH ZKLOH %LJDOWD KDV WKH UHYHUVH VLWXDWLRQ QHLWKHU FRPELQDWLRQ EHLQJ DGYDQWDJHRXV WR WKH SURGXFHU 7KH REMHFWLYH RI WKLV VWXG\ ZDV WR WKRURXJKO\ HYDOXDWH WZR

PAGE 88

SURPLVLQJ LQWURGXFWLRQV DJDLQVW %LJDLWD DQG 5HGDOWD 7KH JRDO ZDV WR REWDLQ D SHUVLVWHQW KLJK TXDOLW\ KLJK \LHOGLQJ OLPSRJUDVV IRU IODWZRRG VLWHV LQ )ORULGD DQG WKH VRXWKHDVW 0DWHULDOV DQG 0HWKRGV 7ZR H[SHULPHQWDO DUHDV ZHUH XVHG LQ WKLV VWXG\ 7KH H[SHULn PHQWDWLRQ ZDV FRQGXFWHG RQ IRXU ZHOO HVWDEOLVKHG SODQWLQJV RI OLPSRJUDVV RQ D :DFKXOD VDQG D SRRUO\ GUDLQHG VLOLFHRXV K\SHUWKHUPLF XOWLF KDSORTXRG VRLO DW WKH %HHI 5HVHDUFK 8QLW RI WKH 8QLYHUVLW\ RI )ORULGD $QRWKHU ILHOG ZDV HVWDEOLVKHG LQ E\ YHJHWDWLYH SURSDJDWLRQ IURP PDWHULDO WDNHQ IURP WKH H[SHULPHQWDO DUHD 7KH QHZ SODQWLQJ ZDV LQ DQ $GDPVYLOOH VDQG 7KH OLPSRJUDVVHV XVHG ZHUH 3, 3, 5HGDOWD DQG %LJDOWD 5HGDOWD LV GLSORLG Q f DQG WKH RWKHU JUDVVHV XVHG KHUH DUH WHWUDSORLGV Q f (VWDEOLVKPHQW 7KH H[SHULPHQWDO DUHD ZDV [ P DQG VXSSRUWHG D PDWXUH VWDQG RI U\H 6HFDOH FHUHDOH /f SULRU WR FXOWLYDWLQJ ZLWK D *URXQG +DZJ URWRWLOOHU RQ 0D\ 7KH ILHOG ZDV EORFNHG LQWR WKUHH VHFWLRQV DQG ZLWKLQ HDFK VHFWLRQ IRXU PDLQ SORWV JHQRW\SHVf ZHUH PDUNHG WR PHDVXUH [ P ZLWK P DOOH\V EHWZHHQ HDFK PDLQ SORW 7KH DOOH\V ZHUH SODQWHG WR n$UJHQWLQHn EDKLDJUDVV :DLVW KLJK VWDQGV RI HDFK OLPSRJUDVV JURZLQJ LQ WKH H[SHULn PHQWDO DUHD ZHUH PRZHG ZLWK D VLFNOH EDU VHW DW FP KHLJKW UDNHG DQG FDUULHG WR DQ DSSURSULDWHO\ DVVLJQHG UDQGRP ORFDWLRQ ZLWKLQ HDFK EORFN 7KH KHUEDJH ZDV HYHQO\ GLVWULEXWHG RYHU WKH VRLO OLJKWO\ GLVNHG WR FP DQG WKH HQWLUH ILHOG ZDV UROOHG XVLQJ D FXOWLSDFNHU VHHGHU WR

PAGE 89

DVVXUH JRRG VRLO WR VWHP FRQWDFW 7KH ILHOG ZDV LUULJDWHG ZKHQ QHFHVn VDU\ WR LQVXUH HVWDEOLVKPHQW DQG IHUWLOL]HG RQ -XQH DQG $XJXVW ZLWK NJKD RI 1A2A)2M2f FRQWDLQLQJ NJKD RI D PLFURQXWULHQW PL[ 6RLO WHVW UHVXOWV IURP -XQH VKRZHG D S+ RI NJKD SKRVSKRUXV 3f DQG NJKD SRWDVVLXP .f 7KH ILHOG ZDV IODLO FKRSSHG RQ 'HFHPEHU WR FP KHLJKW WR UHPRYH IURVWHG IRUDJH ([SHULPHQWDWLRQ :LWKLQ HDFK HVWDEOLVKHG OLPSRJUDVV VZDUG LQ WKH H[SHULPHQWDO DUHD SORWV ZHUH PDUNHG RXW IRU D FOLSSLQJ VWXG\ 7KH EORFNV ZHUH XQUHSOLFDWHG KHQFH WKH GHVLJQ ZDV D VSOLW SORW GHVLJQ ZLWKRXW UHSOLFDn WLRQ RI PDLQ SORWV 7KH FOLSSLQJ WUHDWPHQWV ZLWKLQ D OLPSRJUDVV EORFN ZHUH UDQGRPO\ DVVLJQHG DQG UHSOLFDWHG WKUHH WLPHV 7KH JHQRW\SH r WUHDWPHQW UHSf WHUP ZDV XVHG WR WHVW JHQRW\SH FOLSSLQJ WUHDWPHQW DQG LQWHUDFWLRQ HIIHFWV IRU WKH UHVSRQVHV WKDW ZHUH VWXGLHG 7KH FOLSSLQJ H[SHULPHQW ZDV LQLWLDWHG E\ PRZLQJ DOO SORWV WR FP RQ -XO\ DQG KDUYHVWLQJ DW DQG ZHHN LQWHUYDOV XQWLO 2FWREHU 7KH OLPSRJUDVVHV ZHUH IHUWLOL]HG RQ 0D\ DQG -XO\ ZLWK NJKD RI 1A2A,A2f FRQWDLQLQJ SHUFHQW RI D PLFURn QXWULHQW PL[ 6RLO WHVW UHVXOWV IURP VDPSOHV WDNHQ RQ 'HFHPEHU VKRZHG D S+ RI NJKD 3 DQG NJKD 7KH ZHHN LQWHUYDO WUHDWPHQWV ZHUH FOLSSHG DW FP KHLJKW RQ $XJXVW $XJXVW 6HSWHPEHU DQG 2FWREHU XVLQJ D -DUL VLFNOH EDU PRZHU
PAGE 90

2FWREHU 7KH ZHHN WUHDWPHQW ZDV KDUYHVWHG DQG VDPSOHG IRU ,920' \LHOG DQG &3 RQO\ RQ 2FWREHU ([SHULPHQWDWLRQ 6RLO WHVW UHVXOWV IURP WKH QHZO\ HVWDEOLVKHG ILHOG RI OLPSRJUDVV UHYHDOHG D S+ RI NJKD 3 DQG NJKD 2Q 0DUFK WKH ILHOG ZDV XQLIRUPO\ PRZHG WR FP KHLJKW DQG WUHDWPHQWV FRQVLVWLQJ RI ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH FXWWLQJ IUHTXHQn FLHV )f ZHUH LPSRVHG RQ WKH PDLQ SORWV 7KH WZR IDFWRUV 1 DQG )f ZHUH W K V FRPELQHG WR UHSUHVHQW RI D [ FRPSOHWH IDFWRULDO )LJXUH VKRZV D [ IDFWRUDO DQG WKH VSDWLDO DUUDQJHPHQW RI WZR IDFWRUV DQG )LJXUH VKRZV D FHQWUDO FRPSRVLWH GHVLJQ LQ WKUHH IDFWRUV 7KH WUHDWPHQW FRPELQDWLRQV XVHG LQ WKLV VWXG\ DUH VKRZQ JUDSKLFDOO\ LQ )LJXUH DQG WKH GHVLJQ SRLQWV DUH SORWWHG DV JULG FRRUGLQDWHV LQ )LJXUH 7KH DUUDQJHPHQW RI WUHDWPHQW FRPELQDWLRQV LQ )LJXUH VKRZV WKH XQGHUO\LQJ GLIIHUHQFH EHWZHHQ WKH FODVVLFDO FHQWUDO FRPSRVLWH GHVLJQ )LJXUH f DQG WKH PRGLILHG FHQWUDO FRPSRVLWH GHVLJQ GHVFULEHG E\ /LWWHOO DQG 0RWW f 1RWH LQ )LJXUH KRZ LQIRUPDWLRQ LV FRQFHQn WUDWHG LQ ZKDW ZDV FRQVLGHUHG WKH PDQDJHPHQW UHDOP NJKD 1f DQG WR ZHHN GHIROLDWLRQ IUHTXHQFLHV $OVR QRWH WKH JHRPHWULF VFDOLQJ RI 1 DQG WKH DULWKPHWULF SURJUHVVLRQ IRU FXWWLQJ IUHTXHQF\ ZLWK D VNLS DW WKH ZHHN OHYHOf $V WKH GRWWHG OLQHV VXJJHVW WKH GHVLJQ FRXOG EH FRQVLGHUHG D VXSHULPSRVLWLRQ RI WZR FRPSOHWH IDFWRULDOV DQG f &RQVHTXHQWO\ WKH JUDVVHV ZHUH UHSOLFDWHG DV PDLQ SORWV LQ D VSOLW SORW GHVLJQ DQG WKH WUHDWPHQWV SURYLGHG D UHVSRQVH VXUIDFH ZLWK DOO SRLQWV UHSOLFDWHG WKUHH WLPHV (DFK PDLQ SORW FRQWDLQHG SORWV

PAGE 91

)LJXUH 6SDWLDO DUUDQJHPHQW RI D WKUHH OHYHO IDFWRULDO GHVLJQ LQ WZR YDULDEOHV )LJXUH 6SDWLDO DUUDQJHPHQW RI D FHQWUDO FRPSRVLWH GHVLDQ LQ WKUHH YDULDEOHV

PAGE 92

75($70(17 0$75,; p p _ac p 1 k )(57,/,=$7,21 NJKD\Uf r p k _@ k k p k k p &877,1* )5(48(1&< ZNVf )LJXUH 7UHDWPHQW PDWUL[ VKRZLQJ WUHDWPHQW QXPEHUV LQVLGH HDFK DFWXDO FLUFOHVf DQG DEVHQW VTXDUHVf WUHDWn PHQW FRPELQDWLRQ 1 )(57,/,=$7,21 NJKD\Uf $&78$/ 25,(17$7,21 2) )$&725 63$&( ZNVf )LJXUH 7UHDWPHQW FRPELQDWLRQV RI QLWURJHQ 1f DQG FXWWLQJ IUHTXHQF\ SORWWHG DV JULG FRRUGLQDWHV ZLWK WKH RULJLQ DW NJKD\U 1 DQG ZHHNV FXWWLQJ IUHTXHQF\

PAGE 93

[ Pf IRU WUHDWPHQWV 7KH HQWLUH ILHOG KDG WRWDO SORWV UHSOLFDWLRQV [ JUDVVHV [ WUHDWPHQWVf )LJXUH VKRZV WKH WUHDWPHQW OHYHOV RI 1 DQG ) WKH WUHDWPHQW QXPEHU DQG WKH KDUYHVW ),f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f FRQWDLQLQJ PLFURHOHPHQWV SULRU WR WKH VWDUW RI WKH H[SHULPHQW DQG DW WKH KDOIZD\ SRLQW ODWH LQ -XO\ 7KH VDPSOLQJ VFKHGXOH LV VKRZQ E\ V\PEROV LQ )LJXUH IRU ,90' &3 &3 [ 1f '0 \LHOG GHWHUPLQDWLRQ DQG 71& PHDVXUHPHQW )RUDJH FXW IRU \LHOG DQG TXDOLW\ ZDV RYHQ GULHG DW & 7KH GULHG IRUDJH ZDV JURXQG LQ D :LOH\ PLOO WR SDVV D PP VFUHHQ DQG DQDO\]HG IRU SHUFHQW 1 E\ WKH .MHOGDKO SURFHGXUH 7KH ,90' GHWHUPLQDWLRQ XVHG D PRGLILFDWLRQ RI WKH 7LOOH\ DQG 7HUU\ WHFKQLTXH 0RRUH HW DO f 7RWDO QRQVWUXF WXUDO FDUERK\GUDWH 71&f VDPSOHV ZHUH UDQGRPO\ VHOHFWHG GXJ ZDVKHG IUHH RI VRLO DQG DUUDQJHG LQ FRWWRQ EDJV 7KH FRWWRQ EDJV ZHUH WKHQ SDFNHG LQ D SODVWLF EDJ DQG SXW RQ LFH 8SRQ UHDFKLQJ WKH ODE WKH VDPSOHV ZHUH GULHG LQ D 7KHOFR IRUFHGDLU RYHQ DW & IRU PLQXWHV DIWHU ZKLFK WKH WHPSHUDWXUH ZDV ORZHUHG WR & XQWLO WKH GULHG VDPSOHV ZHUH UHPRYHG KRXUV ODWHU 6PLWK Df 7KH VWHP EDVHV ZHUH VFUDSHG IUHH RI URRWV OHDI VKHDWKV DQG VDQG E\ XVLQJ D ZLUH

PAGE 94

<,(/' $1' 48$/,7< $ $1' 71& N 6$03/,1* ,1 7UW 1R 1 ) NJKD\Uf ZNVf 5HSV ,, +$59(67 '$7( $1' :((. +2 +, + + + + + + + + +,2 +OO +, 0DU $SU 0D\ 0D\ -XQ -XO -XO $XJ 6HS 6HS 2FW 1RY 1RY $ƒ$$$$$ƒ8ƒ$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ $$$$$$$$$$$$$ )LJXUH
PAGE 95

EXIILQJ ZKHHO DWWDFKHG WR WKH H[WHQGHG VKDIW RI D VPDOO HOHFWULF PRWRU : 530f 7KH ERWWRP FP RI WKH VWHP FURZQVf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f RQ DQ $PGDKO 9 ZLWK 26096 5HOHDVH DQG -(61-( 5HOHDVH &RPSXWLQJ ZDV SHUIRUPHG DW WKH 'HSDUWPHQW RI $JULFXOWXUDO (QJLQHHULQJ DQG WKH 1RUWKHDVW 5HJLRQDO 'DWD &HQWHU RI WKH 6WDWH 8QLYHUVLW\ 6\VWHP RI )ORULGD ORFDWHG RQ WKH FDPSXV RI WKH 8QLYHUVLW\ RI )ORULGD LQ *DLQHVYLOOH 0HDQV ZHUH DQDO\]HG XVLQJ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW $QDO\VLV RI YDULDQFH DQG UHJUHVVLRQ ZHUH XVHG WR WHVW UHVSRQVH

PAGE 96

VXUIDFH PRGHOV IRU DGHTXDWH ILW $ FRPSOHWH H[SODQDWLRQ RI WKH 6$6 PHWKRGRORJ\ IRU FRQVWUXFWLQJ VXUIDFH SORWV LV JLYHQ LQ &KDSWHU 5HVXOWV 'U\ 0DWWHU
PAGE 97

7DEOH 7RWDO GU\ PDWWHU '0f \LHOGV IRU IRXU OLPSRJUDVVHV FOLSSHG DW WKUHH GLIIHUHQW IUHTXHQFLHV IRU ZHHNV HQGLQJ RQ 2FWREHU /LPSRJUDVV &OLSSLQJ IUHTXHQF\ ZNVf '0 NJKDf 3, D DE D n5HGDOWDn D F G n%LJDLWD D D E 3, D E F &OLSSLQJ WUHDWPHQWV ZLWKLQ D OLPSRJUDVV JHQRW\SH DUH FRPSDUHG XVLQJ WKH XQGHUOLQLQJ WHFKQLTXH DQG OHWWHUV ZLWKLQ D FROXPQ FRPSDUH OLPSRJUDVVHV IRU D WUHDWPHQW $Q\ YDOXHV VKDULQJ D FRPPXQ XQGHUOLQH RU OHWWHU DUH QRW GLIIHUHQW 3 f XVLQJ WKH :DOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW

PAGE 98

727$/ '0 352'8&7,21 )25 )285 /,032*5$66(6 ,1 3, n5HGDOWDn n%LJDOWDn 3, )LJXUH 7RWDO GU\ PDWWHU '0f \LHOG IRU IRXU OLPSRJUDVVHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI GHIROLDWLRQ WKURXJKRXW WKH JURZLQJ VHDVRQ DW WKH %HHI 5HVHDUFK 8QLW QHDU *DLQHVYLOOH )ORULGD

PAGE 99

7DEOH (IIHFW RI QLWURJHQ 1f UDWHV DQG FXWWLQJ IUHTXHQF\ )f RQ GU\ PDWWHU '0f \LHOG LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f FUXGH SURWHLQ &3f KDUYHVW RI SURWHLQ DQG7HUW77<]HU 1 HIILFLHQF\ )HUW 1 NJKDf 'HIROLDWLRQ /OPSRJUDVV 7RWDO '0 \LHOG P WRQKD\Uf '0 \LHOG DGGHG E\ 1 P WRQKDf +DUYHVW RI SURWHLQ NJKD\Uf 1 KDUYHVWHG DSSOLHG 1 ;f -XO\ 1RYHPEHU 6WDQG : )UHT ) ZNVf 7RWDO QR FD 7RWDO '0 \LHOG P WRQKDf ,920' rf &3 bf '0 \LHOG P WRQKDf ,920' bf &3 ;f 3, 5HGDOWD f§ f§ %LJDOWD f§ f§ 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, f§ 5HGDOWD f§ f§ %LJDOWD f§ f§ 3, f§ f§

PAGE 100

7DEOH &RQWLQXHG )HUW 1 NJKDf 'HIROLDWLRQ /LPSRJUDVV 7RWDO '0 \LHOG P WRQKD\Uf '0 \LHOG DGGHG E\ 1 P WRQKDf +DUYHVW RI SURWHLQ NJKD\Uf 1 KDUYHVWHG DSSOLHG 1 rf -XO\ '0 \LHOG P WRQKDf ,90' rf &3 ;f 1RYHPEHU '0 \LHOG ,90' P WRQKDf bf &3 ;f 6WDQG bf )UHT ) ZNVf 7RWDO QR 3HU FXW 7RWDO 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD f§ f§ %LJDOWD f§ f§ 3, 3, 5HGDOWD %LJDOWD 3, 3, 5HGDOWD %LJDOWD 3, r 7RWDO '0 SUHGLFWLRQ HTXDWLRQV JDYH '0 YDOXHV IRU QLWURJHQ DW DQG ZHHN GHIROLDWLRQV

PAGE 101

VOLJKW \LHOG DGYDQWDJH LQ WKH WUHDWPHQW FXW HYHU\ ZHHNV LQ DV ZDV VLPLODUO\ VKRZQ LQ IRU %LJDOWD FXW HYHU\ ZHHNV 7DEOH f :KDW LV FOHDUO\ VKRZQ LQ 7DEOH KRZHYHU LV WKH WUHPHQGRXV DELOLW\ RI DOO WKH OLPSRJUDVVHV WR UHVSRQG WR 1 DQG DFFXPXODWH '0 HVSHFLDOO\ DW ORQJHU FXWWLQJ LQWHUYDOV 7DEOH DOVR VKRZV WKH LQFUHPHQW RI WRWDO '0 \LHOG FDXVHG E\ 1 IHUWLOL]DWLRQ %URZQ f WHUPHG SURGXFWLYLW\ RI ELRPDVV SHU XQLW RI 1 DSSOLHG DV QLWURJHQ XVH HIILFLHQF\ 7KH '0 SURGXFWLRQ DW ]HUR 1 LQ HDFK FXWWLQJ IUHTXHQF\ ZDV VXEWUDFWHG IURP WKH SURGXFWLRQ REWDLQHG ZLWK 1 DSSOLHG LQ RUGHU WR GHWHUPLQH WKLV VWDWLVWLF ,W LV VKRZQ WKDW WKH PD[LPXP '0 UHVSRQVH WR 1 ZDV REWDLQHG ZLWK KLJK 1 UDWHV DQG ORQJ FXWWLQJ IUHTXHQFLHV r r r DQG rf 1R FOHDU FXW DGYDQWDJH ZDV REVHUYHG IRU DQ\ RQH OLPSRJUDVV DFURVV DOO WUHDWPHQWV 7KH PRVW SURWHLQ KDUYHVWHG LQ KHUEDJH ZDV REWDLQHG IURP WKH r WUHDWPHQW DV VKRZQ LQ 7DEOH 7KH DFWXDO HIILFLHQF\ RI QLWURJHQ XSWDNH ZDV DOVR FDOFXODWHG U 1 LQ SODQWV 1 LQ SODQWV UHFHLYLQJ ]HUR 1 r OQQ r 1 IHUWLOL]HU DSSOLHG /LPSRJUDVV 3, ZDV EHVW IRU DOPRVW DOO WUHDWPHQWV LQ WKLV VWDWLVWLF EXW LW ZDV QRW DV HIIHFWLYH DW WDNLQJ XS 1 DW ]HUR 1 FRQGLWLRQV 7DEOH f 'U\ PDWWHU '0f \LHOG GLVWULEXWLRQ LV DV LPSRUWDQW DV WRWDO SURGXFn WLRQ LQ )ORULGD 7KH ZHHNORQJ H[SHULPHQW ZDV GLYLGHG LQWR WKUHH ZHHN VHFWLRQV WKH ILUVW LQWHUYDO WHUPHG VSULQJ DQG WKH ODVW ZHHN LQWHUYDO FDOOHG DXWXPQ 2QO\ WKH DQG ZHHN WUHDWPHQWV ZHUH LQFOXGHG LQ VSULQJ DQG DXWXPQ EHFDXVH WKH DQG ZHHN FXWWLQJ

PAGE 102

IUHTXHQFLHV GLG QRW PDWFK LQ KDUYHVW VHTXHQFH 7KH JUDSK LQ )LJXUH VKRZV '0 SORWWHG DJDLQVW WUHDWPHQW FRPELQDWLRQ 6SULQJ SURGXFWLRQ ZDV JUHDWHU WKDQ DXWXPQ SURGXFWLRQ HVSHFLDOO\ LQ WKH WUHDWPHQWV FXW HYHU\ ZHHNV 7KH GHFOLQH LQ DXWXPQ \LHOG IRU ZHHN WUHDWPHQWV ZLWK LQFUHDVLQJ 1 OHYHOV H[HPSOLILHG WKH HIIHFW RI IUHTXHQW GHIROLDWLRQ DQG KLJK 1 UDWHV RQ OLPSRJUDVV ,W LV LQWHUHVWLQJ WR QRWH WKDW '0 \LHOG LQ WKH r WUHDWPHQW IRU %LJDOWD SURGXFHG P WRQV LQ VSULQJ EXW WKH VXEVHTXHQW VWUHVV SHUPLWWHG RQO\ P WRQV WRWDO '0 \LHOG 7DEOH f &OHDUO\ 3, ZDV WKH KLJKHVW LQ VSULQJ SURGXFWLRQ DW WKH r WUHDWPHQW FRPELQDWLRQ )LJXUH f &DOFXODWLRQV XVLQJ WKH r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f ZKR GHPRQVWUDWHG WKH QHFHVVLW\ RI HDUO\ $XJXVW VWDJLQJ IRU DGHTXDWH '0 DFFXPXODWLRQ IRU VWRFNSLOLQJ $W DQ\ VWDJLQJ GDWH WKH VHFRQG KLJKHVW 1 UDWH SURGXFHG QHDUO\ DV PXFK RU PRUH IRUDJH DV WKH KLJKHVW 1 UDWH ,Q WKH H[SHULPHQW 3, \LHOGHG P WRQV IRU WKH VLQJOH ZHHN FXWWLQJ LQWHUYDO VWDJHG IURP -XO\ WR 2FWREHU 7DEOH f 1LWURJHQ ZDV DSSOLHG RQ -XO\ DW D UDWH RI NJKD ,Q WKH H[SHULPHQW NJKD 1 ZDV DSSOLHG RQ -XO\ IRU WKH

PAGE 103

) 75($70(17 &20%,1$7,21 )LJXUH 6SULQJ DQG DXWXPQ VHDVRQDO GLVWULEXWLRQ RI GU\ PDWWHU '0f \LHOG IRU IRXU OLPSRJUDVVHV LQ

PAGE 104

'0
PAGE 105

r WUHDWPHQW DQG LW VXEVHTXHQWO\ \LHOGHG P WRQV )LJXUH f ,Q 4XHVHQEHUU\ DQG 5XHONH f DSSOLHG NJKD 1 WR 3, RQ $XJXVW DQG PHDVXUHG P WRQV VWRFNSLOHG '0 RQ 'HFHPEHU 7KLV YDOXH LV QHDUO\ LGHQWLFDO WR WKH P WRQV '0 PHDVXUHG LQ )LJXUH f IRU WKH r WUHDWPHQW NJKD 1 DSSOLHG -XO\ f 'U\ 0DWWHU
PAGE 106

5(6,'8$/ 75($70(17 ())(&76 21 '0 <,(/' ,1 )520 &877,1* $1' )(57,/,=$7,21 ,1 7+( (;3(5,0(17$7,21 GDWHV WRWDO -XQH $XJXVW )LJXUH 5HVLGXDO FOLSSLQJ DQG IHUWLOL]DWLRQ HIIHFWV RQ GU\ PDWWHU '0f \LHOG IRU 3, KDUYHVWHG HYHU\ ZHHNV LQ -XQH $XJXVW DQG 2FWREHU QRW VKRZQff DQG WKH WRWDO SURGXFWLRQ IRU WKH \HDU &DXWLRQ '0 \LHOG XQLWV YDU\ DPRQJ JUDSKV

PAGE 107

KLJKHU '0 P WRQVKDf IROORZHG E\ 3, P WRQVKDf 3, P WRQVKDf DQG 5HGDOWD P WRQVKDf &UXGH 3URWHLQ ([SHULPHQW 7DEOH UHYHDOV WKH PRVW GLVDSSRLQWLQJ FKDUDFWHULVWLF RI OLPSRJUDVV ORZ &3 FRQWHQW ,Q WKH H[SHULPHQW WKH SORWV ZHUH IHUWLOL]HG RQ 0D\ DQG -XO\ ZLWK NJKD 1 EXW WKLV ZDV QRW VXIILFLHQW WR UDLVH WKH &3 OHYHOV LQ WKH DQG ZHHN WUHDWPHQWV WR WKH SHUFHQW YDOXH GHHPHG QHFHVVDU\ IRU PDLQWHQDQFH LQ DQLPDOV 0LOIRUG DQG 0LQVRQ f %LJDOWD KDG WKH KLJKHVW YDOXHV IRU &3 &UXGH 3URWHLQ ([SHULPHQW ,Q )LJXUH WKH &3 YDOXHV DUH VKRZQ IRU WKH PLGGOH -XO\f DQG HQG 1RYHPEHUf RI WKH H[SHULPHQW )RU -XO\ QRWH KRZ TXLFNO\ &3 GURSV IURP WKH KLJKHVW YDOXHV DW WKH OHIW IURQW HGJH RI WKH SORW ZKHUH 1 UDQJHV IURP NJKD\U LQ DVVRFLDWLRQ ZLWK D ZHHN FXWWLQJ IUHTXHQF\f 1RW D VLQJOH YDOXH IURP VDPSOHV ROGHU WKDQ ZHHNV RI DJH ZDV DERYH SHUFHQW &3 7DEOH f 7KH GLSORLG 5HGDOWDf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

PAGE 108

7DEOH 3HUFHQW FUXGH SURWHLQ &3f IRU IRXU LPSRJUDVVHV FOLSSHG DW WKUHH GLIIHUHQW IUHTXHQFLHV IRU ZHHNV HQGLQJ RQ 2FWREHU 6DPSOHV WDNHQ RQ 2FWREHU ZHUH DQDO\]HG IRU &3f /LPSRJUDVV &OLSSLQJ IUHTXHQF\ ZNVf &UXGH SURWHLQ bf r 3, F D E n5HGDOWDn EH D D n%LJDOWD D D E 3, E D E &OLSSLQJ WUHDWPHQWV ZLWKLQ D OLPSRJUDVV JHQRW\SH DUH FRPSDUHG E\ WKH XQGHUOLQLQJ WHFKQLTXH DQG OHWWHUV ZLWKLQ D FROXPQ FRPSDUH OLPSRJUDVV YDOXHV IRU D WUHDWPHQW $Q\ YDOXHV VKDULQJ D FRPPRQ XQGHUOLQH RU OHWWHU DUH QRW GLIIHUHQW 3 f XVLQJ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULVRQ 7HVW

PAGE 109

&3 ,1 7:2 /,032*5$66(6 )25 7:2 +$59(676 ,1 3, n5HGDOWDn )LJXUH &RPSDULVRQ RI FUXGH SURWHLQ &3f SHUFHQWDJHV LQ WKH WLVVXH RI 3, DQG n5HGDOWD IRU WZR GDWHV LQ LQ UHVSRQVH WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI FOLSSLQJ &DXWLRQ 7KH &3 VFDOH LV QRQXQLIRUPO\ DQG QRQOLQHDUO\ VFDOHG RQ 1RYHPEHU

PAGE 110

ZHHN OHYHO RI FXWWLQJ IUHTXHQF\ DQG EH\RQG WKH ULGJH WKH PDWXULW\ RI WKH IRUDJH LQFUHDVHG DQG &3 YDOXHV IHOO 7DEOH VKRZV WKH VXEVWDQWLDO LPSURYHPHQW LQ SHUFHQW &3 EHWZHHQ -XO\ DQG 1RYHPEHU IRU DOO WKH JUDVVHV DQG DOO WUHDWPHQWV H[FHSW WKH r WUHDWPHQW DQG WKH ZHHN ROG VDPSOHV 2Q WKH 1RYHPEHU VDPSOLQJ GDWH RQO\ IRXU RI PHDQV IRU DQG ZHHN ROG VDPSOHV ZHUH EHORZ SHUFHQW &3 7DEOH f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f KDYH VKRZQ WKH ORZ GLJHVWLELOLW\ RI GLSORLGV FRPSDUHG WR WHWUD SORLG OLPSRJUDVVHV 5HGDOWD LV D GLSORLG ZKLOH WKH UHPDLQLQJ WKUHH JUDVVHV LQ WKLV VWXG\ DUH WHWUDSORLGV 7KH 1RYHPEHU SORWV VKRZHG D VPDOO EXW SHUFHSWLYHO\ KLJKHU ,920' DQG VRPHZKDW JUHDWHU XQGXODWLRQ KRZHYHU D WDEXODU SUHVHQWDWLRQ ZDV QHFHVVDU\ IRU LQFUHDVHG XQGHUVWDQGLQJ ,Q 7DEOH IRU DOO WUHDWPHQWV WKH UDQNLQJ RI OLPSRJUDVVHV IRU SHUFHQW ,920' ZDV %LJDOWD 3, 3, 5HGDOWD $ JODQFH DW WKH

PAGE 111

7DEOH 7KH PDLQ HIIHFW RI OLPSRJUDVV DQG FOLSSLQJ WUHDWPHQW RQ SHUFHQW LMQ YL WUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f RI WLVVXH VDPSOHG 2FWREHU ,920' f§ b /LPSRJUDVV 3, E n5HGDWD F n%LJDOWDn D 3, E &OLSSLQJ IUHTXHQF\ ZNVf D E E fN 9DOXHV VKDULQJ D FRPPRQ OHWWHU ZLWKLQ WKH FROXPQ DUH QRW GLIIHUHQW 3 f XVLQJ WKH :DOOHU'XQFDQ 0XOWLSOH &RPSDULn VRQ 3URFHGXUH

PAGE 112

,920' ,1 )285 /,032*5$66(6 )25 7:2 +$59(676 ,1 3, )LJXUH &RPSDULVRQ RI LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f IRU IRXU OLPSRJUDVVHV DW WZR KDUYHVW GDWHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHn TXHQFLHV )f RI FOLSSLQJ

PAGE 113

YDOXHV DFURVV DOO WUHDWPHQWV LQ WKH -XO\ VDPSOLQJ H[SODLQV WKH IODWn QHVV RI WKH VXUIDFHV LQ )LJXUH ,Q WKH 1RYHPEHU VDPSOLQJ WKHUH ZDV D JUHDWHU UDQJH RI ,920' YDOXHV DFURVV WUHDWPHQWV EXW WKH VXUIDFH SORWV LQ )LJXUH ZHUH WRSRJUDSKLFDOO\ WRR IODW IRU HDV\ GHWHFWLRQ RI GLIIHUHQFHV E\ WKH H\H 7KH LQFUHDVH LQ OLPSRJUDVV TXDOLW\ EHWZHHQ -XO\ DQG 1RYHPEHU ZDV UHIOHFWHG LQ WKH 7DEOH ,920' YDOXHV ,W ZDV GHFLGHG WR H[DPLQH WKH ,920' YDOXHV IRU %LJDOWD DQG 3, PRUH FORVHO\ LQ 7DEOH LQ RUGHU WR FRPSDUH WKH LQFUHDVHV LQ GLJHVWLELOLW\ IURP KDUYHVW WR KDUYHVW + WR +,f )RU 3, WKH TXDOLW\ LQFUHDVHG LQ DOO EXW WKH ZHHN WUHDWPHQWV DQG QHDUO\ LGHQWLFDO WUHQGV ZHUH VKRZQ LQ %LJDOWD 7KH ODUJHVW GLIIHUHQFHV EHWZHHQ + DQG +, ZHUH IRXQG LQ WKH DQG ZHHN ROG WLVVXH LQ ERWK JUDVVHV 'LIIHUHQFHV LQ ,920' EHWZHHQ %LJDOWD DQG 3, ZHUH SUHVHQWHG LQ 7DEOH LQ RUGHU WR GHPRQVWUDWH WKH FRQVLVWHQWO\ KLJKHU ,920' RI %LJDOWD 7KH DGYDQWDJH IRU %LJDOWD RYHU 3, ZDV PXFK PRUH GUDPDWLF RQ WKH 1RYHPEHU +,f VDPSOLQJ GDWH WKDQ RQ -XO\ +f 7RWDO 1RQVWUXFWXUDO &DUERK\GUDWH DV 5HODWHG WR ,920' +HDYLO\ IHUWLOL]HG OLPSRJUDVV LQ WKH DXWXPQ DSSHDUHG YHU\ OXVK FRPn SDUHG WR WKH VWLIIHU VWHPV IURP PLGVHDVRQ DQG LW ZDV WKRXJKW WKDW DFFXPXODWHG VWDUFK PLJKW H[SODLQ VRPH RI WKH DXWXPQ ,920' LQFUHDVHV LQ WKLV VWXG\ *DUUDUG DQG :HVW f DQG &DUWHU DQG *DUUDUG f VKRZHG WKDW VWDUFK DFFXPXODWHG LQ WKH OHDYHV RI GLJLWJUDVV GXULQJ FRRO QLJKWV 5HVSRQVH VXUIDFHV IRU VKRRW 71& ZHUH FRQVWUXFWHG IRU %LJDOWD 5HGDOWD DQG 3, IURP GDWD FROOHFWHG RQ -XO\ DQG 1RYHPEHU )LJXUHV DQG f ,Q )LJXUH D ULVLQJ SODQH LV VKRZQ IRU 3,

PAGE 114

0,'6800(5 3(5&(17 71& )281' ,1 7+( 6+227 7,668( 2) 7+5(( /,032*5$66(6 3, n5HGDOWDn n%LJDOWD b 71& )LJXUH 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VKRRW WLVVXH RI WKUHH OLPSRJUDVVHV RQ -XO\ VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH GHIROLDWLRQ IUHTXHQFLHV )f

PAGE 115

/$7( $87801 3(5&(17 71& )281' ,1 7+( 6+227 7,668( 2) 7+5(( /,032*5$66(6 n5HGDOWDn 3, n%LJDOWDn )LJXUH $ FRPSDULVRQ RI WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VKRRWV RI WKUHH OLPSRJUDVVHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI GHIROLDWLRQ VDPSOHG RQ 1RYHPEHU

PAGE 116

5HGDOWD KDV D ULVLQJ SODQH RI D PXFK VKDOORZHU VORSH DQG RI D ORZHU PDJQLWXGH WKDQ 3, 7KH %LJDOWD VXUIDFH SORW URVH SUHFLSLWRXVO\ WR D ULGJH DW WKH ZHHN FXWWLQJ IUHTXHQF\ DQG WKHQ WKH 71& LQ VKRRWV IHOO EDFN WR D ORZHU YDOXH DW WKH UHDU HGJH RI WKH FXEH ZHHN FOLSn SLQJ WUHDWPHQWVf ,Q )LJXUH 5HGDOWD DQG 3, ORRNHG DOPRVW WKH VDPH DV IRU WKH -XO\ VXUIDFHV %LJDOWD RQ WKH RWKHU KDQG LQFUHDVHG FDUERK\GUDWH FRQWHQW IRU ZHHN WUHDWPHQWV IURQW OHIW RI WKH SORWf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n VKLS LI DQ\ EHWZHHQ 71& ,920' DQG &3 ,Q 7DEOH WKH 71& DFFXPXODn WLRQ SDWWHUQV IRU 3, EHWZHHQ + DQG +, VKRZHG DQ LQFUHDVH LQ WKH FURZQV DQG D GHFUHDVH LQ WKH VKRRWV IRU DOO H[FHSW WKH r WUHDWn PHQW FRPELQDWLRQ )RU 3, LW ZDV ORJLFDO WR FRQFOXGH WKDW 71& SURGXFHG LQ VKRRWV LQ WKH VXPPHU ZDV WUDQVORFDWHG WR WKH FURZQV IRU VWRUDJH LQ WKH DXWXPQ ,Q %LJDOWD D GLIIHUHQW WUHQG HPHUJHG 7KH VKRRW 71& YDOXHV GHFUHDVHG IURP + WR +, KRZHYHU FRQWUDU\ WR WKH VLWXDWLRQ LQ 3, WKH %LJDOWD FURZQ SHUFHQW 71& DOVR GHFUHDVHG

PAGE 117

7DEOH %DODQFH VKHHW RI LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f DQG WRWDO QRQVWUXFWXUDO FDUERn K\GUDWHV 71&f LQ WZR OLPSRJUDVVHV 7UHDWPHQW /LPSRJUDVV ,920' +DUY 3, %LJDOWD GLIIHUHQFH 1FXW NJKDfA 7RWDO 1 &XW NJKD \U IUHT 1R RI 71& bf ,940' 71& bf ,920' %LJDOWD PL QXV KDUYHVWV 6KRRW &URZQ A 6KRRW &URZQ bfF 3, f + +, + +, + +, r + +, &2 + +,

PAGE 118

7DEOH &RQWLQXHG 7UHDWPHQW /LPSRJUDVV ,920' +DUY 3, %LJDLWD GLIIHUHQFH 1FXW NJKDA 7RWDL 1 &XW NJKD \U IUHT 1R RI 71& bf ,920' 71& ^bf ,920' %LJDOWD PLQXV KDUYHVWV 6KRRW &URZQ bf 6KRRW &URZQ bf 3, f + +, + +, + +, + r r +, r r + r r +, r r

PAGE 119

7DEOH &RQWLQXHG 1FXW NJKD A 7UHDWPHQW 7RfI &XW NJKD \U IUHT 1R RI KDUYHVWV +DUY QR 3, 71& bf 6KRRW &URZQ /LPSRJUDVV ,90' 71& A 6KRRW %LJDLWD bf &URZQ ,90' GLIIHUHQFH %LJDLWD ,90' PLQXV bf 3, f + +L + +, + +,

PAGE 120

RU UHPDLQHG WKH VDPHf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n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f 6HDVRQDO 71& 7UHQGV 7KH SULPDU\ REMHFWLYH RI XVLQJ WKH 71& DQDO\VHV ZDV LQ VWXG\LQJ VWRUDJH RI FDUERK\GUDWHV LQ OLPSRJUDVV 7KH PHDQV IRU DOO OLPSRJUDVVHV

PAGE 121

IRU DQ\ JLYHQ KDUYHVW GDWH DQG WUHDWPHQW DUH VKRZQ LQ $SSHQGL[ % 7KH RYHUDOO VHDVRQDO WUHQGV FDQ EH FRQVWUXFWHG JLYHQ WKLV LQIRUPDWLRQ 7KH WUHQGV ZHUH f KLJKHVW 71& LQ WKH VSULQJ f ORZHVW 71& LQ PLGVXPPHU f DQ LQFUHDVH LQ 71& IURP $XJXVW WR PLG6HSWHPEHU f D GHFUHDVH LQ 71& WR 1RYHPEHU DQG f D VKDUS LQFUHDVH DJDLQ IURP 'HFHPEHU XQWLO VSULQJ 7KH 71& LQFUHDVHV LQ WKH DXWXPQ ZHUH PRVW SUHGRPLQDQW IRU 3, 7KH GHFUHDVH RI VWRUHG 71& LQ FURZQV LQ VXPPHU PD\ EH DVVRFLDWHG ZLWK UDSLG JURZWK UDWHV DQG PD[LPXP GD\ DQG QLJKW WHPSHUDWXUHV FDXVLQJ D VHYHUH GHPDQG RQ 71& LQ WKH VWRUDJH RUJDQV ,Q VXPPHU 71& ZDV DOVR KLJKHVW LQ WKH VKRRWV EHFDXVH RI DFWLYH JURZWK DQG SKRWRV\QWKHVLV 0F&DUW\ %URZQ DQG %OD]HU %ODVHU HW D f $V QLJKW WHPSHUDWXUHV FRROHG RII LQ HDUO\ DXWXPQ WKH UHVSLUDWRU\ GHPDQG PD\ KDYH GHFUHDVHG DQG DFWLYH SKRWRV\QWKHVLV LQ WKH GD\ DOORZHG D QHW VWRUDJH RI 71& :DUGODZ f VWDWHG WKDW *URZWK RI HVWDEOLVKHG VKRRWV KDV SULRULW\ RYHU URRW DQG EXG JURZWK XQGHU FRQGLWLRQV RI DVVLPLODWH GHILFLHQF\ S f $IWHU UHVHUYHV EHJDQ WR LQFUHDVH DJDLQ LQ HDUO\ DXWXPQ WKH WLOOHUV DQG URRWV EHJDQ DFWLYH JURZWK KRZHYHU WKLV QHZ JURZWK ZRXOG UHTXLUH VXEVWUDWH ZKLFK PD\ H[SODLQ WKH VPDOO GHFUHDVH RI 71& LQ FURZQV EHWZHHQ 6HSWHPEHU DQG 1RYHPEHU ,Q WKLV VWXG\ 71& PD\ KDYH EHHQ WUDQVORFDWHG WR VLWHV RI URRW DQG WLOOHU V\QWKHVLV LQVWHDG RI EHFRPLQJ ORFDOL]HG LQ WKH VKRRWV DV REVHUYHG E\ *DUUDUG DQG :HVW f DQG &DUWHU DQG *DUUDUG f &KDWWHUWRQ HW DO f VWXGLHG WLOOHUn LQJ DV D YDULDEOH RQ WKH HIIHFW RI FRRO QLJKW WHPSHUDWXUHV RQ 3DQJOD GLJLWJUDVV 7KH\ VKRZHG WKDW DFWLYHO\ WLOOHULQJ SODQWV DFFXPXODWHG QR VWDUFK ZKLOH QRQWLOOHULQJ SODQWV DFFXPXODWHG VWDUFK GXULQJ WKH GD\ WKDW ORFDOL]HG LQ WKH FKORURSODVWV GXULQJ FRRO QLJKWV

PAGE 122

:DUGODZ f VWDWHG WKDW WHPSHUDWXUH DIIHFWV JURZWK LH GHYHORSPHQWf JUHDWHU WKDQ SKRWRV\QWKHVLV DQG WUDQVORFDWLRQ 7KH ILQDO VXUJH RI FDUERK\GUDWH VKRZQ LQ OLPSRJUDVV FURZQV GXULQJ 'HFHPEHU DQG EH\RQG ZDV SRVVLEO\ FDXVHG E\ WKH SKRWRV\QWKHVL]LQJ EDVDO WLOOHUV GXULQJ ZDUP GD\V LQ WKH FRRO VHDVRQ 7RWDO 1RQVWUXFWXUDO &DUERK\GUDWHV DQG 3HUVLVWHQFH $QRWKHU LQWHQWLRQ RI XVLQJ 71& DQDO\VLV ZDV WR VWXG\ GLIIHUHQFHV LQ SHUVLVWHQFH REVHUYHG LQ WKH LPSRJUDVVHV ,Q )LJXUH WKH VDPSOLQJ VFKHGXOH IRU 71& VKRZV WKDW FRPSOHWH VDPSOLQJ IRU DOO WUHDWPHQWV ZDV EHJXQ RQ 6HSWHPEHU +f 8QWLO WKDW SRLQW WKH WUHDWPHQWV ZHUH DOORZHG WR WDNH WKHLU HIIHFW ,Q )LJXUH GDWD REWDLQHG IURP FURZQV KDUn YHVWHG RQ 6HSWHPEHU DQG 1RYHPEHU IURP 3, DQG %LJDOWD DUH SORWWHG $ SUHGRPLQDQW ULGJH DORQJ WKH IURQW OHIW HGJH RI DOO SORWV ZDV H[SODLQHG E\ WKH FKDQJH LQ PRUSKRORJ\ HIIHFWHG E\ FOLSSLQJ HYHU\ ZHHNV 3ODQWV ZHUH SURVWUDWH KDG PRUH WLOOHUV SHU XQLW DUHD DQG KHQFH KLJK SKRWRV\QWKHWLF DFWLYLW\ ZDV DEOH WR SURYLGH DVVLPLODWHV IRU VWRUDJH LQ WKH VWRORQV DQG VWHP EDVHV $ WURXJK RU EDVLQ VKRZQ IRU ERWK 3, DQG %LJDOWD LQ WKH 6HSWHPEHU SORWV ZDV FDXVHG E\ WKH ZHHN WUHDWn PHQWV LQ WKLV SRUWLRQ RI WKH UHVSRQVH VXUIDFHV 7KHVH WUHDWPHQW FRPn ELQDWLRQV ZHUH QRW VKRUW HQRXJK WR SURPRWH D SURVWUDWH KDELW DQG QRW ORQJ HQRXJK WR DOORZ FRPSOHWH VWHP HORQJDWLRQ (ORQJDWLRQ UHTXLUHV FDUERK\GUDWH FRPSRXQGV DQG EHIRUH WKH VKRRWV ZHUH DEOH WR UHVXSSO\ WKH VWRUDJH RUJDQV ZLWK SKRWRV\QWKDWH LW ZDV GHIROLDWHG DQG WKH F\FOH EHJDQ DJDLQ %H\RQG WKH ZHHN OHYHO RI FXWWLQJ IUHTXHQF\ WKH 71& FRQFHQWUDn WLRQ LQ FURZQV LQFUHDVHG JUDGXDOO\ IRU 6HSWHPEHU VDPSOHV ZKHUHDV

PAGE 123

QR 71& ,1 67(0 %$6(6 2) 7:2 /,032*5$66(6 )25 7:2 +$59(676 ,1 3, n%LJDOWD 6(37(0%(5 )LJXUH 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VWHP EDVH FURZQVf RI WZR LPSRJUDVVHV RQ WZR GDWHV DV DIIHFWHG E\ QLWURJHQ 1f IHUWLOL]DWLRQ DQG GHIROLDWLRQ IUHTXHQF\ )f

PAGE 124

P WKH LQFUHDVH ZDV PRUH GUDPDWLF IRU 1RYHPEHU HQGLQJ ZLWK D SODWHDX DORQJ WKH UHDU HGJH RI WKH VXUIDFH SORWV ,Q )LJXUH WKH 71& SHUFHQW RI WKH ZHHN WUHDWPHQWV LV SORWWHG DFURVV WKH VHDVRQ IRU 3, DQG %LJDLWD 7KH HIIHFW RI 1 FDXVHG DQ LQFUHDVH LQ 71& LQ WKH FURZQV HDUO\ LQ WKH \HDU ZKHQ VWRUHG FDUERK\GUDWH FRPSRXQGV ZHUH DEOH WR SURYLGH WKH FDUERQ VNHOHWRQV IRU SURWHLQ DQG WLVVXH V\QWKHVLV 7KH UDSLGO\ JURZLQJ WLVVXH SKRWRV\QWKH VL]HG HQRXJK FDUERK\GUDWH LQ %LJDOWD LQ WKH r WUHDWPHQW WR PDLQWDLQ PRUH VWRUHG VXEVWUDWH WKDQ WKH RWKHU OHYHOV RI 1 XQWLO WKH IRXUWK KDUYHVW $W WKLV WLPH WKH 71& YDOXHV GURSSHG LQ ERWK OLPSRJUDVVHV DV UHVSLUDWLRQ DQG JURZWK H[HUWHG D KHDY\ GHPDQG RQ VWRUHG 71& )URP PLGVXPPHU RQZDUG LQ )LJXUH WKH 1 OHYHO FDXVHG D GHSUHVn VLRQ LQ VWRUHG 71& IRU WKH ZHHN WUHDWPHQWV 7KH VKDUS ULVH RI 71& DW + LQGLFDWHG WKDW HQYLURQPHQWDO VWUHVV KDG GLPLQLVKHG LQ FRPSDULVRQ WR PLGVXPPHU 1HZ URRW DQG WLVVXH V\QWKHVLV ZDV PRVW OLNHO\ WKH FDXVH RI WKH ++ 71& GHFOLQH 7KH 71& LQFUHDVHG DJDLQ LQ WKH 1RYHPEHU VDPSOn LQJV RI 3, EXW %LJDOWD GLG QRW UHVSRQG LQ WKH VDPH PDQQHU DV ZDV VKRZQ LQ 1RYHPEHU VDPSOHV IURP WKH H[SHULPHQW &KDSWHU 7DEOH f 6DPSOLQJ ORJLVWLFV SUHYHQWHG DQ DQDO\VLV RI DFWXDO TXDQWLWLHV RI 71& SHU XQLW DUHD RI SORW 2QO\ SHUFHQW 71& ZDV PHDVXUHG DQG DV VKRZQ LQ )LJXUH WKHUH ZHUH QR ODUJH GLIIHUHQFHV EHWZHHQ WKH SHUVLVWHQW 3, f DQG QRQSHUVLVWHQW %LJDOWDf OLPSRJUDVVHV LQ SHUFHQW 71& 9LVXDO HVWLPDWHV RI SHUFHQW OLPSRJUDVV KRZHYHU GLG SHUPLW DQ DQDO\VLV RI SHUVLVWHQFH )LJXUH f )LJXUH VKRZV KRZ WKH 1 OHYHO FDXVHG UDSLG JURZWK HDUO\ LQ WKH VHDVRQ DV LQGLFDWHG E\ WKH KLJK SHUFHQW RI OLPSRJUDVV LQ ERWK VWDQGV XQWLO + 0LGVXPPHU VWUHVV FDXVHG E\ IUHTXHQW GHIROLDWLRQ KLJK

PAGE 125

)LJXUH $ FRPSDULVRQ RI WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f SHUFHQW LQ WKH FURZQV RI WZR OLPSRJUDVVHV VXEMHFWHG WR ZHHN GHIROLDWLRQ IUHTXHQFLHV )f DQG IHUWLOL]HG DW WKUHH OHYHOV RI QLWURJHQ 1f GXULQJ WKH JURZLQJ VHDVRQ

PAGE 126

9,68$/ (67,0$7,216 ([SHULPHQW )LJXUH 9LVXDO HVWLPDWLRQV RI SHUFHQW n%LJDOWDn DQG 3, IRU WUHDWPHQWV KDYLQJ ZHHN GHIROLDWLRQ IUHTXHQFLHV )f DQG WKUHH OHYHOV RI QLWURJHQ 1f LQ WKH JURZLQJ VHDVRQ

PAGE 127

IHUWLOL]DWLRQ DV ZHOO DV 71& GHPDQGV IURP JURZWK DQG UHVSLUDWLRQ FDXVHG D WUHPHQGRXV ORVV RI OLPSRJUDVV LQ SORWV 7KH ORVV RI %LJDOWD ZDV PXFK JUHDWHU EHFDXVH WKLV JUDVV KDG IHZHU VWRORQV WR VHUYH DV D FDUERK\GUDWH UHVHUYRLUV ZKHQ 1 ZDV DSSOLHG 7LVVXH GHJUDGDWLRQ DQG FRPSHWLWLRQ IURP VXPPHU ZHHGV IXUWKHU GHSUHVVHG SHUFHQW %LJDOWD DQG 3, XQWLO + ZKHQ WKH VXPPHU ZHHGV GLHG DQG WKH OLPSRJUDVV FRQWLQXHG JURZLQJ WKHUHn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f WKH DFWXDO WHPSHUDWXUH VWUHVVHV RQ WKH OLPSRJUDVVHV SUREDEO\ GLPLQLVKHG SULRU WR 2FWREHU :DWHU GHILFLWV KDYH EHHQ UHSRUWHG WR FDXVH LQFUHDVHV LQ VWRUHG 71& EXW WKH ZHHN SHULRG EHIRUH + KDG PRUH UDLQ FPf WKDQ PRVW RI WKH ZHHN SHULRGV LQ WKH KDUYHVW VHDVRQ )LJXUH f +HQFH ODFN RI ZDWHU FDQQRW EH XVHG WR H[SODLQ WKH 71& LQFUHDVH DW + ,Q )LJXUH WKH SHUFHQW OLPSRJUDVV EHJDQ WR ULVH DJDLQ EHWZHHQ + + $XJXVW 6HSWHPEHUf IRU WKH r DQG r WUHDWPHQWV LQ 3, DQG LQ WKH r %LJDOWD WUHDWPHQW 7KLV LQFUHDVH LQ OLPSRJUDVV SHUFHQW

PAGE 128

+$59(67 GDWH )LJXUH 5DLQIDOO DQG WHPSHUDWXUH GDWD IRU WKH JURZLQJ VHDVRQ WDNHQ DW WKH %HHI 5HVHDUFK 8QLW QHDU *DLQHVYLOOH )ORULGD

PAGE 129

RI VWDQGV PDWFKHV WKH DXWXPQ 71& SHDN DW + LQ )LJXUH )XUWKHU LQFUHDVHV ZHUH PDGH LQ SHUFHQW OLPSRJUDVV LQ VWDQGV RQFH WKH DQQXDO VXPPHU ZHHGV GLHG RXW DW + %HWZHHQ +, DQG +, WKH DYHUDJH PLQLPXP GDLO\ WHPSHUDWXUH GURSSHG EHORZ & )LJXUH f DQG WKH OLPSRJUDVVHV JUHZ YHU\ VORZO\ *UHHQ WLVVXH ZDV SUHVHQW WKURXJKRXW WKH ZLQWHU SHULRG KRZHYHU HYHQ DIWHU IUHH]LQJ WHPSHUDWXUHV ,Q )LJXUH WKH YLVXDO HVWLPDWLRQV IRU DOO OLPSRJUDVVHV EHWZHHQ -XQH DQG 6HSWHPEHU + + +f ZHUH SORWWHG DV UHVSRQVH VXUIDFHV 1RWLFH KRZ OLPSRJUDVV GRPLQDWHG VZDUG FRPSRVLWLRQ RQ -XQH EXW E\ -XO\ KDG GLPLQLVKHG DSSUHFLDEO\ IRU WKH ZHHN FXWWLQJ IUHTXHQF\ WUHDWPHQWV %\ 6HSWHPEHU WKH QRQSHUVLVWHQW OLPSRJUDVVHV %LJDOWD DQG 3, f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f KDG FRQWLQXHG KLV PHDVXUHPHQW RI \LHOG EH\RQG $XJXVW KLV %LJDOWD \LHOGV PLJKW KDYH LQFUHDVHG EH\RQG WKH P WRQV KH UHSRUWHG %\ $XJXVW WKH OLPSRJUDVVHV LQ WKLV VWXG\ SURGXFHG P WRQV %LJDOWD DQG 5HGDOWD SURGXFHG P WRQVf

PAGE 130

9,68$/ (67,0$7(6 2) 6:$5' &20326,7,21 2) )285 /,032*5$66 *(127<3(6 )25 7+5(( '$7(6 ,1 3, n5HGDOWDn n%LJDOWDn 3, )LJXUH 9LVXDO HVWLPDWHV RI IRXU OLPSRJUDVVHV VXEMHFWHG WR ILYH OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH IUHTXHQFLHV )f RI FOLSSLQJ DFURVV WKUHH VXPPHU KDUYHVWV

PAGE 131

7KH \LHOG SRWHQWLDO RI OLPSRJUDVV ZRXOG EH XQLPSUHVVLYH LI LW ZHUH QRW IRU LWV DXWXPQ DQG VSULQJ SURGXFWLRQ 7KHUH LV QR ODFN RI SRRU TXDOLW\ ELRPDVV LQ )ORULGD LQ WKH VXPPHU &RROVHDVRQ SURGXFWLRQ LV D JUHDWHU SUREOHP LQ QRUWK )ORULGD WKDQ IXUWKHU VRXWK &ROGHU WHPSHUDn WXUHV LQ QRUWK )ORULGD UHVWULFW WKH JURZWK RI VRPH LPSURYHG IRUDJHV WKDW GR ZHOO LQ VRXWK )ORULGD /LPSRJUDVV LV ZHOO DGDSWHG 2DNHV f 6WRFNSLOHG OLPSRJUDVV PD\ QRW EH WKH DQVZHU IRU OLPSRJUDVV XWLOL]DWLRQ HYHQ WKRXJK LW LV DQ LQH[SHQVLYH WHFKQLTXH /LPSRJUDVV LV VXSSRVHGO\ GLIILFXOW WR FXUH DV KD\ EXW WKHUH LV QRWKLQJ LQ WKH OLWHUDWXUH WR VXJJHVW DQ\RQH KDV WULHG 7KH LQFUHDVLQJ DFFHSWDQFH RI URXQG EDOHV DV ZHOO DV KD\ DPPRQLILFDWLRQ SURFHGXUHV FRXOG LPSURYH WKH XWLOL]DWLRQ RI OLPSRJUDVV WR WDNH DGYDQWDJH RI WKH WUHPHQGRXV SRWHQWLDO IRU '0 DFFXPXODWLRQ 5XHONH DQG 4XHVHQEHUU\ f DUH QRZ VWXG\LQJ WKH HDUO\ DSSOLFDWLRQ RI 1 WR SURGXFH ODUJH TXDQWLWLHV RI OLPSRJUDVV IRUDJH LQ WKH VSULQJ %LJ EDOHV FRXOG SUHVHUYH WKLV PDWHULDO EHIRUH VXPPHU UDLQV LQ -XQH $PPRQLILFDWLRQ VRXQGV LQWXLWLYHO\ PRUH DSSHDOLQJ WKDQ ODWH 1 IHUn WLOL]DWLRQ %OXH HW DO .UHWVFKPHU f DV D ZD\ RI LQFUHDVLQJ WKH 1 FRQWHQW RI PDWXUH OLPSRJUDVV $QRWKHU PHWKRG WR LQFUHDVH 1 LQ OLPSRJUDVV IRUDJH LV WR PL[ OHJXPHV LQ WKH VWDQG :LWK QR 1 DSSOLHG WKH OLPSRJUDVV FDQRS\ LV YHU\ RSHQ DQG PLJKW EH FRQGXFLYH WR WKH DVVRFLDWLRQ RI D OHJXPH VXFK DV MRLQW YHWFK $HVFK\QRPHQH DPHULFDQD /f :RUN LV SUHVHQWO\ LQLWLDWHG LQ WKLV UHJDUG 7KH OLPSRJUDVVHV KDYH UHFHQWO\ EHHQ DWWULEXWHG DQ DOOHRSDWKLF HIIHFW 5XHONH DQG 4XHVHQEHUU\
PAGE 132

/LPSRJUDVV 3, KDV WKH SRWHQWLDO WR DGG IOH[LELOLW\ WR WKH IODWZRRGV IRUDJH V\VWHP ,I SURGXFHUV KDYH EDKLDJUDVV SDVWXUHV WKH\ FDQ JUD]H 3, LQ VSULQJ DQG DXWXPQ 2FXPSDXJK f %DKLDJUDVV FRXOG EH JUD]HG LQ WKH VXPPHU DQG OLPSRJUDVV FRXOG DFFXPXODWH '0 XQWLO WKH VXPPHU UDLQV HQG DW ZKLFK WLPH LW FRXOG EH EDOHG LI DPPRQLILFDWLRQ ZRUNVf RU HQVLOHG DQG WKHQ VWDJHG IRU VWRFNSLOLQJ XQWLO 2FWREHU ZKHQ WKH EDKLDJUDVV VWRSV LWV JURZWK :KHQ ZLQWHU HQVXHV WKH OLPSRJUDVV VLODJH RU KD\ FRXOG EH IHG 7DEOH LV SUHVHQWHG DW WKLV SRLQW WR VKRZ WKH WUHDWPHQW FRPELQDn WLRQV UHSUHVHQWLQJ VHYHUDO FRPELQDWLRQV RI \LHOG DQG TXDOLW\ LQ WKH SUHVHQW VWXG\ ,W H[HPSOLILHV WKH KLJK TXDOLW\ DWWDLQHG LQ 3, ZLWKRXW WKH SURGXFWLRQ RI '0 RU WKH KLJK \LHOG ZLWKRXW DGHTXDWH TXDOLW\ &OHDUO\ D SURWHLQ VXSSOHPHQW PXVW EH LQMHFWHG LQWR D OLPSRJUDVV IHHGLQJ SURJUDP WKDW XWLOL]HV VRPH RI WKH FRPELQDWLRQV VKRZQ LQ 7DEOH )ORULGD KDV ERWK D VXPPHU DQG ZLQWHU VOXPS 0RWW DQG 0RRUH f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f PRGLn ILFDWLRQ WR WKH 7LOOH\ DQG 7HUU\ WZR VWDJH WHVW IRU ,920' VR\EHDQ PHDO LV IHG WR WKH ILVWXODWHG VWHHU D VKRUW WLPH SULRU WR FROOHFWLQJ WKH

PAGE 133

7DEOH 7UHDWPHQW FRPELQDWLRQV LQ 3, UHSUHVHQWLQJ WKH EHVW FRPSURPLVH EHWZHHQ \LHOG DQG TXDOLW\ LQ WKH H[SHULPHQW )(57 1 'HIROLDWLRQ 'DWH RI KDUYHVW 9 DUr 8UO 7RWDO 3HU 7RWDO )5(4 1R FXW RI RI 0D\ 0D\ -XQ -XO -XO $XJ 6HS 6HS 2FW 1RY 1RY NJKDf NJKDf '0 ,920' &3 '0 ,920' &3 '0 ,920' &3 '0 ,920' &3 r'0 P WRQKDf ,920' bf &3 bf

PAGE 134

UXPHQ IOXLG IRU WKH WHVW ,W LV SRVVLEOH WKDW WKH VR\EHDQ PHDO GLG QRW VXSSO\ VXIILFLHQW 1 WR WKH UXPHQ EDFWHULD LQ YLWUR WKHUHIRUH FDXVLQJ D OLPLWDWLRQ RQ WKH QRUPDO GHJUHH RI FHOO ZDOO GHJUDGDWLRQ -ROOLII HW DO f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n0DQQHWMH f 7KH VHDVRQDO IOX[ RI 71& LQ WKLV VWXG\ ZDV VLPLODU WR WKDW VKRZQ E\ ZLUHJUDVV $ULVWLGD VWULFWD 0LFK[f LQ WKH ZHVWHUQ VDQG KLOOV RI )ORULGD :RRGV HW DO f &DUERK\GUDWH UHVHUYHV LQ ZLUHJUDVV URRWV ZHUH LQ OHDVW VXSSO\ GXULQJ PLG-XO\ DQG ZHUH KLJKHVW DURXQG WKH ILUVW RI )HEUXDU\ ,Q $XVWUDOLD :LOVRQ DQG Wn0DQQHWMH f VKRZHG D 71& GHFUHDVH LQ VXPPHU DQG KLJKHU OHYHOV LQ WKH VSULQJ IRU OHDYHV RI EXIIHO JUDVV &HQFKUXV FLOLDULV FY n%LRORHODnf DQG JUHHQ SDQLF 3DQLFXP PD[LPXP YDU WULFKRJOXPH FY n3HWULHnf KRZHYHU WKH PDJQLWXGH RI VHDVRQDO 71& IOXFWXDWLRQV ZHUH IDU OHVV SHUFHQWf WKDQ REVHUYHG KHUH IRU OLPSR JUDVV SHUFHQW LQ 3, rf 2FWREHU WR SHUFHQW LQ %LJDOWD rf $SULOf 7KH 71& VWXGLHV GLG QRW HQWLUHO\ H[SODLQ WKH ORVV RI SHUVLVWHQFH LQ %LJDOWD DQG 3, 7KH OLWHUDWXUH ZDUQHG RI WKH SLWIDOOV RI SHUFHQW GDWD LQ 71& VWXGLHV +XPSKUH\V 0D\ f EXW WKH ORJLVWLFDO

PAGE 135

SUREOHPV ZHUH WRR LPPHQVH WR FRQVLGHU WKH H[WUD ZRUN LQYROYHG LQ SURGXFn LQJ 71& GDWD RQ DFWXDO DPRXQWV RI FDUERK\GUDWH SHU XQLW DUHD RU SHU VKRRW 9LVXDO HVWLPDWLRQV RI SHUFHQW OLPSRJUDVV ZHUH WDNHQ RQ DOO SORWV SULRU WR HDFK KDUYHVW E\ WKH VDPH H[SHULPHQWHUV 7KLV PHWKRG ZDV UHOLn DEOH HQRXJK WR FKDUDFWHUL]H WKH WUHDWPHQW HIIHFWV RQ WKH OLPSRJUDVV SHUVLVWHQFH 7KH UHVXOWV RI WKLV VWXG\ VKRZHG WKDW %LJDOWD ZDV VXSHULRU LQ TXDOLW\ DQG LV DQ H[FHOOHQW JUDVV ZKHQ SURSHUO\ PDQDJHG 7KLV FRQWHQn WLRQ LV ERUQH RXW E\ WKH IDFW WKDW LW UHFRYHUHG VR TXLFNO\ LQ IURP EHORZ SHUFHQW RI VWDQGV DIWHU VWUHVVIXO WUHDWPHQWV LQ 5HGDOWD KDG WKH ZRUVW TXDOLW\ DQG KDV D OLPLWHG IXWXUH &DUHIXO PDQDJHPHQW LV DOVR QHFHVVDU\ LQ RUGHU IRU 3, WR SHUVLVW 2I DOO OLPSRJUDVVHV 3, KDG KLJKHU \LHOG DQG SHUVLVWHQFH WKDQ %LJDOWD DQG 3, KLJKHU \LHOG DQG TXDOLW\ WKDQ 5HGDOWD DQG 3, DQG KDV EHHQ UHFRPn PHQGHG IRU FXOWLYDU UHOHDVH LQ &RQFOXVLRQV 7KH IRXU OLPSRJUDVVHV LQFOXGHG LQ WKLV VWXG\ KDG GLIIHUHQW DGYDQWDJHV LQ \LHOG SRWHQWLDO DW YDULRXV OHYHOV RI 1 DQG ) %LJDOWD SURGXFHG ZHOO DW LQWHUPHGLDWH FXWWLQJ IUHTXHQFLHV ZKLOH 3, SURn GXFHG PRUH '0 ZLWK KLJK QLWURJHQ DQG ORQJ FXWWLQJ IUHTXHQFLHV 7KH IRXU OLPSRJUDVVHV GLIIHUHG LQ WKHLU VHDVRQDO GLVWULEXWLRQ RI '0 \LHOG %LJDOWD ZDV ORZHVW LQ VSULQJ SURGXFWLRQ DQG 3, ZDV KLJKHVW 6WRFNSLOHG IRUDJH IURP -XO\ WR 1RYHPEHU ZDV JUHDWn HVW IRU 3, 9DOXHV IRU ,920' ZHUH ORZHU WKDQ VKRZQ LQ RWKHU SXEOLVKHG UHSRUWV HVSHFLDOO\ IRU VDPSOHV WDNHQ GXULQJ PLGVXPPHU 7KH ,920'

PAGE 136

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f 6WDSI HW &( +XEEf KDV LGHQWLILHG WKH SRWHQWLDO RI WKLV JUDVV IRU IODWZRRG IRUDJH OLYHVWRFN V\VWHPV GXH WR EHWWHU TXDOLW\ DQG GU\ PDWWHU \LHOG SRWHQWLDO

PAGE 137

WKDQ EDKLDJUDVV 3DVSDOXP QRWDWXP )OLLJJHf $ FXUUHQW SUREOHP LV WKDW WKH OLPSRJUDVV FXOWLYDUV %LJDLWD DQG n5HGDOWDn KDYH DWWULEXWHV WKDW PDNH WKHP OHVV WKDQ LGHDO IRUDJHV 5HFHQW HYDOXDWLRQV RI 3, VXJJHVW WKDW WKLV JUDVV DGHTXDWHO\ FRPELQHV WKH GHVLUDEOH FKDUDFWHULVn WLFV RI \LHOG SHUVLVWHQFH TXDOLW\ DQG DQLPDO DFFHSWDELOLW\ 7KH SUHVHQW VWXG\ ZDV GHVLJQHG WR WKRURXJKO\ XQGHUVWDQG ZK\ KRZ DQG LQ ZKDW FDWHJRULHV WKH QHZ OLPSRJUDVV ZDV VXSHULRU WR WKH DOUHDG\ UHOHDVHG FXOWLYDUV 5HVXOWV IURP DQ H[WHQVLYH ILHOG VWXG\ DQG WKRURXJK ODERUDWRU\ DQDO\VHV VKRZHG WKDW 3, GLG LQGHHG SRVVHVV EHWWHU DWWULEXWHV IRU GU\ PDWWHU '0f \LHOG DQG SHUVLVWHQFH WKDQ %LJDLWD EHWWHU GLJHVWLn ELOLW\ DQG \LHOG WKDQ 5HGDOWD KRZHYHU LW IHOO VKRUW RI %LJDLWD LQ FUXGH SURWHLQ &3f DQG LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f 6HDVRQDO '0 GLVWULEXWLRQ DQG WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f IOX[ ZHUH DOVR VWXGLHG DV LQIOXHQFHG E\ D FRPSOHWH DUUD\ RI QLWURJHQ IHUWLOL]DWLRQ DQG NJKD\Uf DQG D ZLGH UDQJH RI GHIROLDWLRQ IUHTXHQFLHV DQG ZHHNVf LQ D PRGLILHG FHQWUDO FRPSRVLWH UHVSRQVH VXUIDFH GHVLJQ 6SULQJ '0 SURGXFWLRQ ZDV KLJKHVW IRU DQRWKHU QHZ OLPSRJUDVV 3, f WKDW \LHOGHG P WRQV LQ WKH ZHHN SHULRG SULRU WR -XQH %LJDOWD KDG WKH ORZHVW VSULQJ \LHOGV DQG ZDV WKH PRVW IURVW VXVn FHSWLEOH LQ WKH DXWXPQ /LPSRJUDVV 3, SURGXFHG P WRQV RI '0 ZKHQ VWDJHG WR JURZ DV D VWRFNSLOHG IRUDJH IURP -XO\ WR 1RYHPEHU DQG DOVR SURGXFHG WKH PRVW DQQXDO '0 P WRQVf 7RWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ WKH VWRUDJH RUJDQV GHFUHDVHG IURP D KLJK LQ 0DUFK WR D ORZ LQ -XO\ ,Q YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f DQG &3 YDOXHV LQ WLVVXH RI HTXLYDOHQW

PAGE 138

FKURQRORJLFDO DJH YDULHG VHDVRQDOO\ ZLWK UHJDUG WR TXDOLW\ &UXGH SURWHLQ LQ VDPSOHV WDNHQ GXULQJ WKH VXPPHU ZHUH H[WUHPHO\ ORZ LQ DOO WKH LPSRJUDVVHVf§ ZHOO EHORZ WKH SHUFHQW QHFHVVDU\ WR PDLQWDLQ ERG\ ZHLJKW LQ UXPLQDQW DQLPDOV 7KH PLGVXPPHU SHULRG ZDV PDUNHG E\ ORZHVW TXDOLW\ KLJKHVW VWUHVV RQ WKH UHVHUYHV DQG PRVW PDUNHG GHFUHDVHV LQ SHUFHQW OLPSRJUDVV LQ WKH VWDQGV

PAGE 139

&+$37(5 67$7,67,&$/ $1$/<6,6 6<67(0 6$6f 0(7+2'2/2*< )25 &216758&7,1* 5(63216( 685)$&( *5$3+,&6 ,QWURGXFWLRQ 7KH FKDUDFWHUL]DWLRQ DQG TXDQWLILFDWLRQ RI UHODWLRQVKLSV EHWZHHQ LQGHSHQGHQW DQG GHSHQGHQW YDULDEOHV ZDV WUDGLWLRQDOO\ DSSURDFKHG E\ WHVWLQJ RQH IDFWRU DW D WLPH ZKLOH KROGLQJ RWKHU IDFWRUV FRQVWDQW ,W ZDV UHFRJQL]HG WKDW LQIRUPDWLRQ RI PXFK ZLGHU JHQHUDOLW\ FRXOG EH SURn GXFHG LI VHYHUDO IDFWRUV ZHUH LQYHVWLJDWHG VLPXOWDQHRXVO\ +DGHU HW DO f &RPSOHWH IDFWRUDOV DUH RQH VWDWLVWLFDO DSSURDFK WR XQGHUVWDQGLQJ FRPSOH[ UHODWLRQVKLSV DPRQJ H[SHULPHQWDO IDFWRUV EXW WKHVH GHVLJQV FRQn VXPH SURIOLJDWH DPRXQWV RI WLPH UHVRXUFHV DQG PRQH\ EHFDXVH RI WKH H[SRQHQWLDO H[SORVLRQ RI WUHDWPHQW FRPELQDWLRQV ZLWK JUHDWHU WKDQ WKUHH IDFWRUV )UDFWLRQDO IDFWRULDOV UHSUHVHQW VRPH LPSURYHPHQW LQ UHGXFLQJ WKH VL]H RI WKH H[SHULPHQW EXW WKH\ UHTXLUH H[WUHPH FDUH LQ DYRLGLQJ FRQIRXQGHG HIIHFWV EHFDXVH RI SUDWLDO LQWHUQDO UHSOLFDWLRQ &RFKUDQ DQG &R[ f %R[ DQG :LOVRQ f ZHUH WKH ILUVW WR GHYHORS DQG GHVFULEH WKH UHVSRQVH VXUIDFH GHVLJQ 5HVSRQVH VXUIDFH PHWKRGRORJ\ GHFUHDVHV WKH ODUJH QXPEHU RI WUHDWPHQW FRPELQDWLRQV QHFHVVDU\ LQ IDFWRULDO H[SHULPHQWV 7KH ILUVW DSSOLFDWLRQV ZHUH LQ WKH FKHPLVWU\ DQG HQJLQHHU LQJ ILHOGV %\ WKH OLWHUDWXUH ZDV IXOO RI GLYHUVH DSSOLFDWLRQV LQ D YDULHW\ RI ILHOGV +LOO DQG +XQWHU f

PAGE 140

$ IDFWRULDO DUUDQJHPHQW RI GHVLJQ SRLQWV IRU D WZR IDFWRU H[SHULn PHQW DUH VKRZQ LQ )LJXUH DQG %R[ DQG :LOVRQnV f FRPSRVLWH GHVLJQ LQ WKUHH IDFWRUV LV JUDSKLFDOO\ UHSUHVHQWHG LQ )LJXUH 7KH ODWWHU KDV WKH SURSHUW\ RI V\PPHWU\ ZKHQ YLHZHG IURP DQ\ D[LV $FFRUGLQJ WR +DGHU HW DO f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f ,Q VRPH LQGXVWULDO DSSOLFDWLRQV RI UHVSRQVH VXUIDFHV D VHULHV RI VKRUW H[SHULPHQWV FRXOG EH SODQQHG VR DV WR XVH WKH UHVXOWV RI WKH ILUVW WR SODQ WKH WUHDWPHQWV IRU WKH VHFRQG $ SHUSHQGLFXODU FRXUVH XS RU GRZQ WKH FRQWRXUV ZRXOG KHDG WKH H[SHULPHQWHU WRZDUGV WKH GHVLUHG RSWLPXP 7KLV SDWK RI VWHHSHVW DVFHQW RU GHVFHQW FRXOG EH XVHG WR IL[ WKH RULJLQ LQ WKH VHFRQG DQG VXEVHTXHQW H[SHULPHQWV &RFKUDQ DQG &R[ 0\HUV %R[ HW DO f &DQRQLFDO DQDO\VLV LV DQRWKHU PHWKRG XVHG LQ VHTXHQWLDO WHVWLQJ WKDW FRQVLVWV RI VKLIWLQJ WKH RULJLQ WR D QHZ SRLQW DQG URWDWLQJ WKH D[HV VR WKH\ FRUUHVSRQG WR WKH D[HV RI WKH FRQWRXUV :KHQ WKH UHVSRQVH VXUIDFH LV RULHQWHG WR WKH QHZ VHW RI D[HV WKH VHFRQG RUGHU HTXDWLRQV DUH JUHDWO\ VLPSOLILHG DQG WKHLU

PAGE 141

WRSRJUDSKLFDO QDWXUH EHFRPHV PRUH REYLRXV %R[ HW DO 0\HUV f ,Q OLQHV RI ZRUN ZKHUH DQ H[SHULPHQW PXVW H[WHQG RYHU D ORQJ SHULRG RI WLPH LQ RUGHU WKDW WKH WUHDWPHQWV SURGXFH WKHLU HIIHFWV WKH QDWXUDO VWUDWHJ\ LV WR WU\ WR GLVFRYHU WKH RSWLPXP FRPELQDWLRQV DW WKH HQG RI D VLQJOH H[SHULPHQW &RFKUDQ DQG &R[ f 6RPH H[SHULPHQWHUV KDYH PRGLILHG UHVSRQVH VXUIDFH GHVLJQV LQ RUGHU WR KDYH WUHDWPHQW FRPELQDn WLRQV WKDW EHWWHU ILW WKH QHHGV RI WKHLU VLWXDWLRQ 7KH PRGLILFDWLRQV DUH XVXDOO\ LQ WKH IRUP RI DGGLQJ GHVLJQ SRLQWV RXWVLGH WKH H[SHULPHQWDO UHJLRQ RI WKH IRUPDO GHVLJQ RU WR VKLIW SRLQWV DURXQG ZLWKLQ WKH H[SHULPHQWDO UHJLRQ WR FRQFHQWUDWH LQIRUPDWLRQ LQ FHUWDLQ DUHDV RI LQWHUHVW /LWWHOO DQG 0RWW f 0RWW Df KDV UHIHUUHG WR GHVLJQV RI WKLV VRUW DV PRGLILHG FHQWUDO FRPSRVLWH GHVLJQV /LWWHOO DQG 0RWW f GHVFULEHG WKH XVH RI FRQWRXU GLDJUDPV XVLQJ 6$6 +HQGHUVRQ DQG 5RELQVRQ f SUHVHQWHG EXW GLG QRW GHVFULEH WKH IRUPDWLRQ RI 6$6 VXUIDFH SORWV 6FKRQH\ HW DO f GHVFULEHG FRQWRXU DQG UHVSRQVH VXUIDFHV IURP FRPSXWHU JUDSKLFV EXW WKHLU 8QLYHUVLW\ RI :LVFRQVLQ0DGLVRQ FRPSXWLQJ SDFNDJH LV QRW DV ZLGHO\ DYDLODEOH DV 6$6 7KH REMHFWLYH RI WKLV SDSHU LV WR VKRZ D FRRNERRN SURFHGXUH XVLQJ 6$6 &RXQFLO DQG +HOZLJ f DQG 6$6*5$3+ &RXQFLO DQG +HOZLJ f LQ SORWWLQJ YDULRXV VXUIDFH SORWV IRU DJURQRPLF YDULDEOHV LQ D PRGLILHG FHQWUDO FRPSRVLWH GHVLJQ 7HVWV IRU ODFN RI ILW /2)f SUHFDXWLRQV DQG VRPH SLWIDOOV DUH GLVFXVVHG 0DWHULDOV DQG 0HWKRGV 9HJHWDWLYH VSULJJLQJ ZDV XVHG WR HVWDEOLVK IRXU OLPSRJUDVV JHQRn W\SHV RQ DQ $GDPVYLOOH VDQG D SRRUO\ GUDLQHG VLOLFHRXV K\SHUWKHUPLF

PAGE 142

XOWLF KDSODTXRG VRLO DW WKH %HHI 5HVHDUFK 8QLW RI WKH 8QLYHUVLW\ RI )ORULGD 7KH JUDVVHV ZHUH 3, 3, n5HGDOWD DQG n%LJDOWDn )ORULGD ([SHULPHQWDO 6WDWLRQ DFFHVVLRQ QXPEHUV ZHUH DQG UHVSHFWLYHO\ (VWDEOLVKPHQW
PAGE 143

OHYHOV RI QLWURJHQ 1f IHUWLOL]DWLRQ DQG ILYH FXWWLQJ IUHTXHQFLHV )f ZHUH LPSRVHG RQ WKH PDLQ SORWV 7KH WUHDWPHQWV UHSUHVHQWHG WK RI D [ IDFWRULDO DV GHSLFWHG LQ )LJXUH *UDVVHV ZHUH EORFNHG RU UHSOLFDWHGf DV PDLQ SORWV LQ D VSOLW SORW GHVLJQ DQG WKH WUHDWPHQWV SURYLGHG D UHVSRQVH VXUIDFH ZLWK DOO SRLQWV UHSOLFDWHG WKUHH WLPHV (DFK PDLQ SORW FRQWDLQHG SORWV [ Pf IRU WUHDWPHQWV 7KH HQWLUH ILHOG KDG WRWDO SORWV WKUHH EORFNV [ IRXU JUDVVHV [ WUHDWPHQWVf 7KH WUHDWPHQW FRPELQDWLRQV UHSUHVHQWHG D PRGLILHG FHQWUDO FRPSRVLWH UHVSRQVH VXUIDFH GHVLJQ WKDW PD[LPL]HG LQIRUPDWLRQ LQ ZKDW ZDV FRQVLGn HUHG WKH PDQDJHPHQW UHDOP NJKD\U 1 DQG ZHHN GHIROLDWLRQ IUHTXHQFLHVf 1RWH WKH JHRPHWULF VFDOLQJ RI 1 DQG WKH DULWKPHWLF SURn JUHVVLRQ IRU FXWWLQJ IUHTXHQF\ ZLWK D VNLS DW WKH ZHHN OHYHOf LQ )LJXUH $V WKH GRWWHG OLQHV LQ )LJXUH VXJJHVW WKH GHVLJQ FRXOG EH FRQVLGHUHG D VXSHULPSRVLWLRQ RI WZR FRPSOHWH IDFWRULDOV DQG f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f 1 DQDO\VLV GU\ PDWWHU '0f \LHOG GHWHUn PLQDWLRQ DQG WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f PHDVXUHPHQW

PAGE 144

)XOO LQIRUPDWLRQ RQ DOO WUHDWPHQWV LV QHFHVVDU\ IRU WKH IRUPDWLRQ RI VXUIDFH SORWV DQG VWHP EDVH FURZQf GDWD REWDLQHG IURP 71& VDPSOHV WDNHQ RQ 6HSWHPEHU VDWLVILHG WKLV UHTXLUHPHQW LQ WKH H[DPSOH WR IROORZ 7KH 71& DQDO\VLV LV IXOO\ GHVFULEHG LQ $SSHQGL[ $ 'DWD ZHUH DQDO\]HG XVLQJ 6$6 RQ DQ $PGDKO 9 ZLWK 26096 5HOHDVH DQG -(61-( 5HOHDVH &RPSXWLQJ ZDV SHUIRUPHG DW WKH 1RUWKHDVW 5HJLRQDO 'DWD &HQWHU DQG WKH $JULFXOWXUDO (QJLQHHULQJ 'HSDUWn PHQW RI WKH 6WDWH 8QLYHUVLW\ 6\VWHP RI )ORULGD ORFDWHG RQ WKH FDPSXV RI WKH 8QLYHUVLW\ RI )ORULGD LQ *DLQHVYLOOH $Q DQDO\VLV RI /2) SUHFHGHG VXUIDFH SORWWLQJ 7KH IROORZLQJ 6$6 FRPPDQGV ZHUH XVHG WR JHQHUDWH WKH PRGHOV QHFHVVDU\ WR WHVW IRU /2) LQ WKLV H[DPSOH $ FXELF PRGHO GHVFULEHG WKH 71& UHVSRQVH DIWHU D TXDn GUDWLF PRGHO IDLOHGf &RPSOHWH PRGHO /,1( OLPSRJUDVV JHQRW\SHf 352& $129$ &/$66 /,1( 757 5(3 02'(/ 71& 5(3 /,1( /,1(r5(3 757 /,1(r757 )UDFWLRQDWHG PRGHO 352& */0 &/$66 /,1( 5(3 02'(/ 71& 5(3 /,1( 5(3r/,1( 1 1r1 1r1r1 ) )r) )r)r) 1r) 1r1r) 1r)r) /,1(r1 /,1(r1r1 /,1(r1r1r1 /,1(r) /,1(r)r) /,1(r)r)r) /,1(r1r) /,1(r1r1r) /,1(r1r)r)62/87,21 7KH WHVW RI /2) ZDV DFFRPPRGDWHG E\ DQ )WHVW RI WKH WZR SUHYLRXVO\ GHVFULEHG PRGHOV 6XPV RI 6TXDUHV 66f 'HJUHHV RI )UHHGRP ')f 0HDQ 6TXDUH (UURU 06(ff

PAGE 145

U 66 HUURU IUDFWLRQDWHGf B U 66 HUURU FRPSOHWHf )/2)f ') IUDFWLRQDWHGf a ') FRPSOHWHf n n 06( FRPSOHWHf 2IWHQ DQ HOLPLQDWLRQ SURFHGXUH IRU QRQVLJQLILFDQW WHUPV LV XVHIXO IRU DGGLQJ 66 DQG ') EDFN LQWR WKH HUURU WHUP IRU WKH IUDFWLRQDWHG PRGHO :KHQ D PRGHO ZDV ILQDOO\ GHHPHG DFFHSWDEOH WKH FRPSXWHU SURn YLGHG HVWLPDWHV WKDW ZHUH VROYHG IRU WKH LQGLYLGXDO OLPSRJUDVV LQWHUn FHSWV DQG SDUDPHWHUV 1 DQG )f 7KH HTXDWLRQ IRU RQH OLPSRJUDVV OLQH 3, RU )ORULGD QXPEHU f ZDV XVHG WR GHPRQVWUDWH WKH SURJUDP IRU VXUIDFH SORWWLQJ 7DEOH f 7KH SURJUDP KDV WKH IROORZLQJ VWHSV f ORDG WKH UDZ GDWD IURP WKH 71& DQDO\VHV DQG FUHDWH PHDQV f ORDG WKH SUHYLRXVO\ GHVFULEHG VXUIDFH HTXDWLRQV f DVVLJQ WUHDWPHQW QXPEHUV )LJXUH f IRU WKH SUHn GLFWHG WUHDWPHQW FRPELQDWLRQV f XVH WKH VXUIDFH HTXDWLRQ WR VROYH IRU 71& DW WKH PLVVLQJ OHYHOV )LJXUH f RI 1 DQG ) f PHUJH DFWXDO SOXV SUHGLFWHG GDWD f DVVLJQ 1 DQG ) OHYHOV IRU WKH DFWXDO GDWD f WULFN WKH FRPSXWHU WR VWDUW DW ]HUR RQ WKH 71& D[LV DQG f SORW WKH VXUIDFH XVLQJ 352& *' )LJXUH f 5HVXOWV DQG 'LVFXVVLRQ ,W ZLOO EH DVVXPHG WKH UHDGHU KDV D PLQLPXP NQRZOHGJH RI 6$6 EHFDXVH D GHWDLOHG H[SODQDWLRQ RI WKH SURJUDP ZRXOG EH ERWK ODERULRXV DQG RXW RI SODFH 7DEOHV DQG DUH VKRZQ LQ RUGHU WR DOORZ WKH UHDGHU WR WUDFH WKH YDOXHV IURP WKH FRPSXWHU RXWSXW WR WKH WHVW IRU /2) LQ 7DEOH ,Q WKLV H[DPSOH /2) LV GHWHUPLQHG ZLWK DQ )WHVW RI PRGHOV 7KH FRPSOHWH PRGHO FRQWDLQV WKH PLQLPXP SRVVLEOH VXPV RI VTXDUHV EHFDXVH LW UHSUHVHQWV WKH EHVW H[SODQDWLRQ RI WUHDWPHQW

PAGE 146

7DEOH $ 6$6 SURJUDP XVLQJ D SUHYLRXVO\ ILWWHG UHJUHVVLRQ WR REWDLQ YDOXHV IRU WKH GHSHQGHQW YDULDEOH WR ILOO WKH KROHV LQ WKH WUHDWPHQW PDWUL[ LQ RUGHU WR PHUJH ZLWK DFWXDO GDWD DQG XVH 352& *' 71& -2% fn&+5,67,$16(1&/$66 r3$66:5' 6&277 r587( 35,17 5(027( (;(& 6$65(*,1 .3/7 *237,216 '(9,&( *28/' '$7$ &5 ,1387 12 0 < /,1( 757 5(3 ,1& '(9 :7 3, ',/1 3 ' ' '$7( 0'<0'
PAGE 147

7DEOH &RQWLQXHG ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, ** ++ ,, &200(17 71& &52:1 $$ %% r1 (r1r1 (r1r1r1 r) r) r) (r)r)r) r1r) ( r1r1r) (r1r)r) 287387 (1' (1' (1' '$7$ '800< 6(7 '80% ,) 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 25 757 352& 6257 %< 757 352& 35,17 2% /,1( 1 ) 757 71& 2% /,1( 1 ) 757 71& r P P P '$7$ 685)$&( 0(5*( 1(: '800< %< 757 '$7$ 5(63216( 6(7 685)$&( 0$&52 && Lf§ aQ f§ f§ ,, b

PAGE 148

7DEOH &RQWLQXHG 0$&52 '' 7+(1 '2 1 b 0$&52 (( ) b 0$&52 )) (1' b && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) && '' (( )) ,) 757 7+(1 71& QR ,) 1 $1' ) 7+(1 71& 7,7/( $&78$/ 3/86 35(',&7(' 71& 7,7/( b 71& )25 6(37(0%(5 ,1 3, &52:16 352& 6257 352& 35,17 $&78$/ 35(',&7(' 757 71& 1 ) 757 71& 1 ) QUR 352& *' 3/27 1r) 71& (;(& 3;3/27 r(2)

PAGE 149

b 71& )256(37(0%(5 ,1 3, &52:16 )LJXUH $Q H[DPSOH RI WKH UHVSRQVH VXUIDFH SORWWHG XVLQJ 6$6*5$3+ FRPSXWHU DVVLVWDQFH

PAGE 150

7DEOH $ 6$6 DQDO\VLV RI YDULDQFH SURFHGXUH IRU D PRGHO WKDW TXDQWLILHV WKH PD[LPXP WUHDWPHQW 757f VXP RI VTXDUHV 66f IRU WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f LQ KDUYHVW RI WKH H[SHULPHQW ,1'(3(1'(17 9$5,$%/( 6285&( 71& ') 680 2) 648$5(6 0($1 648$5( ) 9$/8( 3 ) 5648$5( &9 02'(/ (5525 &255(&7(' 727$/ 67' '(9 71& 0($1 6285&( ') $129$ 66 ) 9$/8( 35 ) 5(3 /,1( /,1(r5(3 757 /,1(r757 7(67 2) +<327+(6(6 86,1* 7+( $129$ 06 )25 /,1(r5(3 $6 $1 (5525 7(50 6285&( ') $129$ 66 ) 9$/8( 35 ) /,1(

PAGE 151

7DEOH 757 $ FXELF 6$6 JHQHUDO OLQHDU PRGHO WKDW IUDFWLRQDWHV WKH WUHDWPHQW VXP RI VTXDUHV 66f LQWR 66 H[SODLQDEOH E\ QLWURJHQ 1f IHUWLOL]DWLRQ DQG IUHTXHQF\ )f RI GHIROLDWLRQ IRU WRWDO QRQ VWUXFWXUDO FDUERK\GUDWH 71&f LQ KDUYHVW RI WKH H[SHULPHQW 6285&( ') 680 2) 648$5(6 0($1 648$5( ) 9$/8( 3 ) 5648$5( &9 02'(/ (5525 &255(&7(' 727$/ 67' '(9 71& 0($1 6285&( ') 7<3( 66 ) 9$/8( 35 ) 7<3( ,9 66 ) 9$/8( 35 ) 5(3 /,1( /,1(r5(3 1 1r1 1r1r1 ) )r) )r)r) 1r) 1r1r) 1r)r) &2 &2

PAGE 152

&/

PAGE 153

7DEOH $Q )WHVW RI PRGHOV IRU GHWHUPLQLQJ D VWDWLVWLFDOO\ VLJQLILFDQW UHVSRQVH ODFN RI ILW /2)f DQG DQ HOLPLQDWLRQ PHWKRG IRU QRQVLJQLILFDQW WHUPV 7(67 )25 /2) )/2)f 66 FRPSOHWH PRGHO 66 IUDFWLRQDWHG PRGHO Y ') FRPSOHWH PRGHO ') IUDFWLRQDWHG PRGHO n 66 FRPSOHWH PRGHO n ') FRPSOHWH PRGHO )/2)f )WDE f f )WDE f f &21&/8'( QV /2) (/,0,1$7,21 0(7+2'2/2*< 6285&( ') 7<3( 66 3 ) (5525 1r1r)r/,1( QrIrIrOLQH IORIf )WDE f f )WDE f f &21&/8'( r/)3 f

PAGE 154

YDULDELOLW\ 7KH FXELF PRGHO ZLWK 1 DQG ) SDUDPHWHUV H[SODLQHG HQRXJK RI WKH WUHDWPHQW YDULDELOLW\ WR JLYH D QRQVWDWLVWLFDO /2) 7KH HOLPLQDWLRQ SURFHGXUH GHSLFWHG LQ 7DEOH PXVW EH XVHG ZLWK DGKHUHQFH WR SDUDPHWHU KLHUDUFK\ LH LI WKH 1r1 WHUP IRU H[DPSOH LV QRQVLJQLILFDQW EXW WKH /,1(r1r1 WHUP LV VLJQLILFDQW WKHQ WKH IRUPHU FDQQRW EH UHPRYHG IURP WKH PRGHO GXH WR WKH GHSHQGHQFH RI WKH ODWWHU (YDOXDWLRQ RI 3 ) WHUP LV EHJXQ IURP WKH ERWWRP SDUDPHWHU 7<3( 66f XSZDUGV XQWLO D VLJQLILFDQW WHUP LV UHDFKHG DQG DW WKDW WLPH 3 ) GHFLn VLRQ LV EDVHG XSRQ 7<3( ,9 66 7KH UHDVRQ IRU WKH DERYH LV EHFDXVH 7<3( 3 ) LV FDOFXODWHG IRU HDFK WHUP SULRU WR WKH DGGLWLRQ RI VXEn VHTXHQW WHUPV 7<3( ,9 3 ) JLYHV DQ DQDO\VLV RI HDFK WHUP JLYHQ WKDW DOO RWKHU WHUPV DUH SUHVHQW $ ILQDO JHQHUDO UXOH LQ WKH DWWHPSW IRU D PRUH SDUVLPRQLRXV PRGHO LQ WKH SURFHVV RI HOLPLQDWLQJ QRQVLJQLILFDQW WHUPV WU\ WR SLFN WKH SDUDPHWHUV ZLWK VPDOO 66 DQG ODUJH ') EHFDXVH WKLV ZLOO IXUWKHU UHGXFH WKH IUDFWLRQDWHG PRGHO 06( 66')f ZKHQ WKH 66 DQG ') DUH DGGHG EDFN LQWR WKLV SRUWLRQ RI WKH QXPHUDWRU VKRZQ LQ 7DEOH $ UHGXFHG QXPHUDn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

PAGE 155

7DEOH 7KH 6$6 JHQHUDO OLQHDU PRGHOV SURFHGXUH VROXWLRQ IRU HVWLPDWHV RI LQWHUFHSL OLPSRJUDVV OLQHf DQG UHS SDUDPHWHUV 3$5$0(7(5 (67,0$7( ,17(5&(37 5(3 [ /,1( f f r5(3 [ [ ; [ DGMXVWHG IRU WKH ,17(5&(37 )25 /,032*5$66 /,1(

PAGE 156

7DEOH 7KH 6$6 JHQHUDO OLQHDU PRGHOV SURFHGXUH VROXWLRQV IRU HVWLPDWHV RI QLWURJHQ 1f DQG FXWWLQJ IUHTXHQF\ )f WUHDWPHQW SDUDPHWHUV DGMXVWHG IRU HDFK OLPSRJUDVV OLQHf 75($70(17 3$5$0(7(5 (67,0$7( 1 $f 1r1 ( %f 1r1r1 ( &f ) 'f )r) (f )r)r) )f 1r) *f 1r1r) ( +f 1r)r) ( ,f 75($70(17 3$5$0(7(56 $'-867(' )25 /,1( (67,0$7(6 1r/,1( $f 1r1r/,1( ( ( ( %f ( ( r1 r1r1

PAGE 157

7DEOH &RQWLQXHG 75($70(17 3$5$0(7(56 $'-867(' )25 /,1( (67,0$7(6 1r1r1r/,1( ( ( ( &f ( r1r1r1 ( ( ( )r/,1( 'f r) )r)r/,1( (f r)r) )r)r)r/,1( ( ( )f rSrSrS 1r)r/,1( *f r1r) 1r1r)r/,1( ( ( ( ( +f ( ( r1r1r) ( 1r)r)r/,1( ( ( 6( ,f ( r1r)r) ( ( ( 7+( ),1$/ (48$7,216 )25 71& ,1 /,032*5$66 71& r1 (r1r1 (r1r1r1 r) r)r) (r)r)r) r1r) (r1r1r) (r1r)r)

PAGE 158

GDWD SURYLGHV D YHULILFDWLRQ RI WKH PRGHO HTXDWLRQ LI WKH DFWXDO DQG SUHGLFWHG SRLQWV EOHQG HYHQO\ 7KLV ZDV FRQVLGHUHG LPSRUWDQW EHFDXVH VRPH PRGHOV WKDW ZHUH SORWWHG ZHUH VWDWLVWLFDOO\ DGHTXDWH EXW QRQH WKH OHVV IDLOHG WR ILW LQ VRPH UHJLRQV RI WKH VXUIDFH SORW 7KH 7DEOH SURJUDP ZLOO EH GLVFXVVHG IURP WRS WR ERWWRP XVLQJ WKH SURJUDP OLQH QXPEHU IRU HDV\ UHIHUHQFH 2Q OLQH WKH LQIRUPDWLRQ DVVRn FLDWHG ZLWK WKH UDZ GDWD LV LQSXWHG )RU D PRUH WKRURXJK LQYHVWLJDWLRQ LQWR WKH PHFKDQLFV XVHG LQ SURFHVVLQJ WKH 71& GDWD FRQVXOW $SSHQGL[ $ 2Q OLQHV WKH 71& PHDQV DUH FUHDWHG 7KH '$7$ '80% OLQH f LV WKH GDWD VHW FRQWDLQLQJ WKH SUHYLRXVO\ GHWHUPLQHG VXUIDFH HTXDWLRQ DQG LQVWUXFWLRQV WR FRGH IRU WUHDWPHQW 1 DQG ) 2QO\ WKH PLVVLQJ WUHDWn PHQWV VKRZQ LQ )LJXUH DUH JLYHQ WUHDWPHQW QXPEHUV DV ZHOO DV RQH PRUH 757 f ZKLFK ZLOO EH XVHG ODWHU WR WULFN WKH FRPSXWHU LQWR VWDUWLQJ WKH 71& D[LV DW ]HUR 7KH VXUIDFH HTXDWLRQ LV XVHG WR VROYH IRU 71& DW WKH VSHFLILHG OHYHOV RI 1 DQG ) OLQHV f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

PAGE 159

7KH VPRRWKLQJ WHFKQLTXHV HPSOR\HG E\ 6FKRQH\ HW DO f DUH FOHDUO\ PRUH DGYDQFHG WKDQ WKH PHWKRGRORJ\ GHVFULEHG KHUH KRZHYHU 6$6*5$3+ LV QRW \HW VR VRSKLVWLFDWHG 7KH PDMRU SUHFDXWLRQ WKLV SDSHU GHOLYHUV LV WKDW EHIRUH WKH VPRRWK ILWWHG IXQFWLRQ LV XVHG LQ 6$6 VXUn IDFH SORWV DQG WKHLU FRUUHVSRQGLQJ FRQWRXU SORWVf WKH SUHGLFWHG SRLQWV VKRXOG EH SORWWHG DPLGVW WKH DFWXDO GDWD IRU D YLVXDO WHVW RI /2) $OVR EHZDUH RI WKH LQWHUSUHWDWLRQ RI D VXUIDFH EDVHG XSRQ D QRQOLQHDU VFDOLQJ RI WKH = D[LV &RPSXWHU JHQHUDWHG JUDSKLFV KDYH EHHQ SURPRWHG E\ &DG\ DQG )XOOHU f DQG LW LV EHOLHYHG WKDW WKHVH JUDSKLFV ZLOO EH PRUH ZLGHO\ XVHG LQ WKH IXWXUH 7KURXJK JUDSKLFV WKH GLVWULEXWLRQ SK\VLFDO UHVSRQVH DQG LQWHUDFWLRQ FKDUDFWHULVWLFV RI D ODUJH YROXPH RI GDWD FDQ EH UHDGLO\ DQDO\]HG 6FKRQH\ HW DO S f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

PAGE 160

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f IHUWLOL]DWLRQ DQG GHIROLDn WLRQ IUHTXHQF\ )f RQ IRXU OLPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf JHQRW\SHV 7RWDO QRQVWUXFWXUDO FDUERK\GUDWHV 71&f IRUDJH TXDOLW\ SHUVLVWHQFH DQG \LHOG ZHUH PHDVXUHG DQG IURP WKRVH UHVSRQVHV D FXELF PRGHO GHVFULELQJ 71& LQ WKH FURZQV ORZHVW FP RI VWHPf RI 3, OLPSRJUDVV KDUYHVWHG 6HSWHPEHU ZDV VHOHFWHG WR H[HPSOLI\ WKH FRPSXWHU SURFHGXUHV QHFHVVDU\ IRU FRQVWUXFWLQJ VXUIDFH SORWV 7KH SORWWLQJ SURFHGXUHV LQ 6$6*5$3+ FDQQRW DFFHSW KROHV LQ WKH WUHDWPHQW FRPELQDWLRQ PDWUL[ $ SURFHGXUH ZDV SUHVHQWHG VKRZLQJ KRZ WR ILOO WKH KROHV ZLWK SUHGLFWHG QXPEHUV GHULYHG IURP D UHJUHVVLRQ HTXDn WLRQ ILWWLQJ WKH VXUIDFH UHVSRQVH &RPSDULVRQ RI VXUIDFHV LV GLIILFXOW EHFDXVH QR XQLIRUP SORWWLQJ SURJUDP LV \HW DYDLODEOH LQ 6$6*5$3+ 'XH WR WKH OLPLWDWLRQ RI QRQ XQLIRUP VFDOLQJ DQG VRPHWLPHV QRQOLQHDU VFDOLQJ RI WKH = D[LV H[WUHPH FDXWLRQ LV DGYLVHG LQ WKH LQWHUSUHWDWLRQ RI VXUIDFH SORWV

PAGE 161

6800$5< $1' &21&/86,216 7KH LQWURGXFWLRQ RI OLPSRJUDVV LQWR WKH 8QLWHG 6WDWHV RFFXUUHG LQ DQG VLQFH WKDW WLPH LW KDV JURZQ SRSXODU LQ WKH VRXWKHDVW DV DQ LPSURYHG IRUDJH IRU WKH IODWZRRG VLWH 5HVHDUFK SULRU WR FXOPLn QDWHG LQ WKH LGHQWLILFDWLRQ FKDUDFWHUL]DWLRQ DQG FXOWLYDU UHOHDVH RI n%LJDOWDn n5HGDOWDn DQG n*UHHQDOWDn 0RUH OLPSRJUDVV JHUPSODVP EHFDPH DYDLODEOH IRU HYDOXDWLRQ DIWHU SODQW H[SORUDWLRQV LQ DQG DQG VHYHUDO RI WKH SODQW LQWURGXFWLRQV ZHUH VXVSHFWHG WR KDYH DJURQRPLF FKDUDFWHULVWLFV WKDW ZHUH VXSHULRU RU HTXDO WR WKH UHOHDVHG FXOWLYDUV ,Q WKLV VWXG\ %LJDOWD DQG 5HGDOWD ZHUH FRPSDUHG WR 3, DQG 3, LQ D VHULHV RI H[SHULPHQWV GXULQJ %LJDOWD LV D KLJKO\ GLJHVWLEOH WHWUDSORLG OLPSRJUDVV WKDW KDV QRW SHUVLVWHG XQGHU PDQDJHPHQW SUDFWLFHV LQYROYLQJ IUHTXHQW GHIROLDWLRQ DQGRU KHDY\ DSSOLn FDWLRQV RI QLWURJHQ IHUWLOL]HU 5HGDOWD D GLSORLG KDV H[FHOOHQW SHUn VLVWHQFH EXW LWV RYHUDOO QRQGLJHVWLELOLW\ LV XQGHVLUDEOH 7KH SURPLVLQJ OLPSRJUDVVHV 3, DQG 3, ZHUH VXVSHFWHG WR KDYH KLJKHU SHUVLVWHQFH WKDQ %LJDOWD DORQJ ZLWK KLJKHU TXDOLW\ WKDQ 5HGDOWD :RUNLQJ ZLWK HVWDEOLVKHG VWDQGV RI OLPSRJUDVV LQ WKH OLPSRn JUDVVHV ZHUH IUDFWLRQDWHG DQG DQDO\]HG WR GHWHUPLQH WKH VLWH RI WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f VWRUDJH $W WKH FRQFOXVLRQ RI WKH SODQW SDUW VDPSOLQJ ILYH LQWHUYDOV RI FXWWLQJ ZHUH LPSRVHG LQ RUGHU WR GLIIHUHQWLDOO\ GUDLQ WKH VWRUDJH RUJDQV RI WKHLU FDUERK\GUDWH UHVHUYHV

PAGE 162

&DUERK\GUDWH OHYHOV ZHUH WHVWHG FKHPLFDOO\ DQG E\ WKH XVH RI UHJURZWK LQGDUNQHVV WHFKQLTXHV :HLJKW GDWD RQ HWLRODWHG VKRRWV VWXEEOH DQG URRWV SURYLGHG WKH EDVLV IRU D PRUSKRORJLFDO FRPSDULVRQ RI JHQRW\SHV 3RVLWLYH FRUUHODWLRQV EHWZHHQ ZHLJKW RI VWXEEOH DQG 71& VWRUHG LQ WKH ERWWRP FP RI VWHPV VXJJHVWHG WKDW ODUJH VWXEEOH V\VWHPV 5HGDOWD DQG 3, f PD\ DLG LQ SODQW SHUVLVWHQFH ,Q DQ H[WHQVLYH ILHOG H[SHULPHQW ZDV FRQGXFWHG ZLWK ILYH OHYHOV RI 1 IHUWLOL]DWLRQ NJKD\Uf DQG ILYH IUHn TXHQFLHV RI GHIROLDWLRQ DQG ZHHNVf 'U\ PDWWHU '0f \LHOGV LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,90'f FUXGH SURWHLQ &3f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

PAGE 163

5HVXOWV IURP WKLV VWXG\ VKRZ WKDW WKH '0 \LHOG SURGXFWLRQ LQ VSULQJ FRXOG EH H[SORLWHG LQ PDQDJHPHQW V\VWHPV /LPSRJUDVVHV PD\ EH XWLOL]HG IRU JUD]LQJ LQ WKH VSULQJ RI WKH \HDU XQWLO EDKLDJUDVV EHJLQV LWV JURZWK 7KH VXPPHU SHULRG ZDV PDUNHG E\ DFWLYH OLPSRJUDVV JURZWK ORZ SHUn FHQWDJHV RI VWRUHG 71& DQG YHU\ ORZ YDOXHV IRU &3 DQG ,920' 7KH VXPPHU SHULRG DOVR VHHPHG WR EH WKH WLPH ZKHQ OLPSRJUDVV ZDV PRVW VXVn FHSWLEOH WR VWUHVV DQG ORVV RI VWDQG 7KH UHFRPPHQGDWLRQ WR UHVW WKH OLPSRJUDVV GXULQJ WKH VXPPHU KDV EHHQ VXJJHVWHG KRZHYHU WKH DFFXPXn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f VKRZHG WKDW FDUERK\GUDWH OHYHOV LQ WKH WLVVXH PD\ EH DGHTXDWH IRU VLODJH PDNLQJ (YHQ LI WKH\ ZHUH QRW WKH WHFKQRORJ\ H[LVWV IRU WKH DGGLWLRQ RI FDUERK\GUDWH WR VLODJH /LPSRJUDVV VWXGLHG DV VWRFNSLOHG IRUDJH GLG QRW HIIHFWLYHO\ XWLOL]H IHUWLOL]HU 1 DSSOLHG DW UDWHV DERYH NJKD 7KH $XJXVW VWDJLQJ GDWH ZDV FULWLFDO IRU VXIILFLHQW '0 DFFXPXODWLRQ KRZHYHU WKH VWDQGLQJ IRUDJH VKRXOG EH XWLOL]HG EHIRUH IURVW LQ RUGHU WR WDNH PD[LPXP DGYDQn WDJH RI WKH UHODWLYHO\ JRRG IRUDJH TXDOLW\ SUHVHQW LQ DXWXPQ OLPSRJUDVV KHUEDJH

PAGE 164

)LQDOO\ LI D SUHVHUYDWLRQ WHFKQLTXH ZHUH VWXGLHG DQG HOXFLGDWHG WKH OLPSRJUDVV FRXOG EH IHG ZLWK VXSSOHPHQWDO SURWHLQ DQG PLQHUDOV GXULQJ WKH ZLQWHU ,Q FRQFOXVLRQ LW ZRXOG DSSHDU WKDW OLPSRJUDVV FDQ DGG PXFK IOH[LELOLW\ WR WKH IODWZRRG IRUDJH V\VWHP DQG FRQWLQXHG HIIRUWV LQ WKLV GLUHFWLRQ VHHP MXVWLILHG

PAGE 165

$33(1',&(6

PAGE 166

$33(1',; $ 727$/ 1216758&785$/ &$5%2+<'5$7( 71&f 352&('85(

PAGE 167

727$/ 1216758&785$/ &$5%2+<'5$7( 71&f 352&('85( 81,9(56,7< 2) )/25,'$ $*52120< '(3$570(17 *$,1(69,//( )/ $8*867 )2857+ 5(9,6,21 6FRWW &KULVWLDQVHQ %RRWH DQG ( % %OD]H\ 3UHOLPLQDU\ 6WHSV 6DPSOH $FTXLVLWLRQ $IWHU WKRURXJK FRQVLGHUDWLRQ RI H[SHULPHQWDO GHVLJQ WKH VDPSOLQJ IUHTXHQF\ DQG SODQW SDUW WR EH DQDO\]HG PD\ EH GHWHUPLQHG 7RWDO QRQ VWUXFWXUDO FDUERK\GUDWH 71&f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n WLRQ ZKLFK ERWK UHVXOW LQ ORVV RI FDUERK\GUDWHV

PAGE 168

32%R[ 6DQ &LW\ $= )HEUXDU\ 7R :KRP &RQFHUQV 7KLV QRWH LV LQ UHJDUG WR WKH DYDLODELOLW\ RI .\L DVH HQ]\PH VXJJHVWHG IRU XVH LQ :LVFRQVLQ $JULH ([S WD 5HV 5HSRUW 5& OOf HQWLWOHG 5HPRYLQJ t DQDO\]LQJ WRWDO QRQVWUXFWXUnDO FDUERK\n GUDWHV IURP SODQW WLVVXHn ,Q WKH ILUVW OLQHV RI FROXPQ RQ SDJH } LW VWDWHV WKDW WKH FKHPLFDO FRPSDQ\ SURGXFLQJ 0\ODVH LV LQ 'HV 3ODLQHV ,/ 7KH FRPSDQ\ KDV PRYHG IURP ,OOLQRLV WR &KDUORWWH 12 DQG WKH\ FDQ EH FRQWDFWHG DV RXWOLQHG EHORZ *% )HUPHQWDWLRQ ,QGXVWULHV ,QF 7HO 3 %R[ &KDUORWWH 1& N-X X X L & M \ 6LQFHUHO\ 'DO H P L WK (KHULWXV 3URIHVVRU RI $JURQRP\ 8QLY RI :L VFRQ VLQ0DGL VRQ

PAGE 169

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

PAGE 170

H[FOXGH DQ\ UHPDLQLQJ DLU VSDFH 7KLV FRDUVHO\ JURXQG WLVVXH VKRXOG VXEVHTXHQWO\ EH UHJURXQG WKURXJK D VPDOO :LOH\ PLOO D EDOO PLOO RU D 8'< &\FORQH PLOO WR SDVV PP RU D PHVK VFUHHQ 7KH VHFRQG JULQGn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n SXWHU FDUGV RU D WHUPLQDO 6HH 7DEOH $O IRU DQ H[DPSOH RI D MRE VXEn PLWWHG XVLQJ 6$6 DW 1(5'&f 3UHSDUH WKH HQ]\PH PL[ +DYH 0B DFHWDWH EXIIHU PDGH SULRU WR WKLV RSHUDWLRQf 7KH HQ]\PHV QHFHVVDU\ IRU 71& DQDO\VLV DQG VXSSOLHU LQIRUPDWLRQ PD\ EH IRXQG LQ DQRWKHU VHFWLRQ RI WKLV SURFHGXUH $GG PO RI GLVWLOOHG ZDWHU WR HUOHQPH\HU IODVNV KROGLQJ VDPSOHV LQWHUQDO SODQW VWDQGDUGV JOXFRVH IRU GHWHUPLQDWLRQ RI JOXFRVH

PAGE 171

7DEOH $O 6$6 MRE DV VXEPLWWHG RQ FDUGV WR WKH FRPSXWHU 71& -2% f86(,)$6)50&/$66 $ r3$66:5' (;(& 6$6 '$7$ 3$576f &200(17 7+( &+$//(1*( 2) /($51,1* 6$6 ,6 <285 352%/(0 ,1387 0 < /,1( $*( 37 53 ,1& '9 :7 3, '1 3 ' ' '523 '' '$7( 0'<0'
PAGE 172

7DEOH $ 6WDWLVWLFDO DQDO\VLV V\VWHP 2% 0 < /,1( $*( 37 53 ,1& '9 :7 3, '1 3 -8/$1 3 2'$ 0,&*0 0*&+2 &+2 $'&+2 :. 52 5 :. &5 & :. 67 6 :. 6+ 7 :. 52 5 :. &5 & :. 67 6 :. 6+ 7 :. 52 5 :. &5 & :. 67 6 :. 6+ 7 :. 52 5 :. &5 & :. 67 6 :. 6+ 7

PAGE 173

UHFRYHU\ DQG WKH HPSW\ IODVNV XVHG WR GHWHUPLQH WKH 71& FRQWHQW RI WKH HQ]\PH $ &RUQZDOO SLSHW ZRUNV ZHOO IRU WKLV DGGLWLRQ &ODPS WKH IODVNV LQWR WKH VDPSOH EHG DQG WRS ZLWK WKH VWRSSHUSROLFHPHQ $GMXVW WKH ZDWHU OHYHO LQ WKH ERLOLQJ EDWK KLJKHU WKDQ WKH OHYHO RI WKH OLTXLG LQ WKH IODVNV ,PPHUVH WKH VDPSOH EHG LQWR WKH ERLOLQJ &f ZDWHU IRU PLQXWHV 8VH D WLPLQJ FORFN 5HPRYH WKH VDPSOH EHG IURP WKH ERLOLQJ ZDWHU EDWK DQG DGG PO 2I 0 DFHWDWH EXIIHU WR HDFK IODVN ZKHQ WKH\ KDYH FRROHG 7KH &RUQn ZDOO SLSHW LV XVHIXO IRU WKLV RSHUDWLRQ DOVR $GG PO RI HQ]\PH PL[ WR WKH IODVNV IRU VDPSOHV JOXFRVH UHFRYHU\ SODQW VWDQGDUGV DQG HQ]\PH EODQNV XVLQJ WKH 2[IRUG WULSOH UDQJH PLFURSLSHW VDPSOHU $GG PO RI ZDWHU LQVWHDG RI HQ]\PH WR WKH JOXFRVH IODVNV 7KH HQ]\PH VOXUU\ VKRXOG EH YLJRURXVO\ VWLUUHG RQ D VWLUULQJ SODWH GXULQJ WKLV VWHS WR LQVXUH D KRPRJHQRXV PL[f 5HSODFH VWRSSHUSROLFHPHQ WR PLQLPL]H HYDSRUDWLRQ 5HWXUQ WKH VDPSOH EHG ZLWK WKH VDPSOHV WR WKH VKDNHU XQLW 7KH ZDWHU LQ WKH EDWK VKRXOG EH DW & DQG VHW DW D UHDVRQDEOH VKDNLQJ RVFLOODWLRQ $OORZ D GLJHVWLRQ WLPH RI ZKDWHYHU KDV EHHQ GHWHUPLQHG QHFHVVDU\ LQ SUHOLPLQDU\ H[SHULPHQWV 0DNH VXUH WKH GLVWLOOHG ZDWHU OHYHO LQ WKH VKDNLQJ ZDWHU EDWK UHPDLQV DERYH WKH OHYHO RI OLTXLG LQ WKH IODVNV 3HUFK IXQQHOIROGHG ILOWHU SDSHU GLVNV FP GLDPHWHUf RQ WKH ULP RI PO EHDNHUV $UUDQJH WKH PO EHDNHUV RQ D WUD\ DQG HPSW\ WKH FRQWHQWV RI WKH HUOHQPH\HU IODVNV LQWR WKH ILOWHU SDSHU IXQQHOV ZKHQ WKH VDPSOH VOXUU\ LV FRRO 0DLQWDLQ VDPSOH LGHQWLW\ 7KH WHVW WXEH UDFNV ZLOO KROG OLSOHVV FXOWXUH WXEHV [ PPf DUUDQJHG LQ D [ FRQILJXUDWLRQ ,Q WKH ILUVW GLOXWLRQ HDFK

PAGE 174

UDFN KROGV WKH WHVW WXEHV QHHGHG IRU VDPSOHV 5HPRYH WZR DOLTXRWV IURP WKH ILOWUDWH LQ WKH PO EHDNHUV DQG SODFH WKH OLTXLG LQ WHVW WXEHV ZLWK WKH FRUUHVSRQGLQJ QXPEHUV )LJXUH $Of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f )RU H[DPSOH ZLWK OLPSRJUDVV DJDLQ D PO DOLTXRW LV WDNHQ IURP WKH WXEHV LQ WKH GLOXWLRQ UDFN DQG SODFHG LQ WKH WXEHV IRU GLOXWLRQ $GG PO RI DONDOLQH UHDJHQW WR HDFK WXEH 9RUWH[ HDFK UDFN DV D XQLW UDWKHU WKDQ E\ LQGLYLGXDO WXEHV ZLWKLQ D UDFN /RDG WKH UDFN IRU GLOXWLRQ RQWR WKH WUD\ GHVLJQHG IRU GLSSLQJ WKH VDPSOHV LQWR WKH ERLOn LQJ ZDWHU EDWK :KHQ WKH ZDWHU KDV UHDFKHG D ERLO &f ORZHU WKH

PAGE 175

)LJXUH $O 'LOXWLRQ RQH )LJXUH $ 'LOXWLRQ WZR

PAGE 176

GLSSLQJ WUD\ LQWR WKH EDWK IRU PLQXWHV +DYH WKH ZDWHU DGMXVWHG WR EH DERYH WKH OHYHO RI VDPSOHV LQ WKH WXEHV 0RGHUDWH WKH JDV IRU WKH EXUQHU VR WKDW VSODVKHV IURP WKH ERLO DUH NHSW IURP HQWHULQJ WKH WXEHV $IWHU PLQXWHV UDLVH WKH GLSSLQJ WUD\ DQG WUDQVIHU LW WR WKH FRROLQJ EDWK 5HPRYH WKH UDFNV RXW RI WKH GLSSLQJ WUD\ ZKHQ WKH VDPSOHV DUH FRROHG DQG DUUDQJH RQ WKH ODE EHQFK DV EHIRUH $GG PO RI DUVHQR PRO\EGDWH UHDJHQW WKH FRORU IRUPLQJ UHDJHQWf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nV ODZ UHODWLRQVKLSSOXV PO RI DONDOLQH UHDJHQW DQG PO RI DUVHQRPRO\EGDWH JLYLQJ D WRWDO RI PO LQ WKH WXEH 7KHUHIRUH PO RI GLVWLOOHG ZDWHU LV DGGHG WR REWDLQ WKH ILQDO YROXPH RI PO LQ WKHVH WXEHV :DUP XS WKH &ROHPDQ -U ,, VSHFWURSKRWRPHWHU IRU PLQXWHV DQG VHW DW QP 0DNH VXUH WKH GLVSOD\ LV SXVKHG WR WKH ULJKW VLQFH LW

PAGE 177

7 LV ORRVHf 7KH FXYHWWH VKRXOG EH VFUXSXORXVO\ FOHDQ 3RXU D VPDOO DPRXQW RI GLVWLOOHG ZDWHU LQWR WKH IXQQHO RI WKH YDFXYHWWH FHOO )LOO WKH YDFXYHWWH FHOO ZLWK WKH UHDFWHG ZDWHU EODQN GHVFULEHG LQ WKH VHFn WLRQ RQ JOXFRVH VWDQGDUGVf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n WLRQ PDNHV WKH GDWD HDVLHU WR SXQFK RQWR FRPSXWHU FDUGV DQG GRHV QRW FKDQJH WKH UHODWLRQVKLS ZLWK WKH VWDQGDUGV *OXFRVH 6WDQGDUGV &DUHIXOO\ ZHLJK RXW DQG J DQK\GURXV 'JOXFRVH LQWR / YROXPHWULF IODVNV DQG ILOO WR WKH PDUN ZLWK WKH GLVWLOOHG ZDWHU 7KH JOXFRVH VKRXOG EH NHSW LQ D GHVLFn FDWRU SULRU WR ZHLJKLQJ 3RXU WKH DPRXQW RI VWDQGDUG QHHGHG IRU WKH GXUDWLRQ RI WKH H[SHULPHQWDWLRQ LQWR WZR RU PRUH VHWV RI SODVWLF ERWWOHV DQG IUHH]H WKHP %HIRUH XVLQJ WKH VWDQGDUGV WKDZ DQG DOORZ WKH VROXn WLRQ WR UHDFK URRP WHPSHUDWXUH 6XJDU VWDQGDUGV HQWHU WKH SURFHGXUH DW 6WHS DIWHU DOO VDPSOH WXEHV KDYH EHHQ EURXJKW WKURXJK WKH VHFRQG GLOXWLRQ 6XJDU VWDQGDUGV

PAGE 178

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f 6XSSOLHU 6LJPD &KHPLFDO &R 3 %R[ 6W /RXLV 0 7ROO IUHH f &XVWRPHU 6HUYLFH 3KRQH f FDOO FROOHFWf % ,QYHUWDVH FRQFHQWUDWH IURP \HDVW LQ JO\FHUROf 6WRFN ( PO f 6XSSOLHU *DOODUG 6FKOHVVLQJHU &KHP 0IJ &RUS 0LQHOD $YHQXH &DUOH 3ODFH 1< 3KRQH f 127( 7KH HQ]\PHV VKRXOG EH NHSW LQ WKH IUHH]HU 7KH DP\ORJOXFRVLGDVH VKRXOG EH NHSW GHVLFFDWHG ZKLOH LQ WKH IUHH]HU

PAGE 179

(Q]\PH PL[ PO GLVWLOOHG A PO 0 $FHWDWH EXIIHU PO ,QYHUWDVH FRQFHQWUDWH J DP\ORJOXFRVLGDVH J WK\PRO 0DNH HQ]\PH PL[ IUHVK GDLO\ DQG VWLU RQ D PDJQHWLF VWLUUHU ZKHQ UHPRYLQJ D PO DOLTXRW IRU DGGLWLRQ WR WKH VDPSOH 7KH EXIIHU DQG ZDWHU WDNH WKH UHGXFLQJ VXJDUV LQWR VROXWLRQ 7KH LQYHUWDVH K\GURO\]HV WKH ERQG EHWZHHQ JOXFRVH DQG IUXFWRVH LQ VXFURVH DQG WKH DP\ORJOXFRVLn GDVH K\GURO\]HV WKH VWDUFK PROHFXOHV LQWR PRQRPHUV 7KH UHVXOWLQJ PRQRPHUV DUH PHDVXUHG E\ WKH 1HOVRQn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

PAGE 180

:HLJK RXW PJ RI JOXFRVH DQG WUHDW LW DV \RX ZRXOG D GU\ VDPSOH H[FHSW DGG PO RI ZDWHU LQVWHDG RI HQ]\PH VROXWLRQ $ PO DOLTXRW GLOXWHG XS WR PO DGG PO RI GLVWLOOHG ZDWHUf LQ WKH ILUVW GLOXWLRQ ZLOO UHVXOW LQ D PLGUDQJH RSWLFDO GHQVLW\ XVLQJ D PO DOLTXRW LQ WKH VHFRQG GLOXWLRQ :HLJK RXW J JOXFRVH DQG GLVVROYH LQ / GLVWLOOHG ZDWHU $W I WKH EHJLQQLQJ RI HDFK UXQ SXW PO RI VWRFN PJPOf LQWR WKH HU\OHQ PH\HU IODVNV $GG RQO\ PO RI ZDWHU WR JLYH PO LQ WKH LQLWLDO ERLO VR WKDW YROXPHV EHWZHHQ JOXFRVH UHFRYHU\ IODVNV DUH FRPSDUDEOH WR YROXPH RI VDPSOH IODVNV 8VH WKH VDPH GLOXWLRQV DV GHVFULEHG DERYH $IWHU VXEWUDFWLQJ WKH TXDQWLW\ RI FDUERK\GUDWH DGGHG E\ WKH HQ]\PH IURP SODQW VWDQGDUGV DQG VDPSOHV DQ DGMXVWPHQW LV PDGH WR SHUFHQW JOXFRVH UHFRYHU\ VHH FDOFXODWLRQVf %HHUnV /DZ $ UHJUHVVLRQ HTXDWLRQ PXVW EH XVHG WR REWDLQ WKH UHODWLRQVKLS EHWZHHQ VXJDU FRQWHQW RI WKH VDPSOHV DQG WKH RSWLFDO GHQVLW\ UHDG DW QP 7KH VWDQGDUGV KDYH D NQRZQ TXDQWLW\ RI FDUERK\GUDWH DQG VHUYH DV WKH UHIHUHQFH DQG PXVW EH LQFOXGHG LQ HYHU\ VHW GXH WR D PXOWLWXGH RI IDFWRUV WKDW FDQ FKDQJH WKH FRORU GHYHORSPHQW IURP UXQ WR UXQ ,W LV IDVWHU WR GHWHUPLQH WKH UHJUHVVLRQ HTXDWLRQ E\ KDQG FDOFXODn WRU VR WKDW WKH FRPSXWHU PD\ EH IHG WKH UHPDLQGHU RI WKH LQIRUPDWLRQ LQ 6$6 LQSXW WR FRPSOHWH WKH FRPSXWDWLRQV $ IHZ ZRUGV RQ WKHRU\ DUH QHFHVVDU\ :H DVVXPH WKH JOXFRVH VWDQn GDUGV WR EH DFFXUDWH WKXV LW ZLOO EH WKH LQGHSHQGHQW YDULDEOH ;f 7KH RSWLFDO GHQVLW\ YDOXHV ZLOO WKHUHIRUH EH GHSHQGHQW YDOXHV
PAGE 181

WKH VWDQGDUGV DUH DVVXPHG FRUUHFW WKHQ RXU DQDO\VLV UHYHDOV YDULDELOLW\ LQ WKH WHFKQLTXH )LJXUH $f :H DUH QRW LQWHUHVWHG LQ SUHGLFWLQJ RSWLFDO GHQVLWLHV WKHUHIRUH WKH HTXDWLRQ LV UHDUUDQJHG WR JHW WKH LQYHUVH UHODWLRQVKLS &RQVLGHU WKH IROORZLQJ VHW RI GDWD IRU VWDQGDUGV 3XQFK LQWR WKH 7, FDOFXODWRU 4MJ JOXH X*f@ >RSWLFDO GHQVLW\ 23f@ GLVSOD\ [ W \ ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, ,, 3UHVV QG 6/23( QG ,17&3 IRU WKH HTXDWLRQ 2' P\*f E 2' r \J f f B 2' \J X r QQ XX \J r &DOFXODWLRQV 6WDQGDUGV ZHUH PDGH E\ SODFLQJ J RI 'JOXFRVH LQ / YROXPHWULF IODVNV / PO 22OJ PJ / J PJPO

PAGE 182

I 2SWLFDO 'HQVLW\ DEV[ f < 9DULDELOLW\ LQ < IRU ; RU !n R! R LJ *OXFRVH 6WDQGDUG ; )LJXUH $ 7KH UHJUHVVLRQ RI RSWLFDO GHQVLW\ 2'f YHUVXV JOXFRVH

PAGE 183

7KH UHJUHVVLRQ HTXDWLRQ GHWHUPLQDWHG LQ WKH ODVW VHFWLRQ ZDV EDVHG RQ \J PJPO RU PJWXEH r f§A \J JOXFRVH (Q]\PH EODQNV JOXFRVH UHFRYHU\ DQG LQWHUQDO SODQW VWDQGDUG YDOXHV DUH GHWHUPLQHG DV D IXQFWLRQ RI WKH %HHUnV ODZ UHJUHVVLRQ IRU WKH VXJDU VWDQGDUGV 7KH HQ]\PH EODQN FDQ EH GHWHUPLQHG ILUVW LI LW KDV DQ LGHQWLFDO GLOXWLRQ VHULHV DV WKH XQNQRZQ VDPSOH HJ f \J IRU HQ]\PH EODQNf r 2' r 6DPSOH FDUERK\GUDWH FRQWHQW VXEWUDFWLQJ WKH HQ]\PH EODQNf HJ f \J IRU VDPSOH RU SODQW VWDQGDUGf r 2' r 6XEWUDFWLRQ RI HQ]\PH &+2 \J IRU VDPSOH \J IRU HQ]\PH EODQN &RQYHUW WR PJ FDUERK\GUDWH 127( ,I D GLIIHUHQW GLOXWLRQ ZDV XVHG IRU WKH HQ]\PH EODQN EULQJ ERWK VDPSOH DQG HQ]\PH EODQN WR WKH OHYHO RI PJ &+2 DQG WKHQ VXEWUDFW RXW WKH PJ &+2 IRU WKH HQ]\PH EODQN LH PJ &+2 PJ &+2 VDPSOH HQ]\PHf PJ &+2 HQ]\PH EODQNff PJ &+2 \J r ATTTA r VW GLOXWLRQf r QG GLOXWLRQf

PAGE 184

LF r r PO r PO PO POn &RQYHUW WR b &+2 DVVXPH VDPSOH ZHLJKW PJf b &+2 r ZW b *OXFRVH 5HFRYHU\0HWKRG DVVXPH ZW PJ JOXFRVH f X PJ &+2 r 2' r r PJ r PO r PO J PO PO DVVXPH ZHLJKW RI JOXFRVH f b RI JOXFRVH UHFRYHU\ r b $GMXVW VDPSOH YDOXHV WR SHUFHQW JOXFRVH UHFRYHU\ $'&+b VDP3OH YDOXH HQ]\PH EODQN VXEWUDFWHG RXWf JOXFRVH UHFRYHU\ B b 5HDJHQWV $ONDOLQH UHDJHQW 'LVVROYH J DQK\GURXV VRGLXP FDUERQDWH A&2Jf J SRWDVVLXP VRGLXP WDUWUDWH .1D&A+ +f J

PAGE 185

VRGLXP ELFDUERQDWH 1D+&2Af DQG J DQK\GURXV VRGLXP VXOIDWH 1D6Af LQ PO GLVWLOOHG ZDWHU DQG WKHQ GLOXWH WR PO LQ D YROXPHWULF IODVN 'LVVROYH J FXSULF VXOIDWH SHQWDK\GUDWH &X62A f +f LQ PO GLVWLOOHG ZDWHU IROORZHG E\ RQH GURS RI FRQFHQWUDWHG VXOIXULF DFLG )A62Af &RPELQH DQG PL[ WKH WZR VROXWLRQV )ROORZ WKHVH GLUHFWLRQV $UVHQRPRO\EGDWH UHDJHQW 'LVVROYH J DPPRQLXP PRO\EGDWH WHWUDK\GUDWH >1+AfH02A2AA2f@ LQ PO GLVWLOOHG ZDWHU WKHQ DGG PO FRQFHQWUDWHG VXOIXULF DFLG ZKLOH VWLUULQJ 'LVVROYH J VRGLXP DUVHQDWH 1DA$V2A f +f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‘ + SHU OLWHU RI GLVWLOOHG ZDWHU

PAGE 186

)RU 0 EXIIHU FRPELQH PO 0 DFHWLF DFLG ZLWK PO VRGLXP DFHWDWH 7LWUDWH WKH ILQDO EXIIHU VROXWLRQ WR S+ E\ DGGLWLRQ RI HLWKHU VWRFN VROXWLRQ 5HIULJHUDWH 0 $FHWDWH %XIIHU 'LOXWH WKH 0 EXIIHU DFFRUGLQJO\

PAGE 187

$33(1',; % 727$/ 1216758&785$/ &$5%2+<'5$7( 71&f 5(68/76

PAGE 188

7DEOH $ 3HUFHQW WRWDO QRQVWUXFWXUDO FDUERK\GUDWH 71&f PHDQV IRU DOO WUHDWPHQW GDWHV DQG OLPSRJUDVVHV LQ +2 +, + + + + + + + + +, +, +, 0DU $SU 0D\ 0D\ -XQ -XO -XO $XJ 6HS 6HS 2FW 1RY 1RY 3, P

PAGE 189

7DEOH $ &RQWLQXHG +2 +, + + + + + + + + +, +, +, 0DU $SU 0D\ 0D\ -XQ -XO -XO $XJ 6HS 6HS 2FW 1RY 1RY 5HGDOWD f %LJDOWD P P P V X P

PAGE 190

7DEOH $ &RQWLQXHG 2% 1r +2 +, + + + + + + + + +, +, +, Srr 0DU $SU 0D\ 0D\ -XQ -XO -XO $XJ 6HS 6HS 2FW 1RY 1RY 3, f D f f f r1LWURJHQ NJKD\Uf rr)UHTXHQF\ ZHHNVf

PAGE 191

5()(5(1&(6 $GDPV $OWR -U 'HYHORSPHQW RI WKH %UDIRUG %UHHG 3URF VW $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS $GHJEROD $ 3UHOLPLQDU\ REVHUYDWLRQV RQ WKH UHVHUYH FDUERK\n GUDWH DQG UHJURZWK SRWHQWLDO RI WURSLFDO JUDVVHV 3URF WK ,QW *UDVVOG &RQJU +HOVLQNL )LQODQG SS $GHJEROD $ $ DQG & 0 0F.HOO D (IIHFW RI QLWURJHQ IHUWLOLn ]DWLRQ RQ WKH FDUERK\GUDWH FRQWHQW RI &RDVWDO EHUPXGDJUDVV &\QRGRQ GDFW\O RQ /f 3HUVf $JURQ $GHJEROD $ $ DQG & 0 0F.HOO E 5HJURZWK SRWHQWLDO RI &RDVWDO EHUPXGDJUDVV DV UHODWHG WR SUHYLRXV QLWURJHQ IHUWLOL]DWLRQ $JURQ $JULFXOWXUH LQ DQ 8UEDQ $JH $*8$f 5HSRUW ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH $OEHUGD 7K 7KH LQIOXHQFH RI UHVHUYH VXEVWDQFHV RQ GU\ PDWWHU SURGXFWLRQ DIWHU GHIROLDWLRQ 3URF WK ,QW *UDVVOG &RQJU +HOVLQNL )LQODQG SS $OH[DQGHU & : DQG ( 0F&ORXG ,QIOXHQFH RI WLPH DQG UDWH RI QLWURJHQ DSSOLFDWLRQ RQ SURGXFWLRQ DQG ERWDQLFDO FRPSRVLWLRQ RI IRUDJH $JURQ $PLQRII : : %LQNOH\ 5 6FKDIIHU DQG 5 : 0RZU\ $QDO\WLFDO PHWKRGV IRU FDUERK\GUDWHV -ML 7KH &DUERK\GUDWHV 9RO ,% $FDGHPLF 3UHVV 1HZ
PAGE 192

$UUR\R$JXLOX $ DQG &RZDUG/RUG E 5HODWLRQVKLSV EHWZHHQ DQG ZLWKLQ SK\VLFDO DQG FKHPLFDO FRQVWLWXHQWV DQG LQ YLWUR WUXH GLJHVWLELOLW\ LQ WURSLFDO IRUDJH JUDVVHV $JULF7 8QLY 3XHUWR 5LFR $XGD +DPLG 5 ( %ODVHU DQG 5 + %URZQ 7LOOHULQJ DQG FDUERn K\GUDWH FRQWHQWV RI RUFKDUGJUDVV DV LQIOXHQFHG E\ HQYLURQPHQWDO IDFWRUV &URS 6FL %DNHU ) 6 -U )LQLVKLQJ FDWWOH LQ QRUWK )ORULGD 3UHVHQW DQG IXWXUH 3URF WK $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS %DNHU ) 6 -U /HWnV SURGXFH VODXJKWHU EHHI LQ )ORULGD 3URF WK $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULn FXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS %DNHU + DQG ( $ *DUZRRG 6WXGLHV RQ WKH URRW GHYHORSPHQW RI KHUEDJH SODQWV 9 6HDVRQDO FKDQJHV LQ IUXFWRVDQ DQG VROXEOH VXJDU FRQWHQWV RI FRFNVIRRW KHUEDJH VWXEEOH DQG URRWV XQGHU WZR FXWWLQJ WUHDWPHQWV %ULW *UDVVOG 6RF %HDW\ ( 5 + 7DQ 5 $ 0F&UHHU\ DQG 3RZHOO
PAGE 193

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f %U\DQ (UZLQ -U 3UREOHPV DQG RSSRUWXQLWLHV LQ PDUNHWLQJ FRZV DQG IHHGORW EHHI SURGXFHG LQ )ORULGD DQG WKH VRXWKHDVW 8QLWHG 6WDWHVf§3DFNHU YLHZV 3URF WK $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS %XUULV 6 5 + %URZQ DQG 5 ( %ODVHU (YDOXDWLRQ RI UHVHUYH FDUERK\GUDWHV LQ 0LGODQG EHUPXGDJUDVV &\QRGRQ GDFW\ORQ /f &URS 6FL %XUWRQ : DQG ( -DFNVRQ $ PHWKRG IRU PHDVXULQJ VRG UHVHUYHV $JURQ &DG\ ) % DQG : $ )XOOHU 7KH VWDWLVWLFVFRPSXWHU LQWHUIDFH LQ DJURQRPLF UHVHDUFK $JURQ &DUSHQWHU : SHUVRQDO FRPPXQLFDWLRQf $JULFXOWXUDO 7HFKQLFLDQ 'HSDUWPHQW RI $JURQRP\ ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH &DUWHU / DQG / $ *DUUDUG (IIHFWV RI &QLJKW WHPSHUDn WXUH DQG OHDI DVVLPLODWH OHYHOV RQ SKRWRV\QWKHVLV RI 'L JLWDUL D GHFXPEHQV 6WHQW 3URF 6RLO &URS 6FL )OD &KDQJ & ) $ 6X]XNL 6 .XPDL DQG 6 7DPXUD &KHPLFDO VWXGLHV RQ nFORYHU VLFNQHVVn 3DUW ,, %LRORJLFDO IXQFWLRQV RI LVRIODYLQRLGV DQG WKHLU UHODWHG FRPSRXQGV $JULH %LRO &KHP

PAGE 194

&KDWWHUWRQ 1 ( &DUOVRQ : ( +XQJHUIRUG DQG 5 /HH (IIHFW RI WLOOHULQJ DQG FRRO QLJKWV RQ SKRWRV\QWKHVLV DQG FKORURSODVW VWDUFK LQ 3DQJOD &URS 6FL &KDYHV 0 0 & ) DQG 0RUHLUDA 0WRGRV SDUD D GHWHUPLQDGR GH KLGUDWRV GH FDUERQR WRWDLV QDR HVWUXWXUDLV $JURQRPLD /XVLW f &KLSSHQGDOO / $ $ JXLGH WR WKH LGHQWLILFDWLRQ RI JUDVVHV LQ 6RXWK $IULFD B,QB 0HUHGLWK HGf 7KH *UDVVHV DQG 3DVWXUHV RI 6RXWK $IULFD &HQWUDO 1HZV $JHQF\ 3DURZ & 3 &DSH 7LPHV /7' SS &KULVWLDQVHQ 6FRWW & 5XHONH DQG 5 O\QFK 5HJURZWK LQ GDUNQHVV DV LQIOXHQFHG E\ SUHYLRXV FXWWLQJ WUHDWPHQW RI IRXU OLPSRJUDVV JHQRW\SHV 3URF 6RLO &URS 6FL 6RF )OD &RFKUDQ : DQG 0 &R[ ([SHULPHQWDO 'HVLJQV -RKQ :LOH\ DQG 6RQV ,QF 1HZ
PAGE 195

&XJQDF $QWRLQH GH 5HVHDUFK RQ JOXFRVLGHV RI WKH *UDPLQHDH $QQ 6FL 1DW GD6LOYHLUD $ ) ) )HLWRVD 7HOHV DQG : 6WXOO $ UDSLG WHFKQLTXH IRU WRWDO QRQVWUXFWXUDO FDUERK\GUDWH GHWHUPLQDWLRQ RI SODQW WLVVXH $JULH )RRG &KHP f 'DYLGVRQ / DQG ) / 0LOWKRUSH &DUERK\GUDWH UHVHUYHV LQ WKH UHJURZWK RI FRFNVIRRW 'DFW\OLV JORPHUDWD /f %ULW *UDVVG 6RF 'DYLGVRQ / DQG ) / 0LOWKRUSH D 7KH HIIHFW RI GHIROLDWLRQ RQ WKH FDUERQ EDODQFH LQ 'DFW\OLV JORPHUDWD $QQ %RW 16 'DYLGVRQ / DQG ) / 0LOWKRUSH E /HDI JURZWK LQ 'DFW\OLV JORPHUDWD IROORZLQJ GHIROLDWLRQ $QQ %RW 16 'LHVFKH = &RORULPHWULF PHWKRGV LQ FDUERK\GUDWH FKHPLVWU\ ,QWURGXFWLRQf -QA 0HWKRGV LQ &DUERK\GUDWH &KHPLVWU\ 9RO 5 / :KLVWOHU DQG 0 / :ROIURP HGVf $FDGHPLF 3UHVV 1HZ
PAGE 196

)RUG & : DQG : 7 :LOOLDPV ,Q YLWUR GLJHVWLELOLW\ DQG FDUERK\GUDWH FRPSRVLWLRQ RI 'LJLWDUL D GHFXPEHQV DQG 6HWDULD DQFHSV JURZQ DW GLIIHUHQW OHYHOV RI QLWURJHQRXV IHUWLOL]HU $XVW $JULH 5HV *DOODKHU 5 1 DQG 5 + %URZQ 6WDUFK VWRUDJH LQ & YV & JUDVV OHDI FHOOV DV UHODWHG WR QLWURJHQ GHILFLHQF\ &URS 6FL *DUUDUG / $ DQG 6 + :HVW 6XERSWLPDO WHPSHUDWXUH DQG DVVLPLODWH DFFXPXODWLRQ LQ OHDYHV RI n3DQJODn GLJLWJUDVV 'LJLWDULD GHFXPEHQV 6WHQWf &URS 6FL *DVNLQV 0 + DQG $ 6OHSHU 3KRWRVHQVLWLYLW\ RI VRPH WURSLFDO IRUDJH JUDVVHV LQ )ORULGD 3URF 6RLO &URS 6FL 6RF )OD *RHULQJ + DQG 3 9DQ 6RHVW )RUDJH ILEHU DQDO\VHV DSSDUDWXV UHDJHQWV SURFHGXUHV DQG VRPH DSSOLFDWLRQVf 86'$ $JULH +DQGERRN *UDEHU / ) )RRG UHVHUYHV LQ UHODWLRQ WR RWKHU IDFWRUV OLPLWn LQJ WKH JURZWK RI JUDVVHV 3ODQW 3K\VLRO *UDEHU / ) 1 7 1HOVRQ : $ /HXNHO DQG : % $OEHUW 2UJDQLF IRRG UHVHUYHV LQ UHODWLRQ WR WKH JURZWK RI DOIDOID DQG RWKHU SHUHQQLDO KHUEDFHRXV SODQWV :LVFRQVLQ $JULH ([S 6WD %XOO *UHHQILHOG 3 / 'DOH 6PLWK DQG -RVH $ (VFDODGD &RPSRVLWLRQ RI VWDUFK DFFXPXODWHG LQ WKH OHDI ODPLQHDH RI WLPRWK\ VRUJKXP DQG OXFHUQH 6FL )RRG $JULH *UHHQILHOG 6 % DQG 'DOH 6PLWK 'LXUQDO YDULDWLRQV LQ WKH LQGLYLGXDO SDUWV RI VZLWFKJUDVV VKRRWV DW DQWKHVLV 5DQJH 0JPW f *UHHQZRRG & 7 6WDUFK DQG JO\FRJHQ 6WDUFK ,Q 7KH &DUERK\GUDWHV 9RO ,% : 3LJPDQ DQG +RUWRQ HGV-7 $FDGHPLF 3UHVV 1HZ
PAGE 197

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f +HQGHUVRQ 0 6 DQG / 5RELQVRQ (QYLURQPHQWDO LQIOXHQFHV RQ ILEHU FRPSRQHQW FRQFHQWUDWLRQV RI ZDUPVHDVRQ SHUHQQLDO JUDVVHV $JURQ +LOO : DQG : +XQWHU $ UHYLHZ RI UHVSRQVH VXUIDFH PHWKRGRORJ\ $ OLWHUDWXUH VXUYH\ 7HFKQRPHWULFV f +RGJH ( DQG % 7 +RIUHLWHU 'HWHUPLQDWLRQ RI UHGXFLQJ VXJDUV DQG FDUERK\GUDWHV ,QB 0HWKRGV LQ &DUERK\GUDWH &KHPLVWU\ 9RO 5 / :KLVWOHU DQG 0 / :ROIURP HGVf $FDGHPLF 3UHVV 1HZ
PAGE 198

+XPSKUH\V / 5 DQG $ 5 5RELQVRQ ,QWHUUHODWLRQV RI OHDI DUHD DQG QRQVWUXFWXUDO FDUERK\GUDWH VWDWXV DV GHWHUPLQDQWV RI WKH JURZWK RI VXEWURSLFDO JUDVVHV 3URF WK ,QW *UDVVOG &RQJU +HOVLQNL )LQODQG SS +XQWHU + / +DUULV DQG $ 6 -HQVHQ ,QWHJUDWHG 1DWXUDO 5HVRXUFH 0DQDJHPHQW %XOO )ORULGD &RRSHUDWLYH ([WHQn VLRQ 6HUYLFH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS +XQWHU 5 $ % / 0F,QW\UH DQG 5 0FOOUR\ :DWHUVROXEOH FDUERK\GUDWHV RI WURSLFDO SDVWXUH JUDVVHV DQG OHJXPHV 6FL )RRG $JULH ,NHUG -RKQ %HHI LQGXVWU\ SUREOHPV LQ WKH n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n3DQJODn DQG n7UDQVYDODn GLJLWJUDVV DQG n&RDVWFURVVn EHUPXGDJUDVV 3URF 6RLO DQG &URS 6FL 6RF )OD .LOOLQJHU % /LPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEEf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

PAGE 199

/HXNHO : $ DQG 0 &ROHPDQ *URZWK EHKDYLRU DQG PDLQWHQn DQFH RI RUJDQLF IRRGV LQ EDKLDJUDVV $JULFXOWXUDO ([SHULPHQW 6WDWLRQ %XOO ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS /LQGDKO 5 ( 'DYLV DQG : 6KHSDUG 7KH DSSOLFDWLRQ RI WKH WRWDO FDUERK\GUDWH PHWKRG WR WKH VWXG\ RI FDUERK\GUDWH UHVHUYHV RI VZLWFK FDQH $UXQGLQDUD WHFWDf 3ODQW 3K\VLRO /LWWHOO 5 & DQG 0RWW &RPSXWHU DVVLVWHG GHVLJQ DQG DQDO\VLV RI UHVSRQVH VXUIDFH H[SHULPHQWV LQ DJURQRP\ 3URF 6RLO &URS 6FL 6RF )OD 0DUVKDOO & DQG 5 6DJDU 7KH LQIOXHQFH RI GHIROLDWLRQ RQ WKH GLVWULEXWLRQ RI DVVLPLODWHV LQ /ROLXP PXOWLIORUXP /DP $QQ %RW 0DWFKHV $ ,QIOXHQFH RI FXWWLQJ KHLJKW LQ GDUNQHVV RQ PHDVXUHPHQW RI HQHUJ\ UHVHUYHV RI WDOO IHVFXH $JURQ 0D\ / + 7KH XWLOL]DWLRQ RI FDUERK\GUDWH UHVHUYHV LQ SDVWXUH SODQWV DIWHU GHIROLDWLRQ +HUEDJH $EVWUDFWV f 0D\ / + DQG / 'DYLGVRQ 7KH UROH RI FDUERK\GUDWH UHVHUYHV LQ UHJHQHUDWLRQ RI SODQWV $XVW $JULH 5HV 0F&DUW\ ( & 6HDVRQDO PDUFK RI FDUERK\GUDWHV LQ (O\PXV DPELJXV DQG 0XKOHQEHUJLD JUDFLOLV DQG WKHLU UHDFWLRQ XQGHU PRGHUDWH JUD]LQJ XVH 3ODQW 3K\VLRO 0F&DUW\ ( & 7KH UHODWLRQ RI JURZWK WR WKH YDU\LQJ FDUERK\n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f %XWWHUZRUWKV /RQGRQ 0LVOHY\ 3 % :DVKNR DQG +DUULQJWRQ 3ODQW PDWXULW\ DQG FXWWLQJ IUHTXHQF\ HIIHFW RQ WRWDO QRQVWUXFWXUDO FDUERK\GUDWH SHUFHQWDJHV LQ WKH VWXEEOH DQG FURZQ RI WLPRWK\ DQG RUFKDUGJUDVV $JURQ

PAGE 200

0RRUH ( 0RWW 'XQKDP DQG 5 : 2PHU /DUJH FDSDFLW\ LQ YLWUR RUJDQLF PDWWHU GLJHVWLRQ SURFHGXUH $QL 6FL $EVWUDFWf 0RRUH ( & 5XHONH & ( 5LRV DQG ( )UDQNH 1XWULWLYH HYDOXDWLRQ RI 3HQVDFROD KDELDJUDVV KD\V 3URF 6RLO &URS 6FL 6RF )OD 0RRUH ( 0 $ :RUUHOO 6 0 $EUDPV : 5 2FXPSDXJK DQG 0RWW 4XDOLW\ RI WURSLFDO SHUHQQLDO JUDVV KD\V )OD %HHI &DWWOH 5HV 5HSRUW ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS 0RWW D (YDOXDWLRQ RI SDVWXUH JHUPSODVP XQGHU GLIIHUHQW JUD]LQJ PDQDJHPHQW DOWHUQDWLYHV XQSXEOLVKHG SDSHUf 'HSDUWPHQW RI $JURQRP\ ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH 0RWW E 0DQDJHPHQW RI IRUDJHV IRU EHHI SURGXFWLRQ LQ )ORULGD 3URF VW $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS 0RWW DQG ( 0RRUH ([LVWLQJ DQG SRWHQWLDO V\VWHPV RI ILQLVKLQJ FDWWOH RQ IRUDJHV RU OLPLWHG JUDLQ UDWLRQV LQ WKH WURSLn FDO UHJLRQ RI WKH VRXWK ,QB )RUDJHIHG %HHI 3URGXFWLRQ DQG 0DUNHWLQJ $OWHUQDWLYHV LQ WKH 6RXWK $ 6WXHGHPDQQ HGf %XOO 6RXWKHUQ &RRSHUDWLYH 6HULHV SS 0\HUV 5 + 5HVSRQVH 6XUIDFH 0HWKRGRORJ\ $OO\Q DQG %DFRQ ,QF %RVWRQ SS 1HOVRQ 1 $ SKRWRPHWULF DGDSWDWLRQ RI WKH 6RPRJ\L PHWKRG IRU WKH GHWHUPLQDWLRQ RI JOXFRVH %LRO &KHP 1REOH $ DQG ) /RZH $FRKROVROXEOH FDUERK\GUDWHV LQ YDULRXV WURSLFDO DQG WHPSHUDWH SDVWXUH VSHFLHV 7URS *UDVVOGV f 2DNHV $ +HPDUWKULD FROOHFWLRQ IURP 6RXWK $IULFD 7XUULDOED f 2DNHV $ 5HVLVWDQFH LQ +HPDWKULD VSHFLHV WR WKH \HOORZ VXJDU FDQH DSKLG 6LSKD IODYD )RUEHV7 7URS $JULH 7ULQLGDGf f 2DNHV $ :LQWHU KDUGLQHVV LQ OLPSRJUDVV +HPDUWKULD DOWLVVLPD 3URF 6RLO &URS 6FL 6RF )OD 2DNHV $ DQG & )R\ $ ZLQWHU KDUG\ DOXPLQXP WROHUDQW SHUHQQLDO SDVWXUH JUDVV IRU PLQH VSRLO UHFODPDWLRQ $JURQ $EVWUDFWV S

PAGE 201

2FXPSDXJK : 5 $ QHZ +HPDUWKULD VHOHFWHG IRU SHUVLVWHQFH DQG TXDOLW\ 3URF VW $QQXDO %HHI &DWWOH 6KRUW &RXUVH ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS 2FXPSDXJK : 5 DQG + 4XHVHQEHUU\
PAGE 202

4XHVHQEHUU\ + $ 2DNHV DQG 'RURWK\ 6 -HVVRS &\WRORJLFDO DQG JHRJUDSKLFDO FKDUDFWHUL]DWLRQV RI +HPDUWKULD (XSK\WLFD LQ SUHVVf 4XHVHQEHUU\ + DQG : 5 2FXPSDXJK )RUDJH TXDOLW\ DQG \LHOG RI OLPSRJUDVV DV D FRQVHUYHG IRUDJH IRU ZLQWHU JUD]LQJ $JURQ $EVWUDFWV S 4XHVHQEHUU\ + DQG : 5 2FXPSDXJK 3HUVLVWHQFH \LHOG DQG GLJHVWLELOLW\ RI OLPSRJUDVV JHQRW\SHV XQGHU FOLSSLQJ DQG JUD]LQJ $JURQ $EVWUDFWV S 4XHVHQEHUU\ + DQG : 5 2FXPSDXJK &UXGH SURWHLQ ,920' DQG \LHOG RI VWRFNSLOHG OLPSRJUDVVHV $JURQ 4XHVHQEHUU\ + DQG : 5 2FXPSDXJK 0LQHUDO FRPSRVLWLRQ RI DXWXPQZLQWHU VWRFNSLOHG OLPSRJUDVV 7URS $JULH 7ULQLGDGfLQ SUHVVf 4XHVHQEHUU\ + : 5 2FXPSDXJK DQG & 5XHONH +HPDUWKLD DOWLVVLPD $ SDVWXUH JUDVV IRU WKH WURSLFV 3URF WK ,QW *UDVVOG &RQJU /H[LQJWRQ .< LQ SUHVVf 5£HVH 7 DQG $ 0 'HFNHU
PAGE 203

5XHONH & + 4XHVHQEHUU\ DQG $ 6OHSHU &RPSDULVRQ RI JUHHQKRXVH YV ILHOG SORW WHFKQLTXHV IRU HYDOXDWLQJ QHZ JHUP SODVP RI OLPSRJUDVV +HPDUWKULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEE $JURQ $EVWUDFWV S 6FKDQN 6 & &KURPRVRPH QXPEHUV LQ HOHYHQ QHZ +HPDUWKULD OLPSRJUDVVf LQWURGXFWLRQV &URS 6FL 6FKDQN 6 & 0 $ .ORFN DQG ( 0RRUH /DERUDWRU\ HYDOXDn WLRQ RI TXDOLW\ LQ VXEWURSLFDO JUDVVHV ,, *HQHWLF YDULDWLRQ DPRQJ +HPDUWKULDV LQ LQ YLWUR GLJHVWLRQ DQG VWHP PRUSKRORJ\ $JURQ 6FKRQH\ 5 $ 7 ) %D\ DQG ) 0RQFULHI 8VH RI FRPSXWHU JUDSKLFV LQ WKH GHYHORSPHQW DQG HYDOXDWLRQ RI UHVSRQVH VXUIDFHV $JURQ 6KHDUG 5 : 2UJDQLF UHVHUYHV DQG SODQW UHJURZWK ,BQ 7KH &KHPLVWU\ DQG %LRFKHPLVWU\ RI +HUEDJH 9RO : %XWOHU DQG 5 : %DLOH\ HGVf $FDGHPLF 3UHVV 1HZ
PAGE 204

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nV FDWWOH IHHGn LQJ LQGXVWU\ 6WDII 3DSHU U )RRG DQG 5HVRXUFH (FRQRPLFV 'HSDUWPHQW ,QVWLWXWH RI )RRG DQG $JULFXOWXUDO 6FLHQFHV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH SS 6XOOLYDQ 7 DQG 9 6SUDJXH &RPSRVLWLRQ RI WKH URRWV DQG VWXEEOH RI SHUHQQLDO U\HJUDVV IROORZLQJ SDUWLDO GHIROLDWLRQ 3ODQW 3K\VLRO 7DPXUD 6 & &KDQJ $ 6X]XNL DQG 6 .XPDL ,VRODWLRQ DQG VWUXFWXUH RI D QRYHO LVRIODYHQ GHULYDWLYH LQ UHG FORYHU $JULH %LRO &KHP 7DQJ & 6 DQG & &
PAGE 205

7UOLFD 0 -U DQG & : &RRN &DUERK\GUDWH UHVHUYHV RI FUHVWHG ZKHDWJUDVV DQG 5XVVLDQ ZLOGU\H DV LQIOXHQFHG E\ GHYHORSPHQW DQG GHIROLDWLRQ 5DQJH 0JPW 7URXJKWRQ $ &KHPLFDO FRPSRVLWLRQ ,QB 7KH 8QGHUJURXQG 2UJDQV RI +HUEDJH *UDVVHV $ 7URXJKWRQ HGf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n K\GUDWHV LQ SODQWV %RW 5HY :DWVRQ 9 + DQG & < :DUG ,QIOXHQFH RI LQWDFW WLOOHUV DQG KHLJKW RI FXW RI UHJURZWK DQG FDUERK\GUDWH UHVHUYHV RI GDOOLVJUDVV 3DVSDOXP GLODWDWXP 3RLUf &URS 6FL :HLHU / 5 :LOVRQ DQG 5 :KLWH $ VHPLDXWRPDWHG SURFHGXUH IRU HVWLPDWLQJ WRWDO QRQVWUXFWXUDO FDUERK\GUDWHV LQ JUDVVHV DQG FRPSDULVRQ ZLWK WZR RWKHU SURFHGXUHV &6,5 $XVW 'LY 7URS &URSV 3DVW 7HFK 3DS :HLQPDQQ + 'HWHUPLQDWLRQ RI WRWDO DYDLODEOH FDUERK\GUDWHV LQ SODQWV 3ODQW 3K\VLRO :HLQPDQQ + &DUERK\GUDWH UHVHUYHV LQ JUDVVHV 3URF WK ,QW *UDVVOG &RQJU &ROOHJH 3DUW 3$ 86$ SS :HLQPDQQ + 7KH FKHPLVWU\ DQG SK\VLRORJ\ RI JUDVVHV ,QB 7KH *UDVVHV DQG 3DVWXUHV RI 6RXWK $IULFD 0HUHGLWK HGf &HQWUDO 1HZV $JHQF\ 3DURZ & 3 &DSH 7LPHV /7' SS :HLQPDQQ + 7RWDO DYDLODEOH FDUERK\GUDWHV LQ JUDVVHV DQG OHJXPHV +HUEDJH $EVWUDFWV f

PAGE 206

:HLQPDQQ + DQG ( 3 *ROGVPLWK 8QGHUJURXQG UHVHUYHV RI &\QRGRQ GDFW\O RQ %HWWHU WXUI E\ UHVHDUFK $IULFDQ ([SORVLYHV DQG &KHPLFDO ,QGXVWULHV /WG DQG 6 $IULFDQ 7XUI 5HVHDUFK )XQG -RKDQQHVEXUJ SS :HVWKDIHU 0 $ 7 /DZ -U DQG 7 'XII &DUERK\GUDWH TXDQWLILFDWLRQ DQG UHODWLRQVKLSV ZLWK 1 QXWULWLRQ LQ FRROVHDVRQ WXUIJUDVV $JURQ :KLWH / 0 &DUERK\GUDWH UHVHUYHV RI *UDVVHV $ UHYLHZ 5DQJH 0JPW f :LOPV + : &DUPLFKDHO DQG 6 & 6FKDQN &\WRORJLFDO DQG PRUSKRORJLFDO LQYHVWLJDWLRQV RQ WKH JUDVV +HPDUWKDULD DOWLVVLPD 3RLUf 6WDSI HW &( +XEE &URS 6FL :LOVRQ 5 DQG & : )RUG 7HPSHUDWXUH LQIOXHQFHV RQ WKH LQ YLWUR GLJHVWLELOLW\ DQG VROXEOH FDUERK\GUDWH DFFXPXODWLRQ RI WURSLFDO DQG WHPSHUDWH JUDVVHV $XVW $JULH 5HV :LOVRQ 5 DQG / nW 0DQQHWMH 6HQHVFHQFH GLJHVWLELOLW\ DQG FDUERK\GUDWH FRQWHQW RI EXIIHO JUDVV DQG JUHHQ SDQLF OHDYHV LQ VZDUGV $XVW $JULH 5HV :RRGV ) : + & +DUULV DQG 5 ( &DOGZHOO 0RQWKO\ YDULDn WLRQV RI FDUERK\GUDWHV DQG QLWURJHQ LQ URRWV RI VDQGKLOO RDNV DQG ZLUHJUDVV (FRORJ\
PAGE 207

%,2*5$3+,&$/ 6.(7&+ 6FRWW &KULVWLDQVHQ ZDV ERUQ WKH VRQ RI &DUO $XJXVW &KULVWLDQVHQ -U DQG 0DU\ -HDQ %RLOHUf &KULVWLDQVHQ RQ -XQH DW SP LQ %XIIDOR 1HZ
PAGE 208

, FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ /$ &KDUOHV 5XHONH &KDLUPDQ 3URIHVVRU RI $JURQRP\ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ nn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

PAGE 209

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


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ERTACKWW9_ID7W0Q INGEST_TIME 2017-07-11T21:49:55Z PACKAGE UF00101080_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


ENERGY RESERVES AND AGRONOMIC CHARACTERISTICS OF
FOUR LIMPOGRASSES (Hemarthria altissima (Poir)
Stapf et C.E. Hubb) FOR FLORIDA'S FLATWOODS
BY
SCOTT CHRISTIANSEN
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1982

ACKNOWLEDGEMENTS
I would like to acknowledge the people most responsible for my
personal progress to date and those aiding me in the accomplishment of
my Ph.D. degree. First I thank my mother, Jean, whose compassion taught
me how to be a human being and my father, Red, who gave me the guts to
be a man. Thank you Candy and Tracy for giving me the love only a
sister and brother can give.
I am grateful for the understanding and intelligent guidance of my
friend and mentor Dr. J. C. Winters who introduced me to higher
education. I wish to acknowledge my agronomic mentors Dr. Jesse M.
Scholl, Dr. Dale W. Smith, and Dr. 0. Charles Ruelke, and thank the
balance of my Ph.D. committee, Dr. William R. Ocumpaugh, Dr. Kenneth J.
Boote, Dr. John E. Moore, and Dr. Kenneth H. Quesenberry, for their
individual help, care, and support.
And I thank my friends in Florida who participated most closely
with my "science project":
Carrie Kitts Christiansen for her love and affection;
Bernard P. Monahan for his analytical skills and altruism;
Wendy J. Carpenter for the confidence we shared in each other;
Susan E. Sladden for her trust and diligence;
Hector Urbistondo for his root scraping;
Abelardo J. Saldivar for his composure, consideration, and care;
Findlay M. Pate for his advice, guidance, and generosity;
Paul Robin Harris and Paul W. Lane for their fun loving, pragmatic,
and joyfully sarcastic attitudes;
Deborah Vinci for her overwhelming sensitivity;

Jim and Janet Dean for visitation hours when I needed them the
most; and
Janet Eldred for her intelligent and tolerant approach to the task
of typing this dissertation.
iii

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS ii
LIST OF TABLES vii
LIST OF FIGURES ix
ABSTRACT xii
INTRODUCTION 1
LITERATURE REVIEW 2
Flistory, Potential, Limitations, and Development
of Florida's Forage-Livestock Industry 2
Development of Grazing Land in Florida 3
History and Evolution of Florida's Grazing Research ... 5
Grazing Animal Improvement 6
Florida's Role in the Forage-Livestock Industry 7
The Florida Opportunity for Finished Beef 10
Barriers to Forage Production in Florida 12
The Climate in North Central Florida 13
Flatwoods and Their Potential for Grassland
Productivity 13
The Need for Research on Limpograss 15
Limpograss Literature to Present: An Overview 16
Origin, Distribution, and Description 16
Characterization 19
Pest Screening 23
Small Plot Trials and Forage Response to
Grazing 25
Animal Response and Systems Management 29
Nonstructural Carbohydrates in Forages 30
The Characterization of Carbohydrate Reserves 32
A Progression of Nonstructural Carbohydrate Methods
Used in Agriculture 33
Organic Substances Used in Regrowth 38
Location, Seasonal, and Daily Fluctuation of Reserve
Foods 39
Management Factors Affecting TNC: An Integrated
Approach 41
IV

Page
CHAPTERS
1 REGROWTH IN DARKNESS AS INFLUENCED BY PREVIOUS
CUTTING TREATMENT OF FOUR LIMPOGRASS GENOTYPES 45
Introduction 45
Materials and Methods 46
Experiment 1 47
Experiment 2 48
Residual Effects 48
Results and Discussion 48
Experiment 1 48
Experiment 2 52
Residual Effects 54
Conclusions 54
Summary 55
2 CUTTING FREQUENCY EFFECTS ON LIMPOGRASS MORPHOLOGY
AND TOTAL NONSTRUCTURAL CARBOHYDRATE RESERVES 57
Introduction 57
Materials and Methods 58
Analysis of TNC in Plant Parts 59
TNC as Related to Clipping, Season, and
Genotype 59
Correlation of TNC versus Regrowth-in-Darkness ... 60
Results and Discussion 61
Analysis of TNC in Plant Parts 61
TNC as Related to Clipping, Season, and
Genotype 63
Correlation of TNC versus Regrowth-in-Darkness ... 65
Conclusions 68
Summary 68
3 DRY MATTER YIELD, CRUDE PROTEIN, IN VITRO ORGANIC
MATTER DIGESTIBILITY, TOTAL NONSTRUCTURAL CARBO¬
HYDRATE, AND PERSISTENCE IN TWO PROMISING AMD TWO
RELEASED LIMPOGRASSES: EFFECTS DUE TO NITROGEN
FERTILIZATION AMD CUTTING FREQUENCY 71
Introduction 71
Materials and Methods 75
v

Page
1979 Establishment 75
1979 Experimentation 76
1980 Experimentation 77
1981 Experimentation 82
Computing 82
Results 83
Dry Matter Yields: 1979 Experimentation 83
Dry Matter Yields: 1980 Experimentation 83
Dry Matter Yields: 1981 Experimentation 92
Crude Protein: 1979 Experiment 94
Crude Protein: 1980 Experiment 94
IVOMD: 1979 Experiment 97
IVOMD: 1980 Experiment 97
Total Nonstructural Carbohydrate as Related to
IVOMD 100
Seasonal TNC Trends . 107
Total Nonstructural Carbohydrates and
Persistence 109
Discussion 116
Conclusions 122
Summary 123
4 STATISTICAL ANALYSIS SYSTEM (SAS) METHODOLOGY FOR
CONSTRUCTING RESPONSE SURFACE GRAPHICS 126
Introduction 126
Materials and Methods 128
Establishment Year: 1979 129
Experimental Year: 1980 129
Results and Discussion 132
Conclusions 146
Summary 147
SUMMARY AND CONCLUSIONS 148
APPENDICES
A TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) PROCEDURE 154
B TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) RESULTS 174
REFERENCES 177
BIOGRAPHICAL SKETCH 193
VI

LIST OF TABLES
Table Page
1 Average DM yields of etiolated "shoots'1 of four
limpograsses following 3 weeks of growth in
darkness 49
2 Comparison of etiolated "shoot" DM yields and
associated "root" and "stubble" components for
four limpograsses averaged for all cutting treat¬
ments following 3 weeks of growth in darkness 51
3 Average DM yields of etiolated regrowth from
Experiment 2 following 3 weeks of growth in
darkness 53
4 The main effects of limpograss genotype and clip¬
ping treatment on the percent total nonstructural
carbohydrates (TNC) in the bottom 2 cm of stem
base ("crown") 64
5 Total dry matter (DM) yields for four limpo¬
grasses clipped at three different frequencies
for 10 weeks ending on 4 October 1979 84
6 Effect of nitrogen (N) rates and cutting frequency
(F) on dry matter (DM) yield, in vitro organic
matter digestibility (IVOMD), crude protein (CP),
harvest of protein, and fertilizer N efficiency 86
7 Percent crude protein (CP) for four limpograsses
clipped at three different frequencies for 10
weeks ending on 4 October 1979 (Samples taken on
4 October were analyzed for CP.) 95
8 The main effect of limpograss and clipping treat¬
ment on percent in vitro organic matter digesti¬
bility (IVOMD) of tissue sampled 4 October 1979 98
9 Balance sheet of in vitro organic matter digesti¬
bility (IVOMD) and total nonstructural carbo¬
hydrates (TNC) in two limpograsses 104
vi i

Table Page
10 Treatment combinations in PI 364888 representing
the best compromise between yield and quality in
the 1980 experiment 120
11 A SAS program using a previously fitted regression
to obtain values for the dependent variable to
fill the holes in the treatment matrix in order to
merge with actual data and use PROC G3D 133
12 A SAS analysis of variance procedure for a model
that quantifies the maximum treatment (TRT) sum of
squares (SS) for total nonstructural carbohydrate
(TNC) in harvest 9 of the 1980 experiment 137
13 A cubic SAS general linear model that fractionates
the treatment sum of squares (SS) into SS explain¬
able by nitrogen (N) fertilization and frequency
(F) of defoliation for total nonstructural carbo¬
hydrate (TNC) in harvest 9 of the 1980 experiment .... 138
14 An F-test of models for determining a statistically
significant response lack of fit (L0F) and an
elimination method for nonsignificant terms 140
15 The SAS general linear models procedure solution
for estimates of intercept adjusted for the limpo-
grass (line) and rep parameters 142
16 The SAS general linear models procedure solutions
for estimates of nitrogen (N) and cutting frequency
(F) treatment parameters adjusted for each limpo-
grass (line) 143
A-l SAS job as submitted on cards to the computer 157
A-2 Statistical analysis system 158
A-3 Percent total nonstructural carbohydrate (TNC)
means for all treatment, dates, and limpograsses
in 1980 174
v i i i

LIST OF FIGURES
Figure Page
1 Total nonstructural carbohydrate (TNC) percent of
dry matter of four plant parts in four limpograsses
sampled at two maturities 62
2 Total nonstructural carbohydrate (TNC) percentages
in the "crown" correlated against "stubble" weights
(left) and etiolated regrowth percent of whole
plants (right) for four limpograss genotypes sub¬
jected to three clipping frequencies 66
3 Spatial arrangement of a three level factorial
design in two variables 78
4 Spatial arrangement of a central composite design
in three variables 78
5 Treatment matrix showing treatment numbers inside
each actual (circles) and absent (squares) treat¬
ment combination 79
6 Treatment combinations of nitrogen (N) and cutting
frequency plotted as grid coordinates with the
origin at 120 kg/ha/yr N and 9 weeks cutting
frequency 79
7 Yield, quality (IVOMD and CP), and total nonstruc¬
tural carbohydrate (TNC) sampling schedule for
1980 (open symbols indicate no sample) 81
8 Total dry matter (DM) yield for four limpograsses
subjected to five levels of nitrogen (N) fertili¬
zation and five frequencies (F) of defoliation
throughout the 1980 growing season at the Beef
Research Unit near Gainesville, Florida 85
9 Spring and autumn seasonal distribution of dry
matter (DM) yield for four limpograsses in 1980 90
10Dry matter (DM) yields for PI 364888 at three
levels of nitrogen (N) fertility and staged at
three different dates in the autumn of 1980 91
ix
-

Figure
Page
11 Residual clipping and fertilization effects on
dry matter (DM) yield for PI 364888 harvested
every 9 weeks in 1981 (3 June, 3 August, and
7 October (not shown)) and the total production
for the year 93
12 Comparison of crude protein (CP) percentages in
the tissue of PI 364888 and 'Redalta' for two
dates in 1980 in response to five levels of
nitrogen (N) fertilization and five frequencies
(F) of clipping 96
13 Comparison of in vitro organic matter digesti¬
bility (IVOMD) for four limpograsses at two
1980 harvest dates subjected to five levels of
nitrogen (N) fertilization and five frequencies
(F) of clipping 99
14 Percent total nonstructural carbohydrate (TNC)
in the "shoot" tissue of three limpograsses on
27 July 1980 subjected to five levels of
nitrogen (N) fertilization and five defoliation
frequencies (F) 101
15 A comparison of total nonstructural carbohydrate
(TNC) in the "shoots" of three limpograsses
subjected to five levels of nitrogen (N) fertili¬
zation and five frequencies (F) of defoliation
sampled on 30 November 1980 102
16 Percent total nonstructural carbohydrate (TNC) in
the stem base ("crowns") of two limpograsses on
two dates as affected by nitrogen (N) fertiliza¬
tion and defoliation frequency (F) 110
17 A comparison of total nonstructural carbohydrate
(TNC) percent in the "crowns" of two limpograsses
subjected to 3 week defoliation frequencies (F)
and fertilized at three levels of nitrogen (N)
during the 1980 growing season 112
18 Visual estimations of percent 1Bigalta' and
PI 364888 for treatments having 3 week defolia¬
tion frequencies (F) and three levels of nitrogen
(N) in 1980 growing season 113
19 Rainfall and temperature data for the 1980 grow¬
ing season taken at the Beef Research Unit near
Gainesville, Florida 115
x

Figure Page
20 Visual estimates of four limpograsses subjected
to five levels of nitrogen (N) fertilization
and five frequencies (F) of clipping across
three summer harvests 117
21 An example of the response surface plotted
using SAS/GRAPH computer assistance 136
A-l Dilution one 161
A-2 Dilution two 161
A-3 The regression of optical density (0D) versus
glucose 168
xi

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
ENERGY RESERVES AND AGRONOMIC CHARACTERISTICS OF
FOUR LIMPOGRASSES (Hemarthria altissima (Poir)
Stapf et C.E. Hubb) FOR FLORIDA'S FLATWOODS
By
Scott Christiansen
December, 1982
Chairman: 0. Charles Ruelke
Major Department: Agronomy
Higher quality forages are needed for the four million hectares of
flatwoods that are the base of the forage-livestock industry in Florida.
The improved limpograss (Hemarthria altissima (Poir) Stapf et C.E. Hubb)
cultivars 'Bigalta' and 'Redalta' were compared with PI 364888 and
PI 349753 to find better yield, quality, and persistence, as well as to
investigate morphological differences and total nonstructural carbo¬
hydrate (TNC) physiology as they relate to persistence.
Preliminary work in 1979 used regrowth-in-darkness studies that
quantified the morphological differences in the four 1impograsses.
Larger stubble systems were positively correlated to TNC percentages in
the stem bases. Frequent defoliation treatments in the field prior to
the darkness study caused a shift in habit from upright to prostrate
growth that permitted increases in axillary tiller formation with con¬
comitant increases in TNC accumulation as measured by chemical means.
XT 1

A field experiment was then initiated with nitrogen (N) fertiliza¬
tion added as a new variable at five levels which was combined with five
defoliation frequencies using a response surface design. Dry matter
(DM) yields, in vitro organic matter digestibility (IVOMD), crude
protein (CP), and persistence were measured.
The best yield of all the genotypes was produced by PI 364888
(29 m ton/ha/yr) with two harvests on 27 July and 30 November, represent¬
ing 18 week cutting frequencies. Nitrogen was applied at 240 kg/ha per
harvest interval and yields were 18 and 11 m ton/ha for the first and
second harvests, respectively.
Percentages of IVOMD and CP were very low, especially for CP, in
the mid-summer period, but the quality increased substantially in the
autumn when limpograss growth rates declined. Bigalta had superior
quality among genotypes throughout the year.
Stored TNC decreased from a high in March to a low in July. The
period of lowest percent TNC in stem bases corresponded to the duration
of greatest limpograss losses in plots that were greatly stressed by
frequent clipping and high N fertilization.
Limpograss PI 364888 represents an improvement in yield and per¬
sistence over Bigalta, an improvement in yield and quality over Redalta,
and is scheduled for cultivar release in 1983.
XT 1 1

INTRODUCTION
Limpograss (Hemarthria altissima (Poir) Stapf et C. E. Hubb) is a
stoloniferous grass that has adaptability as an improved forage for
flatwoods in Florida and the southeast. Of the cultivars now released,
'Bigalta' has high digestibility and low persistence while 1 Reda 1ta'
has high persistence and low digestibility. The goal of this project
was to evaluate two new limpograss genotypes, PI 364888 and PI 349753,
based on their characteristics of yield, digestibility, and persistence,
as well as to investigate the nonstructural carbohydrate physiology
that releates to persistence.
A long term objective in Florida is to replace the low quality
forages that commonly grow on flatwoods with higher quality forages
possibly including limpograss. The flatwoods are the most important
ecological site for the forage-livestock industry in the state.
Knowledge of the total nonstructural carbohydrate (TNC) content of
a plant is necessary to understand how a plant grows. Studies of TNC,
morphology, and observations of plant reactions to management can yield
practical recommendations to be passed to agricultural producers.
This study was undertaken to thoroughly explore the flexibility of
limpograss under a wide range of management practices to clearly
identify its strengths and weaknesses as a forage plant for the
flatwoods.
1

LITERATURE REVIEW
History, Potential, Limitations, and
Development of Florida's Forage-Livestock Industry
In 1513 Ponce de Leon set out to find the land with the magic
waters of Indian lore and discovered the mainland near St. Augustine.
He named the place Florida.
One of the early victims of the reputed wealth of Florida was
Hernando de Soto who had whetted his appetite for gold while accompany¬
ing Pizarro to Peru. He outfitted an expedition and landed in Florida
in 1539. The next four years he spent marching through the southern
states seeking gold, until in 1541 he came to a vast river, now known as
the Mississippi. As the party returned from the present states of
Arkansas and Louisiana, de Soto died. Although he dissipated his
fortune in a vain quest for riches, he accomplished a far worthier
result in opening a vast territory for those who followed (Collins,
1946).
The Indian inhabitants of Florida were hunters. Of the four major
groups, the Calusa occupied the lower west coast, the Tesesta inhabited
the lower east coast, the Timucua were located in the north, and the
Apalache--west of the Aucilla River. Fire was used to chase out game,
and hence, these people were responsible for the creation of large
expanses of grasslands (Hunter et al., 1979).
2

3
The Spaniards soon became pragmatic and recognized Florida's
natural resources after the quest for gold fizzled. Large expanses of
the landscape were grazed after Ponce de Leon landed cattle on the
western coast of Florida in 1521.
The Spanish period lasted until the early 1700's, and during that
time cattle herds utilized the natural prairies such as Paynes Prairie
in Alachua County. The Spaniards left the landscape mostly undisturbed.
Much of their effort went into establishing the port city of
St. Augustine in 1565 which is the oldest permanent European settlement
in the United States.
It was under British rule that St. Augustine became an important
port for ship timbers. It has been said that almost all merchantable
live oak within a few hauling miles of a navigable stream was cut by
1823. This harvesting of wood increased the potential area for grazing.
Furthermore, in those days, 70 percent of the woodlands burned annually
so the sight and smell of wood's fire smoke was as much a part of the
Florida scene as pine trees and thunderstorms (Hunter et al., 1979).
Development of Grazing Land in Florida
The production practices in Florida for growing range animals
remained essentially the same for over 200 years until the automobile
replaced the horse and wagon. A problem arose with cattle confronting
cars on the Florida highways, resulting in ever-increasing injury and
loss. In 1948 the state legislature passed the "no fence law" that
required livestock owners to fence in their animals.
This was an important departure from the roaming range practices of
the past. Florida has serious mineral deficiencies, and fencing helped

4
ranchers to form a mental image of the mineral problems on their land.
The cattlemen were aware of the nutrient rich areas on their open range;
the areas were referred to as "hospital farms" where a nutrient defi¬
cient animal could graze and convalesce before being returned to the
herd (Becker, personal communication).
In 1930, before mineral supplements were generally used, a study of
7,100 cattle on 44,516 hectares of range was conducted in central
Florida. On flatwoods the calf crop averaged 34 percent, and 68 percent
of the newborn calves died prior to 30 months of age. The market value
was $10.33 per breeding cow per year or $2.54 per hectare (Becker,
personal communication; Henderson, 1956).
Improved forages and better management have helped to alleviate
many nutritional deficiencies. Supplementation with vitamins, minerals,
protein, and energy has been an important aspect of management.
McDowell et al. (1980) surveyed mineral status of beef herds on four
soil types in Florida and concluded that mineral deficiencies were area
specific. They indicated that phosphorus, selenium, and zinc deficien¬
cies were present in all four of the regions studied. Protein, vitamin
A, potassium, sodium, copper, and cobalt deficiencies were found in
certain regions and were related to seasonality. Magnesium deficiency
was most often associated with cows grazing winter pastures.
In general, the most satisfactory way of providing minerals to
grazing animals is through the use of one complete mineral mixture
offered free choice. The mix should contain a minimum of 6-8 percent
phosphorus, and ideally, the calcium level should not exceed twice the
level of phosphorus. Iron, zinc, manganese, copper, and cobalt should

5
provide 50-100 percent of the daily requirements in 2 ounces of the
mixture (Ammerman, 1979).
History and Evolution of Florida's Grazing Research
The first forage crop specialist at the University of Florida began
work in 1917, and the Agronomy Department was established in 1921.
Since then research on improved pastures has expanded tremendously. The
grazing experimentation was founded upon the observations made by
centuries of cattlemen who noted that lack of sufficiently high quality
forage during the winter months limited the carrying capacity of the
range, retarded the development of immature animals, and affected the
performance of mature animals. Average weight changes of mature cows
on flatwoods rangeland from June 1933 to March 1938 were as follows:
March to June, 34 kg gain; June to September, 10 kg gain; September to
December, 11 kg loss; and December to March, 38 kg loss (Henderson,
1956).
The first grazing trials in Florida were conducted in 1929 for the
purpose of evaluating four perennial summer grasses for use on well
drained sands. The grasses used and the annual beef gains in kg/ha were
as follows: carpetgrass (Axonopus affinis Chase), 195; common bermuda-
grass (Cynodon dactyl on (L.) Pers.), 202; common bahiagrass (Paspalurn
notatum Flügge), 216; and centipedegrass (Eremochloa ophiuroides (Munro)
Hack.), 248. The popularity of carpetgrass among cattlemen in the flat-
woods area and discovery in 1937 of the requirements for successful pro¬
duction of clovers led to grazing trials during 1942-1945 comparing
carpetgrass alone, unfertilized; carpetgrass alone, fertilized;
carpetgrass-lespedeza (Lespedeza striata (Thunb.) H. & A.); and

6
carpetgrass-white clover (Trifolium repens L.). Average annual gains in
kg/ha were as follows: carpetgrass, 84; fertilized carpetgrass, 168;
carpetgrass-lespedeza, 246; and carpetgrass-white clover, 695,
Meanwhile, superior grasses became available through plant introduction
and breeding. In grazing trials conducted between 1943-1947, three of
the new grasses--'Pangóla1 digitgrass (Digitaria decumbens Stent.),
'Pensacola' bahiagrass, and 'Coastal' bermudagrass--ferti1ized with 227
kg of 6-6-6 annually in the spring, produced yearly beef gains averaging
46 percent higher than gains produced by carpetgrass in earlier grazing
trials (Henderson, 1956).
The importance of forages in Florida is obvious. From 1950 to 1973
the forage-livestock industry grew by 70 percent. One-third of the
total range and pasture land in Florida is now planted to improved
forages. In 1973, 270 million kg of beef were produced, totaling $223
million in agricultural income. By 1985 production is expected to reach
425 million kg of beef (Agriculture in an Urban Age (AGUA) Report, 1974).
Grazing Animal Improvement
Adams (1982) describes the rancher's attitude and general percep¬
tion of his world "back in the old days."
In the 40's we had a lot of trouble raising cattle in Florida.
Cattle were cheap and wild. Pastures were large and the small
boned cows could outrun some of the horses. Salt sickness was
prevalent and we knew little about curing animal deficiencies.
Tick fever had been eradicated but the screw worms were eating
the cattle alive. Half the year cattle were bogged down try¬
ing to find a drink of water, and the other half they were
standing in water up to their sides. Our main interest was
survival; for the cattle and ourselves. There was no source
of breeding stock available that had the heat tolerance and
quality to meet the needs of a cowman. (Adams, 1982, p. 91)

7
In the past 30 years, the selection of animals has revolved around
adaptation to the Florida forages and climate. Breeders have sought to
combine the adaptability of Zebu, Brahman, or native animals with the
performance capabilities of European stock.
According to Crockett (1982), it became popular in the 1960‘s to
use sires representing breeds developed in Europe to increase beef pro¬
duction, and this led to diverse breeding combinations in the subtropi¬
cal zone of the Atlantic coastal plain and the Gulf coast states.
Crockett feels that breeding progress can be made in subtropical
environments using Brahman-derivative breeds such as the Beefmaster,
Braford, Brangus, or Santa Gertrudis. Fields and Hentges (1979) state
that the genetic base of Florida cattle is primarily of Brahman
extraction.
The characteristics needed in a commercial Florida herd are as
follows: (1) high fertility, (2) tolerance to the temperature and
humidity, (3) good foraging ability, (4) good maternal ability, (5)
satisfactory feedlot performance, and (6) satisfactory carcass quality
(Koger, 1982).
Florida's Role in the Forage-Livestock Industry
What must exasperate Florida producers is that there is an enormous
appetite for beef in the state filled by producers from outside the
state. Only 26 percent of the beef consumed annually by 10 million
Florida residents and their visitors is produced in Florida. The
remainder represents about 340 million kg of beef shipped in by other
means. Approximately 300 million kg of the beef introduced to Florida
is trucked from the high plains area of the United States (Baker, 1980).

8
A recent survey by Spreen and Shonkwiler (1982a) identifies Texas,
Kansas, Nebraska, and Iowa as the largest suppliers of finished beef to
Florida.
A most baffling feature of the Florida beef market is that cattle¬
men are selling calves to stockers 2250 to 3550 kilometers distant who
then sell to feeders who sell to packinghouses who in turn spend close
to $28 million annually to truck finished beef to Florida (Baker, 1980).
The question arises, "Why does Florida have a cow-calf industry,
and what prevents producers from breaking into the stocker and finishing
phases?" This is not an easily answered question, but one fact is that
Florida does not produce the quantity and quality of feed required to
keep calves in Florida until they reach the consumer. The nagging
thought, however, is that people simply do not persist in doing things
in a way that is less profitable than some logical alternative.
There are inherent cost advantages in keeping a geographically
dispersed cattle production system. Baker (1980) reasoned that by
virtue of escalating transportation expenses an economic incentive would
slow the exodus of calves out of Florida. Stegelin and Simpson (1980),
Spreen and Shonkwiler (1982b), and Ikerd (1981) all gave an economist's
explanation to the contrary. In short, the index of transportation
services in mid 1981 was about 300 percent of the 1974 level. But,
during this same period, the total marketing cost index has increased by
about 290 percent since 1968 and has increased 215 percent since 1974.
Therefore, costs of transportation services have increased only 3-5
percent more than other marketing costs over the past 10-15 years in
spite of rapidly rising fuel costs (Ikerd, 1981).

9
What are these locational advantages of geographical dispersion for
different phases of beef production? The cow-calf phase is land exten¬
sive, utilizing the less productive land suited to pastures but little
else, as in Florida. Stocker operations are located on areas with high
quality pastures such as the wheat pasture areas of Oklahoma, Texas, and
Kansas. Major cattle feeding areas are located near the major feed pro¬
ducing areas, and slaughter plants for cattle are located in the areas
where cattle are fed (Ikerd, 1981). Ikerd's cost analysis gives a $32
per head advantage for shipping cattle to the plains states and trucking
the meat back for consumption.
The economies of scale have allowed more cost efficiency for larger
feedlots and packinghouses making establishment difficult for indepen¬
dent, small competitors. Indeed, it is more likely in the next 20 years
for a continued trend toward fewer and larger specialized operations in
all phases of the system. The areas where cow-calf and stocker opera¬
tions are dominant are likely to remain dominant because of unique land
and pasture requirements that preclude significant changes in the struc¬
ture of these industries. The Florida cattle inventory reflects this
trend. Between 1955 and 1980 there was a 62 percent increase in Florida
cattle numbers compared to a 25 percent increase for the entire United
States; however, in 1960 less than 20 percent of calves were outshipped,
whereas in 1980 over 80 percent were outshipped (Spreen and Shonkwiler,
1982b).
Beef cattle production is capital intensive. An estimate of costs
for land and livestock investment is currently $5,000 per cow in the
United States (Ikerd, 1981). When tight money supply and rising inter¬
est rates squeeze marketing margins, packers and retailers buy less

10
cattle and carcasses. Demand is low, supply is high, and cattle prices
decrease. The cattle feeders make up their losses by offering less
money for stocker animals. The stocker operators, in turn, pass on the
higher interest costs to the cow-calf man. There will be little expan¬
sion in the cattle business with high interest rates and low priced
calves (Ikerd, 1981).
The Florida Opportunity for Finished Beef
The southeastern United States has the largest surplus of stocker-
feeder cattle in the country; the seven states in this area provide
about 5.5 million more calves each year than are needed for herd
replacement or fed in southeastern feedlots (Baker, 1979).
If the calf crop is to be fed until slaughter in Florida, economics
will necessitate fast, continuous, efficient growth from weaning until
slaughter. The calves must be grown on forage (pasture and/or silage)
with or without grain supplement with a finishing phase in the feedlot
on a high energy ration.
Immediately, it is obvious to wonder, "Where will the grain come
from?" The entire southeast is a grain deficient area, and corn will
continue to cost $20/ton more in Florida than in major feeding states
(Baker, 1979).
The unpredictability of the Florida climate makes corn production a
risky business. Florton et al. (1981) and Horton and Mislevy (1981) grew
corn and sorghum silages in south Florida, and their results showed that
both corn and sorghum can be grown successfully and economically using
multiple cropping. The viability of an intensive beef cattle industry

11
in Florida will depend on the ability to produce these locally grown,
high energy feeds.
Provided the necessary feeding systems were developed, more cattle
would remain on Florida ranches. The question of whether increased
cattle numbers should precede the development of feedlot and packing
industries is "putting the cart in front of the horse." Efficient feed¬
ing systems must evolve first, the cattle build up will follow, and
count on the feedlot and packing people to fend for themselves. These
businesses simply want a consistent, dependable supply of feed and
animals (Kaplan, 1981).
As Erwin Bryan Jr. of the Central Packing Co. put it, "It is my
feeling that we already have the know-how to feed cattle in Florida.
The weakest point in the chain, and always has been, is grazing of the
feeder calves until they are big enough to go into the feedlot" (Bryan,
1981, p. 46).
In conclusion, it would appear that 20 years from now a totally
enclosed, self-reliant beef industry in Florida could be operational.
The most likely stimuli will be (1) the development of new forage
species and management systems that permit a shift from grain-fed to
forage-fed livestock operations; (2) increased irrigation costs on the
western high plains, decreasing grain availability; (3) solution of the
agronomic problems for year round feed availability in Florida; and (4)
lower interest rates to allow economic growth in the cattle business.
These factors could shift the geographical advantage to the southeastern
United States.

12
Barriers to Forage Production in Florida
There are three phases of the forage system-production, harvest¬
ing, and utilization. By identifying the present weaknesses in these
phases, combinations of native range and improved pastures and/or
supplemental pastures will be developed in order to provide a distribu¬
tion of forage throughout the year. An even distribution of forages,
more than any other factor, will ultimately be responsible for increased
beef production in Florida (Mott, 1982b).
The biggest limitation to well distributed forage production is
probably seasonality. Most forage production occurs from April to
September, and two-thirds of the yearly growth of perennials occurs
between July 1 and October 1 (Mott, 1982b).
Uneven distribution of forage production throughout the year makes
management difficult, and supplemental feeding on pasture is often
needed to efficiently utilize the forage produced. Sloan Baker summa¬
rizes Florida's current forage production problems. He states that cool
season forages--rye (Secale cereale L.), ryegrass (Lolium multiflorum
Lam.), and clovers (Trifolium spp.)--are excellent, yielding good gains
of 0.5 to 0.9 kg/day for 90 to 160 days. Dry matter yields, however,
are sometimes low due to weather. Warm season forages are more produc¬
tive but give half the daily gains. Dry weather often limits production
in spring and autumn; while in summer, rainfall and humidity are high
enough to make hay curing difficult. Results with grass silage have
been disappointing due to insufficient quality for good gains with young
cattle (Baker, 1979).

13
The Climate in North Central Florida
The climate ranges from subtropical-oceanic in south Florida to a
typical low-altitude, continental frontal pattern in the north (Hunter
et al., 1979). The average year may be divided into two seasons: a
warm, rainy season receiving about 60 percent of the annual rain and a
cooler, dry season. The warm, rainy season runs from about the middle
of June to the end of September. The cooler, dry season dominates the
remainder of the year.
The summer rain occurs as afternoon thunderstorms, generated by
strong surface heating and fed by a double sea breeze convergence.
During the winter months, the differential cooling of land and sea, the
occasional presence of stagnated high pressure cells, and the formation
of low level inversions caused by nocturnal cooling act to maintain a
high degree of atmospheric stability, suppressing frontal activity. A
decrease in frequency of frontal movement across northern Florida is one
cause of periodic drought on the average of once every 7 years
(Dohrenwend, 1978).
The Bermuda high is common throughout the year, centered in the
Caribbean with its strongest effect during winter. If it were not for
large bodies of warm water on either side of the peninsula, Florida
would be as arid as the great subtropical deserts at the same latitudes
(Dohrenwend, 1978).
Flatwoods and Their Potential for
Grassland Productivity
There are 14,236,908 hectares of land area in Florida. At present,
there are approximately 1.21 million hectares of improved pasture, 1.62

14
million hectares of native range, and 2.02 million hectares of forest¬
land that provide some grazing.
Seventeen major natural vegetation types are recognized in Florida.
The most important ecotype from the standpoint of hectarage and poten¬
tial for animal production from forages is the flatwood site. Flatwoods
occupy 4.05 million hectares or almost 30 percent of Florida's land
area.
Most of the state's timber production also occurs on flatwoods.
Open woodland consists predominantly of one to three species of pines:
longleaf (Pinus palustris Mill.), slash (P_. elliottii Engelm.), and pond
(P_. serótina Michx.). Understory produces many grasses such as chalky
bluestenr (Andropogon capillipes Nash), broomsedge bluestem (Andropogon
virginicus L.), paspa!urns (Paspalum spp.), wiregrass (Aristida stricta
Michx.), indiangrass (Sorghastrum secundum (Ell.) Nash), and panicums
(Panicum spp.). Associated forbs include grassleaf goldaster
(Chrysopsis graminifolia (Michx.) Ell.), partridge pea (Cassia
fasciculata Michx.), and beggerweed (Desmodium sp.). Shrubs are pre¬
dominantly saw palmetto (Serinoa repens (Bartram) Small), wax myrtle
(Myrica cerífera L.), blackberry (Rubus sp.), and gall berry (Ilex glabra
(L.) Gray).
Amidst the flatwood areas are small hardwood forests, many cypress
ponds, prairies, marshes, and bay tree swamps. The wet areas support
very desirable forages such as maidencane (Panicum hemitomon Schult.)
and little blue maidencane (Amphicarpum muhlenbergianum (Schult.)
Hitchc.). The flatwoods are inhabited by deer and hogs, quail, gray
squirrels, and turkey; hence, this is also the most important community
type in Florida for hunting.

1 5
The soils are acid (4.0 - 5.5 pH), of the Spodosol Order, with a
4 cm surface horizon colored grey to grey-black by the presence of
organic matter. The 10 to 45 cm stratum is dominated by white leached
sands and from 45 to 55 cm a hard pan is found--fine particles cemented
together by sesquioxides and other compounds. From 55 to 90 cm depth a
brown, tannin-stained sand exists followed by white sand with further
increases in the profile.
Due to the hardpan formed on Spodosols in Florida, a perched water
table is created that is beneficial in reducing the rapid rate of
percolation through the sands but is a problem when rainfall is so great
as to cause standing water above the soil surface.
The acid soil condition of the flatwoods makes it unsuitable for
crop production; however, acid and wet tolerant species like limpograss
(Hemarthria altissima (Poir.) Stapf et C.E. Hubb) are well adapted to
these lands.
The Need for Research on Limpograss
The days of quantum leaps in beef gains, such as those achieved on
improved forages over native range, are over. An excellent inventory of
improved forages is now available, and possibly the old problem of
seasonal production can be partially solved by incorporating limpograss
into the yearly forage system. Limpograss is a stoloniferous, C^,
perennial, summer-growing grass that has good cool-season production.
Limpograss is adapted to the flatwoods habitat, can produce an abundance
of biomass, and can be stockpiled, ensiled, or made into hay.
Research evaluating the first four limpograss introductions from
1964 culminated in the release of cultivars 'Redalta', 'Greenalta', and

16
'Bigalta' in 1978. Since that time many limpograsses were collected and
evaluated in hopes of further exploiting the germplasm. Beneficial
characteristics of quality, yield, persistence, morphology, winter¬
hardiness, and adaptability to flatwoods have given researchers evidence
of the potential of limpograss as a forage-producing grass for the
future.
Limpograss Literature to Present:
An Overview
The literature on limpograss species to date will be grouped into
five phases: (1) origin, distribution, and description; (2) characteri¬
zation; (3) pest screening; (4) small plot trials and forage response to
grazing; and (5) animal response and systems management. Limpograss
research has just recently reached phase five.
Origin, Distribution, and Description
There are probably no more than 12 described species of limpograss
(Kretschmer and Synder, 1979), and the one of most agrnonomic importance
is Hemarthria altissima (Poir.) Stapf et C.E. Hubb. Agrostologists
guess Hemarthrias1 origin to be in tropical Africa, but it is also found
in Madras, Burma, Malaysia, Malay, Siam, Turkey, Nigeria, Italy,
Ethiopia, Tanzania, Ceylon, Northern India, Southeast Asia, and
Argentina (Bor, 1960; Bogdan, 1977; Chippindall, 1955). The species are
desirable, robust perennial fodder plants found in wet habitats such as
river banks, seasonally flooded river valleys, seasonal swamps, and so
forth.

17
Due to plant exploration and redistribution, Hemarthria introduc¬
tions are now found in Australia, Brazil, Bolivia, Columbia, Ecuador,
Hawaii, Malawi, Mexico, New Zealand, Paraguay, Uruquay, Venezuela, and
the Virgin Islands (Quesenberry et al., 1978; Quesenberry et al. , 1981;
and Oakes, 1980).
Overwintering of H_. altissima in the United States occurs in
Alabama, Mississippi, Texas, Tennessee, and as far north as Beltsville,
Maryland (Oakes, 1973; Oakes and Foy, 1980). Later, Oakes (1980)
reported the survival of one limpograss accession from the 1964 collec¬
tion (PI 299039 from Rhodesia) for three winters at Pullman, Washington.
This diversity of winterhardiness may be used to extend the production
range and grazing period for limpograss in the United States.
Some common names for grasses of the genus Hemarthria are limpo¬
grass, teagrass (Florida), Capim gamalote (Brazil), Pasto clavel,
Gramilla canita (Argentina), Baksha, Panisharu (India), Swamp Couch, and
Rooikweek (South Africa) (Bogdan, 1977; Chippendall, 1955).
A description of the genus follows: Member of the
Andropogoneae tribe. Perennials with short rhizomes and long
spreading, decumbent, branched culms that root at the nodes;
the upper part of the stems erect or suberect reaching 150 cm
height but usually 30-80 cm. Leaves up to 20 cm long and 6
mm wide with membraneous ligules. The spike-like racemes are
compressed, 6-10 cm long with spikelets appearing opposite--
each pair composed of a bisexual sessile spikelet, 5-6 mm
long, and a smaller, pedicelled male spikelet. Glumes nearly
equal. First glume flat, 2-keeled, leathery; second glume
coriaceous, fused to the hollowed-out face of the rachis
internode (Bodgan, 1977; Hall, 1978).
Quesenberry et al. (1982) surveyed the USDA collection of
Hemarthria (containing 76 species) and found 72 percent diploid, 27 per¬
cent tetraploid, and 1 percent hexaploid. Thirteen of 16 tetraploids

18
were originally found north of 24 S latitude in Africa. Diploids were
more winterhardy than tetraploids.
The 1964, 1971, and 1976 USDA plant exploration missions to Africa
provided the germplasm for vigorous limpograss experimentation in the
United States for the past 15 years. Jack Oakes did most of the
collecting. Three cultivars were cooperatively released on 19 April
1978 from the Institute of Food and Agricultural Sciences, University of
Florida, and the Soil Conservation Service, USDA. The limpograsses are
Redalta, Greenalta, and Bigalta, previously Pi's 299993, 299994, and
299995, respectively (Quesenberry et al., 1979).
Redalta and Bigalta are two of the four limpograsses used in this
dissertation. Oakes obtained these genotypes from the Rietondale
Research Station near Pretoria, South Africa, in 1964 (Oakes, 1973).
Redalta originated near the Pienaars River in west central Transvaal,
and Bigalta is from an unspecified location near the Transvaal. Redalta
is a diploid (2n = 18), and Bigalta is a tetraploid (2n = 36). Also
included in this dissertation are PI 364888, collected in 1971 from a
small island in the Luvuvhu River several kilometers above its conflu¬
ence with the Limpopo River in Kruger National Park, and PI 349753, from
Mt. Mbya, Kenya. These two limpograsses are tetraploids (Oakes, 1973).
There are many morphological differences among limpograsses.
Leafiness, internode length, anthocyanin content, bunchiness, tillers
per unit area, and stem thickness vary. These inherent differences aid
in maintaining genotype identification.
Cytological investigations (Wilms et al., 1970) characterized the
four limpograsses collected from Africa in 1964, and reported color dif¬
ferences for the anthers of Redalta (purple), Greenalta (brown), and

19
Bigalta (yellow). Schank (1972) characterized chromosome numbers,
pollen stainability, and seed set for 11 more Hemarthrias in preparation
for an intrageneric breeding program. Breeding efforts to date have not
produced any limpograsses with better agronomic attributes than the
vegetatively propagated plant introductions. The logistics of low seed
set (Schank, 1972) may partially discourage plant breeders in this
regard, as well as the difficulty in getting limpograss to flower in the
greenhouse (Quesenberry, personal communication).
Schank et al. (1973) discovered that tetraploid limpograsses had
higher in vitro organic matter digestibility (IVOMD) than diploid
accessions. The mean decrease of IVOMD of the tetraploid, from 68.4
percent at 5 weeks, to 66 percent in mature plants, suggested a slower
decline of quality than for most tropical grasses. Cross-sections of
stems revealed significantly lower vascular bundle area in the tetra¬
ploid, as well as fewer sclerenchyma fiber cells.
Characterization
One of the tangential aspects of the incipient research on limpo¬
grass was that it was noted as having a tea-like odor and flavor.
Killinger (1971) after collaboration with the USDA Northern Utilization
Research Laboratory, filed for a beverage use patent on 25 January 1971.
Thus, the name "teagrass" came into usage first promulgated, in-so-far
as can be determined, by Eldridge D. Lee of the University of Florida
Agronomy Farm. Killinger and Beckham obtained United States Patent
3,709,694 for Hemarthria beverage rights in 1973.
Oakes and Foy (1980) recommended limpograss for revegetating mine
spoils due to an excellent tolerance to aluminum toxicity. Oakes (1973),

20
Oakes and Foy (1980), and Oakes (1980) also reported a wide diversity of
cold tolerance, as mentioned earlier.
A recent finding was an alleopathic effect of some limpograsses
(Ruelke and Quesenberry, 1981; Young and Bartholomew, 1981; Tang and
Young, 1982). The alleopathic compound was suspected by Ruelke and
Quesenberry (1981) in their limpograss mixtures with red clover and the
action was attributed to the clover. The red clover isoflavinoids
(Tamura et al., 1967; Chang et al., 1969) and the phenolic compounds
from limpograss roots (Young and Bartholomew, 1981; Tang and Young,
1982) may have both been active. Bigalta was more growth depressive on
'Greenleaf1desmodium (Desmodium intortum (Mill.) Urb.) than was Greenalta
limpograss. Bigalta root exudates also depressed the growth of
Greenalta.
A series of four papers characterized 10 tropical forage grasses in
Puerto Rico. Limpograss was included in these studies. The main intent
was to survey: (1) fibrous carbohydrate fractions, (2) proximate
nutrient composition, (3) mineral composition, and (4) the decline of
in vitro true digestibility with advancing maturity of 10 tropical
forage grasses. The 10 grasses were guineagrass (Panicum maximum
Jacq.), Pangóla digitgrass, congograss (Brachiaria ruziziensis Germain &
Evrard.), African crabgrass (_D. swazilandensis Stent.), Venezuelan
elephantgrass (Pennisetum setosum (Swartz) L. Rich, in Pers.), giant
Pangóla digitgrass (D. valida Stent.), signalgrass (13. Brizantha
(Hochst. ex A. Rich.) Stapf), buffelgrass (Cenchrus ciliaris L.),
jaragua (Hyparrhenia rufa (Nees) Stapf), and limpograss. The identity
of the Hemarthria was not reported, hence, care must be exercised in

21
extrapolating these results to other limpograsses (acknowledging varia¬
tions among and within ploidy levels).
Paper No. 1 (Coward-Lord et al., 1974a) on carbohydrate fiber frac¬
tions discussed neutral-detergent fiber (NDF), acid-detergent fiber
(ADF), acid-detergent lignin (ADL), hemicellulose, cellulose, and
silica. The NDF fraction represents the total fiber fraction, its dif¬
ference from 100 being the neutral-detergent solubles (NDS), or soluble
nutrients. The ADF content is a measure of the 1igno-cel1ulose
fraction. The difference between NDF and ADF is an estimate of
hemicellulose. Acid-detergent lignin (by the permanganate method) is an
acid treatment of the ADF which leaves cellulose and silica. Ashing
leaves silica.
Results showed that limpograss was one of three grasses with the
highest levels of NDF, among the four lowest in ADF content, the highest
in ADL, the highest in hemicellulose, among the lowest in cellulose, and
the lowest in silica. In other words, among the tropical grasses
studied, limpograss had a high percentage of total fiber-comprised of
high quantities of lignin and hemicellulose relative to cellulose.
In Paper No. 2, Coward-Lord et al. (1974b) studied nutrient compo¬
sition including crude protein (CP), dry matter, crude fiber (CF), ether
extract, ash, and nitrogen-free extract (NFE). This methodology is
slowly losing favor over the Goering and Van Soest (1970) methodology
for fractionating feedstuffs for ruminant value. The definition of CF
as a chemically uniform, non-nutritive substance cannot be reconciled
because CF represents almost all the potentially digestible cellulose
and also includes some lignin and hemicellulose. The imperfect CF
methodology has allowed most of the liginin and hemicellulose to be

22
included in the NFE, which is supposed to represent available
carbohydrate. In some cases the CF can be more digestible than the NFE,
which is clearly incongruous with the aim of the fractionation.
Nevertheless, results showed that limpograss had the highest mean NFE
value of all 10 species at 53.3 percent over 180 days. The CP values
dropped to 5.7 percent by the 90-day stage.
The third Puerto Rican Paper (Arroyo-Aqui1u and Coward-Lord, 1974a)
covered mineral composition. The ranges as percent of dry matter over
180 days for all 10 species and the averages for limpograss (in paren¬
theses) were calcium, 0.11-0.43 (0.15); phosphorus, 0.08-0.39 (0.13);
magnesium, 0.15-0.46 (0.20); and potassium, 0.68-7.33 (1.85).
Limpograss had the lowest phosphorus percent of all grasses at 180 days
(0.08)--clearly deficient for ruminants; magnesium and potassium were
lowest at 30 days, 0.25 and 3.0 percent, respectively.
The final Puerto Rican paper (Arroyo-Aquilu and Coward-Lord, 1974b)
discussed quality decline among the 10 tropical grasses. The mean rate
of in vitro true digestibility decline was 24,1 units from 30-180 days.
The largest decline (12.3 units) occurred between 30-60 days as compared
to declines of 4.8, 3.9, 1.3, and 1.8 units between 30 day intervals
from 60-180 days. All the grasses reacted similarly. This suggests
that the tropical grasses studied may best be utilized between 30-60
days of growth.
Hodges and Martin (1975) included three limpograsses in a study of
23 perennial sub tropical grasses and reported that Cynodons and
Di gitari as were better cool season producers than the limpograsses at
Ona, Florida. Seasonal yield distribution of numerous tropical grasses
was studied by Taylor et al. (1976b) in New Zealand. In this study, a

23
tetraploid limpograss (identity not reported) yielded poorly during the
low rainfall summer of the year. The total warm-season yield for limpo¬
grass nearly doubled under trickle irrigation, and the moist, cool-
season yields were twice as great as the unirrigated warm-season yields.
The digestibility data for 22 grasses were analyzed in another
Taylor et al. (1976a) publication to compare nutritive quality of the
grasses grown at Kaitaia, New Zealand (35 S latitude), with the same
grasses grown in more equatorial environments. The results for grasses
grown at Kaitaia, which has a mean warm-season temperature of 18.4 C,
were compared to the digestibility data for the same species grown at
Lawes, Australia, with a mean temperature of 24.3 C. Slower rates of
maturation in the cooler environment frequently resulted in higher
tissue digestibilities.
The non-flowering limpograss used in the New Zealand studies pro¬
duced tissue with moderate protein (15.7 percent in leaves, 6.1 percent
in stems), moderate fermentable carbohydrate (7.2 percent in leaves, 6.0
percent in stems), and excellent digestibility (66.8 percent in leaves,
77.2 percent in stems). The stem digestibility was the highest of all
the grasses studied.
Pest Screening
A limited number of references are available on limpograss suscep¬
tibility to nematodes and aphids. Boyd and Perry (1969) screened
Redalta, Greenalta, and Bigalta resistance to sting nematodes
(Belonolaimus longicaudatus Rau). This nematode is among the most
persistent, serious pests in Florida's improved pasture--Pangola digit-
grass, Pensacola bahiagrass, and Coastal bermudagrass are all

24
susceptible. The limpograsses had moderately low nematode counts but
the authors attributed this to lack of roots rather than significant
nematode resistance. Chlorosis was observed in Bigalta, and this geno¬
type had higher counts than the two diploid limpograsses. In 1970, Boyd
and Perry reported additional information on sting nematode damage to
17 pasture grasses. Redalta, Greenalta, and Bigalta were the three most
favorable hosts.
Boyd et al. (1972) studied the interaction of soil temperature and
sting nematodes and found that Greenalta grew best between 20 and 38 C
in uninfested soils and best between 30 and 38 C in infested soils.
Above 38 C, nematodes were reduced but the soil was too hot for good
growth of the grass. Later, Quesenberry and Dunn (1977) received 54
more limpograsses which they screened for response to the sting nematode
in the greenhouse. No available limpograss lines approached immunity,
but a few introductions had greater tolerance. Of the 10 best lines, 50
percent were tetraploid or hexaploid, while all the least tolerant were
diploid. The most tolerant introductions were collected from the
islands of Mauritius in the Indian Ocean.
Pest problems in limpograss were reported by Oakes (1978) who
studied resistance in Hemarthria to the yellow sugar-cane aphid Sipha
flava (Forbes). Variable resistance was found for 54 PL altissima
accessions. Two of the introductions included in this dissertation were
evaluated in the Oakes study--PI 364888 was most susceptible and PI
349753 was moderately resistant, while all the limpograsses had better
resistance to the sugar-cane aphid than found in Digitari a.

25
Small Plot Trials and Forage Response to Grazing
The second and third collections of limpograss were initiated in
response to previous experimental results that recommended an extended
search within Hemarthrias for favorable pasture grass attributes.
Ruelke et al. (1976) evaluated 53 limpograsses in both greenhouse and
small plot trials and this study satisfied their exploratory curiosity
with respect to the identification of superior genotypes.
In 1976-77, Quesenberry and Ocumpaugh (1977) grew Redalta,
Greenalta, and Bigaita as conserved forages. They presented their
results at the American Society of Agronomy meetings at Los Angeles in
1977, and with more detail (Quesenberry and Ocumpaugh, 1980; Ocumpaugh
and Quesenberry, 1980) following a second year of data. Stockpiling is
an inexpensive way of filling the forage-deficient months of November-
February in northern Florida. Stockpiled yields were greatest for
Redalta in 1976-77 (over 10 m tons/ha). Yields were similar for the
three cultivars in 1977-78 and lower, averaging 6 m tons/ha (Quesenberry
and Ocumpaugh, 1980). Based on these results from the Green Acres Unit
near Gainesville, Florida, stockpiling should begin by the beginning of
August to allow 6-8 weeks of growth before frost.
The above study did not include data on animal acceptance; however,
the authors observed satisfactory consumption of mature Bigalta. Other
producers (Wendy J. Carpenter, personal communication) have reported
animal rejection of similarly aged Bigalta.
Quesenberry and Ocumpaugh (1982) presented data on the tissue
sampled from the stockpiling experiments. Potassium decreased from a
high of 2.5 percent to below the National Research Council (NRC)

26
recommended level for ruminants (0.65-0.80) by early November.
Phosphorus dropped below the NRC minimum for ruminants (0.16-0.24) by
mid-October and in the second year was never above the minimum level.
Magnesium was not considered a nutritional problem, while the mean
calcium percent for both years (0.28) was adequate for mature pregnant
beef cows (NRC = 0.16) but barely sufficient for lactating beef cows
(NRC = 0.27). The authors' recommendation was to supplement with
potassium, phosphorus, calcium, and protein after mid-October.
Quesenberry et al. (1978) coalesced the pertinent limpograss pro¬
duction data through 1978 for six sites in Florida. At Ft. Pierce,
Bigaita had higher production in November and December than Redalta or
Greenalta, but produced less in spring. At Gainesville, the late season
production of limpograss was slightly less than that of the digitgrasses
and one bermudagrass. Bigalta, however, had the highest total season
production. Frequent clipping defoliation at Ona produced weed inva¬
sions and the limpograsses had intermediate yields compared to other
tropical, perennial grasses. Results from Jay demonstrated slow estab¬
lishment rates compared to Coastal bermudagrass, 'Transvala' digitgrass,
and one bahiagrass. From work done at Quincy, it was concluded that
limpograsses are not adapted to the dry, upland soils of the Florida
panhandle. At Belle Glade, Coleman and Pate (Quesenberry et al., 1978)
found good digestibility and acceptability of Bigalta by beef animals;
however, St. Augustinegrass (Stenetaphrum secundatum Kuntz) was better
adapted to the muckland soils at this south Florida site. Bigalta did
not persist under heavy grazing. This led to the recommendation of
rotational grazing for Bigalta.

27
Rotational grazing is too intensive for most Florida producers. It
is fortunate that some of the new limpograss introductions will persist
under continual use because this is an important criterion in a ranchers
mind (Pate, personal communication).
Ruelke (1978) studied Redalta, Greenalta, and Bigalta and found
significant yield responses to nitrogen up to 330 kg/ha/yr; however,
severe losses occurred following frequent defoliation at high nitrogen
rates, especially for Bigalta.
Bigalta's high digestibility combined with poor persistence was
disconcerting. Quesenberry et al. (1981) found that PI 364888 was a
digestible, persistent tetraploid. This accession was soon included in
the limpograss studies with Bigalta and Redalta, while Greenalta was
omitted from experimentation due to its similarity to Redalta.
Ruelke et al. (1978) included the promising accession in a limpo¬
grass establishment study. Denser stands, earlier production, and
higher second year dry matter yields were obtained when planting
material was sprigged, followed by disking to partially cover the stems,
and cultipacked for firm contact between sprigs and soil. In the year
after establishment, PI 364888 outyielded both Redalta and Bigalta.
Kretschmer and Synder (1979) compared growth of Redalta, Greenalta,
and Bigalta to Transvala digitgrass, Pangóla digitgrass, and
1Coastcross-11 bermudagrass and found that a 2 week cutting interval
severely decreased the ability of limpograss to accumulate dry matter.
Frequent clipping also diminished the efficiency of nitrogen usage.
Delaying autumn fertilization at Ft. Pierce, Florida, from 17 September
or 1 October to 29 October resulted in greatly reduced forage production
when harvested on 17 December. The later nitrogen fertilization,

28
however, resulted in a better combination of yield and quality by rais¬
ing the inherently low protein content in mature limpograss. Bigalta
responded to cool-season nitrogen fertilization but gave way to weed
encroachment during the summer which led to the recommendation that it
should be rested sometime in the warm season to maintain plant
populations.
Quesenberry and Ocumpaugh (1979) studied clipping and grazing
defoliation methods for three years. The mob grazing method shortened
the time necessary to advance the new limpograss germplasm through the
early phases of agronomic evaluation. Quesenberry et al. (1981) summa¬
rized the events leading to the clip/graze experimentation. After
preliminary testing of 53 clones in greenhouses and small plot clipping
trials, 22 were selected for evaluation by clipping and 27 by grazing.
Eight of the best genotypes were then evaluated at four frequencies of
grazing (3, 5, 7, and 9 weeks). Ocumpaugh et al. (1981) identified
PI 364888 as superior to Bigalta under grazing due to comparable
digestibilities and higher persistence.
Meanwhile, Ruelke and Quesenberry (1982) obtained more data on
seasonal productivity for PI 364888. They found nitrogen fertilization
increased early spring growth; however, the responses they obtained were
limited due to effects of spring drought and cold temperatures. In the
autumn they studied deferred forage characteristics and suggested that
after 10 weeks of age forage quality would decline to a maintenance
level, and by 20 weeks maturity the forage would be rejected. This
statement may reflect a change in attitude with respect to stockpiling.
The differences in persistence among limpograss genotypes led
researchers to ask "why?" They reasoned that a knowledge of

29
nonstructural carbohydrate metabolism should contribute to an under¬
standing of limpograss behavior. Christiansen et al. (1981) studied
etiolated regrowth as an indicator of the stored energy in four
1impograsses. Some of the results indicated that morphological differ¬
ences were related to observations of agronomic performance. Frequent
cutting treatments caused a drain of the energy reserves of Bigalta but
not PI 364888. This suggested greater reserve energy storage for
PI 364888.
Animal Response and Systems Management
Hodges and Pitman (1981) studied Bigalta limpograss, 1 Cal lie*
bermudagrass, 'Sarasota1 stargrass (Cynodon nlemfuensis Vanderyst), and
'Ona' stargrass under year long grazing. During the cool season,
Bigalta produced average daily gains of 0.20 kg as compared to 0.23,
0.21, and 0.13 kg for Ona, Sarasota, and Callie, respectively. Warm
season average daily gains were similar for the four grasses and twice
as high as the gains produced in the cool season. The stargrasses had
much higher yearly production as reflected by animal grazing days per
hectare: Sarasota, 1089; Ona, 1072; Callie, 911; and Bigalta, 783.
Beef production over both seasons averaged 585, 558, 450, and 431 kg/ha
for Ona, Sarasota, Bigalta, and Callie, respectively.
Ocumpaugh (1982) is presently conducting the second year of a three
year animal production study comparing beef gains on Pensacola bahia-
grass and PI 364888 limpograss. Preliminary results from 1981 indicate
better average gain per yearling heifer on limpograss (78.6 vs. 57.7
kg/yr) and a 70 calendar day advantage in grazing days compared to

30
Pensacola bahiagrass. Ocumpaugh states that PI 364888 is being con¬
sidered for cultivar release in June 1983.
Nonstructural Carbohydrates in Forages
Plants are the primary source of carbohydrates. Cellulose is the
most prevalent organic compound on earth and is man's most important
industrial carbohydrate. The staple grains are predominantly starch
which is the chief carbohydrate in the human diet (Greenwood, 1970).
This review will be confined to a summary of the carbohydrates important
in forages. Structural carbohydrates are beta-linked molecules that are
degraded into more utilizable forms by rumen microflora. These simpler
substrates are then converted into energy and animal products useful to
man. The nonstructural plant carbohydrates can be completely utilized
in animal diets and also provide the energy necessary for bacterial
preservation of silage.
In plants, nonstructural carbohydrates are used in growth and
respiration and have been called "food reserves." The main objective of
this review will be to synopsize the evolution of non-structural carbo¬
hydrate research in pasture plants during the past 60 years.
Quantities of nonstructural carbohydrates in plants fluctuate
during each day as well as during the season. Variations in geography,
environment, taxonomy, anatomy, and management (or experimental condi¬
tions) all add to the dynamics of carbohydrate flux.
Reactions to management vary from year to year depending on
environment, species, and stress; hence, there are conflicts in the
literature due to incongruities between imminent and long term results,
between animal and pasture requirements, and between applied and

31
basic objectives. Three generations of researchers have often perpetu¬
ated a benign acceptance of conventional wisdoms.
The overwhelming quantity of carbohydrate studies created a need
for review articles to summarize the research findings. Many theories
were proposed to consolidate research on different species grown under
different conditions for different purposes. The most important eluci¬
dations were (1) the definition of what organic substances should be
considered reserves, (2) what methods were best to satisfactorily frac¬
tionate carbohydrate components, (3) what role carbohydrates play in
regrowth mechanisms, (4) where plant foods accumulated, (5) how non-
structural carbohydrates varied by day and season, and (6) how carbo¬
hydrates in storage organs were affected by management. The reader is
referred to the following publications as a chronological guide: Graber
(1931), Weinmann (1952, 1955, 1961), Troughton (1957), May and Davidson
(1958), Hunter et al. (1970), Sheard (1973), Smith (1973b), White
(1973), and Noble and Lowe (1974). For work prior to the 1930's, the
reader is referred to the literature citations of Graber et al. (1927)
and Graber (1931).
The important findings of Cugnac (1931) are alluded to by nearly
all carbohydrate reviewers. He separated the grasses into two groups--
the fructose accumulating grasses native to temperature climates and
grasses accumulating sucrose and starch that are mostly adapted to warm
regions.
Graber (1931) assessed the condition of "low" or "high" organic
reserves in pasture plants by means of dry matter yields, persistence,
and weed encroachment. He saw that plant growth behavior was related to

32
available nutrients and that quantities of organic food were likewise
correlated.
The Characterization of Carbohydrate Reserves
Leukel and Coleman (1930) in Florida measured carbohydrate frac¬
tions of bahiagrass and claimed that hemicellulose was transformed to
lignin and cellulose under long cutting intervals and reduced to simpler
sugars for use in tissue synthesis with a frequent defoliation regime.
McCarty (1935, 1938) studied seasonal carbohydrate fluctuations in
several range grasses and concluded that sucrose and starch were the
stored foods in California bromegrass (Bromus carinatus Hook, and Arn.).
McCarty (1938) also conjectured that hemicellulose may have been con¬
verted into simpler components.
Sullivan and Sprague (1943) showed no hemicellulose utilization in
the regrowth mechanism of ryegrass (Lolium perenne L.). Fructosan was
the key reserve substance and these results were reinforced by Waite and
Boyd (1953a, b) and Waite (1957, 1985). The concept that hemicellulose
participated in respiration or tissue synthesis was essentially dis¬
carded by the late 1940's; however, fructosan storage characterization
in important northern adapted grasses continued (Sprague and Sullivan,
1950; Waite and Gorrod, 1959; Okajima and Smith, 1964; Smith and
Grotelueschen, 1966; Smith, 1967; Grotelueschen and Smith, 1968; Smith,
1975). Smith (1968) cataloged the carbohydrate storage tendencies of
many North American grasses.
Tropical legume and grass carbohydrate characterization occurred
later and remains an active area of research. Hunter et al. (1970)
found no fructosan accumulation in the tropical plants he studied.

33
Noble and Lowe (1974) showed smaller seasonal variation of alcohol
soluble carbohydrates in tropical grasses than in temperate grasses, and
Wilson and Ford (1973) found that temperate grasses accumulated much
higher concentrations of soluble carbohydrate than the tropical grasses.
A Progression of Nonstructural Carbohydrate
Methods Used in Agriculture
A simple measure of energy reserves is obtained by the regrowth of
a defoliated plant in darkness. The technique involves the removal of
the sod, usually a 15 cm diameter plug, and allowing the defoliated
plants to regrow in darkness with adequate moisture until the energy
producing materials are exhausted. The weight of the clippings produced
during the period give an index of the regrowth potential of the plant.
Sheard (1973) reviewed etiolated regrowth studies, and his oldest refer¬
ence is that of Burton and Jackson (1962); however, the regrowth-in¬
darkness technique goes much further back in time.
According to Smith (personal communication):
The first use of growth in darkness that I know of is the
early work of L. F. Graber (1927). I have always been
intrigued with the technique, probably from being a student of
Dr. Graber's. Where Graber got the idea I do not know, but he
did his Ph.D. work with Dr. Kraus, a botanist at the Univer¬
sity of Chicago, who worked a great deal with carbohydrate/
nitrogen ratios in the growth of tomatoes and who may have
used darkness studies. The first person to use growth in
darkness on grasses was Vance Sprague (Sullivan and Sprague,
1943) at the USDA Pasture Lab in Pennsylvania. He picked up
the technique from Graber when he was his Ph.D. student, and
Blaser got the idea from Sprague (Ward and Blaser, 1961).
Pretreatment in darkness to vary carbohydrate concentration was
used by Davidson and Milthorpe (1966b) and others that used the
regrowth-in-darkness experimentation were Adegbola (1966), Adegbola and

34
McKell (1966b), Alberda (1966), Humphreys and Robinson (1966), Ráese and
Decker (1966), Matches (1969), Watson and Ward (1970), and Christiansen
et al. (1981). Christiansen et al. (1981) found improved sensitivity in
separating treatment differences by using stem base weights as a covari¬
able to control variations in plant size.
Concerns over the unreliability of chemical fractions were due to
the use of acids in early applications of carbohydrate methodology.
Fructosans are water- and ethanol-soluble, as well as readily hydrolyzed
to fructose monomers by acid treatment. Consequently, Sullivan and
Sprague (1943) complained that acid analyses of starch were confounded
by the contribution of fructose from fructosan in with glucose from
starch in tests of reducing power.
In 1947, Weinmann published a method for total available carbo¬
hydrate (TAC) determination in plants. Total available carbohydrate was
defined as "all those carbohydrates which can be used in the plant body
as a source of energy or as a building material, either directly or
indirectly after having been broken down by enzymes" (p. 279). In
Weinmann's method, small samples of finely ground air-dry material are
digested by takadiastase in water resulting in the breakdown of starch,
dextrins, and maltose to glucose, while other sugars and fructosan are
solubilized at the same time. The latter compounds are converted to
hexose sugars by acid hydrolysis, following which the reducing power of
the cleared, neutralized hydrolysate is determined (Weinmann, 1947).
Lindahl et al. (1949) made slight modifications in the Weinmann
TAC method and hailed the efficacy of the procedure in agronomic
applications. Their major modification was to change the enzyme from
dialyzed takadiastase to "clarase"--a highly purified and concentrated

35
form of takadiastase containing invertase, maltase, and amylase. This
procedure satisfied most agricultural needs for TAC determination
throughout the 1950's; however, Smith continued to examine acid tech¬
niques from his lab in Wisconsin. The acid methods were faster, requir
ing 8 hours to complete analysis compared to 12 hours working time plus
44 hours incubation for the enzyme method.
Most of the hindrance of using an acid method was in finding the
proper concentrations. Smith et al. (1964) varied sulfuric acid concen
trations from 0.2 N_ - 0.8 N_ and compared the results to takadiastase as
a standard. The 0.2 N_ F^SO* method most nearly duplicated the
takadiastase extraction while higher concentrations degraded
hemicellulose. The results were supported by Burris et al. (1967) who
added that the chances of obtaining erroneous data with acids were
greatest when they analyzed bermudagrass during rapid growth phases and
with tissue high in starch.
Grotelueschen and Smith (1967) qualified earlier work (Smith
et al., 1964) after examining alfalfa (Medicago sativa L.), high in
starch content. For fructosan accumulating tissue such as timothy
(Phleum pratense L.), containing little starch, dilute acid (0.005 N_
F^SO^) procedures may be used for nonstructural carbohydrate
extraction. On the other hand, with tissue high in starch such as
legumes and tropical grasses, acid strengths beginning at 0.2 N_ F^SO^
degraded hemicellulose and incompletely hydrolyzed starch as well.
Greub and Wedin (1969) supported Grotelueschen and Smith's 1967
findings and also warned that above 0.2 N F^SO^ free fructose or
fructose liberated from fructosan was being destroyed. The enzymatic

36
method was more accurate, demonstrated again at a later date, in
Portugal by Chaves and Moreira (1977).
In 1969 Smith wrote his widely accepted version of the Weinmann/
Lindahl procedure. It was and still is the most popular titrimetric
nonstructural carbohydrate method. Smith suggested the term total non-
structural carbohydrate (TNC) as a more clearly definable term than TAC
to both plant and animal investigators. The advantage of the Smith
method was its organized, complete, "cookbook" presentation which
included an information review, appendices, criteria for selecting a
method, and sample preparation. Smith (1981b) revised the TNC method
with minor modifications including an enzyme change. It was found that
Mylase 100 has rapid saccharogenic activity and will completely digest
the starch in tissues containing 30 percent or less of TNC in 20 hours.
Advances in carbohydrate methodology in the past 10 years have
been centered around the shift toward colorimetry to increase speed,
sample number, and efficiency (Haslemore and Roughan, 1976; Weier
et al., 1977; daSalveira et al., 1978; Westhafer et al., 1982).
In colorimetric methods, specific color reactions for portions of
carbohydrate molecules are due to formation of furfural or furfural
homologues in strong acids, especially following heating. These furans
or their reaction products are derived from oxidation, reduction, or
condensation processes in strong acid and can form colored products upon
reaction with sugars. There are treatmendous variabilities associated
with the reactions depending on the sugar, reagent concentration, tem¬
perature, and time of heating (Dische, 1962). The search for specific,
quantitative, and reproducible colorimetric assays for the determination
of carbohydrates is an important area of research.

37
Colorimetric tests have advantages over titrimetric procedures due
to their speed and equivalent precision. Two distinct steps involved
in colorimetric reactions are (1) formation of a chromocen from the
sugar, and (2) development of the color by a condensation of the
chromogen with a specific reagent (Aminoff et al., 1970). It is impor¬
tant to recognize that most colorimetric methods are empirical. Since
little is known about the reaction mechanism or the exact nature of the
chromogen involved, absolute stochiometry is not often obtained. As
long as the results obey Beer's Law within appropriate limits of concen¬
tration, the problem can be solved by using appropriate internal
standards and blanks.
In quantitative analysis, two divergent objectives must be
considered: (1) an overall analysis of all the sugars present by a very
general reaction, and (2) the selective determination of one sugar in
the presence of others. For example, Westhafer et al. (1982) found
sucrose levels in turfgrass root tissue to have the most dramatic
response to nitrogen treatment; thus, a TNC method would be less sensi¬
tive measure of carbohydrate concentrations changes than a test of
sucrose.
The principal chemical methods for quantitative measurement of
sugars use the action of sugar reduction on alkaline solutions of the
salts of certain metals. The most extensively used metal in sugar
analysis is copper. Most of the agricultural needs for quantitative
sugar analysis have been satisfied with the titrimetric method of
Somogyi (1945, 1952) and the following colorimetric techniques: phenol-
sulfuric (Dubois et al. , 1956), anthrone, and Nelson's (1944) test for

38
reducing sugars. An excellent discussion of reducing sugar techniques
is given by Hodge and Hofreiter (1962).
Organic Substances Used in Regrowth
Perhaps the biggest controversy in the history of nonstructural
carbohydrate research was conclusively proving that food reserves were
used in the synthesis of new growth. May and Davidson (1958) stated
that a general acceptance of the importance of carbohydrate reserves in
regeneration seemed unjustifiable and argued that no causal role in
shoot regrowth was suggested by decreases of carbohydrate in storage
organs. May (1960) dismissed as conventional wisdom the Graber and
Weinmann definitions of "reserves" on the basis that the term had become
semi-technical and no longer subject to criticism. May (1960) also
cited Archbold (1945) and Bernatwoicz (1958) who discarded the idea of
stored sugars as a "purposive" reserve--"reserve" connotating provision
for the future and as such signifying teleological thinking. A better
definition was "accumulate" since it was non-committal concerning pur¬
pose or intent.
Carbon balance and labeling studies put an end to the controversy
(Marshall and Sagar, 1965; Alberda, 1966; Davidson and Milthorpe, 1966a;
Ehara et al., 1966; Wardlaw, 1968; Watson and Ward, 1970; Sheard, 1973).
Davidson and Milthorpe (1966a) demonstrated that nitrogenous organic
compounds were also mobilized under heavy stress to the plant; there¬
fore, carbohydrates were only part of the labile pool used for respira¬
tion and new growth.

39
Location, Seasonal, and Daily Fluctuation
of Reserve Foods
Nonstructural carbohydrates may be temporarily stored in all plant
parts. Troughton (1957) concluded that the major storage regions were
in the underground organs. Many other studies, however, have shown that
major storage parts are stem bases, including stolons, corms, and
rhizomes (Sullivan and Sprague, 1943; Baker and Garwood, 1961). Waite
and Boyd (1953a, b) and Smith (1967) found percentages of fructosans and
most sugars to be higher in the stems than in leaves in temperate
grasses. For three tropical species, Hunter et al. (1970) found a
higher concentration of reducing and total sugars in the stems than leaf
blades plus sheath. Perry and Moser (1974) reported on the TNC content
of eight range grasses and stressed the importance of locating the
specific storage organ(s) of a grass before proceeding with TNC analysis.
The products of photosynthesis may be held in the leaf blades as
sucrose or starch. Since sucrose is the main transíocatable sugar and
starch is stored in chloroplasts, it explains the frequent measurement
of high amounts of these substances in leaves (Greenfield et al., 1974),
especially with cool temperatures (Garrard and West, 1972; Carter and
Garrard, 1976).
The environment governs seasonal assimilate distribution and carbo¬
hydrate metabolism is greatly affected by temperature. The accumulation
of carbohydrates at low temperatures indicates that growth rates are
more affected than are photosynthetic rates, and there is no conclusive
evidence that reduced translocation is ever the primary cause of limit¬
ing growth under low temperatures (Wardlaw, 1968). Hence, for both
temperate and tropical species, carbohydrate accumulation occurs in the
autumn or cool season. However, in temperate areas autumn turns to

40
freezing winters and TNC reserves are necessary for survival until
spring. Once perennial forages begin growth in the spring, TNC levels
generally increase through vegetative stages to anthesis and later the
carbohydrate reserves decline slowly through the summer when hot night
temperatures cause high rates of respiration. Studies to support the
above hypothesis for temperate grasses were published by Waite and Boyd
(1953b), Baker and Garwood (1961), Trlica and Cook (1972), Smith (1975),
and Mislevy et al. (1978).
For many tropical growing grasses carbohydrate accumulation occurs
during the cool or dry season and carbohydrate drain is most intense
during the summer due to high night temperatures and very active growth.
It could be speculated that plant survival is dependent upon carbohy¬
drate accumulations during the cool seasons. In northern Florida, for
instance, frost will kill most above ground herbage but warm day tem¬
peratures and adequate day lengths permit basal leaf growth in species
such as limpograss (Gaskins and Sleper, 1974). Studies showing the
accumulation of carbohydrate in tropical grass species during the cool
season were reported by Woods et al. (1959), Ferraris (1978), and Wilson
and t'Mannetje (1978).
Daily carbohydrate fluctuations occur for all species but to dif¬
ferent extents. Holt and Hilst (1969) showed that bromegrass (Bromus
inermus Leyss.) utilized almost one-third of the TNC in the herbage
during the night, but diurnal fluctuations were less for other grasses.
Greenfield and Smith (1974) studied switchgrass (Panicum virgatum L.)
and found that the diurnal trend was an increase of total sugars and
starch from 6 am to 6 pm and then a decrease to midnight. Basal sheaths
and internodes tended to increase in percent starch and TNC from 6 am

41
to midnight. Since these are storage parts, carbohydrates were presum¬
ably being translocated continuously from upper parts to these lower
sinks for storage, especially after 6 pm.
Management Factors Affecting TNC:
An Integrated Approach
Applied management of forage plants must optimize yield, quality,
and plant persistence. Plant behavior can be modified by cultural
practices but these procedures must integrate the plant physiology
involved in maintaining vigor. Too often experiments disect plant
response in order to study one variable at a time. A holistic approach
that takes into account environment and plant growth stage will be
necessary for management systems to be of any practical use.
While it is generally known that carbohydrate accumulation varies
inversely with the growth rate of the plant (McCarty, 1935; Brown and
Blaser, 1965; Colby et al., 1965; Blaser et al., 1966), few management
systems are based upon these findings. Carbohydrate analyses alone
cannot unambigously identify a superior management regime because of
confounding variations in residual leaf area, crown structure, axillary
bud number, leaf age, and altered root characteristics (Humphreys,
1966).
Much effort was expended in the understanding of residual leaf
area, light interception, and carbohydrate reserves in explaining
regrowth following defoliation (Ward and Blaser, 1961; Pearce et al.,
1965; Davidson and Milthorpe, 1966b; Humphreys and Robinson, 1966).
When moisture and nutrients are in adequate supply, residual leaf area
is generally more important than food reserves; however, this concept

42
of explaining regrowth is dependent upon the intensity of defoliation,
the presence or absence of buds, the age of leaf tissue, and species
differences. Further, there is no clear relationship between rate of
growth and leaf area index. Hence, this is an oversimplified model on
which to base a management system (Milthorpe and Davidson, 1966).
The role of hormones in releasing apical dominance during frequent
defoliation is often neglected. Certainly some plants have a better
ability to adopt a prostrate growth habit with many small leaf blades
that maintain assimilate supply to the plant. The classical study of
bermudagrass by VJeinmann and Goldsmith (1948) as reviewed by Weinmann
(1961) comes to mind. Close cutting of a well fertilized green of
Cynodon dactyl on 91 times in a season did not result in TAC depletion
due to high residual leaf area; however, complete defoliation by means
of scissors, repeated at weekly intervals, nearly exhausted TAC
reserves. Graber (1931) and Leukel and Coleman (1930) recognized the
ability of plants to defend against frequent defoliation by altered
morphology.
Nitrogen fertilization will stimulate herbage growth and, in
general, will cause a reduction of carbohydrate reserves as they are
used as carbon skeletons for protein synthesis (Waite, 1958; Alexander
and McCloud, 1962; Colby et al., 1965; Adegbola and McKell, 1966a;
Alberda, 1966; Auda et al., 1966; Gallaher and Brown, 1977; Wilson and
t'Mannetje, 1978).
Nitrogen fertilization can cause increases or decreases in carbo¬
hydrate storage depending on the amount applied and time of sampling
(Sprague and Sullivan, 1950). Carbohydrate reserves are only utilized
for a short time following defoliation (2 days--Davidson and Milthorpe,

43
1965, 1966a, b; 6 days—Ehara et al., 1966; 7 days--Sul1ivan and
Sprague, 1943). Moderate nitrogen fertilization promotes growth, photo¬
synthesis, and TNC storage; therefore, sampling too late in non-stressed
plants will not show a carbohydrate decline.
In frequently cut and highly fertilized swards, carbohydrate drain
from storage organs can continue to the detriment of plant persistence.
Alberda (1966) reported the death of tillers following severe defolia¬
tion stress and stated, "It may be supposed that a considerable part of
. . . these tillers is broken down and translocated to the remaining
tillers to be used for new leaf formation, but this has not been proven"
(p. 147).
The rate of degradation of a stressed sward is accelerated once a
state of tiller degradation occurs. As nitrogen fertilization forces
herbage growth and frequent cutting disallows full leaf expansion,
little assimilate is mobilized to roots and buds (Wardlaw, 1968).
Energy reaching the roots is inadequate to meet the needs of respira¬
tion, root growth, and absorption have slowed or stopped, and root degra¬
dation occurs as other substances are scavenged as a last defense
(Davidson and Milthorpe, 1966a).
Further complications in this theoretical example of high plant
stress are imposed by aggressive weeds competing for nutrients, mois¬
ture, and light. Due to the combined effects of all the above factors,
the botanical composition of the desired species declines and the sward
degenerates. In conclusion, when TNC studies are used as a single tool
complemented with other information, they greatly aid in understanding
the total dynamics of plant behavior. However, emphasis must be placed
on a balanced understanding of all factors involved in plant growth if

44
it is the objective of research to lead to sound management
recommendations.

CHAPTER 1
REGROWTH IN DARKNESS AS INFLUENCED BY PREVIOUS CUTTING
TREATMENT OF FOUR LIMPOGRASS GENOTYPES
Introduction
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) is
native to the humid subtropics of Africa. A brief resume of its intro¬
duction into American grassland agriculture has been reported by Oakes
(1973). Agronomic evaluations (Kretschmer and Snyder, 1979; Quesenberry
et al., 1978; Quesenberry and Ocumpaugh, 1980; Ruelke et al., 1978) have
generated information supporting the view that limpograss has forage
potential in subtropical regions and on soils that are intermittently
flooded. This study of etiolated regrowth was performed as a means of
characterizing the energy reserves and morphologies of four 1impograsses.
The regrowth-in-darkness technique has been used by many
investigators (Adegbola, 1966; Adegbola and McKell, 1966b; Burton and
Jackson, 1962; Dovrat and Cohen, 1970; Matches, 1969; Ráese and Decker,
1966; Ward and Blaser, 1961; Watson and Ward, 1970). The objective of
this study was to determine whether frequent clipping in the field could
reduce energy reserves below the critical amounts necessary to maintain
stands through the dormant season. Although chemical tests were not
conducted, much evidence exists substantiating a strong correlation
between etiolated regrowth and nonstructural carbohydrates in storage
organs (Adegbola, 1966; Adegbola and McKell, 1966b; Dovrat and Cohen,
1970; Ráese and Decker, 1966).
45

46
No carbohydrate studies of limpograss could be found; hence, it was
desired to learn the location of carbohydrate storage in these
stoloniferous grasses before full scale chemical analyses were con¬
ducted in later phases of research.
Materials and Methods
A clipping study was conducted on four well-established plantings
of limpograss on a Wachula sand, a poorly drained siliceous hyperthermic
ultic haplaquod soil at the Beef Research Unit of the University of
Florida near Gainesville. The grasses were PI 349753, PI 299995,
PI 299993, and PI 364888. Grasses PI 299995 and PI 299993 have been
released as 1Bigalta1 and 'Redalta1, respectively. Redalta is a diploid
(2n = 18) and the other genotypes used here are tetraploids (2n = 36).
The four blocks of grasses representing genotypes were unreplicated;
hence, the design was a split plot without replication of main plots.
The genotype*treatment (rep) term was used to test genotype, treatment,'
and interaction effects. Within main plots, clipping treatments were
randomly assigned with three replications. Clipping treatments were
initiated by mowing all plots to 5 cm on 27 July 1979 and harvesting at
2.5, 5, and 10 week intervals until 4 October 1979. Additional areas
without replication were reserved to allow limpograss top growth to
reach 15 and 25 weeks of age by 14 November 1979, when the regrowth-in¬
darkness Experiment 1 began.
The limpograsses were fertilized 28 May and 30 July 1979 with 280
kg/ha 17-5-10 (N-P^Og-K^O) containing 1 percent of a microelement mix.
Soil test results from samples taken on 3 December 1979 showed a pH of
6.6, 27 kg/ha phosphorus, and 22 kg/ha of potassium.

47
Experiment 1
On 14 November 1979, 15 cm diameter cores were removed from the
2.5, 5, 10, 15, and 25 week treatments in each of the four limpograss
plantings. The cores were placed in black plastic pots with the plant
material trimmed to 2.5 cm height. Two subsamples were taken from each
of the 2.5, 5, and 10 week field plots, and six samples were taken from
the areas assigned a 15 and 25 week cutting interval. Hence, there were
six pots containing limpograss for each treatment, and they were
arranged on shelves in a dark room using a randomized, complete block
design. A small electric heater was used to maintain the air tempera¬
ture at approximately 30 C. Plants were watered and sprayed with fungi¬
cide when necessary. The study was terminated after 3 weeks when growth
had ceased. "Shoot" regrowth was removed before the remainder of the
plant was washed free of soil and separated into "roots" and "stubble."
All three components were then dried at 60 C and weighed. The dry
matter (DM) yields were subjected to analysis of variance and regression
analysis. The mean yields for treatments were subjected to the Waller-
Duncan Multiple Comparison Test.
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Northeast Regional Data Center of the
State University System of Florida, located on the campus of the
University of Florida in Gainesville.

48
Experiment 2
A second etiolated regrowth experiment was begun on 6 February 1980
using new cores of PI 364888 from each of the five treatments described
in Experiment 1. In this study plants were trimmed to ground level
leaving short "stem bases" below ground. A completely randomized design
was used with nine replications in the dark room. The regrowth period
was 3 weeks in length, and plants were processed as in Experiment 1. In
Experiment 1, various statistical models were employed to explain the
etiolated regrowth as a function of the weight of "stubble" and/or
"roots," and in Experiment 2, the reduced "stubble," i.e., "stem base"
weight, was selected as a covariable in order to increase precision by
accounting for variations in plant size.
Residual Effects
The limpograsses were clipped at a height of 5 cm on 30 April 1980.
The forage was collected and dried at 60 C, weighed, and analyzed for
yield differences created by the 1979 clipping treatments.
Results and Discussion
Experiment 1
The results obtained in Experiment 1 are summarized in Table 1.
Clipping treatments caused etiolated regrowth yields to be different for
PI 349753 and Bigalta (P = 0.026 and P < 0.01, respectively), whereas
Redalta and PI 364888 were not affected by previous cutting treatment in
the field (P = 0.138 and P = 0.157, respectively). For both PI 349753
and Bigalta, a maximum regrowth was observed at the 10 week cutting

49
Table 1. Average DM yields of etiolated "shoots" for four limpograsses
following 3 weeks of growth in darkness
Clipping
frequency
(weeks)
Limpograss
number or name
PI 349753
Bigalta
Redalta
PI 364888
y/^ "
*
2.5
0.74 ab
0.95 ab
0.43 a
0.36 a
5
0.76 ab
0.81 b
0.36 a
0.45 a
10
0.87 a
1.08 a
0.62 a
0.42 a
15
0.52 b
0.34 c
0.65 a
0.45 a
25
0.51 b
0.41 c
0.55 a
0.63 a
★
Values within each column followed by the same letter are not
significantly different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.

50
frequency with lower yields of regrowth for the 15 and 25 week
treatments. Within the field plots representing the two longest cutting
frequencies, new growth was noticed emerging from plant bases below a
dense canopy. It is believed that this light-starved growth in
PI 349753 and Bigaita contributed to a respiratory drain of the energy
reserves. The same trend for lower DM yields in the 25 week treatment
was observed for Redalta but not for PI 364888 which had more resistance
to lodging allowing more light to reach new tillers beneath the canopy.
Weights of "shoot," "root," and "stubble" components were averaged
for all cutting frequencies in order to compare plant form among
1impograsses. As shown in Table 2, Bigalta and PI 349753 had signifi¬
cantly heavier "shoot" weights compared to Redalta and PI 364888.
Bigalta and PI 349753 also had the lowest "stubble" weights. These data
agree with field notes characterizing Bigalta and PI 349753 as having
larger but fewer "shoots" per unit area than Redalta and PI 364888. Of
the four grasses, PI 349753 had the lowest "root" weight. Redalta has
a bunch-type growth habit and a large "root" mass, both of which may
contribute to its excellent persistence.
These results suggest that the large amount of "stubble" in Redalta
and PI 364888 might have provided a reservoir of energy reserves which
buffered the cutting pressure on these two lines. Matches (1969) showed
higher regrowth yields with increasing height of cutting, indicating
an energy reserve sink in the stem bases of tall fescue (Festuca
arundinacea Schreb.).
In this study of limpograss there was a possibility that a lower
cutting height in darkness would reduce sink size and separate treatment
effects in a significant way.

51
Table 2. Comparison of etiolated "shoot" DM yields and associated
"root" and "stubble" components for four limpograsses averaged
for all cutting treatments following 3 weeks of growth in
darkness
Limpograss
Component
"Shoot"
"Root"
"Stubble"
PI 349753
0.68 a*
g/pot
9.66 c
2.46 c
Bigalta
0.72 a
14.47 b
2.33 c
Redalta
0.52 b
17.46 a
3.14 b
PI 364888
0.46 b
12.36 b
4.71 a
â– k
Values within each column followed by the same letter are not
significantly different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.

52
Experiment 2
The etiolated "shoot" weights from Experiment 1 were modeled as a
function of the cutting treatments and "roots"; cutting treatments and
"stubble"; and cutting treatments, "roots," and "stubble." The reduced
models, which explained regrowth yields as a function of treatment and
"stubble" (or treatment and "roots"), described etiolated regrowth as
well as the full model for all four 1impograsses. "Root" weights were
affected (P = 0.006) by treatments in Experiment 1, whereas "stubble"
weights were not (P = 0.712). Covariables are not supposed to be
affected by treatments, and since the "stem bases" from Experiment 2
were actually that portion of the "stubble" below ground, the "stem
base" weight was used as a covariable to explain some residual error in
the analysis of Experiment 2 regrowth.
Analysis of variance of etiolated regrowth in Experiment 2 did not
detect treatment differences (P = 0.180) but analysis of covariance did
(P = 0.026). Table 3 shows the Haller-Duncan Multiple Comparison Test
of average DM yields of "shoots" obtained for each clipping frequency in
PI 364888. "Shoot" yields were lower and the range narrower than found
in Experiment 1 for PI 364888, but the data followed the same trend.
Lower yields can be attributed to the loss of some energy reserves that
could be located in the above ground stem tissue removed prior to the
start of Experiment 2.
Results agree with those of Matches (1969), who suggested that with
shorter heights of cut the ranking order of treatments would remain
nearly the same but the magnitude of difference of energy reserves might
be less. Matches also stated that in etiolated regrowth experiments

53
Table 3. Average DM yields of etiolated regrowth from
Experiment 2 following 3 weeks of growth in darkness
Clipping frequency
(weeks)
PI 364888 "shoots"
(g/pot)
2.5
0.28 c
5
0.32 be
10
0.34 abc
15
0.38 ab
25
0.42 a
★
Values followed by the same letter are not significantly
different (P < 0.05) based on the Waller-Duncan Multiple
Comparison Test.

54
higher cutting heights permit more regrowth and allow greater differ¬
entiation of treatment effects. That statement is further substantiated
by Watson and Ward (1970) who demonstrated a 25 percent reduction in
food reserves when cutting height was reduced from 7.5 to 2.5 cm in
dallisgrass (Paspa!urn dilatatum Poir.).
In this study the results of Experiment 2 show that trimming the
plants to ground level required a covariable ("stem base" weight) to
detect treatment differences. However, reducing rather than increasing
residual plant height together with the small differences in regrowth
yields between Experiment 1 and Experiment 2 inferred that the majority
of food storage occurs in the very basal portions of the culm ("stem
base").
Residual Effects
Yield data obtained from 30 April 1980 harvest of the limpograsses
showed no evidence of a residual treatment effect on the sward. No
sward damage was observed even in the limpograss plots harvested for
four cycles of 2.5 weeks. Hence, the limpograsses were not sufficiently
stressed to cause stand deterioration in 1980.
Conclusions
1. Experiment 1 showed that etiolated regrowth weights were maxi¬
mized for the 10 week clipping treatment in Bigalta (P < 0.01) and
PI 349753 (P = 0.026), but no differences were found due to previous
cutting for Redalta (P = 0.138) or PI 364888 (P = 0.157).

55
2. Following the complete removal of above ground stem tissue at
the start of Experiment 2, the production of considerable etiolated
growth suggested energy reserve storage lower on the culm.
3. "Stem base" weights were used as a covariable in Experiment 2,
and differences (P = 0.026) were found in yield of etiolated regrowth
due to previous cutting pressures on limpograss PI 364888 with longer
cutting intervals allowing greater yields of regrowth.
Summary
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb.)
research has reached a stage where management recommendations are needed
to fully implement limpograss' usefulness for the large hectarage of
improved pastures on Spodosols like Florida's flatwood soils. A pre¬
liminary study of energy reserves in Redalta, Bigaita, and two other
promising introductions was conducted using the regrowth-in-darkness
technique.
Cutting frequencies of 2.5, 5, 10, 15, and 25 weeks on limpograss
swards in the field were imposed to establish various levels of reserves.
The reserve energy pool was then measured by regrowth yields of plant
cores placed in a dark room.
Two regrowth-in-darkness experiments were conducted. Experiment 1
used a randomized, complete block design with six replications harvested
at 2.5 cm "stubble" height. "Stubble" was classified as all the stem
between the cutting height and "roots." The first experiment showed
that etiolated regrowth weights were maximized for the 10 week clipping
treatment in Bigalta and PI 349753 (P < 0.01 and P = 0.026,
respectively), but Redalta and PI 364888, which had higher "stubble"

56
weights, showed no treatment effects on weight of etiolated "shoots."
In Experiment 2, PI 364888, the line with the highest "stubble" weight
from Experiment 1, was evaluated using a completely randomized design
and nine replications with all plant material removed to ground level.
Hence, only the "stem bases" remaining below the soil surface were
responsible for regrowth in Experiment 2. No treatment effects were
found (P = 0.180) until the data were analyzed with covariance tech¬
niques using "stem base" weight as a covariable. Statistical sensi¬
tivity improved, and increases (P = 0.026) in regrowth potential were
detected with clipping treatments of a longer cutting interval. The
covariable analysis represented an improvement in the regrowth-in¬
darkness technique.
Removing all plant material to the soil level prior to the begin¬
ning of the second experiment left 2-3 cm of "stem base" below ground.
The close similarity of regrowth yields in Experiment 1, where plants
were clipped to 2.5 cm above the soil surface, and those of Experiment 2
having no above ground tissue, suggested that the energy reserves were
predominantly located in the bottom 2-3 cm of the stem.

CHAPTER 2
CUTTING FREQUENCY EFFECTS ON LIMPOGRASS MORPHOLOGY
AND TOTAL NONSTRUCTURAL CARBOHYDRATE RESERVES
Introduction
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) has
promise as an adapted, warm-season grass for Florida's vast hectarage of
acid flatwood soils. Unfortunately, 'Bigalta1, most favored by ranchers
because of its high forage quality, is also the least persistent of the
cultivars which have been released (Quesenberry et al., 1978; Ruelke,
1978; Kretschmer and Snyder, 1979; Quesenberry et al., 1981). Other
studies of limpograss have identified a few promising plant introduc¬
tions that have comparable quality and better persistence than Bigalta
(Ruelke, 1978; Quesenberry and Ocumpaugh, 1979; Ocumpaugh and
Quesenberry, 1980; Ocumpaugh et al., 1981; Ocumpaugh, 1982).
In Chapter 1, Christiansen conducted regrowth-in-darkness experi¬
ments with limpograss that indicated the area of compressed, lower nodes
on the stem base as the major site of energy reserve in these robust,
stoloniferous, perennial fodder plants. The regrowth-in-darkness tech¬
nique is a simple way to measure energy reserves by analyzing regrowth
from defoliated plants placed in darkness (Burton and Jackson, 1962;
Sheard, 1973). Perry and Moser (1974) stressed the importance of locat¬
ing the specific carbohydrate storage organ of a grass before proceeding
with total nonstructural carbohydrate (TNC) analyses. Studies of TNC
content were deemed necessary to verify the regrowth-in-darkness results.
57

58
Many researchers have indicated a positive correlation between etiolated
regrowth and TNC in the storage organs (Adegbola, 1966; Adegbola and
McKell, 1966b; Dovrat and Cohen, 1970; Ráese and Decker, 1966).
Therefore, it was of interest to see if this positive correlation held
for limpograss TNC versus etiolated regrowth.
Leukel and Coleman (1930) and Graber (1931) discussed the ability
of various forage plants to transform an upright habit to a prostrate
habit following frequent defoliation. Some limpograsses may be sus¬
pected to have greater flexibility than others in altering their growth
habit. The objective was to study morphological changes induced by
frequent cutting and elucidate subsequent differences in TNC accumula¬
tion for two promising and two released cultivars of limpograss.
Materials and Methods
The experiments to follow were conducted on four well established
plantings of limpograss on a Wachula sand; a poorly drained siliceous
hyperthermic ultic haplaquod soil, at the Beef Research Unit of the
University of Florida. The grasses were PI 349753, PI 299995,
PI 299993, and PI 364888. Both PI 299995 and PI 299993 have been
released as Bigalta and ’Redalta', respectively. Redalta is a diploid
(2n = 18) and the other genotypes used here are tetraploids (2n = 36).
The limpograsses were fertilized 28 May and 30 July 1979, with
280 kg/ha of 17-5-10 (N-P20g-!<20) containing 1 percent of a microelement
mix. Soil test results from samples taken on 3 December 1979 showed a
pH of 6.6, 27 kg/ha phosphorus, and 22 kg/ha of potassium.

59
Analysis of TNC in Plant Parts
Prior to the layout of the clipping experiment, whole plant samples
of limpograss were taken on 3 July and 26 July 1979, representing 6 and
9 week old plant maturities, respectively. Five large plants of each
genotype were randomly selected, dug, washed free of soil, and arranged
in cotton sample bags. The cotton bags were then packed in a plastic
bag and put on ice. Upon reaching the lab, the samples were dried in a
Thelco forced-air oven at 100 C for 30-45 minutes after which the tem¬
perature was lowered to 70 C until the samples were removed 36-48 hours
later (Smith, 1973a).
The plants were carefully fractionated into "shoots," "stubble,"
"crown," and "roots." In the plant part analysis, "crown" was desig¬
nated as the bottom 2 cm of the stem base, and "stubble" was classified
as the immediate 2 cm above the "crown." "Shoots" included all herbage
above the "stubble," and "roots" were severed from the "crown" by knife.
Each component was ground through a 1 mm screen in a small Wiley mill
and then reground through a UDY Cyclone sample mill fitted with a 0.5 mm
screen. The samples were stored in plastic, 20 ml Dilu-vials and then
analyzed for TNC using enzymatic hydrolysis and spectrophotometric
measurement of reducing sugars as described in Appendix A.
TNC as Related to Clipping, Season, and Genotype
A 10 week long clipping study was initiated by mowing all plots to
5 cm on 27 July 1979 and harvesting at 2.5 cm height with a Jari sickle
bar mower at 2.5, 5, and 10 week intervals until 4 October 1979. The
four blocks of grasses representing genotypes were unreplicated; hence,

60
the design was a split plot without replication of the main plots. The
genotype * treatment (rep) term was used to test genotypes, clipping
treatments, and interaction effects. Within main plots the clipping
treatments were randomly assigned and replicated three times.
Samples for TNC analysis were dug prior to clipping on 13 August,
30 August, 17 September, 4 October, and 11 November and processed as
before except that only "crown" (bottom 2 cm of stem base) samples were
saved for TNC analysis. The "crowns" were scraped free of roots, leaf
sheaths, and sand by using a wire buffing wheel attached to the extended
shaft of a small electric motor (20 W, 1525 RPM). The "crowns" were
then ground through a 0.5 mm screen in the UDY mill and stored until
tested for TNC.
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Northeast Regional Data Center of the
State University System of Florida, located on the campus of the Univ¬
ersity of Florida in Gainesville. The plant part analyses and the per¬
cent TNC data for all subsequent samplings were subjected to analysis of
variance and means were compared using the Waller-Duncan Multiple
Comparison Test.
Correlation of TNC versus Regrowth-in-Darkness
Prior to discussing the methods used in the correlation of TNC and
etiolated regrowth, terminology will be reviewed for clarity. The
regrowth-in-darkness experiments were completed before any chemical
analysis took place. Hence, the first use of the term "stubble" was in
Experiment 1 of Chapter 1 and represented the 2.5 cm of stem tissue

61
above the soil level as well as the stem base below ground. Recall that
the stem material above ground level was removed at the start of the
second regrowth-in-darkness study (Experment 2, Chapter 1), and the term
"stem bases" was used to define the 2-3 cm long stem segments from below
ground. In the plant part experimentation an effort was made to more
specifically focus on the location of TNC accumulation. The "stem
bases" from Experiment 2 in Chapter 1 were believed to contain most of
the stored carbohydrate so in the plant part experiment the bottom 2 cm
of the stem was called the "crown." The 2 cm of stem immediately above
the "crown" was called "stubble" corresponding to common agronomic
usage, i.e., "residual above ground tissue following defoliation."
Limpograss samples taken on 11 November 1979 were analyzed for per¬
cent TNC to quantitatively characterize the concentration of reserves
present in the "crown" at the beginning of the regrowth-in-darkness
study (Experiment 1, Chapter 1). The darkness study began on 14
November 1979 and lasted 3 weeks. The dry weights of etiolated
"shoots," as well as weights of "stubble" (2.5 cm of stem above ground
plus stem bases below ground) and "roots" per pot were recorded. The
"crown" TNC data from 11 November 1979 were correlated against "stubble"
and against "shoot" percentages of the "stubble" plus "roots" present
in the pots (("Shoot"/("stubble" + "roots')) * 1Ó0).
Results and Discussion
Analysis of TNC in Plant Parts
Results for TNC analyses of plant parts are shown in Figure 1. The
limpograsses were different (P < 0.01) in their percent TNC; however,
significant two and three way interactions among parts, age, and

_ PI 364888
12i
o
T 8
N
6 ' b ☆
c
REDALTA
☆ a
BIGALTA PI 349753
☆ d
4
O Í—i i-
6 9
AGE OF LIMPOGRASS TISSUE (weeks)
• crowns ☆roots o stubble Dshoots
Figure 1. Total nonstructural carbohydrate (TNC) percent of dry matter of four plant parts in
four 1impograsses sampled at two maturities
(Different letters within an age of limpograss tissue indicate a significant differ¬
ence (P < 0.05) of percent TNC in the plant parts.)

63
genotypes prohibited a comparison of overall means for the four
1impograsses. When each limpograss was analyzed separately, all except
PI 364888 showed different (P < 0.01) accumulation of carbohydrate to
plant parts at 6 weeks as compared to 9 weeks. Between 6 and 9 weeks
Redalta increased TNC to the "roots" at the expense of other parts,
while Bigalta did just the opposite. In PI 349753 "roots" and "shoots"
increased in TNC while "crown" and "stubble" declined between the two
maturities. These results were interesting for Bigalta and Redalta
because of the high or low contribution, respectively, of readily
digested carbohydrates which would be measured in tests of in vitro
organic matter digestibilities (IVOMD). Many studies have shown large
IVOMD differences between these two grasses (Schank et al., 1973;
Quesenberry et al., 1978).
Criteria necessary in selecting a plant part for TNC analysis were
(1) disqualification of "shoots" because they were not perenniating
parts, and (2) rejection of "roots" because of sand and dead tissue
contamination. Figure 1 shows that the "crown" was statistically as
high or higher in percent TNC than the "stubble" at all times. The
bottom 2 cm of the stem base ("crown") was selected for further TNC
analysis by chemical means.
TNC as Related to Clipping, Season, and Genotype
Table 4 shows the TNC results for four 1impograsses subjected to
three cutting frequencies in the 1979 growing season. In a combined
analysis of sample dates, 1impograsses, and cutting treatments, each
factor influenced (P < 0.01) the TNC measured in "crowns." The two
way interactions (P < 0.01) of limpograss and clipping frequency with

64
Table 4. The main effects of limpograss genotype and clipping
treatment on the percent total nonstructural carbo¬
hydrates (TNC) in the bottom 2 cm of stem base
("crown")
Limpograss
1979 sampling date
8/13
8/30 9/17 10/4
11/11
TNC %
PI 364888
10.6b*
13.9a
12.4a
10.1a
13.6a
Redalta
6.6d
6.9d
6.4d
6.0d
10.8b
Bigalta
11.0a
11.8b
8.4b
9.9b
8.3c
PI 349753
9.1c
8.7c
8.0c
8.2c
8.Id
Clipping
frequency
(weeks)
2.5
9.8a
10.5a
10.2a
8.7a
12.6a
5.0
9.9a
10.2a
8.8b
9.2a
10.6b
10.0
8.2a
10.2a
7.5c
7.7a
7.5c
Different
letters
within a
sampling date
represent sig-
nificant (P < 0.
05) differences '
in percent TNC.

65
sampling date suggested that TNC concentration was independently altered
by both cutting treatment and seasonal carbohydrate flux. There was no
three way interaction (P = 0.47) or interaction between limpograss geno¬
types and clipping (P = 0.76); therefore, TNC concentrations were chang¬
ing according to cutting frequency, but the reaction was similar among
limpograsses.
Analysis of percentage TNC by sampling date revealed the rankings
shown in Table 4. Bigalta was higher in TNC than other limpograsses on
13 August. From 30 August to 4 October PI 364888 was highest in TNC.
In the last sampling on 11 November, the percent TNC in Redalta surged
above Bigalta and PI 349753 but not above that of PI 364888. Perhaps
the late autumn surge of TNC seen in PI 364888 and Redalta contributed
to their higher persistence as opposed to the plateau or decline of TNC
in PI 349753 and Bigalta.
Frequent cutting promoted lateral growth of stolons and increased
the number of leaf blades per unit area. This change toward a turf-like
morphology took some time to effect. In the bottom portion of Table 4
no differences in stored TNC were found for any frequency of cut in the
limpograsses until 17 September. On 17 September and 11 November
shorter cutting intervals enhanced TNC in the "crowns," suggesting an
enhanced ability of the plants with a turf-like habit to accumulate TNC.
Correlation of TNC versus Regrowth-in-Darkness
In Figure 2 data from the regrowth study (Experiment 1, Chapter 1)
was used to correlate against TNC in "crowns" of samples taken at the
start of the regrowth experiment. The best correlations were found for

%
TNC
16
14
12
10
8
6
4
9
ri
9
© 5
5 a
©
•
9
a
•
B
O O
cutting
©
freq.
9
r value
(wks.)
9*
0.74**
0 2.5
0.65*
â–¡ 5
m
n.s.
OIO
s ^ ^
H
9
9
dsn
O
B
B
©
a
cutting
9
freq.
©
r value
(wks.)
a %
-0.65*
o 2.5
-0.67*
â–¡ 5
O
•
-0.65*
OIO
> ♦,
S3 ^
S3
PI 364888 a
REDALTA B
BIG ALTA S3
PI 349753 â– 
2 3 4 5
Wt. of Stubble.g
2 4 6 8 10 12
Etiolated Shoot % of
Stubble and Roots
14
CTi
Figure 2. Total nonstructural carbohydrate (TNC) percentages in the "crown" correlated against
"stubble" weights (left) and etiolated regrowth percent of whole plants (right) for
four limpograss genotypes subjected to three clipping frequencies
P < 0.01

67
TNC versus "stubble" and TNC versus percent plant regrowth in darkness
(("shoots"/("stubble" + "roots")) * 100).
In Figure 2 the symbols represent the clipping frequencies used
within each limpograss (genotypes are identified within a symbol by the
coloration pattern). The important relationships are found within each
cutting frequency.
Genotypes having greater "stubble" weight per pot had higher con¬
tent of TNC in the "crowns." Frequent clipping induced a prostrate
habit in all genotypes, but PI 364888 and Redalta provided more sites
for leaf emergence--as reflected by the accumulation of TNC in "crowns"
for the frequently cut treatments.
A negative correlation was observed in Figure 2 when the TNC in
"crowns" were correlated against the "shoot" percentage of "stubble"
plus "roots." Plants with more "stubble" (PI 364888) and "roots"
(Redalta) (see Table 2, Chapter 1) produced small "shoots" that were
greater in number and lower in weight; hence, were a lower percentage
of the "stubble" plus "roots" (1 to 4 percent in Figure 2). Bigalta and
PI 349753 had fewer but larger "shoots," as well as lower weights of
"roots" and "stubble" (see Table 2, Chapter 1); hence, "shoots" were a
larger percentage of the "stubble" plus "roots" in pots (4 to 14 percent
in Figure 2).
When TNC was correlated directly against the dry weight of etio¬
lated "shoots," a negative correlation was also found (-0.63) but not
presented. Other experimenters (Adegbola, 1966; Adegbola and McKell,
1966b; Dovrat and Cohen, 1970; Ráese and Decker, 1966) have found posi¬
tive correlations of TNC versus etiolated regrowth, i.e., higher TNC in
plants having longer rest intervals. These studies, however, used

68
upright instead of stoloniferous grasses, and used a single genotype of
a species instead of four genotypes having variable morphologies.
Conclusions
1. Analysis of limpograss plant parts indicated that the bottom
2 cm of stem base ("crown") was a sight of TNC accumulation.
2. Cutting treatments and harvest date contributed independently
to the significant (P > 0.01) variations seen in limpograss carbohydrate
flux.
3. The more frequently clipped limpograsses effected a prostrate
morphology which allowed more axillary tiller formation and TNC accumu¬
lation for all limpograsses.
4. Redalta and PI 364888 showed an autumn surge of TNC that might
contribute to their better persistence as opposed to a decline or
plateau of TNC for Bigalta and PI 349753.
5. Plants having greater weights of "stubble" also had higher
concentrations of TNC in the bottom 2 cm of the stem base ("crown").
Summary
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) could
significantly contribute to the forage-livestock economy in the south¬
eastern United States; however, cultivars must have good quality and
high persistence before an impact will be made. Of the cultivars now
available, 'Redalta' and 'Greenalta' have low digestibility and adequate
persistence, while 'Bigalta' has excellent digestibility and poor
persistence.

69
Current research efforts have identified PI 364888 limpograss as
both persistent and digestible. The purpose of this study was to com¬
pare PI 364888 and another promising limpograss with Bigalta and Redalta
to try to understand persistence relative to the total nonstructural
carbohydrate (TNC) status of the plants.
The objective of this study was to locate the TNC storage site and
characterize seasonal, genotypic, and clipping frequency effects on the
TNC concentration in two released and two promising 1impograsses.
Secondly, it was desired to compare the chemical data with the regrowth-
in-darkness results obtained in a previous study.
A sampling of 6 and 9 week old limpograss on 3 July and 26 July
1979, fractionated into four plant parts ("root," "shoot," "crown," and
"stubble"), was chemically analyzed for percent TNC using an enzymatic
sugar hydrolysis and spectrophotometric measurement of reducing sugars.
The bottom 2 cm of stem base ("crown") was identified as the primary TNC
storage site and, subsequently, served to compare the carbohydrate
status in the four 1impograsses.
A clipping study was conducted to variously deplete the reserves by
cutting the limpograss field plots every 2.5, 5, or 10 weeks from 27
July to 4 October 1979. Results showed that frequent clipping induced
a prostrate growth behavior and the turf-like condition allowed greater
carbohydrate accumulation in the storage sites. Generally, PI 364888
was highest in stored TNC throughout the study and along with Redalta
showed a surge of TNC in samples taken on 11 November 1979. Bigalta and
PI 349753 did not react in this manner in late autumn, and they were
also observed to be less persistent in previous studies. Limpograss

70
PI 364888 had the largest stolon system, stored the most TNC, and was
very persistent.
The TNC results for 11 November 1979 were correlated against
"shoot" and other plant weight data obtained from a 3 week regrowth-in¬
darkness study initiated on 14 November 1979. The results revealed a
positive relationship (r = 0.56) across limpograsses for stubble weight
versus TNC and a negative correlation (r = -0.56) for etiolated shoots
expressed as a percent of the roots plus stubble. The significant (P <
0.01) regressions were caused by inherent differences in limpograss
morphologies rather than cutting frequency effects within each
1impograss.

CHAPTER 3
DRY MATTER YIELD, CRUDE PROTEIN, IN VITRO ORGANIC
MATTER DIGESTIBILITY, TOTAL NONSTRUCTURAL
CARBOHYDRATE, AND PERSISTENCE IN TWO PROMISING
AND TWO RELEASED LIMPOGRASSES: EFFECTS DUE TO
NITROGEN FERTILIZATION AND CUTTING FREQUENCY
Introduction
The southeastern United States is a subtropical zone that permits
the growth of a wide variety of grasses; only a handful of which are
agronomically important. Bahiagrass (Paspalum notatum Fl'úgge) is the
most widespread improved grass in Florida (Mott and Moore, 1977) because
of its excellent persistence and broad adaptability. Bermudagrass
(Cynodon dactyl on (L.) Pers.) cultivars are economically important as
hay and grazing crops due to successful plant breeding programs; but
these cultivars generally grow better on upland sites. Digitgrasses
(Digitaria decumbens Stent.) and stargrasses (Cynodon nlemfuensis
Vanderyst) are grown more in south and central Florida due to their
lower frost tolerance and winter hardiness. St. Augustinegrass
(Stenotaphrum secundatum (Walt.) Kuntze and paragrass (Brachiaria mutica
(Forsk) Stapf) have special adaptabilities for organic soils but are not
widely used as pasture grasses elswhere in Florida.
Limpograss (Hemarthria altissima (Poir.) Stapf et C.E. Hubb) is a
viable alternative to bahiagrass for flatwood sites. Limpograss can be
equally persistent, but higher in quality than bahiagrass (Moore et al.,
1981). Immature bahiagrass has potentially good quality; however,
71

72
quality rapidly declines in tissue greater than 6 weeks of age (Moore
et al., 1970) and requires heavy utilization for optimum yield and
quality (Beaty et al., 1980).
Ocumpaugh (1982) reported better yearly beef production and similar
average daily gains from PI 364888 limpograss than from 'Pensacola'
bahiagrass because of a 70 day grazing advantage. Gaskins and Sleper
(1974) showed that daylength sensitivity was not a limiting factor for
cool-season growth of limpograss but was for digitgrass and bermuda-
grass. Perhaps bahiagrass also falls into the latter category.
Bahiagrass is primarily used for grazing, not hay, and the major
forage related problems in the southeastern United States are seasonal
forage distribution and lack of preservation practices (Mott, 1982b).
Limpograss is more seasonably flexible and can produce copious amounts
of biomass. Killinger (1971) produced 15 m tons/ha of dry matter (DM)
by 23 May and 23 m tons/ha by 14 August in north central Florida. This
surpassed any of the yearly yields for five hybrid bermudagrasses,
'Pangóla' digitgrass, and Pensacola bahiagrass reported in 1971 by
Ruelke and Prine. Hodges and Martin (1975) studied the warm- and cool-
season (1 November - 15 May) production of 23 subtropical grasses at
Ona, Florida, and found the digitgrasses and Cynodons outyielded limpo-
grasses; however, limpograsses were superior to Pensacola bahiagrass at
all three fertility levels during the cool season. Kretschmer and
Snyder (1979) obtained 21.8 m tons of dry matter (DM) with two harvests
of 'Bigalta' limpograss using a 12 week cutting frequency and 168 kg/ha
N per interval. In a management study, Ruelke obtained yields of 13.5-
23.8 m tons for 'Redalta', 'Greenalta', and Bigalta limpograsses
(Quesenberry et al., 1978).

73
The quality characteristics of tetraploid limpograsses are well
known. Schank et al. (1973) identified high in vitro organic matter
digestibilities (IVOMD) and a slower rate of IVOMD decline in the
tetraploids. Taylor et al. (1976a), in New Zealand, also reported high
in vitro digestibilities (leaves: 66.8 percent; stems: 77.2 percent).
The tetraploid limpograss stem digestibility was the highest of all 25
summer grasses studied.
The above findings encouraged Quesenberry and Ocumpaugh (1977) to
initiate stockpiling experimentation. They reported the beginning of
August as the proper staging period to initiate regrowth of adequate
yields of standing forage. Redalta produced 10.8-11.9 m tons DM but
declined rapidly in quality, whereas Bigalta produced 6.2 m tons and
maintained a 45 percent IVOMD when measured the following March. The
rate of IVOMD decline was similar, but the intercept started 13 units
higher in Bigalta than for Redalta and Greenalta (Quesenberry and
Ocumpaugh, 1980).
Ruelke's management studies using limpograsses (Ruelke, 1978;
Quesenberry et al., 1978) showed that close, frequent defoliation and
high nitrogen fertilization led to the loss of Bigalta stands. Mob
grazing and clipping trials by Quesenberry and Ocumpaugh (1979) also
showed Bigalta's poor persistence. Quesenberry et al. (1981) identified
PI 364888 as having superior persistence and comparable quality to
Bigalta. Ocumpaugh et al. (1981) identified another promising limpo¬
grass (PI 349753) that persisted under mob grazing but that was of a
slightly lower acceptability by animals than PI 364888 or Bigalta.
Total nonstructural carbohydrate (TNC) analysis has for decades
been a useful technique in quantifying the fluctuation of organic

74
reserves in response to season (McCarty, 1935; Waite and Boyd, 1953a),
defoliation (Sullivan and Sprague, 1943; May, 1960; Baker and Garwood,
1961; Marshall and Sagar, 1965), nitrogen fertilization (Adegbola and
McKell, 1966a; Ráese and Decker, 1966; Ford and Williams, 1973), and
persistence (Graber, 1931; Alberda, 1966). The efforts of Dale Smith
(1981a) and others in understanding the carbohydrate metabolism in
alfalfa (Medicago sativa L.) is a prime example of the TNC studies
yielding practical management recommendations. According to Graber
(1931) "under field conditions, the limitations of root growth and the
modifications of the internal environment resulting from low reserves
may reduce the absorptive capacity of a plant so greatly and may so
increase its susceptibility to drought, winter injury, weed encroach¬
ments, insect injury, and other hazards as to jeopardize its permanence"
(p. 47).
Various reports have shown that high rates of nitrogen fertilizer
have led to reduced persistence in digitgrass (Creel, 1957; Ruelke,
1960; Kien et al., 1975) and in orchardgrass (Dactylis glomerata L.)
(Alexander and McCloud, 1962), but no carbohydrate analyses were
presented. No chemical data relating TNC to persistence in limpograss
could be located. Christiansen (Chapter 1) used regrowth-in-darkness
studies as preliminary tests of reserve energy in limpograss. In
Chapter 2 of this report the site of nonstructural carbohydrate accumu¬
lation was determined, as well as clipping effects on TNC percent of
limpograss in 1979.
Redalta has poor quality but excellent persistence, while Bigalta
has the reverse situation; neither combination being advantageous to the
producer. The objective of this study was to thoroughly evaluate two

75
promising introductions against Bigaita and Redalta. The goal was to
obtain a persistent, high quality, high yielding limpograss for flatwood
sites in Florida and the southeast.
Materials and Methods
Two experimental areas were used in this study. The 1979 experi¬
mentation was conducted on four well established plantings of limpograss
on a Wachula sand, a poorly drained siliceous hyperthermic ultic
haploquod soil at the Beef Research Unit of the University of Florida.
Another field was established in 1979 by vegetative propagation from
material taken from the 1979 experimental area. The new planting was in
an Adamsville sand. The limpograsses used were PI 349753, PI 364888,
Redalta, and Bigalta. Redalta is diploid (2n = 18) and the other
grasses used here are tetraploids (2n = 36).
1979 Establishment
The experimental area was 30 x 118 m and supported a mature stand
of rye (Secale cereale L.) prior to cultivating with a Ground Hawg
rototiller on 22 May 1979. The field was blocked into three sections
and within each section four main plots (genotypes) were marked to
measure 7 x 30 m with 3 m alleys between each main plot. The alleys
were planted to 'Argentine' bahiagrass.
Waist high stands of each limpograss growing in the 1979 experi¬
mental area were mowed with a sickle bar set at 7.5 cm height, raked,
and carried to an appropriately assigned, random location within each
block. The herbage was evenly distributed over the soil, lightly disked
to 15 cm, and the entire field was rolled using a cultipacker seeder to

76
assure good soil to stem contact. The field was irrigated when neces¬
sary to insure establishment and fertilized on 29 June and 23 August
1979 with 336 kg/ha of 17-5-10 (N^Og-I^O) containing 3.4 kg/ha of a
micronutrient mix. Soil test results from 7 June 1979 showed a pH of
6.2, 20.3 kg/ha phosphorus (P) and 40 kg/ha potassium (K). The field
was flail chopped on 6 December 1979 to 7.5 cm height to remove frosted
forage.
1979 Experimentation
Within each established limpograss sward in the 1979 experimental
area plots were marked out for a clipping study. The blocks were
unreplicated; hence, the design was a split plot design without replica¬
tion of main plots. The clipping treatments within a limpograss block
were randomly assigned and replicated three times. The genotype *
treatment (rep) term was used to test genotype, clipping treatment, and
interaction effects for the responses that were studied.
The clipping experiment was initiated by mowing all plots to 5 cm
on 27 July 1979 and harvesting at 2.5, 5 and 10 week intervals until
4 October. The limpograsses were fertilized on 28 May and 30 July 1979
with 280 kg/ha of 17-5-10 (N^O^-I^O) containing 1 percent of a micro¬
nutrient mix. Soil test results from samples taken on 3 December 1979
showed a pH of 6.6, 28.2 kg/ha P, and 22.5 kg/ha K.
The 2.5 week interval treatments were clipped at 5 cm height on
13 August, 30 August, 17 September, and 4 October 1979 using a Jari
sickle bar mower. Yields were determined and "shoot" samples (tissue
above cutting height) were taken for IV0MD and crude protein (CP).
The 5 week interval treatments were harvested on 30 August and

77
4 October 1979. The 10 week treatment was harvested and sampled for
IVOMD, yield, and CP only on 4 October 1979.
1980 Experimentation
Soil test results from the newly established field of limpograss
revealed a pH of 6.1, 16.4 kg/ha P, and 53.1 kg/ha K. On 23 March 1980
the field was uniformly mowed to 7.5 cm height and treatments consisting
of five levels of nitrogen (N) fertilization and five cutting frequen¬
cies (F) were imposed on the main plots. The two factors (N and F) were
t h s
combined to represent 13/25 of a 5 x 5 complete factorial. Figure 3
shows a 3 x 3 factoral and the spatial arrangement of two factors, and
Figure 4 shows a central composite design in three factors. The 13
treatment combinations used in this study are shown graphically in
Figure 5 and the design points are plotted as grid coordinates in
Figure 6. The arrangement of treatment combinations in Figure 6 shows
the underlying difference between the classical central composite design
(Figure 3) and the modified central composite design described by
Littell and Mott (1975). Note in Figure 6 how information is concen¬
trated in what was considered the "management realm" (0-120 kg/ha N) and
3 to 9 week defoliation frequencies. Also note the geometric scaling of
N and the arithmetric progression for cutting frequency (with a skip at
the 15 week level). As the dotted lines suggest, the design could be
2 2
considered a superimposition of two complete factorials (2 and 3 ).
Consequently, the grasses were replicated as main plots in a split
plot design, and the 13 treatments provided a response surface with all
points replicated three times. Each main plot contained 13 plots

78
Figure 3. Spatial arrangement of a three level factorial design
in two variables
Figure 4. Spatial arrangement of a central composite desian in
three variables

79
TREATMENT MATRIX
480
®
0
®
|~23¡
®
N 240
0
©
0
®
25
FERTILIZATION
(kg/ha/yr) *20
®
0
©
[Ü]
©
60
0
©
0
®
0
©
0
©
0
®
3
6
9
12
18
CUTTING FREQUENCY
(wks)
Figure 5. Treatment matrix showing treatment numbers inside
each actual (circles) and absent (squares) treat¬
ment combination
N
FERTILIZATION
(kg/ha/yr)
ACTUAL ORIENTATION
OF FACTOR SPACE
(wks)
Figure 6. Treatment combinations of nitrogen (N) and cutting
frequency plotted as grid coordinates with the origin
at 120 kg/ha/yr N and 9 weeks cutting frequency

80
(2.3 x 7 m) for treatments. The entire field had 156 total plots (3
replications x 4 grasses x 13 treatments).
Figure 7 shows the treatment levels of N and F, the treatment
number, and the harvest (FI) number, date, and chronological stage of the
1980 experiment. Sampling for carbohydrate analyses followed visual
estimations of percent limpograss but preceded sampling for herbage
quality and clipping for yield. Fertilization followed clipping, and
the N rates were applied by hand as ammonium nitrate dispersed in sand
to ensure uniform application. The N application was split according to
the number of defoliations; an equal rate applied at the start of the
experiment and after every defoliation except the last. The entire
field was fertilized with 300 kg/ha 0-10-20 (N^O^-I^O) containing
microelements prior to the start of the experiment and at the halfway
point late in July.
The sampling schedule is shown by symbols in Figure 7 for IV0MD,
CP (CP = 6.25 x N), DM yield determination, and TNC measurement. Forage
cut for yield and quality was oven dried at 60 C. The dried forage was
ground in a Wiley mill to pass a 1 mm screen and analyzed for percent N
by the Kjeldahl procedure. The IV0MD determination used a modification
of the Tilley and Terry technique (Moore et al., 1972). Total nonstruc-
tural carbohydrate (TNC) samples were randomly selected, dug, washed
free of soil, and arranged in cotton bags. The cotton bags were then
packed in a plastic bag and put on ice. Upon reaching the lab, the
samples were dried in a Thelco forced-air oven at 100 C for 30-45
minutes after which the temperature was lowered to 70 C until the dried
samples were removed 36-48 hours later (Smith, 1973a). The stem bases
were scraped free of roots, leaf sheaths, and sand by using a wire

Figure
YIELD AND QUALITY J AND TNC L SAMPLING IN 1980
HARVEST,
DATE,
AND WEEK
HO HI H2 H3 H4 H5 H6 H7 H8 H9 HIO Hll HI2
Trt. N F
No. (kg/ha/yr) (wks)
Reps
¿o id 3 ( c.( id b ¿a its a ou
Mar Apr May May Jun Jul Jul Aug Sep Sep Oct Nov Nov
0 3 6 9 12 15 18 21 24 27 30 33 36
1 0
3
3
AAAAAAAAAAAAA
2 120
3
3
AAAAAAAAAAAAA
3 240
3
3
AAAAAAAAAAAAA
4 60
6
3
AAAAAAAAAAAAA
5 240
6
3
AAAAAAAAAAAAA
6 0
9
3
AAAAAAAAAAAAA
7 120
9
3
AAAAAAAAAAAAA
8 480
9
3
AAAAAAAAAAAAA
9 60
12
3
AAAAAAAAAAAAA
10 240
12
3
AAAAAAAAAAAAA
II 0
18
3
AAAAAAAAAAAAA
12 120
18
3
AAAAAAAAAAAAA
13 480
18
3
AAAAAAAAAAAAA
. Yield, quality
(IVOMD and CP), and total nonstructural carbohydrate (TNC)
sampling schedule for 1980 (open symbols indicate no sample)

82
buffing wheel attached to the extended shaft of a small electric motor
(20 W, 1525 RPM). The bottom 2 cm of the stem ("crowns") were severed
and ground through a 0.5 mm screen in a UDY Cyclone mill and stored
until tested for TNC. "Shoot" tissue was analyzed for TNC on the 27
July and 30 November dates. "Shoot" material was also ground through
the UDY mill, and all samples were stored in plastic 20 ml Dilu-vials
and then analyzed for TNC using an enzymatic hydrolysis and spectro-
photometric measurement of reducing sugars as described in Appendix A.
1981 Experimentation
The 1980 experimental area was studied in 1981 for residual effects
of cutting frequency in 1980 on 1981 DM yields. The limpograsses were
harvested for three 9 week cutting intervals on 3 June, 3 August, and
7 October 1981. No fertilizer was applied. Soil test results from
cores taken on 26 February 1981 indicated a pH of 6.2, 14.2 kg/ha P, and
46.5 kg/ha K.
Computing
Data were analyzed using the Statistical Analysis System (SAS) on
an Amdahl 470 V/6-11 with OS/MVS Release 3.8 and JES2/NJE Release 3.
Computing was performed at the Department of Agricultural Engineering
and the Northeast Regional Data Center of the State University System of
Florida, located on the campus of the University of Florida in
Gainesville.
Means were analyzed using the Waller/Duncan Multiple Comparison
Test. Analysis of variance and regression were used to test response

83
surface models for adequate fit. A complete explanation of the SAS
methodology for constructing surface plots is given in Chapter 4.
Results
Dry Matter Yields: 1979 Experimentation
Table 5 shows the DM yields obtained for limpograsses cut and
weighed four times, twice, and once for a 10 week period beginning 27
July 1979 and ending 4 October 1979. Within limpograss genotypes each
treatment had a significantly different DM production. Among grasses,
no differences were found for combined yields from plots cut every 2.5
weeks, but if grasses were allowed to grow for two cycles of 5 weeks,
Bigalta and PI 364888 had higher DM accumulation, as was seen again for
the 10 week treatment yield.
Dry Matter Yields: 1980 Experimentation
The total limpograss DM yields in the 1980 experimentation were
summed and modelled to construct response surface plots in Figure 8.
For total DM yields in 1980 at an actual treatment combination, con¬
sult Table 6. Figure 8 and Table 6 both show DM yields for PI 364888 >
PI 349753 > Bigalta > Redalta for the 480*18 (N*F) treatment. The sur¬
face plots show that Redalta and Bigalta plateaued in the surface region
focused at the 480*18 treatment combination. These two limpograsses
lodged at this combination of N and F, and it is believed that their
decumbent habit caused some loss of DM.
Varying DM yield advantages were shown for limpograsses under dif¬
ferent treatment regimes than 480*18. Bigalta, for instance, had a

84
Table 5. Total dry matter (DM) yields for four limpograsses
clipped at three different frequencies for 10 weeks
ending on 4 October 1979
Limpograss
Clipping frequency
(wks)
2.5 5 10
DM (kg/ha)
PI 364888
1756 a
3261 ab
8327 a
'Redalta'
1718 a
2328 c
4592 d
'Bigaita1
1903 a
3988 a
6616 b
PI 349753
1723 a
3133 b
5425 c
★
Clipping treatments within a limpograss genotype are
compared using the underlining technique and letters within a
column compare limpograsses for a treatment. Any values
sharing a commun underline or letter are not different (P <
0.05) using the Wal1er-Duncan Multiple Comparison Test.

TOTAL DM PRODUCTION FOR FOUR LIMPOGRASSES IN 1980
PI 364888 'Redalta' 'Bigalta' PI 349753
Figure 8. Total dry matter (DM) yield for four limpograsses subjected to five levels of nitrogen
(N) fertilization and five frequencies (F) of defoliation throughout the 1980 growing
season at the Beef Research Unit near Gainesville, Florida

Table 6. Effect of nitrogen (N) rates and cutting frequency (F) on dry matter (DM) yield, in vitro
organic matter digestibility (IVOMD), crude protein (CP), harvest of protein, andTertTTYzer
N efficiency
Fert. N
(kg/ha)
Defoliation
Llmpograss
Total
DM yield
(m ton/ha/yr)
DM yield
added by N
(m ton/ha)
Harvest
of protein
(kg/ha/yr)
N
harvested
applied N
U)
27 July
1980
30 November 1980
Stand
(*)
Freq
F
(wks)
Total
no.
caí Total
DM yield
(m ton/ha)
IVOMD
(%)
CP
(*)
DM yield
(m ton/ha)
IVOMD
(%)
CP
(%)
0 0
3
12
PI 364888
2.7
171
0.5
49.9
7.5
0
51 .9
6.6
92
Redalta
3.3
—
221
—
0.7
41.5
7.4
0
40.0
5.9
92
Bigalta
2.7
—
219
—
0.5
58.3
8.8
0
65.6
10.6
62
PI 349753
3.1
---
215
---
0.3
47.0
9.1
0
52.6
7.5
92
10 120
3
12
PI 364888
4.4
1.7
361
26
0.6
50.7
9.0
0
56.8
9.8
82
Redalta
4.7
1.4
357
18
1.0
39.8
7.6
0
42.1
8.2
90
Bigalta
3.1
0.4
281
8
0 4
58.2
9.2
0
69.5
12.2
50
PI 349753
2.9
0
225
2
0.4
45.0
7.3
0
57.5
10.3
83
40 480
3
12
PI 364888
6.8
4.1
708
18
0.5
51.8
9.7
0
61.8
12.7
78
Redalta
7.6
4.3
744
18
0.8
44.3
9.8
0
47.9
10.4
88
Bigalta
5.6
2.9
666
15
0.1
63.7
13.2
0
72.0
14.6
23
PI 349753
6.1
3.1
643
14
0.4
45.2
9.7
0
61.5
12.5
68
10 60
6
6
PI 364888
7.1
0
384
47
3.5
49.8
4.4
0.2
56.0
7.6
95
Redalta
7.4
1.8
413
37
2.6
40.8
5.1
0.2
44.1
8.3
97
Bigalta
7.4
2.2
448
32
3.7
51.1
4.8
0.1
64.2
9.3
68
PI 349753
6.4
0
371
27
2.6
43.3
5.3
0.2
54.6
8.3
97
60 240
6
6
PI 364888
11.7
4.5
731
35
4.6
48.3
4.7
0.4
60.0
10.3
100
Redalta
8.2
2.6
557
19
3.4
39.4
6.0
0.1
69.4
10.3
82
Bigalta
12.9
7.7
921
39
4.8
55.1
5.3
0.1
69.9
13.4
70
PI 349753
10.7
4.3
704
29
4.1
49.1
5.4
0.2
60.7
10.6
95
0 0
9
4
PI 364888
5.9
—
262
3.8
45.8
2.9
0.1
53.0
6.6
98
Redalta
6.5
—
329
—
2.5
41.6
4.8
0.2
45.9
7.7
98
Bigalta
9.1
—
448
—
5.0
52.2
3.0
0.2
64.3
8.6
92
PI 349753
6.8
—
324
—
4.0
45.8
3.7
0.1
49.9
6.3
98

Table 6. Continued
Fert. N
(kg/ha)
Defoliation
Limpograss
Total
DM yield
(m ton/ha/yr)
DM yield
added by N
(m ton/ha)
Harvest
of protein
(kg/ha/yr)
N
harvested
applied N
(*)
27 July
DM yield
(m ton/ha)
1980
IV0MD
(*)
CP
(X)
30 November 1980
DM yield IV0MD
(m ton/ha) (%)
CP
(X)
Stand
(%)
Freq
F
(wks)
Total
no.
Per
cut
Total
30
120
9
4
PI 364888
12.1
6.2
544
38
7.2
43.3
2.6
0.3
58.5
8.4
98
Redalta
11.6
5.2
592
35
4.4
38.3
3.9
0.4
45.9
9.0
98
Bigalta
9.7
0.7
502
3
5.5
49.4
2.9
0.3
64.6
10.6
73
PI 349753
12.4
5.6
506
38
6.3
41.0
3.5
0.3
56.9
9.6
100
120
480
9
4
PI 364888
24.4
17.5
1449
40
10.0
47.5
4.1
0.5
63.9
12.1
100
Redalta
19.1
12.6
1200
29
6.6
41.2
5.3
0.2
51.0
10.6
100
Bigalta
17.3
8.2
1229
25
7.6
50.0
5.1
0.5
67.8
13.3
93
PI 349753
17.2
10.4
1100
26
7.1
44.0
4.7
0.4
63.9
12.1
100
20
60
12
3
PI 364888
12.3
0.4
429
47
11.6
44.0
1.4
0.8
51.0
4.9
98
Redalta
11.0
0.7
384
24
4.0
38.0
3.2
1.0
40.4
4.6
97
Bigalta
9.9
0
356
0
11.6
48.2
1 .4
1.5
53.1
4.7
98
PI 349753
10.0
0
375
20
6.7
42.0
2.4
0.8
47.5
5.8
93
80
240
12
3
PI 364888
19.1
7.2
756
33
14.4
43.4
1.2
3.1
53.4
6.0
100
Redalta
16.2
5.8
707
28
9.7
36.2
2.9
1.8
42.0
6.8
100
Bigalta
12.7
2.7
584
13
15.7
48.8
1.3
2.9
61.9
7.1
100
PI 349753
20.0
8.9
830
35
10.5
41.0
2.0
3.0
52.6
6.1
100
0
0
18
2
PI 364888
10.4
252
7.2
43.1
2.4
3.2
40.8
2.3
98
Redalta
8.8
—
253
—
5.4
33.7
3.3
3.4
32.7
2.7
96
Bigalta
10.4
—
292
—
6.4
48.8
3.1
4.2
41.3
2.4
100
PI 349753
11.0
...
284
---
7.8
37.2
2.6
3.2
37.3
2.5
98
60
120
18
2
PI 364888
17.4
7.0
366
16
9.0
41.0
2.0
8.4
37.0
2.1
98
Redalta
19.0
10.1
405
21
11.1
31.5
2.1
7.9
29.2
2.2
100
Bigalta
16.0
5.5
334
6
9.5
42.7
1.9
6.5
39.3
2.3
70
PI 349753
18.6
7.5
387
14
11.4
33.3
2.1
7.2
33.9
2.0
98
240
480
18
2
PI 364888
29.3
18.9
992
25
18.3
42.0
2.4
11.0
42.3
4.3
98
Redalta
18.3
9.5
611
12
12.0
33.0
2.7
6.3
33.0
4.2
93
Bigalta
19.7
9.0
620
11
12.1
43.1
2.4
7.6
48.4
4.1
100
PI 349753
26.8
15.7
760
16
18.0
34.6
2.3
8.8
38.9
3.3
98
*
Total DM prediction equations gave DM values for 0 nitrogen at 6 and 12 week defoliations.

88
slight yield advantage in the treatment cut every 6 weeks in 1980, as
was similarly shown in 1979 for Bigalta cut every 5 weeks (Table 5).
What is clearly shown in Table 6, however, is the tremendous ability of
all the limpograsses to respond to N and accumulate DM, especially at
longer cutting intervals.
Table 6 also shows the increment of total DM yield caused by N
fertilization. Brown (1978) termed productivity of biomass per unit of
N applied as "nitrogen use efficiency." The DM production at zero N in
each cutting frequency was subtracted from the production obtained with
N applied in order to determine this statistic. It is shown that the
maximum DM response to N was obtained with high N rates and long cutting
frequencies (480*9, 240*12, 120*18, and 480*18). No clear cut advantage
was observed for any one limpograss across all treatments.
The most protein harvested in herbage was obtained from the 480*9
treatment as shown in Table 6. The actual efficiency of nitrogen uptake
was also calculated:
r N in plants - N in plants receiving zero N * lnn
*- N fertilizer applied
Limpograss PI 364888 was best for almost all treatments in this
statistic, but it was not as effective at taking up N at zero N
conditions (Table 6).
Dry matter (DM) yield distribution is as important as total produc¬
tion in Florida. The 36-week-long 1980 experiment was divided into
three 12 week sections: the first interval termed spring and the last
12 week interval called autumn. Only the 3, 6, and 12 week treatments
were included in spring and autumn because the 9 and 18 week cutting

89
frequencies did not match in harvest sequence. The graph in Figure 9
shows DM plotted against treatment combination. Spring production was
greater than autumn production, especially in the treatments cut every
3 weeks. The decline in autumn yield for 3 week treatments with
increasing N levels exemplified the effect of frequent defoliation and
high N rates on limpograss. It is interesting to note that DM yield in
the 480*3 treatment for Bigalta produced 3 m tons in spring but the
subsequent stress permitted only 5.6 m tons total DM yield (Table 6).
Clearly, PI 349753 was the highest in spring production at the
240*12 treatment combination (Figure 9). Calculations using the 240*12
treatment data from Figure 9 and the total DM data presented in Table 6
allow a determination of summer yield for PI 349753 by difference. In
this manner, PI 349753 was shown to produce 8 m tons of DM in spring,
6.9 m tons in summer, and 2.8 m tons in autumn.
Figure 10 has relevance to the use of limpograss as stockpiled
forage. The PI 364888 limpograss was used in this analysis showing
three staging dates that corresponded to the final intervals of the 9,
12, and 18 week cutting frequencies. A maximum of stockpiled forage was
obtained for the 27 July > 6 September > 28 September staging dates.
These dates agree with those of Quesenberry and Ocumpaugh (1980) who
demonstrated the necessity of early August staging for adequate DM
accumulation for stockpiling. At any staging date the second highest N
rate produced nearly as much or more forage as the highest N rate.
In the 1979 experiment, PI 364888 yielded 8.3 m tons for the single
10 week cutting interval staged from 27 July 1979 to 4 October 1979
(Table 5). Nitrogen was applied on 30 July at a rate of 38 kg/ha. In
the 1980 experiment, 60 kg/ha N was applied on 27 July 1980 for the

90
F 3 3 3 66 12 12 333 66 12 12
TREATMENT COMBINATION
Figure 9. Spring and autumn seasonal distribution of dry matter
(DM) yield for four limpograsses in 1980

91
DM Yields for PI 364888 on 30 November
kg/ha/yr
Figure 10. Dry matter (DM) yields for PI 364888 at three levels of
nitrogen (N) fertility and staged at three different
dates in the autumn of 1980
(Parentheses indicate the N(kg/ha) applied at the beginning of the
interval to give the resultant yield.)

92
120*18 treatment, and it subsequently yielded 8.4 m tons (Figure 10).
In 1981 Quesenberry and Ruelke (1982) applied 75 kg/ha N to PI 364888 on
4 August 1981 and measured 11.1 m tons stockpiled DM on 7 December 1981.
This value is nearly identical to the 11 m tons DM measured in 1980
(Figure 10) for the 480*18 treatment (240 kg/ha N applied 27 July 1980).
Dry Matter Yields: 1981 Experimentation
In 1981, residual effects from the 1980 experiment were analyzed by
measuring DM yields. The N applied later than 28 September must have
remained as residual fertilizer in the soil at the beginning of the 1981
harvest season.
The surface plots shown in Figure 11 for PI 364888 indicate the
highest yields at 480*9 and 480*3 for 3 June 1981. The 480*3 plots had
lower percent limpograss at the close of the 1980 experiment following a
year of frequent defoliation; however, by 3 August 1981 the plots had
completely recovered in all the limpograsses due to the residual N
fertilizer in the spring. Although not shown, the performance of
Redalta during the first harvest interval of 1981 was superior to the
other grasses. Redalta's 480*3 and 480*9 treatments averaged 4.2 m
tons/ha DM, followed by PI 364888 (2.9 m tons/ha), PI 349753 (2.6 m
tons/ha), and Bigalta (2.3 m tons/ha).
The 7 October harvest was not shown due to the similarity to the
second harvest on 3 August. Yields in 1981 for all limpograsses showed
similar response surface plots. The annual yields in all grasses were
highest in the 480*3 and 480*9 treatments (10-12 m tons/ha) due to the
yield advantage attained in the spring. The rest of the treatments were
similar in annual DM production within a grass, but Bigalta yielded

RESIDUAL TREATMENT EFFECTS ON DM YIELD IN 1981 FROM CUTTING
AND FERTILIZATION IN THE 1980 EXPERIMENTATION
dates total
3 June 1981 3 August 1981
Figure 11. Residual clipping and fertilization effects on dry matter (DM) yield for PI 364888
harvested every 9 weeks in 1981 (3 June, 3 August, and 7 October (not shown)) and
the total production for the year
Caution: DM yield units vary among graphs

94
higher DM (7.8-9.9 m tons/ha), followed by PI 349753 (5.9-9.3 m tons/ha),
PI 364888 (6.1-8.2 m tons/ha), and Redalta (4.7-7.4 m tons/ha).
Crude Protein: 1979 Experiment
Table 7 reveals the most disappointing characteristic of limpograss:
low CP content. In the 1979 experiment the plots were fertilized on 28
May and 30 July with 38 kg/ha N, but this was not sufficient to raise
the CP levels in the 5 and 10 week treatments to the 7 percent value
deemed necessary for maintenance in animals (Milford and Minson, 1965).
Bigalta had the highest values for CP.
Crude Protein: 1980 Experiment
In Figure 12 the CP values are shown for the middle (27 July) and
end (30 November) of the 1980 experiment. For 27 July note how quickly
CP drops from the highest values at the left front edge of the plot
(where N ranges from 0-480 kg/ha/yr in association with a 3 week cutting
frequency). Not a single value from samples older than 3 weeks of age
was above 7 percent CP (Table 6). The diploid (Redalta) and tetraploid
limpograsses were alike in this regard. The peculiar aspect of Figure
12 was the prominent differences in CP percent of limpograss between the
midsummer date and the autumn date.
Autumn temperatures allowed very little growth between the 8
November and 30 November sampling interval and, hence, in an effort to
obtain an adequate sample older, residual forage was inadvertently
included with the sample herbage in the 3 week treatments. Therefore,
in the bottom portion of Figure 12, CP values for the 3 week frequency
of cutting may have been unrealistically low. A ridge was formed at the

95
Table 7. Percent crude protein (CP) for four 1impograsses
clipped at three different frequencies for 10 weeks
ending on 4 October 1979 (Samples taken on
4 October were analyzed for CP.)
Limpograss
Clipping frequency
(wks)
2.5
5
10
-- Crude protein (%) --
*
PI 364888
6.2 c
4.9 a
3.1 b
'Redalta'
7.1 be
4.4 a
4.6 a
'Bigalta'
9.4 a
5.2 a
3.2 b
PI 349753
7.3 b
5.0 a
3.5 b
Clipping treatments within a limpograss genotype are
compared by the underlining technique and letters within a
column compare limpograss values for a treatment. Any values
sharing a common underline or letter are not different (P <
0.05) using the Waller-Duncan Multiple Comparison Test.

96
CP IN TWO LIMPOGRASSES FOR
TWO HARVESTS IN 1980
PI 364888 'Redalta'
Figure 12. Comparison of crude protein (CP) percentages in the
tissue of PI 364888 and 'Redalta1 for two dates in
1980 in response to five levels of nitrogen (N)
fertilization and five frequencies (F) of clipping
Caution: The CP scale is non-uniformly and non-linearly scaled
on 30 November 1980

97
9 week level of cutting frequency, and beyond the ridge the maturity of
the forage increased and CP values fell. Table 6 shows the substantial
improvement in percent CP between 27 July and 30 November 1980 for all
the grasses and all treatments except the 0*3 treatment and the 18 week
old samples. On the 30 November sampling date, only four of 32 means
for 3, 6, and 9 week old samples were below 7 percent CP (Table 6).
IVOMD: 1979 Experiment
Table 8 shows IVOMD data obtained from tissue collected on 4
October 1979 for the 10 week study conducted in 1979. The ranking of
the limpograsses is similar to what will be presented later from the
1980 results. The main effect of clipping treatment is also shown in
Table 8 and shows that quality was higher in the younger plant material.
IVOMD: 1980 Experiment
Figure 13 compared the IVOMD in all four limpograsses for 27 July
and 30 November 1980. Treatments did not vary by a great degree as
indicated by the topographical uniformity in the surface plots. One
apparent feature of Figure 13 is the lower IVOMD found in Redalta.
Other investigators (Schank et al., 1973; Quesenberry and Ocumpaugh
1980) have shown the low digestibility of diploids compared to tetra-
ploid limpograsses. Redalta is a diploid while the remaining three
grasses in this study are tetraploids. The 30 November plots showed a
small but perceptively higher IVOMD and somewhat greater undulation;
however, a tabular presentation was necessary for increased understanding.
In Table 6, for all treatments, the ranking of limpograsses for percent
IVOMD was Bigalta > PI 364888 > PI 349753 > Redalta. A glance at the

98
Table 8. The main effect of limpograss and clipping
treatment on percent ijn. vi tro organic matter
digestibility (IVOMD) of tissue sampled 4 October
1979
IVOMD
— % --
Limpograss
PI 364888
50.1 b
'Reda!ta1
43.0 c
'Bigalta'
55.4 a
PI 349753
50.0 b
Clipping frequency (wks)
2.5
51.6 a
5
50.4 b
10
46.9 b
•k
Values sharing a common letter within the column are not
different (P < 0.05) using the Waller-Duncan Multiple Compari¬
son Procedure.

IVOMD IN FOUR LIMPOGRASSES FOR TWO HARVESTS IN 1980
PI 349753
Figure 13. Comparison of in vitro organic matter digestibility (IVOMD) for four limpograsses at two
1980 harvest dates subjected to five levels of nitrogen (N) fertilization and five fre¬
quencies (F) of clipping

100
values across all treatments in the 27 July sampling explains the flat¬
ness of the surfaces in Figure 13. In the 30 November sampling there
was a greater range of IVOMD values across treatments but the surface
plots in Figure 13 were topographically too flat for easy detection of
differences by the eye.
The increase in limpograss quality between 27 July and 30 November
was reflected in the Table 6 IVOMD values. It was decided to examine
the IVOMD values for Bigalta and PI 364888 more closely in Table 9 in
order to compare the increases in digestibility from harvest 6 to
harvest 12 (H6 to HI2). For PI 364888 the quality increased in all but
the 18 week treatments and nearly identical trends were shown in Bigalta.
The largest differences between H6 and HI2 were found in the 6 and 9
week old tissue in both grasses. Differences in IVOMD between Bigalta
and PI 364888 were presented in Table 9 in order to demonstrate the
consistently higher IVOMD of Bigalta. The advantage for Bigalta over
PI 364888 was much more dramatic on the 30 November (HI2) sampling date
than on 27 July 1980 (H6).
Total Nonstructural Carbohydrate as Related to IVOMD
Heavily fertilized limpograss in the autumn appeared very lush com¬
pared to the stiffer stems from mid-season, and it was thought that
accumulated starch might explain some of the autumn IVOMD increases in
this study. Garrard and West (1972) and Carter and Garrard (1976)
showed that starch accumulated in the leaves of digitgrass during cool
nights. Response surfaces for "shoot" TNC were constructed for Bigalta,
Redalta, and PI 364888 from data collected on 27 July and 30 November
(Figures 14 and 15). In Figure 14 a rising plane is shown for PI 364888.

MIDSUMMER PERCENT TNC FOUND IN THE SHOOT TISSUE OF
THREE LIMPOGRASSES
PI 364888 'Redalta' 'Bigalta
% TNC
Figure 14. Percent total nonstructural carbohydrate (TNC) in the "shoot" tissue of three
limpograsses on 27 July 1980 subjected to five levels of nitrogen (N) fertilization
and five defoliation frequencies (F)

LATE AUTUMN PERCENT TNC FOUND IN THE SHOOT TISSUE
OF THREE LIMPOGRASSES
'Redalta' PI 364888 'Bigalta'
Figure 15. A comparison of total nonstructural carbohydrate (TNC) in the "shoots" of three
limpograsses subjected to five levels of nitrogen (N) fertilization and five
frequencies (F) of defoliation sampled on 30 November 1980

103
Redalta has a rising plane of a much shallower slope and of a lower
magnitude than PI 364888. The Bigalta surface plot rose precipitously
to a ridge at the 12 week cutting frequency and then the TNC in "shoots"
fell back to a lower value at the rear edge of the cube (18 week clip¬
ping treatments). In Figure 15 Redalta and PI 364888 looked almost the
same as for the July surfaces. Bigalta, on the other hand, increased
carbohydrate content for 3 week treatments (front left of the plot);
decreased TNC in 6, 9, and 12 week treatments greatly; and decreased the
TNC percent in 18 week treatments very little. These effects inverted
the response in Bigalta at 30 November from what was shown on 27 July
1980. These interpretations were verified by the actual "shoot" TNC
percentages shown in Table 9. The occurrences described above were
incongruous with the hypothesis that autumn increases in "shoot" TNC
would increase percent IVOMD. What was observed was a decrease in
"shoot" TNC and an increase in IVOMD especially in the 6, 9, and 12 week
treatments.
The variation of TNC content in the "crowns" and "shoots" between
27 July and 30 November was studied in hopes of clarifying the relation¬
ship, if any, between TNC, IVOMD, and CP. In Table 9 the TNC accumula¬
tion patterns for PI 364888 between H6 and HI2 showed an increase in the
"crowns" and a decrease in the "shoots" for all except the 480*3 treat¬
ment combination. For PI 364888 it was logical to conclude that TNC
produced in "shoots" in the summer was translocated to the "crowns" for
storage in the autumn. In Bigalta a different trend emerged. The
"shoot" TNC values decreased from H6 to HI2; however, contrary to the
situation in PI 364888, the Bigalta "crown" percent TNC also decreased

Table 9. Balance sheet of in vitro organic matter digestibility (IVOMD) and total nonstructural carbo¬
hydrates (TNC) in two limpograsses
Treatment
Limpograss
IVOMD
Harv PI 364888
Bigalta
difference
N/cut
kgha”^
Total N Cut
kgha yr freq
No. of
TNC (%) IVQMD
TNC (%)
IVOMD
(Bigalta
mi nus
harvests
Shoot Crown ^
Shoot Crown
(%)c
PI 364888)
0
0
3
12
H6
8.1
10.5
49.9
9.9
11.4
58.3
8.4
HI 2
7.0
16.2
51.9
9.6
14.3
65.6
13.7
-1.1
+5.7
+2.0
-0.3
+2.9
+ 7.3
10
120
3
12
H6
7.6
10.9
50.7
9.7
11.3
58.2
7.5
HI 2
7.4
17.1
56.8
10.0
13.9
69.5
12.7
-0.2
+6.2
+6.1
+0.3
+0.6
+ 11.3
40
480
3
12
H6
6.6
10.2
51.8
8.1
9.9
63.7
11.9
HI 2
8.9
13.8
61.8
10.9
11.5
72.0
10.2
+2.3
+3.6
+10.0
+2.8
+1 .6
*8.3
10
60
6
6
H6
11.1
13.0
49.8
11.6
13.6
51.1
1.3
HI 2
CO
14.6
56.0
10.1
13.4
64.2
8.2
-3.3
+1.6
+6.2
-1.5
-0.2
+13.1
60
240
6
6
H6
9.6
13.4
48.3
11.2
9.6
55.1
6.8
HI 2
8.8
14.1
60.0
7.7
7.9
69.9
9.0
-0.8
+0.7
+11.7
-3.5
-1.7
+14.8

Table 9. Continued
Treatment
Limpograss
IVOMD
Harv PI 364888
Bigaita
difference
N/cut
kgha"^
Totai N Cut
kgha yr freq
No. of
TNC (%)
IVOMD
TNC {%)
IVOMD
(Bigalta
minus
harvests
Shoot Crown
(%)
Shoot Crown
(%)
PI 364888)
0
0
9
4
H6
11.8
12.7
45.8
16.1
11.3
52.2
6.4
HI 2
8.9
13.9
53.0
7.6
6.0
64.3
11.0
-2.9
+1.2
+7.2
-8.5
-5.3
+12.1
30
120
9
4
H6
12.4
10.4
43.3
16.2
10.6
49.4
6.1
HI 2
9.4
11.8
58.5
8.1
7.2
64.6
6.1
-3.0
+ 1.4
+15.2
-8.1
-3.4
+15.2
120
480
9
4
H6
12.4
8.3
47.5
16.3
10.4
50.0
2.5
HI 2
9.9
11.7
63.9
7.8
9.3
67.8
3.9
-2.5
+3.4
+16.4
-8.5
-1.1
+17.8
20
60
12
3
H6
13.0
*
44.0
18.9
*
48.2
4.2
HI 2
8.3
16.3
51.0
9.3
12.3
53.1
2.1
-4.5
*
+7.0
-9.6
*
+4.9
80
240
12
3
H6
13.2
*
43.4
18.2
*
48.8
5.4
HI 2
8.9
15.7
53.4
9.7
11.2
61.9
8.5
-4.3
*
+ 10.0
-8.5
*
+13.1

Table 9. Continued
N/cut
kgha ^
Treatment
To‘f " Cut
kgha yr freq
No. of
harvests
Harv
no.
PI 364888
TNC (%)
Shoot Crown
Limpograss
IV0MD TNC
^ Shoot
Bigaita
(%)
Crown
IV0MD
difference
(Bigaita
IV0MD minus
(%) PI 364888)
0
0
18
2
H6
12.4
10.2
43.1
11.1
13.9
48.8
5.7
Hi 2
8.9
16.8
40.8
10.1
11.2
41.3
0.5
-3.5
+6.6
-2.3
-1 .0
-2.7
-7.5
60
120
18
2
H6
12.3
11.9
41.0
11.2
12.2
42.7
1.7
HI 2
9.7
17.9
37.0
9.5
11.4
39.3
2.3
-2.6
+6.0
-4.0
-1.7
-0.8
-3.4
240
480
18
2
H6
15.0
10.7
42.0
10.7
11.1
43.1
1.1
HI 2
9.8
14.7
42.3
10.5
10.2
48.4
6.1
-5.2
+4.0
-0.3
-0.2
-0.9
+5.3

107
(or remained the same) for all treatments except those cut every three
weeks.
The question was, "What was responsible for the increase in quality
seen in all limpograsses between midsummer and late autumn?" From the
above data for limpograss it was postulated that the high levels of TNC
in summer "shoot" tissue along with very low CP contents may have
allowed complete digestion of the nonstructural carbohydrates but
restricted the rumen bacteria from degrading the cell wall materials.
In the summer all the absorbed N would be sequestered by rapid growth
and tissue synthesis, as opposed to autumn when growth rates decline.
Hence, in autumn it was thought that TNC was partially bound up in com¬
pounds such as protein in the stems or leaves that was not measured in
TNC analyses but was readily digested by rumen microorganisms in tests
of IVOMD. The hypothesis was especially attractive for Bigalta because
of the large decreases of TNC in shoots for 6, 9, and 12 week treatments
along with concomitantly large increases in IVOMD and CP for the same
treatments between the two dates. The IVOMD and CP changes between H6
and H12 for all four limpograsses can be studied in Table 6. The same
situation was observed to a lesser degree in PI 364888; however, this
plant appeared to partition more TNC into the storage organs than
Bigalta. This explanation was appealing because of the excellent
persistence reported for PI 364888 and the poor persistence reported for
Bigalta (Quesenberry et al., 1978).
Seasonal TNC Trends
The primary objective of using the TNC analyses was in studying
storage of carbohydrates in limpograss. The means for all limpograsses

108
for any given harvest date and treatment are shown in Appendix B. The
overall seasonal trends can be constructed given this information. The
trends were (1) highest TNC in the spring, (2) lowest TNC in midsummer,
(3) an increase in TNC from August to mid-September, (4) a decrease in
TNC to November, and (5) a sharp increase again from December until
spring. The TNC increases in the autumn were most predominant for
PI 364888.
The decrease of stored TNC in "crowns" in summer may be associated
with rapid growth rates and maximum day and night temperatures causing a
severe demand on TNC in the storage organs. In summer TNC was also
highest in the "shoots" because of active growth and photosynthesis
(McCarty, 1935; Brown and Blazer, 1965; Blaser et a!., 1966). As night
temperatures cooled off in early autumn, the respiratory demand may have
decreased and active photosynthesis in the day allowed a net storage of
TNC. Wardlaw (1968) stated that, "Growth of established shoots has
priority over root and bud growth under conditions of assimilate
deficiency" (p. 86). After reserves began to increase again in early
autumn, the tillers and roots began active growth; however, this new
growth would require substrate which may explain the small decrease of
TNC in "crowns" between September and November. In this study TNC may
have been translocated to sites of root and tiller synthesis instead of
becoming localized in the shoots as observed by Garrard and West (1972)
and Carter and Garrard (1976). Chatterton et al. (1972) studied tiller¬
ing as a variable on the effect of cool night temperatures on Pangóla
digitgrass. They showed that actively tillering plants accumulated no
starch, while non-tillering plants accumulated starch during the day
that localized in the chloroplasts during cool nights.

109
Wardlaw (1968) stated that temperature affects growth (i.e.,
development) greater than photosynthesis and translocation. The final
surge of carbohydrate shown in limpograss "crowns" during December and
beyond was possibly caused by the photosynthesizing basal tillers during
warm days in the cool season.
Total Nonstructural Carbohydrates and Persistence
Another intention of using TNC analysis was to study differences in
persistence observed in the 1impograsses. In Figure 7 the sampling
schedule for TNC shows that complete sampling for all treatments was
begun on 6 September (H8). Until that point the treatments were allowed
to take their effect. In Figure 16 data obtained from "crowns" har¬
vested on 28 September and 30 November from PI 364888 and Bigalta are
plotted. A predominant ridge along the front left edge of all plots was
explained by the change in morphology effected by clipping every 3 weeks.
Plants were prostrate, had more tillers per unit area, and hence, high
photosynthetic activity was able to provide assimilates for storage in
the stolons and stem bases. A trough or basin shown for both PI 364888
and Bigalta in the 28 September plots was caused by the 6 week treat¬
ments in this portion of the response surfaces. These treatment com¬
binations were not short enough to promote a prostrate habit and not
long enough to allow complete stem elongation. Elongation requires
carbohydrate compounds, and before the shoots were able to resupply the
storage organs with photosynthate, it was defoliated and the cycle began
again. Beyond the 9 week level of cutting frequency the TNC concentra¬
tion in "crowns" increased gradually for 28 September samples, whereas

no
TNC IN STEM BASES OF TWO LIMPOGRASSES
FOR TWO HARVESTS IN 1980
PI 364888 'Bigalta
28 SEPTEMBER 1980
Figure 16. Percent total nonstructural carbohydrate (TNC) in the stem
base ("crowns") of two 1impograsses on two dates as affected
by nitrogen (N) fertilization and defoliation frequency (F)

m
the increase was more dramatic for 30 November, ending with a plateau
along the rear edge of the surface plots.
In Figure 17 the TNC percent of the 3 week treatments is plotted
across the 1980 season for PI 364888 and Bigaita. The effect of N
caused an increase in TNC in the "crowns" early in the year when stored
carbohydrate compounds were able to provide the carbon skeletons for
protein and tissue synthesis. The rapidly growing tissue photosynthe-
sized enough carbohydrate in Bigalta in the 480*3 treatment to maintain
more stored substrate than the other levels of N until the fourth
harvest. At this time the TNC values dropped in both limpograsses as
respiration and growth exerted a heavy demand on stored TNC.
From midsummer onward in Figure 17 the 480 N level caused a depres¬
sion in stored TNC for the 3 week treatments. The sharp rise of TNC at
H8 indicated that environmental stress had diminished in comparison to
midsummer. New root and tissue synthesis was most likely the cause of
the H8-H11 TNC decline. The TNC increased again in the November sampl¬
ings of PI 364888 but Bigalta did not respond in the same manner, as was
shown in November samples from the 1979 experiment (Chapter 2, Table 4).
Sampling logistics prevented an analysis of actual quantities of
TNC per unit area of plot. Only percent TNC was measured and, as shown
in Figure 17, there were no large differences between the persistent
(PI 364888) and nonpersistent (Bigalta) limpograsses in percent TNC.
Visual estimates of percent limpograss, however, did permit an analysis
of persistence (Figure 18).
Figure 18 shows how the 480 N level caused rapid growth early in
the season, as indicated by the high percent of limpograss in both
stands until H4. Midsummer stress caused by frequent defoliation, high

112
Figure 17. A comparison of total nonstructural carbohydrate (TNC)
percent in the "crowns" of two limpograsses subjected
to 3 week defoliation frequencies (F) and fertilized
at three levels of nitrogen (N) during the 1980 growing
season

113
VISUAL ESTIMATIONS
1980 Experiment
Figure 13. Visual estimations of percent 'Bigalta' and PI 364888
for treatments having 3 week defoliation frequencies
(F) and three levels of nitrogen (N) in the 1980
growing season

114
fertilization, as well as TNC demands from growth and respiration caused
a tremendous loss of limpograss in plots. The loss of Bigalta was much
greater because this grass had fewer stolons to serve as a carbohydrate
reservoirs when N was applied. Tissue degradation and competition from
summer weeds further depressed percent Bigalta and PI 364888 until H10
when the summer weeds died and the limpograss continued growing; there¬
fore, becoming a greater percent of the vegetation in the plots. This
effect was present for all three treatments.
In Figure 19 the 1980 temperature and rainfall data are presented.
The weather data was averaged for the 3 week periods prior to each
harvest in order to represent the temperature and rainfall regimes under
which the samples were taken. It was hoped that weather data could help
explain the TNC peak centered about H8 in Figure 17. The maximum and
minimum average daily temperatures for the 3 week period prior to H8
were almost as high as in midsummer; however, these data did not reflect
the duration of heat for a summer day as opposed to an early autumn day.
Hence, even though the minimum and maximum average daily temperatures
did not appreciably decline until the beginning of October (H9-H10), the
actual temperature stresses on the limpograsses probably diminished
prior to October.
Water deficits have been reported to cause increases in stored TNC
but the 3 week period before H8 had more rain (14 cm) than most of the
3 week periods in the 1980 harvest season (Figure 19). Hence, lack of
water cannot be used to explain the TNC increase at H8.
In Figure 18 the percent limpograss began to rise again between H7-
H8 (16 August-6 September) for the 0*3 and 120*3 treatments in PI 364888
and in the 0*3 Bigalta treatment. This increase in limpograss percent

115
HARVEST
date
Figure 19. Rainfall and temperature data for the 1980 growing season
taken at the Beef Research Unit near Gainesville, Florida

116
of stands matches the autumn TNC peak at H8 in Figure 17. Further
increases were made in percent limpograss in stands once the annual
summer weeds died out at H10. Between HI0 and HI 1 the average minimum
daily temperature dropped below 10 C (Figure 19) and the limpograsses
grew very slowly. Green tissue was present throughout the winter
period, however, even after freezing temperatures.
In Figure 20 the visual estimations for all limpograsses between
14 June and 6 September (H4, H6, H8) were plotted as response surfaces.
Notice how limpograss dominated sward composition on 14 June but by 27
July had diminished appreciably for the 3 week cutting frequency
treatments. By 6 September the non-persistent limpograsses (Bigalta and
PI 349753) were identified and their persistence declined further at
28 September. Values for percent limpograss at the end of the 1980
experiment on 30 November may be found in Table 6. It was once again
believed that the persistence of both Redalta and PI 364888 was due to a
larger reservoir of carbohydrate imparted by morphological differences
amongst the grasses.
Discussion
The top DM yields obtained for the limpograsses in this study were
excellent. Limpograss PI 364888 produced a total of 29.3 m tons in 1980
and this was not matched in the literature. If Killinger (1971) had
continued his measurement of yield beyond August, his Bigalta yields
might have increased beyond the 27 m tons he reported. By August the
limpograsses in this study produced 12-18 m tons (Bigalta and Redalta
produced 12 m tons).

117
VISUAL ESTIMATES OF SWARD COMPOSITION OF
FOUR LIMPOGRASS GENOTYPES FOR THREE
DATES IN 1980
PI 364888 'Redalta' 'Bigalta' PI 349753
Figure 20. Visual estimates of four limpograsses subjected to five
levels of nitrogen (N) fertilization and five frequencies
(F) of clipping across three summer harvests

118
The yield potential of limpograss would be unimpressive if it were
not for its autumn and spring production. There is no lack of poor
quality biomass in Florida in the summer. Cool-season production is a
greater problem in north Florida than further south. Colder tempera¬
tures in north Florida restrict the growth of some improved forages that
do well in south Florida. Limpograss is well adapted (Oakes, 1980).
Stockpiled limpograss may not be the answer for limpograss utilization
even though it is an inexpensive technique. Limpograss is supposedly
difficult to cure as hay but there is nothing in the literature to
suggest anyone has tried. The increasing acceptance of round bales as
well as hay ammonification procedures could improve the utilization of
limpograss to take advantage of the tremendous potential for DM
accumulation. Ruelke and Quesenberry (1982) are now studying the early
application of N to produce large quantities of limpograss forage in
the spring. Big bales could preserve this material before summer rains
in June.
Ammonification sounds intuitively more appealing than late N fer¬
tilization (Blue et al., 1961; Kretschmer, 1965) as a way of increasing
the N content of mature limpograss. Another method to increase N in
limpograss forage is to mix legumes in the stand. With no N applied the
limpograss canopy is very open and might be conducive to the association
of a legume such as joint vetch (Aeschynomene americana L.). Work is
presently initiated in this regard. The limpograsses have recently been
attributed an alleopathic effect (Ruelke and Quesenberry, 1981; Young
and Bartholomew, 1981; Tang and Young, 1982) which may inhibit
weeds and legumes as well.

119
Limpograss PI 364888 has the potential to add flexibility to the
flatwoods forage system. If producers have bahiagrass pastures they can
graze PI 364888 in spring and autumn (Ocumpaugh, 1982). Bahiagrass
could be grazed in the summer and limpograss could accumulate DM until
the summer rains end, at which time it could be baled (if ammonification
works) or ensiled and then staged for stockpiling until October when the
bahiagrass stops its growth. When winter ensues the limpograss silage
or hay could be fed.
Table 10 is presented at this point to show the treatment combina¬
tions representing several combinations of yield and quality in the
present study. It exemplifies the high quality attained in PI 364888
without the production of DM, or the high yield without adequate quality.
Clearly a protein supplement must be injected into a limpograss feeding
program that utilizes some of the combinations shown in Table 10.
Florida has both a summer and winter slump (Mott and Moore, 1977)
in which cattle perform poorly. The summer slump could be considered
both an animal and plant problem. Midsummer CP and IVOMD values
reported for limpograss on 27 July of this study were as poor as any
that were found for limpograss in the literature. The samples for
quality evaluation were clipped at 5 cm and as such would not represent
what an animal would selectively consume; however, the objective was not
to inject sampling bias by plucking rather than uniformly sampling the
plants. The IVOMD analyses were conducted confounding replications with
laboratory runs and included internal plant standards whose results did
not vary from their accustomed values. In the Moore et al. (1972) modi¬
fication to the Tilley and Terry two stage test for IVOMD, soybean meal
is fed to the fistulated steer a short time prior to collecting the

Table 10. Treatment combinations in PI 364888 representing the best compromise between yield and quality
in the 1980 experiment
FERT
N
Defoliation
Date
of harvest
V ar*
Url
Total
Per
Total
FREQ
No.
3
25
14
7
27
16 6
28
18
11
30
cut
of
of
May
May
Jun
Jul
Jul
Aug Sep
Sep
Oct
Nov
Nov
-(kg/ha) -
(kg/ha)
60
240
6
6
DM
0.6
2.7
4.6
2.1
0.4
0.3
10.7
IVOMD
57.3
46.5
48.4
48.7
52.0
59.7
CP
6.0
3.5
4.7
5.5
7.1
10.2
120
480
9
4
DM
6.1
10.1
6.8
0.4
23.4
IVOMD
49.7
47.6
45.8
63.9
CP
4.5
4.1
4.0
12.1
80
240
12
3
DM
6.7
9.3
3.1
19.1
IVOMD
42.9
39.6
53.5
CP
2.8
3.1
6.0
240
480
18
2
DM
18.3
11.0
29.3
IVOMD
42.0
42.3
CP
2.4
4.3
*DM (m ton/ha); IVOMD (%); CP (%).

121
rumen fluid for the test. It is possible that the soybean meal did not
supply sufficient N to the rumen bacteria in vitro, therefore causing
a limitation on the normal degree of cell wall degradation.
Jolliff et al. (1979) showed that forage of the same chronological
age can vary widely in nutritive value when sampled at different times
of the year. The study, conducted in Texas, had a similar seasonal
temperature regime as present in Gainesville, Florida, and it was found
that percent changes in CP and digestibility in 8 week old forage varied
by as high as 4.6 and 12.7 percentage units within a 30 day period.
These findings support the seasonal changes of percent CP and IVOMD
found between 27 July and 30 November in this study. The higher quality
in the autumn could be due to slower maturation in lower temperature
regimes (Wilson and t'Mannetje, 1978).
The seasonal flux of TNC in this study was similar to that shown by
wiregrass (Aristida stricta Michx.) in the western sand hills of Florida
(Woods et al., 1959). Carbohydrate reserves in wiregrass roots were in
least supply during mid-July, and were highest around the first of
February. In Australia Wilson and t'Mannetje (1978) showed a TNC
decrease in summer and higher levels in the spring for leaves of buffel-
grass (Cenchrus ciliaris cv.'Bioloela') and green panic (Panicum maximum
var. trichoglume cv. 'Petrie'); however, the magnitude of seasonal TNC
fluctuations were far less (8-12 percent) than observed here for limpo-
grass (4.2 percent in PI 349753 (480*9), 18 October to 26.0 percent in
Bigalta (480*3), 16 April).
The TNC studies did not entirely explain the loss of persistence in
Bigalta and PI 349793. The literature warned of the pitfalls of percent
data in TNC studies (Humphreys, 1966; May, 1960), but the logistical

122
problems were too immense to consider the extra work involved in produc¬
ing TNC data on actual amounts of carbohydrate per unit area or per
shoot. Visual estimations of percent limpograss were taken on all plots
prior to each harvest by the same experimenters. This method was reli¬
able enough to characterize the treatment effects on the limpograss
persistence.
The results of this study showed that Bigalta was superior in
quality and is an excellent grass when properly managed. This conten¬
tion is borne out by the fact that it recovered so quickly in 1981 from
below 20 percent of stands after stressful treatments in 1980. Redalta
had the worst quality and has a limited future. Careful management is
also necessary in order for PI 349753 to persist. Of all limpograsses,
PI 346888 had higher yield and persistence than Bigalta and PI 349753;
higher yield and quality than Redalta and PI 349753; and has been recom¬
mended for cultivar release in 1983.
Conclusions
1. The four limpograsses included in this study had different
advantages in yield potential at various levels of N and F. Bigalta
produced well at intermediate cutting frequencies, while PI 364888 pro¬
duced more DM with high nitrogen and long cutting frequencies.
2. The four limpograsses differed in their seasonal distribution
of DM yield. Bigalta was lowest in spring production and PI 349753 was
highest. Stockpiled forage from 27 July 1980 to 30 November was great¬
est for PI 364888.
3. Values for IVOMD were lower than shown in other published
reports, especially for samples taken during midsummer. The IVOMD

123
percents increased in all grasses in the autumn and throughout the year
Bigalta was superior in IVOMD percent than the other limpograss
genotypes.
4. Crude protein was uniformly low in all genotypes, especially
during midsummer. As with the IVOMD values, CP increased in the samples
collected in the autumn. Bigalta was highest in CP content.
5. Studies of TNC revealed a maximum percent of organic food
reserves in stem bases in March and a low in July. As temperature
stressed decreased in September and October, TNC increased and then
decreased again until November when new roots and tillers may have been
using the newly accumulated TNC. After November a slow accumulation of
TNC occurred through the winter.
6. The 3 week cutting frequency and 480 kg/ha/yr N rate caused
limpograss deterioration in the plots. Bigalta and PI 349753 were most
susceptible to the treatment stresses while Redalta and PI 364888 were
most persistent.
7. The period of low TNC reserves in midsummer coincided with the
period of most severe limpograss deterioration in the plots. Perhaps
the high amounts of absorbed N in the plant demanded more carbohydrate
for tissue and protein synthesis than the plants were able to supply.
Weed encroachment and competition for nutrients, light, and water were
contributory factors in the decline of limpograss stands.
Summary
Research on limpograss (Hemarthria altissima (Poir) Stapf et C.E.
Hubb) has identified the potential of this grass for flatwood forage-
livestock systems due to better quality and dry matter yield potential

124
than bahiagrass (Paspalum notatum Fliigge). A current problem is that
the limpograss cultivars 1Bigaita1 and 'Redalta' have attributes that
make them less than ideal forages. Recent evaluations of PI 364888
suggest that this grass adequately combines the desirable characteris¬
tics of yield, persistence, quality, and animal acceptability. The
present study was designed to thoroughly understand why, how, and in
what categories the new limpograss was superior to the already released
cultivars.
Results from an extensive field study and thorough laboratory
analyses showed that PI 364888 did, indeed, possess better attributes
for dry matter (DM) yield and persistence than Bigaita; better digesti¬
bility and yield than Redalta; however, it fell short of Bigaita in
crude protein (CP) and in vitro organic matter digestibility (IVOMD).
Seasonal DM distribution and total nonstructural carbohydrate (TNC)
flux were also studied as influenced by a complete array of nitrogen
fertilization (0, 60, 120, 240, and 480 kg/ha/yr) and a wide range of
defoliation frequencies (3, 6, 9, 12, and 18 weeks) in a modified
central composite, response surface design.
Spring DM production was highest for another new limpograss
(PI 349753) that yielded 8 m tons in the 12 week period prior to 14 June
1980. Bigalta had the lowest spring yields and was the most frost sus¬
ceptible in the autumn. Limpograss PI 364888 produced 11 m tons of DM
when staged to grow as a stockpiled forage from 27 July to 30 November
1980, and also produced the most annual DM (29 m tons).
Total nonstructural carbohydrate (TNC) in the storage organs
decreased from a high in March to a low in July. In vitro organic
matter digestibility (IVOMD) and CP values in tissue of equivalent

125
chronological age varied seasonally with regard to quality. Crude
protein in samples taken during the summer were extremely low in all the
1 impograsses— well below the 7 percent necessary to maintain body weight
in ruminant animals. The midsummer period was marked by lowest quality,
highest stress on the reserves, and most marked decreases in percent
limpograss in the stands.

CHAPTER 4
STATISTICAL ANALYSIS SYSTEM (SAS) METHODOLOGY
FOR CONSTRUCTING RESPONSE SURFACE GRAPHICS
Introduction
The characterization and quantification of relationships between
"independent" and "dependent" variables was traditionally approached by
testing one factor at a time while holding other factors constant. It
was recognized that information of much wider generality could be pro¬
duced if several factors were investigated simultaneously (Hader et al.
1957).
Complete factorals are one statistical approach to understanding
complex relationships among experimental factors but these designs con¬
sume profligate amounts of time, resources, and money because of the
exponential explosion of treatment combinations with greater than three
factors. Fractional factorials represent some improvement in reducing
the size of the experiment, but they require extreme care in avoiding
confounded effects because of pratial internal replication (Cochran and
Cox, 1957).
Box and Wilson (1951) were the first to develop and describe the
response surface design. Response surface methodology decreases the
large number of treatment combinations necessary in factorial
experiments. The first applications were in the chemistry and engineer
ing fields. By 1966 the literature was full of diverse applications in
a variety of fields (Hill and Hunter, 1966).
126

127
A factorial arrangement of design points for a two factor experi¬
ment are shown in Figure 3, and Box and Wilson's (1951) composite
design in three factors is graphically represented in Figure 4. The
latter has the property of symmetry when viewed from any axis.
According to Hader et al. (1957), Box and Hunter thoroughly explored
designs for fitting second order surfaces and eventually produced a new
class called rotatable designs. Rotatable designs are defined as those
whose estimations are equally reliable at any equal distance from the
design origin. Hence, the standard error of the estimated response is
dependent on distance rather than direction.
Rotatability is a reasonable property for exploratory work when
the experimenter does not know in advance how the response surface will
look. Consequently, he/she has no rational basis for specifying that
the standard error of the estimate should be smaller in some directions
than in others (Cochran and Cox, 1957).
In some industrial applications of response surfaces a series of
short experiments could be planned so as to use the results of the first
to plan the treatments for the second. A perpendicular course up or
down the contours would head the experimenter towards the desired
"optimum." This path of steepest ascent or descent could be used to
"fix" the origin in the second and subsequent experiments (Cochran and
Cox, 1957; Myers, 1971; Box et al., 1978). Canonical analysis is
another method used in sequential testing that consists of shifting the
origin to a new point and rotating the axes so they correspond to the
axes of the contours. When the response surface is oriented to the new
set of axes, the second order equations are greatly simplified and their

128
topographical nature becomes more obvious (Box et al., 1978; Myers,
1971).
In lines of work where an experiment must extend over a long period
of time in order that the treatments produce their effects, the natural
strategy is to try to discover the optimum combinations at the end of a
single experiment (Cochran and Cox, 1957). Some experimenters have
modified response surface designs in order to have treatment combina¬
tions that better fit the needs of their situation. The modifications
are usually in the form of adding design points outside the experimental
region of the formal design, or to shift points around within the
experimental region to concentrate information in certain areas of
interest (Littell and Mott, 1975). Mott (1982a) has referred to designs
of this sort as modified central composite designs.
Littell and Mott (1975) described the use of contour diagrams using
SAS; Henderson and Robinson (1982) presented but did not describe the
formation of SAS surface plots. Schoney et al. (1981) described contour
and response surfaces from computer graphics, but their University of
Wisconsin-Madison computing package is not as widely available as SAS.
The objective of this paper is to show a "cookbook" procedure using
SAS (Council and Helwig, 1979) and SAS/GRAPH (Council and Helwig, 1981)
in plotting various surface plots for agronomic variables in a modified
central composite design. Tests for lack of fit (LOF), precautions, and
some pitfalls are discussed.
Materials and Methods
Vegetative sprigging was used to establish four limpograss geno¬
types on an Adamsville sand; a poorly drained siliceous hyperthermic

129
ultic haplaquod soil at the Beef Research Unit of the University of
Florida. The grasses were PI 364888, PI 349753, 'Redalta1, and
'Bigalta'. Florida Experimental Station accession numbers were 297,
886, 553, and 554, respectively.
Establishment Year: 1979
The experimental area was 30 x 118 m and supported a mature stand
of rye (Secale cereale L.) prior to cultivating with a Ground Hawg roto-
tiller on 22 May 1979. The field was blocked into three sections, and
within each third, four main plots were marked to measure 7 x 30 m with
3 m alleys between each main plot. The alleys were planted to
'Argentine' bahiagrass (Paspalum notatum FlLigge).
Waist-high stands of each limpograss were cut at 7.5 cm height,
raked, and carried to an appropriately assigned, random location within
each block. The herbage was evenly distributed over the soil, lightly
disked to 15 cm, and the whole field was rolled using a cultipacker
seeder to assure good soil to stem contact. The field was fertilized on
29 June and 23 August 1979 with 336 kg/ha 17-5-10 (N-P^O^-K^O) contain¬
ing 1 percent of a microelement mix. Soil test results from 7 June
showed a pH of 6.2, 20.3 kg/ha phosphorus (P), and 40 kg/ha potassium
(K). The field was flail chopped on 6 December 1979 to 7.5 cm height in
order to remove frosted forage.
Experimental Year: 1980
Soil test results from samples taken on 6 March 1980 revealed a pH
of 6.1, 16.4 kg/ha P, and 53.1 kg/ha K. On 23 March 1980, the field was
uniformly mowed to 7.5 cm height and treatments consisting of five

130
levels of nitrogen (N) fertilization and five cutting frequencies (F)
were imposed on the main plots. The treatments represented 13/25of a
5x5 factorial as depicted in Figure 5. Grasses were blocked (or
replicated) as main plots in a split plot design, and the treatments
provided a response surface with all points replicated three times.
Each main plot contained 13 plots (2.3 x 7 m) for treatments. The
entire field had 156 total plots (three blocks x four grasses x 13
treatments).
The treatment combinations represented a modified central composite
response surface design that maximized information in what was consid¬
ered the "management realm" (0-120 kg/ha/yr N and 3-9 week defoliation
frequencies). Note the geometric scaling of N and the arithmetic pro¬
gression for cutting frequency (with a skip at the 15 week level) in
Figure 6. As the dotted lines in Figure 6 suggest, the design could be
considered a superimposition of two complete factorials (2 and 3 ).
Figure 7 shows the treatment levels of N and F, the treatment
number, and the harvest number, date, and chronological stage of the
1980 experiment. Sampling for carbohydrate analyses followed visual
estimations of percent limpograss but preceded sampling for herbage
quality and yield. Fertilization followed clipping, and the N was
applied by hand as ammonium nitrate mixed with sand. The N application
as split according to the number of defoliations; an equal portion
applied at the start of the experiment and after every defoliation
except the last.
The sampling schedule is shown by symbols for in vitro organic
matter digestibility (IVOMD), N analysis, dry matter (DM) yield deter¬
mination, and total nonstructural carbohydrate (TNC) measurement.

131
Full information on all treatments is necessary for the formation of
surface plots and stem base ("crown") data obtained from TNC samples
taken on 28 September 1980 satisfied this requirement in the example to
follow. The TNC analysis is fully described in Appendix A.
Data were analyzed using SAS on an Amdahl 470 V/6-11 with OS/MVS
Release 3.8 and JES2/NJE Release 3. Computing was performed at the
Northeast Regional Data Center and the Agricultural Engineering Depart¬
ment of the State University System of Florida, located on the campus of
the University of Florida in Gainesville.
An analysis of LOF preceded surface plotting. The following SAS
commands were used to generate the models necessary to test for LOF in
this example. (A cubic model described the TNC response after a qua¬
dratic model failed.)
Complete model:
(LINE = limpograss genotype)
PROC ANOVA; CLASS LINE TRT REP;
MODEL TNC9=REP LINE LINE*REP TRT LINE*TRT;
Fractionated model:
PROC GLM; CLASS LINE REP;
MODEL TNC9=REP LINE REP*LINE
N N*N N*N*N
F F*F F*F*F
N*F N*N*F N*F*F
LINE*N LINE*N*N LINE*N*N*N
LINE*F LINE*F*F LINE*F*F*F
LINE*N*F LINE*N*N*F LINE*N*F*F/SOLUTION;
The test of LOF was accommodated by an F-test of the two previously
described models: (Sums of Squares (SS); Degrees of Freedom (DF);
Mean Square Error (MSE)).

132
r SS error (fractionated) _ r SS error (complete)
F(LOF) = 1 DF (fractionated) J 1 ~~ DF (complete) J
' ' MSE (complete)
Often an elimination procedure for nonsignificant terms is useful
for adding SS and DF back into the error term for the fractionated
model. When a model was finally deemed acceptable, the computer pro¬
vided estimates that were solved for the individual limpograss inter¬
cepts and parameters (N and F).
The equation for one limpograss line (PI 364888 or Florida number
297) was used to demonstrate the program for surface plotting (Table
11). The program has the following steps: (1) load the raw data from
the TNC analyses and create means; (2) load the previously described
surface equations; (3) assign treatment numbers (Figure 5) for the pre¬
dicted treatment combinations; (4) use the surface equation to solve for
TNC at the "missing" levels (Figure 5) of N and F; (5) merge actual plus
predicted data; (6) assign N and F levels for the actual data; (7)
"trick" the computer to start at zero on the TNC axis; and (8) plot the
surface using PROC G3D (Figure 21).
Results and Discussion
It will be assumed the reader has a minimum knowledge of SAS
because a detailed explanation of the program would be both laborious
and out of place. Tables 12 and 13 are shown in order to allow the
reader to trace the values from the computer output to the test for LOF
in Table 14. In this example, LOF is determined with an F-test of
models. The complete model contains the minimum possible sums of
squares because it represents the best explanation of treatment

133
Table 11. A SAS program using a previously fitted regression to obtain
values for the dependent variable to fill the holes in the
treatment matrix in order to merge with actual data and use
PROC G3D
1 //TNC9 JOB (1001,1401,29,9,0,),'CHRISTIANSEN;CLASS=2
2 /*PASSW0RD 08,SCOTT
3 /*R0UTE PRINT REMOTE!3
4 // EXEC SAS,REGI0N=372K,PL0T=
5 GOPTIONS DEVICE=GOULD;
6 DATA CR80;
7 INPUT NO 1-4 M 6-7 D 9-10 Y 12-13 LINE 15-17 TRT 19-20 REP 22
8 INC 24-25 DEV 27-28 WT 30-34 PI 36-37 DILN 39-40 P2 42-43 0D1 45-47
9 0D2 49-51 0D3 53-55 0D4 57-59;
10 DATE=MDY(M,D,Y); JULIAN=JULDATE(DATE); DROP DATE;
11 IF JULIAN=80272 THEN HARV=9;
12 0DA=(0D1+0D2+0D3+0D4)/4;
13 CARDS;
14
15 DATA CR80; SET CR80;
16 IF DEV=52 THEN MICR0GM=0.1163113*0DA+4.3782886;
17 IF DEV=53 THEN MICR06M=0.1088506*0DA+3.2966179;
18 IF DEV=54 THEN MICR0GM=0.1232033*0DA+4.5434298;
19 MGCH0=(MICR0GM*0.011*DILN)/(P1*P2);
20 CH0=MGCH0*100/WT;
21 IF DEV=52 THEN DO; ADJCH0=CH0/1.0835; IPS=9.46; ADJIPS=8.73; END;
22 IF DEV=53 THEN DO; ADJCH0=CH0/0.9751; IPS=8.64; ADJIPS=8.86; END;
23 IF DEV=54 THEN DO; ADJCH0=CH0/1.0177; IPS=8.66; ADJIPS=8.51; END;
24
25 DATA TNC9; SET CR80;
26 IF LINE=297
27 PROC SORT; BY TRT;
28 PROC MEANS NOPRINT; BY TRT
29 VAR ADJCHO; OUTPUT OUT=NEW MEAN=TNC;
30
31
DATA DUMB!3;
32
DO LINE=297;
33
DO N=0, 60, 120, 240,
480, 480.1;
34
DO F=3, 6, 9, 12, 18,
18.1
35
MACRO AA IF LINE=%
36
MACRO BB THEN TNC=%
37
MACRO GG IF H=%
38
MACRO HH AND F=%
39
MACRO II THEN TRT=%
40
MACRO JJ; ELSE TRT=.;
l
41
GG 480.1 HH
18.1 II
26 JJ
42
GG 0 HH
6 II
16;
43
GG 0 HH
12 II
21;
44
GG 60 HH
3 II
14;

134
Table 11. Continued
45
GG
60
HH
9
II
19;
46
GG
60
HH
18
II
24;
47
GG
120
HH
6
II
17;
48
GG
120
HH
12
II
22;
49
GG
240
HH
3
II
15;
50
GG
240
HH
9
II
20;
51
GG
240
HH
18
II
25;
52
GG
480
HH
6
II
18;
53
GG
480
HH
12
II
23;
54
COMMENT TNC9 CROWN 80 297;
55
AA 297 BB
20.61946875 +
0.
01089278*N +
1.5022467E-05*N*N
56
- 7.
2858312E-08*N*N*N -
2.
78981175*F +
0.25778999*F
*F
57
- 7.
7240594E-03*F*F*F -
0.
00295115*N*F
+ 3.3195286E
-06*N*N*F
58
+ 6.
2476724E-05*N*F*F;
59
OUTPUT; END; END
; END;
60
61
DATA
DUMMY; SET
DUMB13;
62
IF TRT=16
OR TRT
= 21 OR
TRT
= 14 OR
TRT=19
OR TRT=24 OR TRT=
= 17 OR
63
TRT=
22 OR
TRT=15
OR TRT
=20
OR TRT
=25 OR
TRT=18 OR TRT=23
OR TRT=26;
64
65
PROC
SORT
; BY TRT;
66
PROC
PRINT;
67
68
OB
LINE
N
F
TRT
TNC
OB
LINE N
F
TRT
TNC
69
70
1
297
0.0
3.0
14.36
19
297 240.0
3.0
15
15.42
71
2
297
0.0
6.0
16
11.49
20
297 240.0
6.0
11.40
72
3
297
0.0
9.0
10.76
21
297 240.0
9.0
20
9.79
73
4
297
0.0
12.0
21
10.92
22
297 240.0
12.0
#
9.34
74
5
297
0.0
18.0
8.88
23
297 240.0
18.0
25
6.90
75
6
297
0.0
18.1
8.78
24
297 240.0
18.1
6.80
76
7
297
60.0
3.0
14
14.59
25
297 480.0
3.0
13.31
77
8
297
60.0
6.0
11.33
26
297 480.0
6.0
18
9.29
78
9
297
60.0
9.0
19
10.27
27
297 480.0
9.0
.
7.96
79
10
297
60.0
12.0
10.17
28
297 480.0
12.0
23
8.05
80
11
297
60.0
18.0
24
7.81
29
297 480.0
18.0
.
7.50
81
12
297
60.0
18.1
7.71
30
297 480.0
18.1
7.44
82
13
297
120.0
3.0
14.91
31
297 480.1
3.0
13.31
83
14
297
120.0
6.0
17
11.32
32
297 480.1
6.0
*
9.29
84
15
297
120.0
9.0
10.01
33
297 480.1
9.0
m
7.96
85
16
297
120.0
12.0
22
9.72
34
297 480.1
12.0
m
8.04
86
17
297
120.0
18.0
7.19
35
297 480.1
18.0
m
7.50
87
18
297
120.0
18.1
7.08
36
297 480.1
18.1
26
7.44
88
89
DATA
SURFACE; MERGE NEW
DUMMY;
90
BY TRT;
91
DATA
RESPONSE; SET SURFACE
5
92
MACRO CC
i—<
~n
—1
70
—1
II
%

135
Table 11. Continued
93 MACRO DD THEN DO; N=%
94 MACRO EE F= %
95 MACRO FF; END; %
96
CC
1
DD
0
EE
3
FF
97
CC
2
DD
120
EE
3
FF
98
CC
3
DD
480
EE
3
FF
99
CC
4
DD
60
EE
6
FF
100
CC
5
DD
240
EE
6
FF
101
CC
6
DD
0
EE
9
FF
102
CC
7
DD
120
EE
9
FF
103
CC
8
DD
480
EE
9
FF
104
CC
9
DD
60
EE
12
FF
105
CC
10
DD
240
EE
12
FF
106
CC
11
DD
0
EE
18
FF
107
CC
12
DD
120
EE
18
FF
108
CC
13
DD
480
EE
18
FF
109
IF
TRT=26
THEN TNC=0
5
no
IF
N=480.
1 AND F=18.
1 THEN
TNC=0;
111 TITLE 1 ACTUAL PLUS PREDICTED TNC;
112 TITLE 2 % TNC FOR 28 SEPTEMBER 1980 IN PI 364888 CROWNS;
113
114 PROC SORT; PROC PRINT
115 ACTUAL PREDICTED
116
TRT
TNC
N
F
TRT
TNC
N
F
117
1
14.42
0.0
3.0
14
14.59
60.0
nro-
118
2
14.56
120.0
3.0
15
15.42
240.0
3.0
119
3
13.28
480.0
3.0
16
11.49
0.0
6.0
120
4
12.05
60.0
6.0
17
11.32
120.0
6.0
121
5
11.85
240.0
6.0
18
9.29
480.0
6.0
122
6
10.25
0.0
9.0
19
10.27
60.0
9.0
123
7
8.96
120.0
9.0
20
9.79
240.0
9.0
124
8
7.95
480.0
9.0
21
10.92
0.0
12.0
125
9
11.05
60.0
12.0
22
9.72
120.0
12.0
126
10
9.24
240.0
12.0
23
8.05
480.0
12.0
127
11
8.80
0.0
18.0
24
7.81
60.0
18.0
128
12
7.18
120.0
18.0
25
6.90
240.0
18.0
129
13
7.51
480.0
18.0
26
0.00
480.1
18.1
130
131
PROC
G3D; PLOT
N*F=TNC;
132
133 //EXEC PXPLOT
134 /*EOF

136
% TNC FOR28SEPTEMBER 1980
IN PI 364888 CROWNS
Figure 21. An example of the response surface plotted using SAS/GRAPH
computer assistance

Table 12. A SAS analysis of variance procedure for a model that quantifies the maximum treatment (TRT)
sum of squares (SS) for total nonstructural carbohydrate (TNC) in harvest 9 of the 1980
experiment
INDEPENDENT VARIABLE:
SOURCE
TNC9
DF
SUM OF SQUARES
MEAN SQUARE
F VALUE
P > F
R-SQUARE
C.V.
MODEL
59
1234.35
20.92
21.37
0.0001
.93
11.16
ERROR
CORRECTED TOTAL
96
155
93.98
1328.33
0.98
STD DEV
TNC9 MEAN
0.99
8.86
SOURCE
DF
ANOVA SS
F VALUE
PR > F
REP
2
27.67
14.13
0.0001
LINE
3
240.60
81.92
0.0001
LINE*REP
6
17.19
2.93
0.0115
TRT
12
854.82
72.76
0.0001
LINE*TRT
36
94.06
2.67
0.0001
TEST OF HYPOTHESES
USING THE
ANOVA MS FOR LINE*REP
AS AN ERROR
TERM
SOURCE
DF
ANOVA SS
F VALUE
PR > F
LINE
3
240.60
27.99
0.0006

Table 13.
TRT
A cubic SAS general linear model that fractionates the treatment sum of squares (SS) into SS
explainable by nitrogen (N) fertilization and frequency (F) of defoliation for total non-
structural carbohydrate (TNC) in harvest 9 of the 1980 experiment
SOURCE
DF
SUM OF SQUARES
MEAN SQUARE
F VALUE
P > F R-SQUARE C.V.
MODEL
47
1213.99
25.83
24.40
0.0001 0.92 11.61
ERROR
CORRECTED TOTAL
108
155
114.34
1328.33
1.06
STD DEV
TNC 9 MEAN
1.03
8.86
SOURCE
DF
TYPE I SS
F VALUE
PR > F
TYPE IV SS
F VALUE
PR > F
REP
2
27.67
13.07
0.0001
27.67
13.07
0.0001
LINE
3
240.60
75.75
0.0001
18.76
5.91
0.0010
LINE*REP
6
17.19
2.71
0.0173
17.19
2.71
0.0173
N
1
90.74
85.71
0.0001
3.88
3.67
0.0581
N*N
1
29.48
27.84
0.0001
0.38
0.36
0.5502
N*N*N
1
0.01
0.00
0.9634
0.83
0.78
0.3790
F
1
501.73
473.91
0.0001
114.63
108.27
0.0001
F*F
1
164.28
155.17
0.0001
69.92
66.05
0.0001
F*F*F
1
55.37
52.30
0.0001
53.55
50.58
0.0001
N*F
1
2.57
2.43
0.1221
0.01
0.01
0.9237
N*N*F
1
0.62
0.58
0.4466
0.88
0.83
0.3630
N*F*F
1
0.12
0.11
0.7392
0.13
0.12
0.7245
CO
CO

6CL

Table 14. An F-test of models for determining a statistically significant response lack of fit (LOF)
and an elimination method for nonsignificant terms
TEST FOR LOF
F(LOF)
/ SS complete model - SS fractionated model v
DF complete model - DF fractionated model '
/ SS complete model
' DF complete model
F(LOF) = 114.34 - 93.98/108 - 96
93.98/96
= 1.73
Ftab (0.05,12,80) = 1.85
(0.05,12,100) = 2.36
Ftab (0.01,12,80) = 1.83
(0.01,12,100) = 2.33
CONCLUDE: 1.73 n.s. LOF
ELIMINATION METHODOLOGY:
SOURCE
DF TYPE I SS P > F
ERROR
108
N*N*F*LINE
3
N*F*F*LINE
3
114
F(LOF) =
125.51
114.34
6.34 0.1170
4.83 0.2121
125.51
93.98/114 - 96
.98/96
Ftab (0.05,18,80) = 1.73
(0.05,18,100) = 2.18
Ftab (0.01,18,80) = 1.72
(0.01,18,100) = 2.14
1.76
CONCLUDE: *L0F(P < 0.05)

141
variability. The cubic model with N and F parameters explained enough
of the treatment variability to give a nonstatistical LOF.
The elimination procedure depicted in Table 14 must be used with
adherence to parameter hierarchy, i.e., if the N*N term, for example,
is nonsignificant but the LINE*N*N term is significant, then the former
cannot be removed from the model due to the dependence of the latter.
Evaluation of P > F term is begun from the bottom parameter (TYPE I SS)
upwards until a significant term is reached and at that time P > F deci¬
sion is based upon TYPE IV SS. The reason for the above is because
TYPE I P > F is calculated for each term prior to the addition of sub¬
sequent terms. TYPE IV P > F gives an analysis of each term given that
all other terms are present.
A final general rule in the attempt for a more parsimonious model
in the process of eliminating nonsignificant terms: try to pick the
parameters with small SS and large DF because this will further reduce
the fractionated model MSE (SS/DF) when the SS and DF are added back
into this portion of the numerator shown in Table 11. A reduced numera¬
tor will decrease the F value in the test for LOF.
Tables 15 and 16 explain how the estimates are obtained for each
individual line and model parameter. In this example, the cubic 297
limpograss equation has coefficients for parameters that can be easily
traced back to their origin for illustrative purposes. The other limpo-
grasses will not be considered beyond this point to avoid redundancy.
The major objective of the program in Table 11 was to fill "holes" in
the response surface treatment matrix in order to make use of SAS/GRAPH.
The resultant surface shown in Figure 21 represents a combination of 13
actual data points and 12 predicted points. The inclusion of actual

Table 15. The SAS general linear models procedure solution for estimates of intercepi
limpograss (line) and rep parameters
PARAMETER
ESTIMATE
INTERCEPT
18.95925501
+
0.32606522 =19.28532023
REP
1
0.48188699
2
0.49630868
x = 0.32606522
3
0.00000000
LINE
297
0.88462663
+
0.44952189 = 1.33414852
553
1.75620304
+
(-0.04744913) = 1.70875363
+ 19.28532023
554
9.72148784
+
(-0.30779278) = 9.41369506
886
0.00000000
+
0.00000000 = 0.00000000
*REP
297
1
-0.20409839
2
1.55266406
x = 0.44952189
3
0.00000000
553
1
-0.61035359
2
0.46800535
x = -0.04744913
3
0.00000000
554
1
-0.68583224
2
-0.23754609
X = -0.30779278
3
0.00000000
886
1
0.00000000
2
0.00000000
x = 0.00000000
3
0.00000000
adjusted for the
INTERCEPT FOR
LIMPOGRASS LINE
20.61946875
20.99407386
28.69901529
19.28532023

Table 16.
The SAS general linear models procedure solutions for estimates of nitrogen (N) and cutting
frequency (F) treatment parameters adjusted for each limpograss (line)
TREATMENT PARAMETER
ESTIMATE
N
-0.02030927
(A)
N*N
6.3981011E-05
(B)
N*N*N
-9.5484194E-08
(C)
F
-3.19230044
(D)
F*F
0.27822672
(E)
F*F*F
-0.00771670
(F)
N*F
0.00054597
(G)
N*N*F
9.9283955E-07
(H)
N*F*F
-3.5145052E-05
(I)
TREATMENT PARAMETERS
ADJUSTED
FOR LINE ESTIMATES
N*LINE
297
0.03120205
0.01089278
553
-0.01659688
(A)
-0.03690615
554
0.01919086
0.00111841
886
0.00000000
-0.02030927
N*N*LINE
297
-4.8958544E-05
1.5022467E-05
553
3.6548348E-05
+
(B)
1.0052936E-04
554
-0.00014040
-0.00007642
886
0.00000000
6.3981011E-05
*N
*N*N

Table 16. Continued
TREATMENT PARAMETERS
ADJUSTED
FOR LINE ESTIMATES
N*N*N*LINE
297
2.2625882E-08
-7.2858312 E-08
553
8.0519150E-09
(C)
-8.7432279E-08
*N*N*N
554
2.3065701E-07
+
1.3517282E-07
886
0.00000000
-9.5484194E-08
F*LINE
297
0.40248869
-2.78981175
553
0.13348428
(D)
-3.05881616
554
-2.58616658
-5.77846702
*F
886
0.00000000
-3.19230044
F*F*LINE
297
-0.02043673
0.25778999
553
-0.02705211
(E)
0.25117461
*F*F
554
0.23902603
0.51725275
886
0.00000000
0.27822672
F*F*F*LINE
297
-7.3594410E-06
-7.7240594E-03
553
554
0.00109042
-0.00678204
+
(F)
=
-0.00662628
-0.01449874
*p*p*p
886
0.00000000
-0.00771670
N*F*LINE
297
-0.00349712
-0.00295115
553
0.00082619
(G)
0.00137216
*N*F
554
0.00027520
0.00082117
886
0.00000000
0.00054597
N*N*F*LINE
297
2.3266890E-06
3.3195286E-06
553
554
-2.7248721E-06
-1.1295575E-06
+
(H)
=
-1.7320326E-06
-1.3671795E-07
*N*N*F
886
0.00000000
9.9283955E-07
N*F*F*LINE
297
9.7621776E-05
6.2476724E-05
553
554
9.5S97150E-06
8.4502848E-06
+
(I)
=
-2.5545337E-05
-2.6694797E-05
*N*F*F
886
0.00000000
-3.5145052E-05
THE FINAL EQUATIONS
FOR TNC9
IN 297 LIMPOGRASS:
297 TNC9 = 20.61946875 + 0.01089278*N + 1.5022467E-05*N*N - 7.2858312E-08*N*N*N -2.78981175*F
+ 0.25778999*F*F - 7.7240594E-03*F*F*F - 0.00295115*N*F + 3.3195286E-06*N*N*F
+ 6.2476724E-05*N*F*F

145
data provides a verification of the model equation if the actual and
predicted points blend evenly. This was considered important because
some models that were plotted were statistically adequate but none the
less failed to fit in some regions of the surface plot.
The Table 11 program will be discussed from top to bottom using the
program line number for easy reference. On line 7 the information asso¬
ciated with the raw data is inputed. For a more thorough investigation
into the mechanics used in processing the TNC data, consult Appendix A.
On lines 25-29 the TNC means are created. The DATA DUMB13 (line 31) is
the data set containing the previously determined surface equation and
instructions to code for treatment, N, and F. Only the missing treat¬
ments shown in Figure 5 are given treatment numbers as well as one more
(TRT 26), which will be used later to trick the computer into starting
the TNC axis at zero. The surface equation is used to solve for TNC at
the specified levels of N and F (lines 54-59), and on line 89 the actual
and predicted data sets are merged. The actual data is assigned N and F
levels in lines 92-108 and lines 109-110 set the base line for TNC at
zero. The remainder of the procedure is standard plotting in SAS/GRAPH.
The computer output is often not "publication perfect" because
certain additions are generally added more easily by a graphics person.
As yet some details of SAS/GRAPH are yet to be worked out. One problem
is that no uniform scaling routine is yet available; hence, graphical
comparisons of limpograsses must be uniformly scaled by hand. Also,
sometimes the computer uses a non-linear scale for the "Z" axis that is
clearly undesirable. In general, however, by using photoreduction a
graphics person can easily trace the surface plot in the proper size for
use in dissertations, etc., with a minimum of manipulation.

146
The smoothing techniques employed by Schoney et al. (1981) are
clearly more advanced than the methodology described here; however,
SAS/GRAPH is not yet so sophisticated. The major precaution this paper
delivers is that before the smooth, fitted function is used in SAS sur¬
face plots (and their corresponding contour plots), the predicted points
should be plotted amidst the actual data for a visual test of LOF.
Also, beware of the interpretation of a surface based upon a non-linear
scaling of the "Z" axis.
Computer generated graphics have been promoted by Cady and Fuller
(1970), and it is believed that these graphics will be more widely used
in the future. "Through graphics, the distribution, physical response,
and interaction characteristics of a large volume of data can be readily
analyzed" (Schoney et al., 1981, p. 437).
Conclusions
1. Response surface designs are more efficient in space, time, and
cost, and the three dimensional representation of the surface gives an
overview of the entire response while the contour plot gives an easily
perceivable value of "Z" for any level of "X" and "Y,"
2. The F test of models gives a test of LOF SS by the difference
method: The MSE for a fractionated prediction model minus MSE for a
model explaining a maximum amount of treatment variability yields MSE
LOF.
3. Intercept and parameter estimations provided with the /SOLUTION
option gave the intercept and coefficients needed to obtain a prediction
equation.

147
4. A merger of predicted and actual data provided a verification
of the predicted equation for a visual test of LOF.
5. Comparisons of surfaces must be made with extreme caution
because of the non-uniform scaling produced from one graph to the next
and the non-1inearity of the scaling on the "Z" axis.
Summary
Computer assisted graphics look great but too often experimenters
do not fully document the routine necessary to construct these graphs.
Response surface methodology was used in a field experiment conducted
between 1979-1981 that gave an opportunity to use three dimensional
plots in studying the effect of nitrogen (N) fertilization and defolia¬
tion frequency (F) on four limpograss (Hemarthria altissima (Poir.)
Stapf et C.E. Hubb) genotypes. Total nonstructural carbohydrates (TNC),
forage quality, persistence, and yield were measured, and from those
responses a cubic model describing TNC in the "crowns" (lowest 2 cm of
stem) of PI 364888 limpograss harvested 28 September 1980 was selected
to exemplify the computer procedures necessary for constructing surface
plots.
The plotting procedures in SAS/GRAPH cannot accept "holes" in the
treatment combination matrix. A procedure was presented showing how to
fill the "holes" with predicted numbers derived from a regression equa¬
tion fitting the surface response.
Comparison of surfaces is difficult because no uniform plotting
program is yet available in SAS/GRAPH. Due to the limitation of non-
uniform scaling and sometimes non-linear scaling of the "Z" axis, extreme
caution is advised in the interpretation of surface plots.

SUMMARY AND CONCLUSIONS
The introduction of limpograss into the United States occurred in
1964 and since that time it has grown popular in the southeast as an
improved forage for the flatwood site. Research prior to 1978 culmi¬
nated in the identification, characterization, and cultivar release of
'Bigalta', 'Redalta', and 'Greenalta'. More limpograss germplasm
became available for evaluation after plant explorations in 1971 and
1976, and several of the plant introductions were suspected to have
agronomic characteristics that were superior or equal to the released
cultivars.
In this study Bigalta and Redalta were compared to PI 364888 and
PI 349753 in a series of experiments during 1979-1981. Bigalta is a
highly digestible tetraploid limpograss that has not persisted under
management practices involving frequent defoliation and/or heavy appli¬
cations of nitrogen fertilizer. Redalta, a diploid, has excellent per¬
sistence but its overall nondigestibility is undesirable. The promising
limpograsses PI 364888 and PI 349753 were suspected to have higher
persistence than Bigalta along with higher quality than Redalta.
Working with established stands of limpograss in 1979, the limpo¬
grasses were fractionated and analyzed to determine the site of total
nonstructural carbohydrate (TNC) storage. At the conclusion of the
plant part sampling, five intervals of cutting were imposed in order to
differentially drain the storage organs of their carbohydrate reserves.
148

149
Carbohydrate levels were tested chemically and by the use of regrowth-
in-darkness techniques. Weight data on etiolated shoots, stubble, and
roots provided the basis for a morphological comparison of genotypes.
Positive correlations between weight of stubble and TNC stored in the
bottom 2 cm of stems suggested that large stubble systems (Redalta and
PI 364888) may aid in plant persistence.
In 1980 an extensive field experiment was conducted with five
levels of N fertilization (0, 60, 120, 240, 480 kg/ha/yr) and five fre¬
quencies of defoliation (3, 6, 9, 12, and 18 weeks). Dry matter (DM)
yields, in vitro organic matter digestibility (IV0MD), crude protein
(CP), persistence, and TNC in shoots and storage organs were measured.
Bigaita produced adequate DM at intermediate cutting frequencies,
while PI 364888 produced more DM with high N and long cutting
frequencies. Bigalta was lowest in spring forage production and
PI 349753 was highest. Limpograss PI 364888 was the best genotype for
use as a stockpiled forage.
Bigalta1s IV0MD and CP percentages were clearly superior to the
other limpograsses. The spring and autumn forage was of a very high
quality, but the midsummer values for IV0MD and CP were quite low.
Studies of TNC showed a maximum percent of organic food reserves in
stem bases in March and a low in July. As temperature stress decreased
in September and October, the TNC values increased. The period of low
TNC accumulation in midsummer coincided with the period of most severe
limpograss deterioration in the plots. The 3 week cutting frequency in
combination with the 480 kg/ha/yr N rate was the most stress provoking
treatment. Bigalta and PI 349753 were most sensitive to the treatment
stresses while Redalta and PI 364888 were most persistent.

150
Results from this study show that the DM yield production in spring
could be exploited in management systems. Limpograsses may be utilized
for grazing in the spring of the year until bahiagrass begins its
growth.
The summer period was marked by active limpograss growth, low per¬
centages of stored TNC, and very low values for CP and IVOMD. The
summer period also seemed to be the time when limpograss was most sus¬
ceptible to stress and loss of stand. The recommendation to rest the
limpograss during the summer has been suggested; however, the accumu¬
lated forage must be removed, burned, ensiled, or made into hay so that
the field may be clean at the beginning of August for the onset of
stockpiling.
Heavy summer rains may prohibit limpograss utilization as hay;
however, silage making may be feasible. From data presented in this
study it would appear that protein may be limiting in summer-harvested
limpograss. A promising area of research may be the ammonification of
the ensiled limpograss or even big bales of hay. The TNC values from
shoots (studied in Chapter 3) showed that carbohydrate levels in the
tissue may be adequate for silage making. Even if they were not, the
technology exists for the addition of carbohydrate to silage.
Limpograss studied as stockpiled forage did not effectively utilize
fertilizer N applied at rates above 120 kg/ha. The 1 August staging
date was critical for sufficient DM accumulation; however, the standing
forage should be utilized before frost in order to take maximum advan¬
tage of the relatively good forage quality present in autumn limpograss
herbage.

151
Finally, if a preservation technique were studied and elucidated,
the limpograss could be fed with supplemental protein and minerals
during the winter. In conclusion, it would appear that limpograss can
add much flexibility to the flatwood forage system and continued efforts
in this direction seem justified.

APPENDICES

APPENDIX A
TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) PROCEDURE

TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) PROCEDURE
UNIVERSITY OF FLORIDA
AGRONOMY DEPARTMENT
GAINESVILLE, FL 32611
AUGUST, 1982
FOURTH REVISION
Scott Christiansen, K. J. Boote, and E. B. Blazey
Preliminary Steps
Sample Acquisition
After thorough consideration of experimental design, the sampling
frequency and plant part to be analyzed may be determined. Total non-
structural carbohydrate (TNC) samples must be hand plucked if shoots
are to be studied or dug if roots or stem bases are of interest. In
the case of stem bases, they must be dug, selected, and removed from
the basal body of the plant as single shoots to avoid dead residue in
the sample. The samples are then washed and enclosed in a labeled
diaper and placed on ice to avoid excess respiratory losses. Samples
should be kept on ice to minimize respiration but kept relatively dry
to facilitate drying without formation of Maillard products or fermenta¬
tion which both result in loss of carbohydrates.
154

P.O.Box 1400
Sun City, AZ 85372
7 February 1983
To Whom 11 Concerns:
This note is in regard to the availability of Kyi ase 100 enzyme
suggested for use in Wisconsin Agrie. Exp. 3ta. Res. Report R21C7
(l98l) entitled "Removing & analyzing total nonstructur'al carbohy¬
drates from plant tissue'1. In the first 3 lines of column 1 on
page 5» it states that the chemical company producing Mylase 100
is in Des Plaines, IL. The company has moved from Illinois to
Charlotte, NC, and they can be contacted as outlined below:
G.B. Fermentation Industries Inc. Tel. 704-527-9000
P.0. Box 241068 800-438-1361
Charlotte, NC 28244
kJo. íj uu 1 c 1 j y
Sincerely,
Dal e 3m 1 th
Eheritus Professor of Agronomy
Unlv. of Wl scon sin-Madi son

155
Drying
The drying procedure is best effected using a small forced-air
drier such as those made by Thelco. The drier may be preheated to 120 C
so that by the time samples are loaded the temperature will have dropped
to 100 C. Keep samples at this temperature for 30-45 minutes and then
lower to 70 C. Remove when dry, approximately 36-48 hours later.
Storage
As frequently occurs, samples must be stored for some time until
grinding. Keep samples in air-tight plastic bags. This serves a dual
purpose in stablizing humidity and preventing sample damage by insects.
Store in a cool, dry place or freeze if possible. Freeze drying is the
best preservation technique; however, this is often impractical for
large numbers of samples.
Grinding
Grind samples as soon as possible as it reduces sample bulk and it
simplifies the logistics of sample preservation. Samples taken by hand
will usually be of a size that can be ground entirely so as to maintain
total plant representation. Grinding in a large Wiley mill may be used
to pulverize the sample. With mills that are frequently used, blades
are dull or misfitted resulting in a differential grind, i.e., leaves
pass the 1 mm screen easily but lignin and other structural components
remain behind. Hence, it is important to collect all the tissue in the
grinding chamber as well as tissue having passed through the screen.
This material can then be placed in a Whirlpac bag and pressed close to

156
exclude any remaining air space. This coarsely ground tissue should
subsequently be reground through a small Wiley mill, a ball mill, or a
UDY Cyclone mill to pass 0,5 mm or a 40 mesh screen. The second grind¬
ing has the advantage of thoroughly mixing the tissue of each sample.
Remember to reseal the Whirl pac bags without excess internal air space.
Dilu-vials also work well to hold samples.
Carbohydrate Analysis
Procedure
1. Open Whirlpac bags or Dilu-vials containing sample and place in
a 50 C forced-air drier for 6-8 hours. After removal of the sample from
the bags, eliminate air space while reclosing.
2. With an analytical balance, weigh out 0.1000 to 0.1099 g of
sample and place it into a 25 ml erlenmeyer flask. For samples weighed
out ahead of time for analysis, cork the flasks to prevent dust
contamination. Record the weight. (Often it is better to use 80 column
computer forms such as IFAS Form 2625 to log your data instead of a lab
book. This eliminates a step between the analysis and the computer by
putting your numbers in a form that can be punched directly onto com¬
puter cards or a terminal. See Table A-l for an example of a job sub¬
mitted using SAS at NERDC.)
3. Prepare the enzyme mix. (Have 0.2 M_ acetate buffer made prior
to this operation.) The enzymes necessary for TNC analysis and supplier
information may be found in another section of this procedure.
4. Add 5 ml of distilled water to erlenmeyer flasks holding
samples, internal plant standards, glucose for determination of glucose

157
Table A-l. SAS job as submitted on cards to the computer
//TNC JOB (4001 ,1490,5,5,0,),1USEIFASF0RM26251,CLASS=A
/*PASSW0RD
// EXEC SAS
DATA PARTS•
COMMENT THE CHALLENGE OF LEARNING SAS IS YOUR PROBLEM;
INPUT M 2-3 D 5-6 Y 8-9 LINE 11-13 AGE $ 18-24 PT $ 21-22 RP 24 INC 26-
27 DV 29-30 WT 32-36 PI 38-40 DN 41-42 P2 44-45 0D1 48-50 0D2 52-54 0D3
56-58 0D4 60-62; DROP 0D1-0D4; DATE=MDY(M,D,Y); JULAN=JULDATE(DATE);
DROP DATE;
IF PART=1 DO' THEN P=1R1;
IF PART=1CR* THEN P='C';
IF PART=1SH1 THEN P='T‘;
IF PART=1 ST1 THEN P='S' ;
ODA=(0D1+0D2+0D3+0D4)/4;
CARDS;
07
03
79
297
6WKR0
2
24
19
104.3
.5
10
01
250
260
265
255
07
03
79
297
6WKCR
2
24
19
107.7
.5
10
01
428
383
381
448
07
03
79
297
6WKST
2
24
19
104.3
.5
10
01
420
419
421
441
07
03
79
297
6WKSH
2
24
19
104.5
.5
10
01
320
311
395
395
07
03
79
553
6WKR0
2
24
19
106.1
.5
10
01
320
325
317
342
07
03
79
553
6WKCR
2
24
19
106.3
.5
10
01
190
195
192
190
07
03
79
553
6WKST
2
24
19
105.4
.5
10
01
175
171
174
178
07
03
79
553
6WKSH
2
24
19
108.0
.5
10
01
190
199
240
241
07
03
79
554
6WKR0
2
24
19
106.2
.5
10
01
305
309
328
325
07
03
79
554
6WKCR
2
24
19
102.3
.5
10
01
282
280
318
321
07
03
79
554
6WKST
2
24
19
104.8
.5
10
01
408
370
362
390
07
03
79
554
6WKSH
2
24
19
106.1
.5
10
01
478
489
465
463
07
03
79
886
6WKR0
2
24
19
104.1
.5
10
01
345
331
348
362
07
03
79
886
6WKCR
2
24
19
102.8
.5
10
01
411
431
475
440
07
03
79
886
6WKST
2
24
19
104.7
.5
10
01
412
452
469
475
07
03
79
886
6WKSH
2
24
19
106.1
.5
10
01
459
440
498
500
DATA PARTS; SET PARTS; *
IF DEV=19 THEN MICGM=0.106235*0DA+6.5827748;
MGCHO=(MICGM*0.011*DN)/(P1*P2);
CH0=MGCH0*100/WT;
IF DEV=19 THEN ADCHO=CHO/.9703;
PROC PRINT;
/*EOF
•k
If the enzyme has a sugar content subtract it here, e.g., MICROGM=
(0.106235*0DA+6.5827748)-(MICR0GM for the enzyme blank). Careful!
Depends on the same dilution as the unknown sample.
NOTE: See Table A-2 for example of returned job.

Table A-2. Statistical analysis system
OB
M
D
Y
LINE
AGE
PT
RP
INC
DV
WT
PI
DN
P2
JULAN
P
ODA
MICGM
MGCHO
CHO
ADCHO
1
7
3
79
297
6WK
RO
2
24
19
104.3
0.5
10
1
79184
R
257.50
33.94
7.47
7.16
7.38
2
7
3
79
297
WK
CR
2
24
19
107.7
0.5
10
1
79184
C
410.00
50.14
11.03
10.24
10.56
3
7
3
79
297
WK
ST
2
24
19
104.3
0.5
10
1
79184
S
425.25
51.76
11.39
10.92
11.25
4
7
3
79
297
WK
SH
2
24
19
104.5
0.5
10
1
79184
T
355.25
44.32
9.75
9.33
9.62
5
7
3
79
553
WK
RO
2
24
19
106.1
0.5
10
1
79184
R
326.00
41.22
9.07
8.55
8.01
6
7
3
79
553
WK
CR
2
24
19
106.3
0.5
10
1
79184
C
191.75
26.95
5.93
5.58
5.75
7
7
3
79
553
WK
ST
2
24
19
105.4
0.5
10
1
79184
S
174.50
25.12
5.53
5.24
5.40
8
7
3
79
553
WK
SH
2
24
19
108.0
0.5
10
1
79184
T
217.50
39.69
6.53
6.05
6.23
9
7
3
79
554
WK
RO
2
24
19
106.2
0.5
10
1
79184
R
316.75
40.23
8.85
8.33
8.59
10
7
3
79
554
WK
CR
2
24
19
102.3
0.5
10
1
79184
C
300.25
38.48
8.47
8.28
8.53
11
7
3
79
554
WK
ST
2
24
19
104.8
0.5
10
1
79184
S
382.50
47.22
10.39
9.91
10.22
12
7
3
79
554
WK
SH
2
24
19
106.1
0.5
10
1
79184
T
473.75
56.91
12.52
11.80
12.16
13
7
3
79
886
WK
RO
2
24
19
104.1
0.5
10
1
79184
R
346.50
43.39
9.55
9.17
9.45
14
7
3
79
886
WK
CR
2
24
19
102.8
0.5
10
1
79184
C
429.25
53.25
11.71
11.40
11.74
15
7
3
79
886
WK
ST
2
24
19
104.7
0.5
10
1
79184
S
452.00
54.60
12.01
11.47
11.82
16
7
3
79
886
WK
SH
2
24
19
106.1
0.5
10
1
79184
T
474.25
56.96
12.53
11.81
12.17

159
recovery, and the empty flasks used to determine the TNC content of the
enzyme. A Cornwall pipet works well for this addition. Clamp the
flasks into the sample bed and top with the stopper/policemen.
5. Adjust the water level in the boiling bath higher than the
level of the liquid in the flasks. Immerse the sample bed into the
boiling (100 C) water for 10 minutes. Use a timing clock.
6. Remove the sample bed from the boiling water bath and add 5 ml
Of 0.2 M acetate buffer to each flask when they have cooled. The Corn¬
wall pipet is useful for this operation also.
7. Add 1 ml of enzyme mix to the flasks for samples, glucose
recovery, plant standards, and enzyme blanks using the Oxford triple
range micropipet sampler. Add 1 ml of water instead of enzyme to the
glucose flasks. (The enzyme slurry should be vigorously stirred on a
stirring plate during this step to insure a homogenous mix.) Replace
stopper/policemen to minimize evaporation.
8. Return the sample bed with the samples to the shaker unit. The
water in the bath should be at 48 C and set at a reasonable shaking
oscillation. Allow a digestion time of whatever has been determined
necessary in preliminary experiments. Make sure the distilled water
level in the shaking water bath remains above the level of liquid in the
flasks.
9. Perch funnel-folded filter paper disks (11 cm diameter) on the
rim of 100 ml beakers. Arrange the 100 ml beakers on a tray and empty
the contents of the erlenmeyer flasks into the filter paper funnels when
the sample slurry is cool. Maintain sample identity.
10.The test tube racks will hold 40 lipless culture tubes (16 x
150 mm) arranged in a 10 x 4 configuration. In the first dilution each

160
rack holds the 20 test tubes needed for 10 samples. Remove two aliquots
from the filtrate in the 100 ml beakers and place the liquid in test
tubes with the corresponding numbers (Figure A-l).
NOTE: The aliquot volume is dependent upon the dilutions needed in
samples and sugar standards to result in an optical density that can be
read on the absorbance scale of the spectrophotometer. The dilutions
must be determined experimentally for the particular tissue being
analyzed. For example, with limpograss shoot tissue, 0.5 ml of liquor
is transferred using an Oxford triple range pipet. Then, 9.5 ml of
distilled water is added to make 10.0 ml. Therefore, 10.0 ml is the
volume for the first dilution. The Cornwall pipet speeds the addition
of the distilled water.
11. Once all the samples have been transferred and the appropriate
volume of distilled water is added, vortex the tubes to assure an even
concentration of the liquid. Now continue to the second dilution.
12. Sort the racks for the first dilution so that the sample
numbers run sequentially. Taking one rack at a time, pair the racks for
dilution 1, with the racks for dilution 2. Dilution 2 racks hold 40
tubes instead of 20, Remove two aliquots from each of the two originally
diluted tubes and dispense into the four tubes associated with the
dilution 2 rack (Figure A-2). For example, with limpograss again, a 0.5
ml aliquot is taken from the tubes in the dilution 1 rack and placed in
the tubes for dilution 2.
13. Add 1 ml of alkaline reagent to each tube. Vortex each rack as
a unit rather than by individual tubes within a rack. Load the rack for
dilution 2 onto the tray designed for dipping the samples into the boil¬
ing water bath. When the water has reached a boil (100 C) lower the

161
Figure A-l. Dilution one
Figure A-2. Dilution two

162
dipping tray into the bath for 20 minutes. Have the water adjusted to
be above the level of samples in the tubes. Moderate the gas for the
burner so that splashes from the boil are kept from entering the tubes.
14. After 20 minutes, raise the dipping tray and transfer it to the
cooling bath.
15. Remove the racks out of the dipping tray when the samples are
cooled and arrange on the lab bench as before. Add 1 ml of arseno-
molybdate reagent (the color forming reagent) with a manostat pi pet
containing no metal parts. Vortex the entire rack as a unit rather than
one tube at a time.
NOTE: Do not try to use a Cornwell pi pet for this operation
because the arsenomolybdate reagent will be contaminated by contact with
the metal in the Cornwall pi pet.
16. At this point, the samples may be diluted to their final volume
of 10 ml. For example, if you used a 0.5 ml volume of sample, 1 ml of
alkaline reagent, 1 ml of arsenomolybdate, and 7.5 ml of distilled water
is added. Using the Cornwall pipet with the proper technique, you can
direct the stream of water directly at the wall at the surface of the
sample in the tube to get a good mixture. This is important because the
water must get thoroughly mixed with the ingredients in the tube.
NOTE: The sugar standards use an aliquot volume of 1 ml for the
determination of the Beer's law relationship-plus 1 ml of alkaline
reagent and 1 ml of arsenomolybdate giving a total of 3 ml in the tube.
Therefore, 7.0 ml of distilled water is added to obtain the final volume
of 10 ml in these tubes.
17. Warm up the Coleman Jr. II spectrophotometer for 5 minutes and
set at 540 nm. Make sure the display is pushed to the right (since it

T63
is loose). The cuvette should be scrupulously clean. Pour a small
amount of distilled water into the funnel of the vacuvette cell. Fill
the vacuvette cell with the reacted water blank (described in the sec¬
tion on glucose standards) and adjust the instrument to zero absorbance.
Vortex each tube before placing in the cuvette. Read the absorbance of
the samples and standards. Clean the cuvette after using by flushing
with a mild detergent solution and rinsing in distilled water several
times. To avoid aspiration of liquid waste into the pump, the pump
reservoir should be removed and emptied whenever it gets two-thirds full.
Disconnect the pump from the AC line, then unscrew the reservoir from
the pump; be careful not to tip or agitate the assembly in such a way
as to allow liquid to splash or flow into the pump intake.
18. Multiply the absorbance by 1000 for convenience and transfer
the information to the 80 column data sheet. The 1000-fold multiplica¬
tion makes the data easier to punch onto computer cards and does not
change the relationship with the standards.
Glucose Standards
Carefully weigh out 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, and
0.08 g anhydrous D-glucose into 1 L volumetric flasks and fill to the
mark with the distilled water. The glucose should be kept in a desic¬
cator prior to weighing. Pour the amount of standard needed for the
duration of the experimentation into two or more sets of plastic bottles
and freeze them. Before using the standards, thaw and allow the solu¬
tion to reach room temperature.
Sugar standards enter the procedure at Step 13 after all sample
tubes have been brought through the second dilution. Sugar standards

164
and water blanks enter the procedure at Step 13 after all sample, plant
standard, enzyme blank, and glucose recovery tubes have been brought
through the second dilution.
Using the Oxford pi pet add 1 ml of standard to each of the four
tubes in the sample rows for each standard. One ml of alkaline reagent
is added to these tubes and the 1 ml water blank tubes exactly as is
done for the samples. Vortex. The standards are then treated exactly
the same as samples in the remainder of the procedure. Remember to
include at least four water blank tubes to be combined and used to zero
the spectrophotometer.
Enzymes
A. Amyloglucosidase from Rhizopus
Stock #A7255 500 g $50.00 (6/24/81)
Supplier: Sigma Chemical Co.
P. 0. Box 14508
St. Louis, M0 63178
Toll free: (800) 325-3010
Customer Service Phone: (314) 771-5765 (call collect)
B. Invertase concentrate from yeast (in glycerol)
Stock #39020 4E 250 ml $30.00 (6/24/81)
Supplier: Gallard Schlessinger Chem. Mfg. Corp.
584 Mineóla Avenue
Carle Place, NY 11514
Phone: (516) 333-5600
NOTE: The enzymes should be kept in the freezer. The
amyloglucosidase should be kept desiccated while in the freezer.

165
Enzyme mix.
45 ml distilled 1^0
5.0 ml 0.1 M Acetate buffer
2.5 ml Invertase concentrate
1.25 g amyloglucosidase
0,1 g thymol
Make enzyme mix fresh daily and stir on a magnetic stirrer when
removing a 1 ml aliquot for addition to the sample. The buffer and
water take the reducing sugars into solution. The invertase hydrolyzes
the bond between glucose and fructose in sucrose and the amyloglucosi¬
dase hydrolyzes the starch molecules into monomers. The resulting
monomers are measured by the Nelson's reducing sugar test of which this
procedure is an adaptation.
Enzyme bank. By using buffer, invertase, and amyloglucosidase in
this procedure, a small carbohydrate content is added by the enzyme
which should be determined and subtracted out.
One ml of enzyme is added to an erlenmeyer flask having 5 ml of 0.2
M acetate buffer at Step 7. The enzyme blank is subsequently treated
like the samples.
Glucose Recovery
In order to check whether there is any loss or gain of liquid
during incubation, filtering, etc., a known quantity of glucose can be
tested with the assay to check for concentrating or diluting effects.
This may be done in two ways.

166
1. Weigh out 10 mg of glucose and treat it as you would a dry
sample except add 1 ml of water instead of enzyme solution. A 0.5 ml
aliquot diluted up to 10 ml (add 9.5 ml of distilled water) in the first
dilution will result in a mid-range optical density using a 0.5 ml
aliquot in the second dilution.
2. Weigh out 10 g glucose and dissolve in 1 L distilled water. At
f
the beginning of each run put 1 ml of stock (10 mg/ml) into the erylen-
meyer flasks. Add only 4 ml of water to give 5 ml in the initial boil
so that volumes between glucose recovery flasks are comparable to volume
of sample flasks. Use the same dilutions as described above.
After subtracting the quantity of carbohydrate added by the enzyme
from plant standards and samples, an adjustment is made to 100 percent
glucose recovery (see calculations).
Beer's Law
A regression equation must be used to obtain the relationship
between sugar content of the samples and the optical density read at 540
nm. The standards have a known quantity of carbohydrate and serve as
the reference and must be included in every set due to a multitude of
factors that can change the color development from run to run.
It is faster to determine the regression equation by hand calcula¬
tor so that the computer may be fed the remainder of the information in
SAS input to complete the computations.
A few words on theory are necessary. We assume the glucose stan¬
dards to be accurate, thus it will be the independent variable (X). The
optical density values will, therefore, be dependent values (Y). Since
we are using a least squares method of fitting the regression line and

167
the standards are assumed correct, then our analysis reveals variability
in the technique (Figure A-3).
We are not interested in predicting optical densities; therefore,
the equation is rearranged to get the inverse relationship. Consider
the following set of data for standards. Punch into the TI-55
calculator:
Qjg glue (uG)]
[optical density (OP)] display
10
x t y
93
I
20
II
150
II
30
II
222
II
40
II
312
II
50
II
432
II
60
II
469
II
70
II
558
II
80
II
700
II
Press 2nd SLOPE 2nd INTCP, for the equation OD = m(yG) + b.
OD = 8.5119048 * yg + (-16.035714)
„ _ OD + 16.035714
yg 8.5119048
1
2
3
4
5
6
7
8
u9
1 * nn , 16.035714
8.5119048 uu 8.5119048
yg = 0.1174825 * 0D + 1.8839
Calculations
1. Standards were made by placing 0.01-0.08 g of D-glucose in 1 L
volumetric flasks.
1 L
1000 ml
O.Olg + 1000 mg
1 L g
0.01 mg/ml

700
600
500 f-
Optical
Density 400-
(abs.x 1000)
Y
300
200
100
Variability in Y for X =60
.or ¿>' <
.9?
o>
o
10 20 30 40 50 60 70 80
/ig Glucose Standard
X
Figure A-3. The regression of optical density (OD) versus
glucose

169
2. The regression equation determinated in the last section was
based on yg.
0.01 mg/ml or 0.01 mg/tube * —^ -9 = 10 yg glucose
3. Enzyme blanks, glucose recovery, and internal plant standard
values are determined as a function of the Beer's law regression for the
sugar standards. The enzyme blank can be determined first if it has an
identical dilution series as the unknown sample (e.g., 0D = 45).
yg (for enzyme blank) = 0.1174825 * OD + 1.8839
= 0.1174825 * 45 + 1.8839
= 7.1706
4. Sample carbohydrate content (subtracting the enzyme blank)
(e.g., 0D = 175).
yg (for sample or plant
standard) = 0.1174825 * OD + 1.8839
= 0.1174825 * 175 + 1.8839
= 22.4433
Subtraction of enzyme
CHO = yg for sample - yg for enzyme blank
= 22.4433 - 7.1706 = 15.2727
5. Convert to mg carbohydrate. (NOTE: If a different dilution
was used for the enzyme blank bring both sample and enzyme blank to the
level of mg CHO and then subtract out the mg CHO for the enzyme blank,
i.e., mg CHO = mg CHO (sample + enzyme) - mg CHO (enzyme blank).)
mg CHO = 15.2727 yg * ^qqq^ * (1st dilution) * (2nd
dilution).

170
= ic; 9797 * 1.0 * 11.0 ml * 10.0 ml
1000 0.5 ml 0.5 ml'
= 6.72
6. Convert to % CHO (assume sample weight = 105.5 mg)
% CHO = * 100
wt
= 6.37%
7. Glucose Recovery--Method I (assume wt = 10.5 mg D glucose
0D = 180).
u9
mg CHO
0.1174825 * OD + 1.8839
0.1174825 * 180 + 1.8839
23.0308
23.0308 *
1.0 mg * 11.0 ml * 10.0 ml
1000 g 0.5 ml 0.5 ml
10.13
(assume weight of glucose = 10.5)
% of glucose recovery = * 100 = 96.51%
8.Adjust sample values to 100 percent glucose recovery.
ADJ CH0% = samPle value (enzyme blank subtracted out)
glucose recovery
_ 6.72
0.9651
= 6.96%
Reagents
Alkaline reagent. Dissolve 25.0 g anhydrous sodium carbonate
(^COg), 25.0 g potassium sodium tartrate (KNaC^H, -4H20), 20.0 g

171
sodium bicarbonate (NaHCO^), and 200.0 g anhydrous sodium sulfate
(Na2S0^) in 700 ml distilled water and then dilute to 1000 ml in a
volumetric flask. Dissolve 6.0 g cupric sulfate pentahydrate (CuSO^ •
5H20) in 40 ml distilled water followed by one drop of concentrated
sulfuric acid (F^SO^). Combine and mix the two solutions. Follow these
directions!
Arsenomolybdate reagent. Dissolve 25.0 g ammonium molybdate
tetrahydrate [(NH^)(eMO^O^^O)] in 450 ml distilled water, then add 21
ml concentrated sulfuric acid while stirring. Dissolve 3.0 g sodium
arsenate (Na^AsO^ • 7H20) in 25 ml distilled water. Combine and mix the
two solutions and store in a brown bottle for 24 hours at 37 C. The
reagent should be yellow with no green tint and should be remade after
10 days.
NOTE: Arsenomolybdate reagent is easily contaminated by metal
surfaces so avoid metal bottle caps, etc. Follow directions!
Acetate buffer, 0.2 M, pH 4.5. Prepare the following stock
solutions:
0.2 M Acetic acid--To approximately 500 ml distilled water in a
1 liter volumetric flask add slowly and carefully 11.6 ml glacial acetic
acid. Cool and make to 1 liter with distilled water.
0.2 M Sodium Acetate--Dissolve 16.4 g Na^H^ or 27.2 g of
NaC2Hg02 â–  3H20 per liter of distilled water.

172
For 0.2 M buffer, combine 300 ml 0.2 M acetic acid with 200 ml
sodium acetate. Titrate the final buffer solution to pH 4.5 by addition
of either stock solution.
Refrigerate.
0.1 M Acetate Buffer. Dilute the 0.2 M buffer accordingly.

APPENDIX B
TOTAL NONSTRUCTURAL CARBOHYDRATE (TNC) RESULTS

Table A-3. Percent total nonstructural carbohydrate (TNC) means for all treatment, dates, and
limpograsses in 1980
HO
HI
H2
H3
H4
H5
H6
H7
H8
H8
HI 0
HI 1
HI 2
23
16
3
25
14
7
27
16
6
28
18
8
30
Mar
Apr
May
May
Jun
Jul
Jul
Aug
Sep
Sep
Oct
Nov
Nov
PI 364888
1
0
3
15.7
24.
5
14.4
14.9
15.5
12.0
10.5
12.0
19.4
14.4
15.0
12.8
16.2
2
120
3
14.1
25.
5
17.6
20.4
18.9
11.5
10.9
10.8
19.7
14.6
14.3
11.6
17.1
3
480
3
15.6
22.
6
21 .2
21.1
16.5
11.2
10.2
10.8
15.7
13.3
10.3
10.6
13.8
4
60
6
19.1
18.0
14.7
13.0
9.6
13.3
12.0
11.4
10.2
14.6
5
240
6
19.7
18.1
14.4
13.4
9.0
14.2
11.8
9.9
9.0
14.1
6
0
9
,
16.5
.
12.7
10.5
16.0
10.2
11.9
10.2
13.9
7
120
9
16.0
.
.
10.4
8.9
13.9
9.0
6.9
9.2
11.8
8
480
9
18.3
.
8.3
8.3
12.8
7.9
6.2
9.9
11.7
9
60
12
.
13.8
#
.
13.6
11.0
9.5
10.6
16.3
10
240
12
,
13.4
B
#
11.0
9.2
10.1
10.6
15.7
11
0
18
,
.
.
10.3
7.8
15.4
8.8
8.4
9.4
16.7
12
120
18
.
11.9
6.7
12.7
7.2
7.4
8.1
17.9
13
480
18
.
10.7
7.2
13.1
7.5
6.0
7.7
14.7

Table A-3. Continued
HO
HI
H2
H3
H4
H5
H6
H7
H8
H9
HI 0
HI 1
HI 2
23
16
3
25
14
7
27
16
6
28
18
8
30
Mar
Apr
May
May
Jun
Jul
Jul
Aug
Sep
Sep
Oct
Nov
Nov
Redalta
14 0
3
15.4
18.5
12.0
15
120
3
16.2
19.8
13.0
16
480
3
16.5
20.3
14.5
17
60
6
14.9
18
240
6
13.2
19
0
9
20
120
9
21
480
9
22
60
12
23
240
12
24
0
18
25
120
18
26
480
18
•
Bigalta
27 0
3
14.9
22.0
14.6
28
120
3
14.4
22.7
17.0
29
480
3
16.8
22.6
17.7
30
60
6
17.4
31
240
6
m
.
16.8
32
0
9
m
33
120
9
m
13.2
12.8
9.8
8.6
10.5
13.5
13.6
8.1
8.0
8.2
11.1
12.3
6.9
6.4
6.8
15.7
8.3
8.4
6.3
9.9
6.4
6.3
4.8
12.2
.
8.5
6.4
9.0
.
5.7
4.7
8.1
,
6.0
5.8
,
11.4
.
9.2
s
#
.
#
.
9.9
6.2
.
.
.
9.3
4.4
#
u
m
6.3
4.5
14.8
17.0
12.6
11.4
13.4
18.2
20.1
14.1
11.3
13.8
26.0
19.8
13.6
9.9
11.4
#
20.2
12.8
13.6
11.5
19.0
11.8
9.6
9.5
15.0
11.3
10.5
14.0
.
10.6
8.6
11.8
13.5
11.2
13.4
15.6
14.0
11.5
9.3
11.9
12.6
11.4
10.3
8.4
11.6
8.7
9.4
9.1
9.1
9.2
9.0
8.3
6.7
7.1
9.1
6.4
10.1
9.4
7.5
10.4
9.4
8.5
6.5
4.9
10.0
7.4
6.7
6.2
4.4
9.7
7.0
9.6
7.4
6.7
10.2
9.9
6.9
7.0
5.9
8.9
10.7
9.8
8.7
9.2
12.2
14.5
7.3
7.1
7.0
11.0
11.5
6.8
5.1
5.7
9.0
10.3
17.5
15.2
13.7
13.5
14.4
17.1
15.2
11.7
12.8
13.9
15.7
13.4
11.6
13.0
11.5
14.4
10.0
12.4
10.8
13.4
10.4
7.4
9.1
7.3
7.9
11.3
8.3
6.6
8.6
6.0
9.5
6.8
6.2
7.4
7.2

Table A-3. Continued
OB
N*
HO
HI
H2
H3
H4
H5
H6
H7
H8
H9
HI 0
HI 1
HI 2
p**
23
16
3
25
14
7
27
16
6
28
18
8
30
Mar
Apr
May
May
Jun
Jul
Jul
Aug
Sep
Sep
Oct
Nov
Nov
34
480
9
12.7
10.4
9.7
11.1
7.1
6.5
8.8
9.3
35
60
12
14.4
8.9
8.7
8.2
8.4
12.3
36
240
12
13.3
8.3
7.6
8.4
8.6
11.2
37
0
18
13.9
8.8
10.6
7.6
8.4
7.6
11.2
38
120
18
12.2
8.1
10.4
7.5
7.0
7.2
11.4
39
240
18
11.1
7.3
10.4
6.8
7.0
7.5
10.2
PI
349753
40
0
3
12.8
20.8
15.7
18.0
17.3
13.2
9.2
10.1
13.5
12.0
9.6
9.8
8.4
41
120
3
16.7
23.5
16.5
17.3
17.0
11.8
11.5
11.4
12.7
10.2
9.6
8.9
8.7
42
480
3
17.2
25.1
18.9
20.7
16.3
12.8
10.7
10.4
9.8
7.8
7.7
8.8
6.6
43
60
6
17.3
.
19.5
10.6
10.6
8.9
9.5
8.3
7.7
7.4
8.4
44
240
6
16.5
14.1
9.2
10.6
7.7
7.6
6.7
5.3
6.2
6.9
45
0
9
.
17.2
#
11.6
7.9
9.9
7.0
7.5
7.1
10.5
46
120
9
#
15.2
8.8
5.6
8.8
5.9
4.7
6.9
8.5
47
480
9
.
13.5
8.8
6.9
8.8
4.8
4.2
7.3
8.4
48
60
12
#
15.4
8.2
7.0
6.6
7.6
7.9
49
240
12
•
.
10.6
,
6.8
6.5
5.0
7.5
11.4
50
0
18
#
12.9
8.2
11.0
7.1
5.9
8.4
11.4
51
120
18
a
9.8
6.1
8.6
5.1
4.9
6.7
9.4
52
480
18
•
•
•
10.7
5.4
8.2
4.8
4.4
6.1
8.3
^Nitrogen (kg/ha/yr).
**Frequency (weeks).

REFERENCES
Adams, Alto Jr. 1982. Development of the Braford Breed. Proc. 31st
Annual Beef Cattle Short Course, Institute of Food and Agricultural
Sciences, University of Florida, Gainesville 32611, pp. 91-94.
Adegbola, A. 1966. Preliminary observations on the reserve carbohy¬
drate and regrowth potential of tropical grasses. Proc. 10th Int.
Grassld. Congr. Helsinki, Finland, pp. 933-936.
Adegbola, A. A., and C. M. McKell. 1966a. Effect of nitrogen fertili¬
zation on the carbohydrate content of Coastal bermudagrass (Cynodon
dactyl on (L.), Pers.). Agron. J. 58:60-64.
Adegbola, A. A., and C. M. McKell. 1966b. Regrowth potential of
Coastal bermudagrass as related to previous nitrogen fertilization.
Agron. J. 58:145-146.
Agriculture in an Urban Age (AGUA) Report. 1974. Institute of Food
and Agricultural Sciences, University of Florida, Gainesville
32611.
Alberda, Th. 1966. The influence of reserve substances on dry matter
production after defoliation. Proc. 10th Int. Grassld. Congr.
Helsinki, Finland, pp. 140-147.
Alexander, C. W., and D. E. McCloud. 1962. Influence of time and rate
of nitrogen application on production and botanical composition of
forage. Agron. J. 54:521-522.
Aminoff, D., W. W. Binkley, R. Schaffer, and R. W. Mowry. 1970.
Analytical methods for carbohydrates. Jji The Carbohydrates, Vol.
I IB. Academic Press, New York, pp. 740-807.
Ammerman, C. B. 1979. Meeting the mineral and protein needs of Florida
beef cattle. Proc. 28th Annual Beef Cattle Short Course, Institute
of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 108-117.
Archbold, H. K. 1945. Some factors concerned in the process of starch
storage in the barley grain. Nature 156:70-73.
Arroyo-Aguilu, J. A., and J. Coward-Lord. 1974a. Mineral composition
of 10 tropical forage grasses in Puerto Rico. J. Agrie. Univ.
Puerto Rico 58:426-436.
177

178
Arroyo-Aguilu, J. A., and J. Coward-Lord. 1974b. Relationships between
and within physical and chemical constituents and in vitro true
digestibility in tropical forage grasses. J. AgricT Univ. Puerto
Rico 58:437-447.
Auda, Hamid, R. E. Blaser, and R. H. Brown. 1966. Tillering and carbo¬
hydrate contents of orchardgrass as influenced by environmental
factors. Crop Sci. 6:139-143.
Baker, F. S. Jr. 1979. Finishing cattle in north Florida: Present and
future. Proc. 28th Annual Beef Cattle Short Course, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 122-125.
Baker, F. S. Jr. 1980. Let's produce slaughter beef in Florida. Proc.
29th Annual Beef Cattle Short Course, Institute of Food and Agri¬
cultural Sciences, University of Florida, Gainesville 32611, pp.
5-7.
Baker, H. K., and E. A. Garwood. 1961. Studies on the root development
of herbage plants. V. Seasonal changes in fructosan and soluble-
sugar contents of cocksfoot herbage, stubble, and roots under two
cutting treatments. J. Brit. Grassld. Soc. 16:263-267.
Beaty, E. R., K. H. Tan, R. A. McCreery, and J. D. Powell. 1980. Yield
and N content of closely clipped bahiagrass as affected by N
treatment. Agron. J. 72:56-60.
Becker, R. B. (personal communication) Professor Emeritus, Department
of Animal Science, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611.
Bernatowicz, A. J. 1958. Teleology in science teaching. Science 128:
1402-1405.
Blaser, R. E., R. H. Brown, and H. T. Bryant. 1966. The relationship
between carbohydrate accumulation and growth of grasses under dif¬
ferent microclimates. Proc. 10th Int. Grassld. Congr. Helsinki,
Finland, pp. 147-150.
Blue, W. G., N. Gammon Jr., and H. W. Lundy. 1961. Late summer fer¬
tilization for winter forage in north Florida. Proc. Soil Crop
Sci. Soc. Fla. 21:56-62.
Bogdan, A. V. 1977. Tropical Pasture and Fodder Plants. Tropical
Agriculture Series. Longman, Inc., New York, 475 pp.
Bor, N. L. 1960. The Grasses of Burma, Ceylon, India, and Pakistan.
Pergammon Press, New York, 767 pp.
Box, G. E. P., W. G. Hunter, and J. S. Hunter. 1978. Statistics for
Experimenters. Johy Wiley and Sons, Inc., New York, pp. 510-539.

179
Box, G. E. P., and K. B. Wilson. 1951. On the experimental attainment
of optimum conditions. J. Roy. Statist. Soc. Ser. B 13, No. 1.
Boyd, F. T., and V. G. Perry. 1969. Effect of sting nematodes on
establishment, yield, and growth of forage grasses on Florida sandy
soils. Proc. Soil Crop Sci. Soc. Fla. 29:288-300.
Boyd, F. T., and V. G. Perry. 1970. Effect of seasonal temperatures
and certain cultural treatments on sting nematodes in forage grass.
Proc. Soil Crop Sci. Soc. Fla. 30:360-365.
Boyd, F. T., V. N. Schroder, and V. G. Perry. 1972. Interaction of
nematodes and soil temperature on growth of three tropical grasses.
Agron. J. 64:497-500.
Brown, R. H. 1978. A difference in N use efficiency in C3 and C4
plants and its implications in adaption and evolution. Crop Sci.
18:93-98.
Brown, R. H., and R. E. Blaser. 1965. Relationships between reserve
carbohydrate accumulation and growth rate in orchardgrass and tall
fescue. Crop Sci. 5(6):577-582.
Bryan, Erwin Jr. 1981. Problems and opportunities in marketing cows
and feedlot beef produced in Florida and the southeast United
States—Packer views. Proc. 30th Annual Beef Cattle Short Course,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 46-47.
Burris, J. S., R. H. Brown, and R. E. Blaser. 1967. Evaluation of
reserve carbohydrates in Midland bermudagrass (Cynodon dactylon
L.). Crop Sci. 7:22-24.
Burton, G. W., and J. E. Jackson. 1962. A method for measuring sod
reserves. Agron. J. 54:53-55.
Cady, F. B., and W. A. Fuller. 1970. The statistics-computer interface
in agronomic research. Agron. J. 62:599-604.
Carpenter, W. J. (personal communication) Agricultural Technician,
Department of Agronomy, Institute of Food and Agricultural
Sciences, University of Florida, Gainesville 32611.
Carter, J. L., and L. A. Garrard. 1976. Effects of 10 C-night tempera¬
ture and leaf assimilate levels on photosynthesis of Di gitari a
decumbens Stent. Proc. Soil Crop Sci. Fla. 35:3-5.
Chang, C. F., A. Suzuki, S. Kumai, and S. Tamura. 1969. Chemical
studies on 'clover sickness'. Part II. Biological functions of
isoflavinoids and their related compounds. Agrie. Biol. Chem. 33:
398-408.

180
Chatterton, N. J., G. E. Carlson, W. E. Hungerford, and D. R. Lee.
1972. Effect of tillering and cool nights on photosynthesis and
chloroplast starch in Pangóla. Crop Sci. 12:206-208.
Chaves, M. M. C. F., and I. Moreira.^ 1977. Métodos para a determinado
de hidratos de carbono totais nao estruturais. Agronomia Lusit.
38(1):41-56.
Chippendall, L. K. A. 1955. A guide to the identification of grasses
in South Africa. _In_ D. Meredith (ed.), The Grasses and Pastures
of South Africa. Central News Agency, Parow, C. P. Cape Times
LTD, 527 pp.
Christiansen, Scott, 0. C. Ruelke, and R. 0. lynch. 1981. Regrowth in
darkness as influenced by previous cutting treatment of four
limpograss genotypes. Proc. Soil Crop Sci. Soc. Fla. 40:156-159.
Cochran, W. G., and G. M. Cox. 1957. Experimental Designs. John Wiley
and Sons, Inc., New York, 611 pp.
Colby, W. G., Mack Drake, D. L. Field, and G. Kreowski. 1965. Seasonal
pattern of fructosan in orchardgrass stubble as influenced by
nitrogen and harvest management. Agron. J. 57:169-173.
Collins, A. C. 1946. The Story of America in Pictures. Doubleday and
Company, Inc., New York, 475 pp.
Council, K. A., and J. T. Helwig (eds.). 1979. SAS User's Guide.
SAS Institute Inc., P. 0. Box 8000, Cary, North Carolina 27511,
494 pp.
Council, K. A., and J. T. Helwig (eds.). 1981. SAS/GRAPH User's Guide.
SAS Institute Inc., P. 0. Box 8000, Cary, North Carolina 27511,
126 pp.
Coward-Lord, J., J. A. Arroyo-Aguilu, and 0. Garcia-Molinari. 1974a.
Fibrous carbohydrate fractions and in vitro true and apparent
digestibility of 10 tropical forage grasses. J. Agrie. Univ.
Puerto Rico 58:293-304.
Coward-Lord, J., J. A. Arroyo-Agui1u, and 0. Garcia-Molinari. 1974b.
Proximate nutrient composition of 10 tropical forage grasses.
J. Agrie. Univ. Puerto Rico 58:305-311.
Creel, J. M. Jr. 1957. The effect of continuous high nitrogen fertili¬
zation on 'Coastal' bermudagrass and 'Pangolagrass'. Master's
thesis, University of Florida, Gainesville 32611.
Crockett, J. R. 1982. Continental exotics for crossbred beef produc¬
tion in south Florida. Proc. 31st Annual Beef Cattle Short Course,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 11-16.

181
Cugnac, Antoine de. 1931. Research on glucosides of the Gramineae.
Ann. Sci. Nat. 13:1-130.
daSilveira, A. J., F. F. Feitosa Teles, and J. W. Stull. 1978. A rapid
technique for total nonstructural carbohydrate determination of
plant tissue. J. Agrie. Food Chem. 26(3):770-772.
Davidson, J. L., and F. L. Milthorpe. 1965. Carbohydrate reserves in
the regrowth of cocksfoot (Dactylis glomerata L.). J. Brit.
Grass!d. Soc. 20:15-18.
Davidson, J. L., and F. L. Milthorpe. 1966a. The effect of defoliation
on the carbon balance in Dactylis glomerata. Ann. Bot., N.S., 30:
185-198.
Davidson, J. L., and F. L. Milthorpe. 1966b. Leaf growth in Dactylis
glomerata following defoliation. Ann. Bot., N.S., 30:173-184.
Diesche, Z. 1962. Colorimetric methods in carbohydrate chemistry
(Introduction). Jn^ Methods in Carbohydrate Chemistry, Vol. I,
R. L. Whistler and M. L. Wolfrom (eds.). Academic Press, New York,
pp. 477-478.
Dohrenwend, R. E. 1978. The climate of Alachua County, Florida. Bull.
796, Agricultural Experiment Station, Institute of Food and Agri¬
cultural Sciences, University of Florida, Gainesville 32611, 25
pp.
Dovrat, A., and Y. Cohen. 1970. Regrowth potential of Rhodesgrass
(Chloris gayana Kunth.) as affected by nitrogen and defoliation.
Proc. 11th Int. Grassld. Congr. Surfer's Paradise, Australia, pp.
552-554.
Dubois, M., K. A. Gilíes, J. K. Hamilton, P. A. Rebers, and F. Smith.
1956. Colorimetric method for determination of sugars and related
substances. Anal. Chem. 28:350-356.
Ehara, K., N. Maeno, and Y. Yamada. 1966. Physiological and ecological
studies on the regrowth of herbage plants. 4. The evidence of
utilization of food reserves during the early stage of regrowth in
bahiagrass (Paspalum notatum FUigge) with C^ 02- J- Jap. Soc.
Grassld. Sci. 12(1):1-13 (from Herbage Abstracts 37(348):53).
Ferraris, R. 1978. The effect of photoperiod and temperature on the
first crop and ratoon growth of Pennisetum purpureum Schum. Aust.
J. Agrie. Res. 29:941-950.
Fields, M. J., and J. F. Hentges Jr. 1979. Semen traits, testicular
growth and testosterone patterns in brahman versus angus bulls.
Proc. 28th Annual Beef Cattle Short Course, Institute of Food and
Agricultural Sciences, University of Florida, Gainesville 32611,
pp. 62-67.

182
Ford, C. W., and W. T. Williams. 1973. In vitro digestibility and
carbohydrate composition of Digitari a decumbens and Setaria anceps
grown at different levels of nitrogenous fertilizer. Aust. J.
Agrie. Res. 24:309-316.
Gallaher, R. N., and R. H. Brown. 1977. Starch storage in C4 vs C3
grass leaf cells as related to nitrogen deficiency. Crop Sci. 17:
85-88.
Garrard, L. A., and S. H. West. 1972. Suboptimal temperature and
assimilate accumulation in leaves of 'Pangóla' digitgrass
(Digitaria decumbens (Stent.). Crop Sci. 12:621-623.
Gaskins, M. H., and D. A. Sleper. 1974. Photosensitivity of some
tropical forage grasses in Florida. Proc. Soil Crop Sci. Soc. Fla.
33:20-21.
Goering, H. K., and P. J. Van Soest. 1970. Forage fiber analyses
(apparatus, reagents, procedures, and some applications). USDA
Agrie. Handbook 379.
Graber, L. F. 1931. Food reserves in relation to other factors limit¬
ing the growth of grasses. Plant Physiol. 6:43-71.
Graber, L. F., N. T. Nelson, W. A. Leukel, and W. B. Albert. 1927.
Organic food reserves in relation to the growth of alfalfa and
other perennial herbaceous plants. Wisconsin Agrie. Exp. Sta.
Bull. 80.
Greenfield, P. L., Dale Smith, and Jose A. Escalada. 1974. Composition
of starch accumulated in the leaf lamineae of timothy, sorghum, and
lucerne. J. Sci. Food Agrie. 25:1363-1368.
Greenfield, S. B., and Dale Smith. 1974. Diurnal variations in the
individual parts of switchgrass shoots at anthesis. J. Range Mgmt.
27(6):466-469.
Greenwood, C. T. 1970. Starch and glycogen. I. Starch. In The
Carbohydrates, Vol. I IB. W. Pigman and D. Horton (eds.JT Academic
Press, New York, pp. 471-499.
Greub, L. J., and W. F. Wedin. 1969. Effects of fineness of grind and
periodic agitation on total available carbohydrate values obtained
by enzyme saccharification. Crop Sci. 9:691-692.
Grotelueschen, R. D., and Dale Smith. 1967. Determination and identi¬
fication of nonstructural carbohydrates removed from grass and
legume tissue by various sulfuric acid concentrations, takadiastase-,
and water. J. Agrie. Food Chem. 15(6): 1048-1051.
Grotelueschen, R. D., and Dale Smith. 1968. Carbohydrates in grasses.
III. Estimations of the degree of polymerization of the fructosans
in the stem bases of timothy and bromegrass near seed maturity.
Crop Sci. 8:210-212.

183
Hader, R. J., M. E. Harward, D. D. Mason, and D. P. Moore. 1957. An
investigation of some of the relationships between copper, iron,
and molybdenum in the growth and nutrition of lettuce: 1.
Experimental design and statistical methods for characterizing the
response surface. Proc. Soil Sci. Soc. Amer. 21:59-64.
Hall, D. W. 1978. The grasses of Florida. Ph.D. dissertation,
University of Florida, Gainesville 32611.
Haslemore, R. M., and P. G. Roughan. 1976. Rapid chemical analysis of
some plant constituents. J. Sci. Food Agrie. 27:1171-1178.
Henderson, J. R. 1956. Florida pastures flourish with care. Plant
Food Review 2(1):8-10, 31-33.
Henderson, M. S., and D. L. Robinson. 1982. Environmental influences
on fiber component concentrations of warm-season perennial grasses.
Agron. J. 74:573-579.
Hill, W. J., and W. G. Hunter. 1966. A review of response surface
methodology: A literature survey. Technometrics 8(4):571 -590.
Hodge, J. E., and B. T. Hofreiter. 1962. Determination of reducing
sugars and carbohydrates. In_ Methods in Carbohydrate Chemistry,
Vol. I. R. L. Whistler and M. L. Wolfrom (eds.), Academic Press,
New York, pp. 380-394.
Hodges, E. M., and F. G. Martin. 1975. Forage production of perennial
grasses as affected by fertilizer rate and season. Proc. Soil
Crop Sci. Soc. Fla. 34:158-161.
Hodges, E. M., and W. D. Pitman. 1981. Grazing evaluation of perennial
pasture grasses in peninsular Florida. Agron. Abstracts, pp. 132-
133.
Holt, D. A., and A. R. Hilst. 1969. Daily variation in carbohydrate
content of selected forage crops. Agron. J. 61:239-242.
Horton, G. M. J., and P. Mislevy. 1981. Performance of growing steers
fed different proportions of corn silage and forage sorghum silage
in the diet. Florida Beef Cattle Res. Report, Institute of Food
and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 49-52.
Horton, G. M. J., P. Mislevy, and B. E. Melton. 1981. The performance
of cattle on corn and sorghum silages grown in a multi cropping
system. Proc. 30th Annual Beef Cattle Short Course, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 84-93.
Humphreys, L. R. 1966. Pasture defoliation practice: A review. J.
Aust. Inst. Agrie. Sci. 32:93-104.

184
Humphreys, L. R., and A. R. Robinson. 1966. Interrelations of leaf
area and non-structural carbohydrate status as determinants of the
growth of sub-tropical grasses. Proc. 10th Int. Grassld. Congr.
Helsinki, Finland, pp. 113-116.
Hunter, D. H., L. D. Harris, and A. S. Jensen. 1979. Integrated
Natural Resource Management. Bull. 190, Florida Cooperative Exten¬
sion Service, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, 50 pp.
Hunter, R. A., B. L. McIntyre, and R. J. Mcllroy. 1970. Water-soluble
carbohydrates of tropical pasture grasses and legumes. J. Sci.
Food Agrie. 21:400-405.
Ikerd, John. 1981. Beef industry problems in the 1980's: Money,
marketing, and transportation. Proc. Beef Industry Conf.,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 44-64.
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.
Kaplan, Donald. 1981. Problems and opportunities in marketing cows and
feedlot beef produced in Florida and the southeast United States--
Packer views. Proc. 30th Annual Beef Cattle Short Course,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 48-49.
Kien, L. T., 0. C. Ruelke, and H. L. Breland. 1975. Effects of
fertilizer on 'Pangóla' and 'Transvala' digitgrass, and
'Coastcross-1' bermudagrass. Proc. Soil and Crop Sci. Soc. Fla.
35:80-83.
Killinger, G. B. 1971. Limpograss (Hemarthria altissima (Poir) Stapf
et C.E. Hubb.), a promising forage and beverage grass for the
south. Agron. Abstracts, p. 56.
Killinger, G. B., and C. F. Beckham. 1973. Beverage from plants of the
genus Hemarthria. United States Patent 3,709,694.
Koger, Marvin. 1982. Selection and breeding for maximum efficiency of
beef production in Florida. Proc. of the 31st Annual Beef Cattle
Short Course, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, pp. 4-5.
Kretschmer, A. E. Jr. 1965. The effect of nitrogen fertilization of
mature Pangolagrass just prior to utilization in the winter on
yields, dry matter, and crude protein contents. Agron. J. 57:529-
534.
Kretschmer, A. E. Jr., and G. H. Snyder. 1979. Production and quality
of limpograss for use in the subtropics. Agron. J. 71:37-41.

185
Leukel, W. A., and J. M. Coleman. 1930. Growth behavior and mainten¬
ance of organic foods in bahiagrass. Agricultural Experiment
Station Bull. 219, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, 56 pp.
Lindahl, I., R. E. Davis, and W. 0. Shepard. 1949. The application of
the total carbohydrate method to the study of carbohydrate reserves
of switch cane (Arundinara tecta). Plant Physiol. 24:285-294.
Littell, R. C., and G. 0. Mott. 1975. Computer assisted design and
analysis of response surface experiments in agronomy. Proc. Soil
Crop Sci. Soc. Fla. 34:94-97.
Marshall, C., and G. R. Sagar. 1965. The influence of defoliation on
the distribution of assimilates in Lolium multiflorum Lam. Ann.
Bot. 29:365-372.
Matches, A. G. 1969. Influence of cutting height in darkness on
measurement of energy reserves of tall fescue. Agron. J. 61:896-
898.
May, L. H. 1960. The utilization of carbohydrate reserves in pasture
plants after defoliation. Herbage Abstracts 30(4):239-245.
May, L. H., and J. L. Davidson. 1958. The role of carbohydrate
reserves in regeneration of plants. Aust. J. Agrie. Res. 9:767-
777.
McCarty, E. C. 1935. Seasonal march of carbohydrates in Elymus
ambigúus and Muhlenbergia gracilis and their reaction under
moderate grazing use. Plant Physiol. 10:727-738.
McCarty, E. C. 1938. The relation of growth to the varying carbohy¬
drate content in mountain brome. USDA Tech. Bull. 598.
McDowell, L. R., M. Kiatoko, J. Bertrand, H. L. Chapman Jr., F. M. Pate,
and J. H. Conrad. 1980. Mineral Status of Beef Cattle Herds from
Four Soil Type Regions in Florida. Proc. 29th Annual Beef Cattle
Short Course, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, pp. 99-104.
Milford, R., and D. J. Minson. 1965. Intake of tropical pasture
species. Proc. 9th Int. Grassld. Congr. Sao Paulo, Brazil, pp.
815-822.
Milthorpe, F. L., and J. L. Davidson. 1966. Physiological aspects of
regrowth in grasses. The Growth of Cereals and Grasses. F. L.
Milthorpe and J. D. Ivins (eds.), Butterworths, London.
Mislevy, P., J. B. Washko, and J. D. Harrington. 1978. Plant maturity
and cutting frequency effect on total nonstructural carbohydrate
percentages in the stubble and crown of timothy and orchardgrass.
Agron. J. 70:907-912.

186
Moore, J. E., G. 0. Mott, D. G. Dunham, and R. W. Omer, 1972. Large
capacity in vitro organic matter digestion procedure. J. Ani. Sci.
35:232 (Abstract).
Moore, J. E., 0. C. Ruelke, C. E. Rios, and D. E. Franke. 1970.
Nutritive evaluation of Pensacola habiagrass hays. Proc. Soil Crop
Sci. Soc. Fla. 30:211-221.
Moore, J. E., M. A. Worrell, S. M. Abrams, W. R. Ocumpaugh, and G. 0,
Mott. 1981. Quality of tropical perennial grass hays. Fla. Beef
Cattle Res. Report, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, pp. 40-44.
Mott, G. 0. 1982a. Evaluation of pasture germplasm under different
grazing management alternatives (unpublished paper). Department of
Agronomy, Institute of Food and Agricultural Sciences, University
of Florida, Gainesville 32611.
Mott, G. 0. 1982b. Management of forages for beef production in
Florida. Proc. 31st Annual Beef Cattle Short Course, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 77-82.
Mott, G. 0., and J. E. Moore. 1977. Existing and potential systems of
finishing cattle on forages or limited grain rations in the tropi¬
cal region of the south. In_ Forage-fed Beef: Production and
Marketing Alternatives in the South. J. A. Stuedemann (ed,), Bull.
220, Southern Cooperative Series, pp. 419-443.
Myers, R. H. 1971. Response Surface Methodology. Allyn and Bacon,
Inc., Boston, 246 pp.
Nelson, N. 1944. A photometric adaptation of the Somogyi method for
the determination of glucose. J. Biol. Chem. 153:375-380.
Noble, A., and K. F. Lowe. 1974. A1cohol-soluble carbohydrates in
various tropical and temperate pasture species. Trop. Grasslds.
8(3):179-187.
Oakes, A. J. 1973. Hemarthria collection from South Africa. Turrialba
23(1):37-40.
Oakes, A. J. 1978. Resistance in Hemathria species to the yellow sugar
cane aphid, Sipha flava (ForbesT Trop. Agrie. (Trinidad) 55(4):
377-381.
Oakes, A. J. 1980. Winter hardiness in limpograss, Hemarthria
altissima. Proc. Soil Crop Sci. Soc. Fla. 39:86-88.
Oakes, A. J., and C. D. Foy. 1980. A winter hardy, aluminum tolerant
perennial pasture grass for mine spoil reclamation. Agron.
Abstracts, p. 103.

187
Ocumpaugh, W. R. 1982. A new Hemarthria selected for persistence and
quality. Proc. 31st Annual Beef Cattle Short Course, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 83-84.
Ocumpaugh, W. R., and K. H. Quesenberry. 1980. Yield and quality of
limpograss in the autumn-winter. Fla. Beef Cattle Res. Report,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 33-38.
Ocumpaugh, W. R., K. H. Quesenberry, and 0. C. Ruelke. 1981. Grazing
frequency effects on the persistence of eight 1impograsses. Agron.
Abstracts, p. 111.
Okajima, Hideo, and Dale Smith. 1964. Available carbohydrate fractions
in the stem bases and seed of timothy, smooth bromegrass, and
several other northern grasses. Crop Sci. 4:317-320.
Pate, F. M. (personal communication) Professor of Animal Science,
Institute of Food and Agricultural Science, Agricultural Research
and Education Center, Belle Glade, Florida.
Pearce, R. B., R. H. Brown, and R. E. Blaser. 1965. Relationships
between leaf area index, light interception and net photosynthesis
in orchardgrass. Crop Sci. 5(6):553-556.
Perry, L. J. Jr., and L. E. Moser. 1974. Carbohydrate and organic
nitrogen concentrations within range grass parts at maturity. J.
Range Mgmt. 27(4):276-278.
Quesenberry, K. H. (personal communication) Associate Professor of
Agronomy, Institute of Food and Agricultural Science, University of
Florida, Gainesville 32611.
Quesenberry, K. H., L. S. Dunavin Jr., E. M. Hodges, G. B. Killinger,
A. E. Kretschmer Jr., W. R. Ocumpaugh, R. D. Roush, 0. C. Ruelke,
S. C. Schank, D. C. Smith, G. H. Synder, and R. L. Stanley. 1978.
Redalta, Greenalta, and Bigaita limpograss, Hemarthria altissima,
promising forages for Florida. Florida Agricultural Experiment
Station Bull. 802, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611, 18 pp.
Quesenberry, K. H., L. S. Dunavin Jr., E. M. Hodges, G. B. Killinger,
A. E. Kretschmer Jr., W. R. Ocumpaugh, R. D. Roush, 0. C. Ruelke,
S. C. Schank, D. C. Smith, G. H. Synder, and R. L. Stanley. 1979.
Registration of Redalta, Greenalta, and Bigalta limpograss. Crop
Sci. 19:294.
Quesenberry, K. H., and R. A. Dunn. 1977. Differential response of
Hemarthria genotypes to sting nematodes in a greenhouse screening
trial. Proc. Soil Crop Sci. Soc. Fla. 37:58-61.

188
Quesenberry, K. H., A. J. Oakes, and Dorothy S. Jessop. 1982.
Cytological and geographical characterizations of Hemarthria.
Euphytica 31:(in press).
Quesenberry, K. H., and W. R. Ocumpaugh. 1977. Forage quality and
yield of limpograss as a conserved forage for winter grazing.
Agron. Abstracts, p. 119.
Quesenberry, K. H., and W. R. Ocumpaugh. 1979. Persistence, yield, and
digestibility of limpograss genotypes under clipping and grazing.
Agron. Abstracts, p. 108.
Quesenberry, K. H., and W. R. Ocumpaugh. 1980. Crude protein, IVOMD,
and yield of stockpiled limpograsses. Agron. J. 72:1021-1024.
Quesenberry, K. H., and W. R. Ocumpaugh. 1982. Mineral composition of
autumn-winter stockpiled limpograss. Trop Agrie. (Trinidad):(in
press).
Quesenberry, K. H., W. R. Ocumpaugh, and 0. C. Ruelke. 1981. Hemarthia
altissima: A pasture grass for the tropics. Proc. 14th Int.
Grassld. Congr. Lexington, KY 40506 (in press).
Ráese, J. T., and A. M. Decker. 1966. Yields, stand persistence, and
carbohydrate reserves of perennial grasses as influenced by spring
harvest stage, stubble height, and nitrogen fertilization. Agron.
J. 58:322-326.
Ruelke, 0. C. 1960. Fertility, as a limiting factor for pastures in
Florida. Proc. Soil Crop Sci. Soc. Fla. 20:23-28.
Ruelke, 0. C. 1978. Effects of management on the production, quality,
and persistence of various limpograss cultivars. Agron. Abstracts,
p: 104.
Ruelke, 0. C., and G. M. Prine. 1971. Performance of six hybrid
bermudagrasses, Pangóla digitgrass, and Pensacola bahiagrass at
three fertility levels in north central Florida. Proc. Soil Crop
Sci. Soc. Fla. 31:67-71.
Ruelke, 0. C., and K. H. Quesenberry. 1981. Topseeding winter clovers
on limpograss, potentials and problems. Proc. Soil Crop Sci. Soc.
Fla. 40:162-164.
Ruelke, 0, C., and K. H. Quesenberry. 1982. Limpograss for off-season
forage production for beef cattle. Fla. Beef Cattle Res. Report,
Institute of Food and Agricultural Sciences, University of Florida,
Gainesville 32611, pp. 39-42.
Ruelke, 0. C., K. H. Quesenberry, and W. R. Ocumpaugh. 1978. Planting
technique effects on establishment, ground cover, production, and
digestion of Hemarthria altissima (Poir) Stapf et C.E. Hubb. Proc.
Soil Crop Sci. Soc. Fla. 38:40-42.

189
Ruelke, 0. C., K. H. Quesenberry, and D. A. Sleper. 1976. Comparison
of greenhouse vs. field plot techniques for evaluating new germ-
plasm of limpograss, Hemarthria altissima (Poir) Stapf et C.E.
Hubb. Agron. Abstracts, p. 112.
Schank, S. C. 1972. Chromosome numbers in eleven new Hemarthria
(limpograss) introductions. Crop Sci. 12:550-551.
Schank, S. C., M. A. Klock, and J. E. Moore. 1973. Laboratory evalua¬
tion of quality in subtropical grasses. II. Genetic variation
among Hemarthrias in in vitro digestion and stem morphology.
Agron. J. 65:256-258.
Schoney, R. A., T. F. Bay, and J. F. Moncrief. 1981. Use of computer
graphics in the development and evaluation of response surfaces.
Agron. J. 73:437-442.
Sheard, R. W. 1973. Organic reserves and plant regrowth. I_n The
Chemistry and Biochemistry of Herbage, Vol. 2. G. W. Butler and
R. W. Bailey (eds.), Academic Press, New York, pp. 353-377.
Smith, Dale, (personal communication) Adjunct Professor of Plant
Sciences, University of Arizona, Tucson 85721.
Smith, Dale. 1967. Carbohydrates in grasses. II. Sugar and fructosan
composition of the stem bases of bromegrass and timothy at several
growth stages and in different plant parts at anthesis. Crop Sci.
7:62-67.
Smith, Dale. 1968. Classification of several North American grasses as
starch or fructosan accumulators in relation to taxonomy. J. Brit.
Grassld. Soc. 23:306-309.
Smith, Dale. 1969. Removing and analyzing total nonstructural carbo¬
hydrate from plant tissue. Wisconsin Agrie. Exp. Sta. Res. Report
41.
Smith, Dale. 1973a. Influence of drying and storage conditions on non¬
structural carbohydrate analysis of herbage tissue—A review. J.
Brit. Grassld. Soc. 28:129-134.
Smith, Dale. 1973b. The nonstructural carbohydrates. Li The
Chemistry and Biochemstiry of Herbage, Vol. 1. G. W. Butler and
R. W. Bailey (eds.), Academic Press, New York, pp. 105-155.
Smith, Dale. 1975. Trends in nonstructural carbohydrates in the stem
basis of switchgrass. J. Range Mgmt. 28(5):389-391.
Smith, Dale. 1981a. Management of alfalfa. In_ Forage Management in
the North. Kendall/Hunt Publishing Co., Dubuque, Iowa, pp. 89-99.
Smith, Dale. 1981b. Removing and analyzing total nonstructural carbo¬
hydrates from plant tissue. Wisconsin Agrie. Bull. R2107
(Supersedes 1969 Wisconsin Agrie. Exp. Sta. Res. Report 41).

190
Smith, Dale, and R. D. Grotelueschen. 1966. Carbohydrates in grasses.
I. Sugar and fructosan composition of the stem bases of several
northern-adapted grasses at seed maturity. Crop Sci. 6:263-266.
Smith, Dale, G. M. Paulsen, and C. A. Raguse. 1964. Extraction of
total available carbohydrates from grass and legume tissue. Plant
Physiol. 39:960-962.
Somogyi, M. 1945. A new reagent for the determination of sugars. J.
Biol. Chem. 160:61-68.
Somogyi, M. 1952. Notes on sugar determination. J. Biol. Chem. 195:
19-23.
Sprague, V. G., and J. T. Sullivan. 1950. Reserve carbohydrates in
orchard grass clipped periodically. Plant Physiol. 25:92-102.
Spreen, T. H., and G. S. Shonkwiler. 1982a. Origin, grade, and form of
beef consumed in Florida. Unpublished paper, Food and Resource
Economics Department, Institute of Food and Agricultural Sciences,
University of Florida, Gainesville 32611.
Spreen, T. H., and G. S. Shonkwiler. 1982b. The impact of increasing
energy costs on the competitive advantages of finishing cattle in
Florida. Proc. 31st Annual Beef Cattle Short Course, Institute of
Food and Agricultural Sciences, University of Florida, Gainesville
32611, pp. 50-58.
Stegelin, F. E., and J. R. Simpson. 1980. An economic analysis of the
effect of increasing transportation costs on Florida's cattle feed¬
ing industry. Staff Paper 161r, Food and Resource Economics
Department, Institute of Food and Agricultural Sciences, University
of Florida, Gainesville 32611, 31 pp.
Sullivan, J. T., and V. G. Sprague. 1943. Composition of the roots and
stubble of perennial ryegrass following partial defoliation. Plant
Physiol. 18:656-670.
Tamura, S., C. Chang, A. Suzuki, and S. Kumai. 1967. Isolation and
structure of a novel isoflaven derivative in red clover. Agrie.
Biol. Chem. 31:1108-1109.
Tang, C. S., and C. C. Young. 1982. Collection and identification of
alleopathic compounds from the undisturbed root system of Bigalta
limpograss (Hemarthria altissima). Plant Physiol 69:155-160.
Taylor, A. 0., R. M. Haslemore, and M. N. McLeod. 1976a. Potential of
new summer grasses in northland. III. Laboratory assessments of
forage quality. N.Z. J. Agrie. Res. 19:483-488.
Taylor, A. 0., J. A. Rowley, and B. J. Hunt. 1976b. Potential of new
summer grasses in northland. II. A further range of grasses.
N.Z. J. Agrie. Res. 19:477-481.

191
Trlica, M. J. Jr. , and C. W. Cook. 1972. Carbohydrate reserves of
crested wheatgrass and Russian wildrye as influenced by development
and defoliation. J. Range Mgmt. 25:430-435.
Troughton, A. 1957. Chemical composition. In_ The Underground Organs
of Herbage Grasses. A. Troughton (ed.), Commonwealth Bur. Pastures
Field Crops Bull. 44, pp. 49-68.
Waite, R. 1957. The water-soluble carbohydrates of grasses. III.
First and second year growth. J. Sci. Food Agrie. 8:422-428.
Waite, R. 1958. The water-soluble carbohydrates of grasses. IV. The
effect of different levels of fertilizer treatment. J. Sci. Food
Agrie. 9:39-43.
Waite, R., and J. Boyd. 1953a. The water-soluble carbohydrates of
grasses. I. Changes occurring during the normal life-cycle. J.
Sci. Food Agrie. 4:197-204.
Waite, R., and J. Boyd. 1953b. The water-soluble carbohydrates of
grasses. II. Grasses cut at grazing height several times during
the growing season. J. Sci, Food Agrie. 4:257-261.
Waite, R., and A. R. N. Gorrod. 1959. The comprehensive analysis of
grasses. J. Sci. Food Agrie. 10:317-326.
Ward, C. Y., and R. E. Blaser. 1961. Carbohydrate food reserves and
leaf area in regrowth of orchardgrass. Crop Sci. 1:366-370.
Wardlaw, I. F. 1968. The control and pattern of movement of carbo¬
hydrates in plants. Bot. Rev. 34:79-104.
Watson, V. H., and C. Y. Ward. 1970. Influence of intact tillers and
height of cut of regrowth and carbohydrate reserves of dallisgrass
(Paspalum dilatatum Poir.). Crop Sci. 10:474-476.
Weier, K. L., J. R. Wilson, and R. J. White. 1977. A semi-automated
procedure for estimating total nonstructural carbohydrates in
grasses, and comparison with two other procedures. CSIR0 Aust.
Div. Trop. Crops Past. Tech. Pap. 20:1-10.
Weinmann, H. 1947. Determination of total available carbohydrates in
plants. Plant Physiol. 22:279-290.
Weinmann, H. 1952. Carbohydrate reserves in grasses. Proc. 6th Int.
Grassld. Congr. College Part, PA, USA, pp. 655-660.
Weinmann, H. 1955. The chemistry and physiology of grasses. In_ The
Grasses and Pastures of South Africa. D. Meredith (ed.), Central
News Agency, Parow, C. P. Cape Times LTD, pp. 571-600.
Weinmann, H. 1961. Total available carbohydrates in grasses and
legumes. Herbage Abstracts 31 (4):255-261.

192
Weinmann, H., and E. P. Goldsmith. 1948. Underground reserves of
Cynodon dactyl on. Better turf by research. African Explosives and
Chemical Industries Ltd., and S. African Turf Research Fund,
Johannesburg, pp. 56-75.
Westhafer, M. A., J. T. Law Jr., and D. T. Duff. 1982. Carbohydrate
quantification and relationships with N nutrition in cool-season
turfgrass. Agron. J. 74:270-274.
White, L. M. 1973. Carbohydrate reserves of Grasses: A review. J.
Range Mgmt. 26(1):13-18.
Wilms, H. J., J. W. Carmichael, and S. C. Schank. 1970. Cytological
and morphological investigations on the grass Hemartharia altissima
(Poir) Stapf et C.E. Hubb. Crop Sci. 10:309-312.
Wilson, J. R., and C. W. Ford. 1973. Temperature influences on the
in vitro digestibility and soluble carbohydrate accumulation of
tropical and temperate grasses. Aust. J. Agrie. Res. 24:187-198.
Wilson, J. R., and L. 't Mannetje. 1978. Senescence, digestibility,
and carbohydrate content of buffel grass and green panic leaves in
swards. Aust. J. Agrie. Res. 29:503-516.
Woods, F. W., H. C. Harris, and R. E. Caldwell. 1959. Monthly varia¬
tions of carbohydrates and nitrogen in roots of sandhill oaks and
wiregrass. Ecology 40:292-295.
Young, C. C., and D. P. Bartholomew. 1981. Alleopathy in a grass-
legume association. I. Effects of Hemarthria altissima (Poir.)
Stapf and Hubb. root residues on the growth of Desmodiurn intortum
(Mill.) Urb. and Hemarthria altissima in a tropical soil. Crop
Sci. 21:770-774.

BIOGRAPHICAL SKETCH
Scott Christiansen was born the son of Carl August Christiansen,
Jr. and Mary Jean (Boiler) Christiansen on 21 June 1954 at 10:53 p.m.
in Buffalo, New York (43 N 79 W). He led an idealistic youth surrounded
by the love of sibs and parents, and had an all American boy adolescence.
He moved to Madison, Wisconsin, on 9 June 1974, after two semesters
of liberal arts courses at the State University of New York at Buffalo.
In Madison he obtained the B.S. degree in August, 1977, and continued
for a M.S. degree that was conferred in May, 1979. On 5 January 1979,
he moved to Gainesville, Florida, and after four years expects to
receive the Doctor of Philosophy degree in December, 1982. In July,
1982, Scott accepted a research agronomist position with the USDA-ARS
Southwest Livestock and Forage Research Unit in El Reno, Oklahoma.
Scott became engaged to be married on 1 July 1982 to Caroline
Mirabel Kitts and wed 29 August 1982. When the couple moves to Oklahoma
in September, 1982, Carrie will continue her education in music, study¬
ing oboe, piano, and the Suzuki method of music pedagogy.
193

I certify that I have read this study and that in my opinion 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.
lía
0. Charles Ruelke, Chairman
Professor of Agronomy
I certify that I have read this study and that in my opinion 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.
''John E. Moore
Professor of Animal Science
I certify that I have read this study and that in my opinion 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.
/C
-
William R. Ocumpaugh (J
Associate Professor of Agronomy
I certify that I have read this study and that in my opinion 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.
Kenneth J. Boote
Associate Professor of Agronomy

I certify that I have read this study and that in my opinion 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.
Kenneth H. Quesenberry
Associate Professor of Agronomy
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December, 1982
Dean for Graduate Studies and Research



xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4JJQ2A3I_2I0CHB INGEST_TIME 2017-11-09T20:17:15Z PACKAGE UF00101080_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


I certify that I have read this study and that in my opinion 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.
Kenneth H. Quesenberry
Associate Professor of Agronomy
This dissertation was submitted to the Graduate Faculty of the College
of Agriculture and to the Graduate Council, and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
December, 1982
Dean for Graduate Studies and Research