Citation
Response of a tropical legume-grass association to systems of grazing management and levels of phosphorus fertilization

Material Information

Title:
Response of a tropical legume-grass association to systems of grazing management and levels of phosphorus fertilization
Added title page title:
Legume-grass association
Creator:
Santillan, Raul Alonso, 1943-
Publication Date:
Language:
English
Physical Description:
xiv, 170 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Analysis of variance ( jstor )
Dry seasons ( jstor )
Forage ( jstor )
Grasses ( jstor )
Grazing ( jstor )
Legumes ( jstor )
Pastures ( jstor )
Radiocarbon ( jstor )
Rainy seasons ( jstor )
Species ( jstor )
Agronomy thesis Ph. D
Dissertations, Academic -- Agronomy -- UF
Forage plants -- Fertilizers ( lcsh )
Forage plants -- Tropics ( lcsh )
Grasses ( lcsh )
Grazing ( lcsh )
Legumes ( lcsh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1983.
Bibliography:
Bibliography: leaves 161-169.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Raul Alonso Santillan.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
030144015 ( ALEPH )
11073444 ( OCLC )

Downloads

This item has the following downloads:


Full Text












RESPONSE OF A TROPICAL LEGUME-GRASS ASSOCIATION TO SYSTEMS OF GRAZING MANAGEMENT AND LEVELS OF
PHOSPHORUS FERT ILI ZATION















BY

RAUL ALONSO SANTILLAN





















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


1983


































Dedicated to my wife, Maggie, my daughters, Alexandra and Carolina, my father and to the memory of my mother














ACKNOWLEDGMENTS


The author expresses his sincere thanks and appreciation to Dr. Gerald 0. Mott, chairman of the supervisory committee, for his understanding, deep concern and wise guidance throughout the field work and graduate program. Special thanks are given to Dr. W. R. Ocumpaugh, Dr. J. E. Moore, Dr. 0. C. Ruelke, and Dr. L. R. McDowell, who willingly served as members of the committee.

Special appreciation is due to Dr. Raul de la Torre, General Coordinator for Animal Production Research at INIAP-Ecuador, who helped design and advise during the field experiment. The author also expresses his gratitude to Ing. Carlos Cortaza, former Director of E. E. T. Pichilingue and to Ecuador's Instituto Nacional de Investigaciones Agropecuarias for the physical and financial support of this study, especially to the staff of the Programa de Pasto y Ganaderia in the E. E. T. Pichilingue. Outstanding among those whose friendship and physical labors are responsible for this study are Ing. Carlos Cadene, Agr. Jorge Molina, and Senores Cerapio Diaz, Emilio Ortega, Jorge Benitez, Alejo Briones and the late Maximo Soria.

Special appreciation is due to Dr. A. I. Khuri, Dr. Michael

Conlon and to the graduate student Jose M. Gallo for their outstanding contributions and help in the statistical analyses of the results of this research.

Special thanks are offered to both Dr. and Mrs. G. 0. Mott for their generosity to the author and his family throughout their stay in Gainesville.



iii











The author is indebted to graduate students, Mr. Luis R.

Rodriguez and his wife, Teresinha Rodriguez for their friendship and hospitality. Thanks are also due to graduate student Juan Herbs and his wife for their friendship.

The author wishes to thank Mrs. Pat French for typing the final copy of this manuscript.

Special acknowledgment is due to the author's wife, Maggie,

and their two daughters, Alexandra and Carolina, for their invaluable help, love and devotion and finally to my parents and dear sisters, Cecilia and Lucy, for their love and financial help.




































iv













TABLE OF CONTENTS

PAGE

ACKNOWLEDGMENTS ........................................... iii

LIST OF TABLES ............................................ vii

LIST OF FIGURES ........................................... xi

ABSTRACT .................................................. xiii

CHAPTER I INTRODUCTION ................................... 1

CHAPTER II LITERATURE REVIEW .............................. 3

The Tropical Forages .......................... 3
Persistence of Tropical Legume-Grass Associations ......................................... 6
Response to Nutrients ......................... 10
Animal Productivity ........................... 14
Pasture Evaluation ............................ 16
Grazing Systems ............................... 19
Estimates of Dry Matter Production and Yield.. 20 Measuring Botanical Composition ............... 22
Effect of the Grazing Animal on Botanical
Composition ................................... 22
Response Surface Methodology .................. 25

CHAPTERIII MATERIALS AND METHODS ......................... 28

Legume-Grass Mixture .......................... 33
Experimental Variables ........................ 33
Experimental Design ........................... 35
Field Plan of the Experiment .................. 35
Land Preparation and Pasture Establishment .... 40 Construction of Physical Facilities ........... 43
Collection of Data in the Three Experimental
Years ......................................... 43
Pasture Measurements .......................... 44

CHAPTER IV RESULTS AND DISCUSSION ........................ 48

Effect of Lengths of Rest Period and Levels of
Grazing Pressure on Aerial Biomass (DM) ....... 48
Effect of Lengths of Rest Period and Levels
Grazing Pressure on the Available Forage (DM). 57
Effect of Lengths of Rest Period and Levels of
Grazing Pressure on Grass Yield (DM) .......... 65


v










PAGE

Effect of Lengths of Rest Period and Levels of
Grazing Pressure Upon Legume Yield (DM) ........ 76
Effect of Lengths of Rest Period and Levels of
Grazing Pressure on the Yield of Weeds (DM) .... 91 Visual Estimation of Forage Components ......... 103 CHAPTER V SUMMARY AND CONCLUSIONS ........................ 122

APPENDIX .................................................. 126

LITERATURE CITED .......................................... 161

BIOGRAPHICAL SKETCH ....................................... 170









































vi













LIST OF TABLES

TABLE PAGE

1 Soil analysis of experimental site (1978) ........... 34

2 Modified central composite non-rotatable design with
four experimental (X) variables, at five levels each,
and 41 design points ................................ 36

3 Modified central composite non-rotatable design with
four experimental (X) variables, at five levels each,
41 design points, 51 experimental units with their
respective area ..................................... 38

4 Aerial biomass production (DM) by year, season, and
treatment ........................................... 49

5 Available forage (DM) by year, season, and treatment combination ......................................... 58

6 Grass yields (DM) by year, season, and treatment combination .......................................... 66

7 Legume yields (DM) by year, season, and treatment
combination ......................................... 78

8 Yields of weed (DM) by year, season, and treatment
combination ......................................... 92

9 Visual estimation of grass percentage by year, season, and treatment combination ........................... 104

10 Visual estimation of legume percentage by year, season, and treatment combination ................... 113

11 Analysis of variance, regression coeffic ents and probabilities for aeiral biomass (g DM/m ) for the
wet season of 1978 .................................. 126

12 Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m2) for the
dry season of 1978 .................................. 127

13 Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m2) for the
wet season of 1979 .................................. 128

14 Analysis of variance, regression coefficnts and
probabilities for aerial biomass (g DM/m ) for the
dry season of 1979 .................................. 129


vii













TABLE PAGE

15 Analysis of variance, regression coefficients and probabilities for aerial biomass (g 1*1/r2) for the
wet season of 1980 ..................................130

16 Analysis of variance, regression coefficients and probabilities for available forage (g DM/m ) for the
wet season of 1978 ..................................131

17 Analysis of variance, regression coefficients and probabilities for aerial biomass (g DM/m2) for the
dry season of 1978 ..................................132

18 Analysis of variance, regression coefficients and probabilities for available forage (g DM/rn2) for the
wet season of 1979 ..................................133

19 Analysis of variance, regression coefficients and probabilities for aerial biomass (g DM/m ) for the
dry season of 1979 ..................................134

20 Analysis of variance, regression coefficients and probabilities for available forage (g DM/m2) for the
wet season of 1980 ..................................135

21 Analysis of variance, regression coefficients and probabilities for grass yield (g DM/m2) for the wet
season of 1978 ......................................136

22 Analysis of variance, regression coefficients and probabilities for grass yield (g DM/m2) for the dry
season of 1978 ......................................137

23 Analysis of variance, regression coefficients and probabilities for grass yield (g DM/m2) for the wet
season of 1979 ......................................138

24 Analysis of variance, regression coefficients and probabilities for grass yield (g DM/m2) for the dry
season of 1979 ......................................139

25 Analysis of variance, regression coefficients and probabilities for grass yield (g DM/m2) for the wet
season of 1980 ......................................140

26 Analysis of variance, regression coefficients and probabilities for legume yield (g DM/m2) for the
wet season of 1978 ..................................141


viii











TABLE PAGE

27 Analysis of variance, regression coefficients and probabilities for legume yield (9 DM/m ) for the
dry season of 1978 ................................. 142

28 Analysis of variance, regression coefficients and probabilities for legume yield (g DM/m2) for the
wet season of 1979 ................................. 143

29 Analysis of variance, regression coefficients and probabilities for legume yield (g DM/m2) for the
dry season of 1979 ................................. 144

30 Analysis of variance, regression coefficients and probabilities for legume yield (g DM/m2) for the
wet season of 1980 ................................. 145

31 Analysis of variance, regression coefficients and probabilities for yield of weeds (g DM/m2) for the
wet season of 1978 ................................. 146

32 Analysis of variance, regression coefficients and probabilities for yield of weeds (g DM/m2) for the
dry season of 1978 ................................. 147

33 Analysis of variance, regression coeffic ents and probabilities for yield of weeds (g DM/m ) for the
wet season of 1979 ................................. 148

34 Analysis of variance, regression coefficients and probabilities for yield of weeds (g DM/m2) for the
dry season of 1979 ................................. 149

35 Analysis of variance, regression coeffic ents and probabilities for yield of weeds (g DM/m ) for the
wet season of 1980 ................................. 150

36 Analysis of variance, regression coefficients and probabilities for visual estimation grass (%) for
the wet season of 1978 ............................. 151

37 Analysis of variance, regression coefficients and probabilities for visual estimation grass (%) for
the dry season of 1978 ............................. 152

38 Analysis of variance, regression coefficients and probabilities for visual estimation grass (%) for
the wet season of 1979 ............................. 153

39 Analysis of variance, regression coefficients and probabilities for visual estimation grass (%) for
the dry season of 1979 ............................. 154


ix










TABLE PAGE

40 Analysis of variance, regression coefficients and probabilities for visual estimation grass (%) for
the wet season of 1980 ............................. 155

41 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1978 ............................. 156

42 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the dry season of 1978 ............................. 157

43 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1979 ............................. 158

44 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the dry season of 1979 ............................. 159

45 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1980 ............................. 160













LIST OF FIGURES

FIGURE PAGE

1 Profile of three natural regions of Ecuador, showing
the main vegetative zones in relation to rainfall
and temperature ..................................... 29

2 Rainfall recorded at Estacion Experimental Pichilingue during the period 1978-1980 .................. 30

3 Temperature recorded at Estacion Experimental Pichilingue during the period 1978-1980 .................. 31

4 Solar radiation recorded at Estacion Experimental
Pichilingue during the period 1978-1980 ............. 32

5 Field plan of the experimental pastures ............. 41

6 Effect of rest period and grazing pressure upon
grass yield (DM) for the wet season of 1978 ......... 73

7 Contours of grass yield (DM) as affected by length
of rest period and levels of grazing pressure in
the wet season of 1980 .............................. 74

8 Effect of rest period and grazing pressure upon grass
yield (DM) for the wet season of 1980 ............... 75

9 Effect of rest period and grazing pressure upon
legume yield (DM) for the wet season of 1978 ........ 84

10 Contours of legume yield (DM) as affected by length of rest period and levels of grazing pressure in the
wet season of 1980 .................................. 88

11 Effect of rest period and grazing pressure upon legume yield (DM) for the wet season of 1980 ........ 89

12 Effect of rest period and grazing pressure upon yield of weed (DM) for the wet season of 1978 ....... 98

13 Contours of yield of weed (DM) as affected by length of rest period and levels of grazing pressure in the
wet season of 1980 .................................. 100

14 Effect of rest period and grazing pressure upon yield of weed (DM) for the wet season of 1980 ............. 101

15 Effect of rest period and grazing pressure upon grass percentage for the wet season of 1978 ............... 108


xi











FIGURE PAGE

16 Contours of grass percentage as affected by length
of rest periods and levels of grazing pressure in
the wet season of 1980 ............................. 109

17 Effect of rest period and levels of grazing pressure
upon grass percentage for the wet season of 1980 ... 110

18 Effect of rest period and grazing pressure upon
legume percentage for the wet season of 1978 ....... 119

19 Contours of legume percentage as affected by length
of rest period and levels of grazing pressure in
the wet season of 1980 ............................. 120

20 Effect of rest period and grazing pressure upon
legume percentage for the wet season of 1980 ....... 121




































xii













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

RESPONSE OF A TROPICAL LEGUME-GRASS ASSOCIATION
TO SYSTEMS OF GRAZING MANAGEMENT AND LEVELS OF PHOSPHORUS FERTILIZATION

By

RAUL ALONSO SANTILLAN

April 1983

Chairman: Dr. G. 0. Mott
Major Department: Agronomy

A legume-grass pasture composed of glycine [Neonotonia wightii (R. Grah ex Wightii and Amrn.) Lackey], centro (Centrosema pubescens Benth.), guineagrass (Panicum maximum Jacq.), and elephantgrass (Pennisetum purpureum Schumach.) was evaluated in a grazing trial from May 1978 to June 1980 at the Estacion Experimental Tropical Pichilingue, Instituto Nacional de Investigaciones Agropecuarias (INIAP), Quevedo, Ecuador.

The main objectives of the study were (a) to determine the effects of length of grazing period (X1), length of rest period

(X2), grazing pressure (X3), and levels of P fertilization (X4) upon the pasture mixture; (b) to determine the proper grazing management to attain the optimum legume contribution; and (c) to measure the pasture response in terms of dry matter production and botanical composition.

Grazing periods studied were 1, 7, 14, 21, and 28 days; rest periods were 0, 14, 28, 42, and 56 days; grazing pressures were

1.6, 3.3, 5.0, 6.6, and 8.3 kg DM on offer/100 kg body weight; and xiii










levels of fertilizer were 0, 100, 200, 300, and 400 kg ha-1 of superphosphate. To cover the five levels of the complete factorial (5 4 ), a modified non-rotatable central composite design made up of 41 treatment combinations was used.

The response variables included aerial biomass (DM), available forage (DM), grass yield (DM), legume yield (DM), yield of weeds

(DM), percentage grass, and percentage legume. A double-sampling procedure was used for estimating pasture production and botanical composition.

Rest periods and grazing pressures had the greatest effects

on all response variables. Aerial biomass, available forage, grass yield and grass percentage were increased by longer rest periods and by lower grazing pressure. Legume yield and legume percentage were decreased by long rest periods and by low grazing pressure. Short rest periods and high grazing pressures resulted in high yields of weeds. Medium levels of both rest periods and grazing pressure were required for high forage dry matter production and for high legume yield.

The other two variables, days grazing and levels of phosphorus
fertilization, had negligible effects upon the response of the pasture sward.











xiv












CHAPTER I
INTRODUCTION


Ecuador, with an area of 273,670 km 2is located in the northwestern part of the South American continent. The dominant topographical features are two parallel ranges of the lofty Andes mountains that separate the fertile littoral lowland on the west and the more extensive and less fertile lowland of the Amazon Basin on the east.

The diversity of natural features of the littoral region is

very great due to its multiple climatic conditions, soils, forms of vegetation, and settlement patterns which set it apart from the more homogeneous Sierra and Oriente regions.

In the littoral, about 2,500,000 ha are considered as pasture

land, supporting 2,875,000 head of cattle giving a carrying capacity of 1.12 animals ha-1

The seasonal pattern of rainfall distribution and the low soil N levels are factors restricting forage production and quality during the wet-dry seasons. It is well known that cattle production is limited by the feed supply during the dry season, while in the wet season, there is abundance of forage.

Pastures are mainly planted to guineagrass (Panicum maximum Jacq.) or elephantgrass (Pennisetum purpureum Schumach.), with other species making up a very small portion of the total hectarage.

At present, there is a growing interest in the establishment

and utilization of tropical grass-legume mixtures for animal production. There are many advantages to having legume components in the



1







2



pasture. They increase or maintain soil fertility due to their ability to fix N, improve the quality of the diet grazed by the animals, provide better seasonal distribution of the forage throughout the year, especially where alternate wet-dry seasons occur. Therefore, a system which combines adequate grazing management practices with high-yielding grasses and legumes growing in mixtures which provide feed during the whole year is needed to increase carrying capacity and the final output of the land.

The objectives of this research were

1) To determine the response of a tropical legume-grass

mixture to various treatment combinations of grazing

management factors length of grazing period, length of rest period, grazing pressure, and levels of phosphorus

fertilization.

2) To determine the proper management strategy for adapted

forage species to attain the optimum legume contribution

as a component in the mixture, and

3) To measure the results of legume-grass mixture in terms

of aerial biomass, available forage, and botanical composition of the forage on offer.












CHAPTER II
LITERATURE REVIEW


In tropical regions animal production on pastures is dependent

upon the quality and quantity of forage available throughout the year. In the humid tropics the seasonal growth of pastures is greatly influenced by the wet and dry periods. At the beginning of the wet season very rapid growth occurs and frequently a large amount of forage accumulates. As the season progresses the rate of growth decreases rapidly and approaches zero during the dry season. The concentration of nitrogen and minerals is relatively high at the beginning of the wet season and also falls rapidly as the season progresses. In addition there is a progressive increase in fiber and decrease in digestibility from the beginning of the wet season through the remainder of the year (Faladines and de Alba, 1963).

Some tropical pasture species seem to be well adapted to extreme environmental conditions; however, their potential to supply feed for cattle production may be limited not only by seasonal changes but also by soil fertility (Tergas, 1968), mechanisms of adaptation (Gartner et al., 1974), and grazing management (Mott, 1960; Evans, 1970; Stobbs, 1969).


The Tropical Forages


Panicum maximum Jacq. has been described by Humphreys (1980) and Bogdan (1977) as a densely tufted perennial grass with relative drought and poor soil tolerance. It is of high nutritive value when young and combines well with other tropical pasture species (Hudgens, 1973;


3






4




Chavez, 1974; Rolando, 1974). It is indigenous to tropical Africa where it is dominant over large areas, particularly under humid and subhumid conditions. In the coastal area of Ecuador it is known as guinea, cauca, saboya, chilena. Betancourt (1969) indicated that guineagrass was the most widespread pasture grass in the lowlands of Ecuador. Acosta-Soliz (1967) reported that this grass grows from sea level up to 1400 m.

Guineagrass is very popular because of its adaptation to the

dry season, resistance to fire, high production of forage, capability of establishment by seed or by division of plant crowns, and finally due to its ability to persist under heavy use and abuse (INIAP, 1980).

Recently the Instituto Nacional de Investigaciones Agropecuarias (INIAP) of Ecuador has tested 124 introductions of guineagrass in small grazing trials, mainly for adaptation and persistence. New cultivars will be released to the farmers as soon as sufficient seed is available for large scale use (INIAP, 1979).

Pennisetum purpureum Schumach, elephantgrass, is a tall-growing species, up to 5 m in height. It has been described by Bogdan (1977), Correa (1926), and Mcllroy (1972).

Acosta-Soliz (1967) and INIAP (1980) stated that elephantgrass grows from sea level up to 2200 m in the warm valleys of the Sierra region. In Ecuador it is mainly used as a pasture grass with some farmers reporting pastures up to 40 years old still under grazing conditions. Under cutting frequencies of 45 days it produces up to
-1 -l
80 tons DM ha yr when receiving 400 kg of N and irrigation during the dry season (INIAP, 1972). There are three cultivars which are






5



in widespread use in the lowlands of Ecuador. They are 'common' for grazing, 'Hybrid 534' and 'Mexican' for cutting.

Centrosema pubescens Benth, centro, a true tropical legume has been described by Humphreys (1980) as a creeping, twining perennial legume native of South America. Grof (1970) reported that the genus Centrosema contains about 70 species growing naturally in the tropical areas of Central and South America. Moore (1962) stated that this tropical legume grows well in humid areas and must be considered as a basic component of pastures under these conditions. Studies conducted at Pichilingue, Ecuador, byHudgens (1973), Chavez (1974), Rolando (1974) and INIAP (1980) indicated that this legume performs well in association with guineagrass and it is also highly persistent under grazing and produces large amounts of seed (Farfan, 1974). In the last few years 132 native ecotypes of Centrosema have been selected and tested in Ecuador. A few of these have been distributed to farmers and are showing some advantages over the Australian commercial cultivars such as higher DM yield, better tolerance to insects and disease, and better adaptation to Ecuador conditions (INIAP, 1979).

Neonotonia wightii (R. Grah. ex Wight and Arn.) Lackey, glycine,

also known as perennial soybeans, has been described by Humphreys (1980) as a perennial plant, slender, twining, and with long stems having some capacity for rooting at the nodes.

Glycine was first introduced into Ecuador in 1966, but it was not used in pastures until 1973 when it was shown to be one of the best legumes for humid and subhumid areas. Three cultivars have been distributed to farmers and these were selected for persistence,







6


adaptability, seed production and disease and insect resistance. These cultivars are 'Malawi' for lower altitudes, 'Cooper' for medium altitudes up to 1200 m and 'Tinaroo' the highest forage yielder grows well from 50 to 1800 m of altitude, producing large amounts of seed from 800-1500 m (INIAP, 1979).


Persistence of Tropical Legume-Grass Associations


Serrao (1976) suggested that the first requirement for successful use of high-yielding pasture legumes was their adaptation to local climatic conditions. Secondly, nutrient requirements must be met to insure high yield and maintenance. And finally, they must persist under heavy grazing to secure a long lasting beneficial contribution to the companion grass, to the soil, and to the grazing animal. Gomez (1978) has suggested some other factors which can affect the persistence of tropical legumes when they are growing in association with grasses. Among the most important are (1) environmental factors such as light, temperature and moisture; (2) growth habits of each species growing in the mixture; (3) nodulation ability and capacity for nitrogen fixation; (4) edaphic factors such as pH, nutrient availability, form of supply; (5) frequency and intensity of defoliation by grazing animals; (6) ability to survive during long drought periods; (7) seed production capacity; and (8) pest and disease tolerance. Ludlow and Wilson (1970) reported that tropical grasses achieve up to three times the photosynthetic rate when compared with tropical legumes. This characteristic obviously gives ecological advantages to C-4 grasses, affording them the opportunity to grow faster, dominate







7



and even exclude the C-3 legumes from the mixture. Tow (1967) showed that green panic (Panicum, maximum var trichoglume) was much more productive at all light intensities and higher root temperatures than glycine when both species were tested under controlled environmental conditions. Roberts (1974) also studied some of the above factors and included some others which are associated with the stability of legume-grass mixtures. These were palatability of the grass and legume, maximum height of the grass, legume ability to grow under the shade projected by the companion vegetation, and the capacity to withstand trampling. He also suggested that continuous grazing helps the legume to compete more effectively with the grass due to more frequent defoliation than under a rotational grazing system. Kretschmer (1974) reported that in general grasses have a better range of adaptation and also a more vigorous growth habit that allows them to compete always at an advantage over most tropical forage legumes. Lack of legume persistence is attributed to the use of unadapted species and cultivars, improper or no fertilization, incorrect rhizobium and overgrazing (t'Mannetje, 1978).

Growth habit and leaf morphology of species that compose a

mixture are important characteristics which have direct effects upon compatibility and persistence, due especially to light interception ability of each individual species. Santhirasegaram (1976) in the humid tropic of Peru reported that in a well-managed guineagrass-centro pasture, the persistence of the legume was due to its viney growth habit enabling it to climb the stems and leaves of this

tall and aggressive grass. Thus, the legume can intercept sufficient






8



solar radiation. They also suggested that the ideal type of twining tropical legume should have a strong stoloniferous or rhizomatous growth habit in order to withstand frequent defoliation and heavy damage by grazing animals. Whiteman (1969) pointed out that frequent defoliation results in low yields of twining legumes such as glycine and Siratro, whereas slight defoliation or absence of defoliation results in a higher contribution of the legumes to the total pasture production. In Africa, Draolu and Nabusin-Napulu (1980) studied the effect of cutting intervals of 3, 6, 12, and 24 weeks and cutting heights of 3.8, 7.5, 15.0, and 30.0 cm in mixed swards of guineagrass and Stylosanthes guianensis. They concluded that DM production was greatly reduced under the lowest and most frequent cutting treatment. The amount of legume in the mixture also decreased as the cutting intervals were increased and the legume was particularly sensitive to close defoliation. In Australia, Mclvor et al. (1981) found the same linear tendency with Desmodium intortum and Setaria sphacelata mixtures under different cutting heights of 8 and 20 cm and cutting intervals of 3, 6, and 9 weeks. The growth response of desmodium was markedly depressed at the lowest cutting height of 8 cm and the shortest cutting interval of 3 weeks. They concluded that cutting at the height of 20 cm at intervals of 6 to 9 weeks was necessary for the persistence of the desmodium in the mixture. Bryan et al. (1971) indicated that the short growth habit of the tropical legumes Stylosanthes humilis and Lotononis bainesil, which may be shaded by taller companion grasses, benefits from heavy grazing pressure which allows light penetration into the canopy. They also







9



concluded that the dominance of climbing legumes such as glycine and Siratro is generally enhanced by light grazing pressure and also by long intervals between grazing periods.

Grazing experiments conducted in the wet tropics of Ecuador by Berrezueta (1975), Chavez (1974), INIAP (1979) and Zapata (1981) showed that guineagrass-centro pastures and guineagrass-glycine pastures are very persistent and productive mixtures even if heavy grazing pressures are applied. They also observed that rest periods of over 28 days during the wet season favored the companion grass and reduced drastically the amount of legume. Finally they concluded that the guineagrass is dominant, especially at the beginning of the wet season, probably due to the large amount of nitrogen stored in the organic matter during the six to seven months dry season. The high growth rate of this grass decreases slowly and reaches the lowest rate of growth at the end of the dry season while the legumes appeared to be more productive during the dry season. The legumes are the main source of feed for the grazing animals during the dry season. A study of eight grass-legume associations was also made from 1971 through 1973 using grazing animals. Paragrass (Brachiaria mutica)-glycine and guineagrass-centro mixtures were shown to be the most compatible and the best accepted by the grazing animals, while tropical kudzu (Pueraria phaseoloides) and Calopogonium mucunoides were unpalatable species, a characteristic that could determine the dominance of this species over the companion grasses (INIAP, 1974).







10



Response to Nutrients


Legumes are generally more sensitive to soil factors than other pasture plants, particularly grasses. This sensitivity of legumes emphasizes the importance of understanding the effects of these limitations in tropical conditions. Russell (1978) suggested that improving the growth of legumes in low fertility soils could be approached in two ways: (1) by amelioration of soil conditions through the use of fertilizers or amendments, and (2) by the selection of legume cultivars or genera which are more tolerant to the limiting conditions. There seems to be a general consensus that N and P, in that order, are the plant nutrients that are more often deficient in the tropics (Fox, 1979).

The amount of N fixation, nodulation, persistence, and yield of tropical forage legumes may be affected by the soil pH and also by the availability of plant nutrients. Manhaes and Dobereiner (1968)and Fox et al. (1974) reported that for good legume establishment, adequate amounts of available P were required, especially during the nodulation stages. These authors determined that glycine required 60 ppm of P 20 5in the soil solution during the establishment phase. This requirement decreased after the second cutting. In Australia, Andrew and Robins (1969) determined the critical P concentration in the tops associated with maximum plant growth as being the critical levels. These were 0.16 and 0.23% for centro and glycine, respectively. For adequate plant uptake, H 2P0 ions in the soil solution should be between 0.07 to 0.2 ppm. According to Sanchez (1977) some











tropical legumes tolerant to low available soil P either absorb P at a faster rate or are able to tranlocate it to the tops more rapidly than do species not tolerant to low P availability (Salinas and Sanchez, 1976; Andrew, 1978). Considerable yield response to P fertilizers was reported by Jones and Freitas (1970) with four tropical legumes (Stylosanthes guianensis, centro, glycine, and Siratro). Similar results were obtained by Franca and Carvalho (1970) in greenhouse studies, using five tropical legumes (glycine var 'Common,' glycine var 'Tinaroo,' Siratro, centro, and Pueraria phaseoloides var 'Javanica' Benth.). In both cases the P deficiency was reflected in decreased nodule weight and N fixation capability. Snyder and Kretschmer (1974) obtained small linear increases in dry matter yields of Siratro, 'Cook' stylo, centro, and Desmodium heterocarpon (L.) DC when lime was applied in 500 kg ha-I increments up to 3000 kg ha-I without P fertilization. When the same levels of lime were used together with 45 kg ha-I of P the response in yield
-1
was linear up to 2000 kg ha of lime and curvilinear thereafter. Estimation of P requirements for plant growth should be based on the amount of P needed to give at least 95% of maximum growth (Ozanne and Shaw, 1976). Neme and Lovadini (1967) working with glycine found
-i
that a combination of 120 kg of P205 plus 6 metric tons of lime ha gave a large increase in yield. Werner (1979) reported substantially increased yields of centro to P and K fertilization. Palacios (1976) in Ecuador found that centro responded positively to P fertilization during the establishment period, but yield response was not related to the amount of P applied to the soil during the sowing time. Falade






12



(1975), comparing six levels of P (0, 15, 30, 60, 120, and 180 mg/

2 kg of soil/pot), found that P concentration in guineagrass and elephantgrass was increased with the addition of P. The same response was reported by Vicente-Chandler (1975) from Puerto Rico, where 80% of the maximum dry matter production of pearl millet [Pennisetum americanum (L.) K. Schum.] was obtained when soil pH was raised to 5.5 and 115 ppm of P were added. At a pH of 4.8, twice as much P was needed to produce the same forage yield. Fox (1979) reported that the standard P requirement was a relative soil requirement, not an absolute plant requirement. Fox et al. (1974) from Hawaii reported large increases in P uptake by several pasture species, once soils which had high P-fixing capacities were limed to pH 5.0 and 6.0. Phosphorus requirements of soils can range from zero to more than 2220 kg P ha-I.

After N and P, S is considered by many as the next most important element needed for tropical legume growth. Tergas (1977) noted the significance of S on the growth and nodulation of several different tropical forage legumes. Siratro and centro dry matter and nodule weight increased as S was increased. Sanchez (1977) mentioned that S deficiencies are widespread throughout the tropics and that some pasture legumes are more susceptible to S deficiency than most grasses.

Medina (1969) reported S deficiencies in some crops growing in the littoral region of Ecuador. Also, a strong response from guineagrass, paragrass, glycine, and centro was observed when S was applied alone or supplied by the ordinary superphosphate or by ammonium sulphate (INIAP, 1980).







13



In the tropics, where there are well defined wet and dry seasons, rainfall plays an important role in the uptake and nutrient content of tropical forage species. Blue and Tergas (1969) reported a drop in N, P, and K contents during the dry season; likewise a decrease of nutrient content during the wet season was found and it was postulated to be due to translocation of nutrients to the roots. Rapid growth during wet seasons may result in trace mienrals being translocated to plant tops where they are rapidly diluted with aerial tissue causing deficiency symptoms in older tissues (Reuter, 1975).

Micronutrients can play an important role in tropical legume pastures growth, mainly because of their function in several enzyme systems and in N-fixation by rhizobium-legume associations. Werner et al. (1975) studied tropical legume response to the micronutrients Mo, Cu, Zn, B, Mn, and Co in the form of FTE BR-1O and also in the salt form. Using three tropical legumes planted in pots, they observed visible symptoms of Mn toxicity on glycine and B toxicity on stylo. This work emphasizes that there is a narrow range between deficient and toxic levels of some micronutrients.

Medina (1969) found some micronutrient deficiency symptoms in some tropical crops growing in the Quevedo area. He reported that B, Zn, and Fe were the most deficient elements. INIAP (1978) found that at the beginning of the rainy season, the period of most rapid growth for grasses, such as paragrass, Zn deficiency symptoms were evident but these disappear in two or three weeks. Molybdenum also has been recognized as an essential element for legume growth especially for establishment and maintenance. Some authors have suggested that this







14


element is necessary for development of enzymes related to N fixation, nitrate reduction and legume nodulation (Andrew, 1978; Epstein, 1972).


Animal Productivity


In Pichilingue, Ecuador, Zapata (1981) observed that liveweight gain of steers grazing on common guineagrass, improved Guineagrass Brachiaria humidicola, and improved guineagrass-glycine pastures were,
-l -l
in the order given, 0.645, 0.678, 0.741, and 0.857 kg animal day1 and an annual liveweight gain of 338.9, 421.9, 458.3, and 540.4 kg ha-1 respectively. The same author also noted a difference in liveweight
-i
gain due to the breed of animal, being 0.598, 0.691, and 0.903 kg an-1
-i
day-1 for red criollo, braham, and braham x holstein crosses, respectively. Similar results were found at 700 m altitude by Cowan et al. (1974), but in this case milking cows were grazed on guineagrassglycine and kikuygrass (Pennisetum clandestinum) pastures without any
-i -i
supplementation. Milk production averaged 9.06 and 12.54 kg cow-1 day-1 for jersey and holstein cows, respectively. In Australia, Grof and Harding (1970) found that a guineagrass-centro pasture yielded 36% more liveweight than Guineagrass alone, over a two year period. Hall (1970) reported that animal production on unimproved native grass was less than 9 kg ha-1 when compared to guineagrass-Siratro mixture which
-i
yielded 112 kg liveweight ha .

In Costa Rica, Kretschmer (1971) reported that centro in mixtures with guineagrass increased forage yield by 20% during the dry season and by 30% during the wet season, when compared with grass alone. Tergas (1976) reported that centro increased total forage production






15



by 40% when it was grown in mixtures with guineagrass, and the amount
-l -i
of N fixed by the legume was estimated to be 100 kg N ha yr In

northern Australia, Norman (1970) showed that gain per animal grazing legume-grass mixtures was linearly related to the proportion of legume in the pasture. Also, animal liveweight gain during the dry season was related to the number of days during which the animals had been grazing the legumes. He also reported that animals grazing on grass pastures gained 60 kg head- 1, while those allowed to graze grass-legume
-i
pastures gained 280 kg head During the period of 112 days of the
-l
dry season, the first group on grass pastures lost almost 40 kg animal ,
-i
while the second gained 60 kg animal due to the companion legume, Stylosanthes humilis. In Ecuador, Chavez (1974) reported that a guinea-i -i
grass-centro mixture produced 536.5 kg liveweight gain ha year with the beneficial effect of the legume being more apparent during the dry season, when crude protein content of the guineagrass alone was below 7% and crude protein content for the grass-legume pastures over 10%. Similar results were reported by INIAP (1979) on guineagrass-glycine
-i -i
var 'Tinaroo' mixture, which produced 458 kg of liveweight ha year In both cases the results were obtained under rotational grazing using the put-and-take technique developed by Mott (1960).

According to Minson and Milford (1967), tropical legumes maintain adequate nutritive values for a longer period of time than most tropical grasses, when each were under the same management system. The critical level of crude protein required in a pasture before intake is reduced by protein deficiency is estimated to lie between 6.0 and 8.5%. Even highly N-fertilized tropical grasses at late growth stages may have






16



values below these points (Ventura et al., 1975). Tropical forage legumes, on the other hand, retain higher crude protein levels during the dry season, even in advanced maturity stages (Milford and Haydock, 1965). Selectivity by grazing animals is a possible explanation for higher performance of grazing cattle during the dry-season periods in tropical regions, especially when available forage meets or exceeds animal requirements (t'Mannetje, 1974).


Pasture Evaluation


Mott and Moore (1970) developed a five-phase scheme for forage

evaluation. Such a scheme involves quantity and quality determinations. Phase: I. Introduction and breeder's lines,

II. Small plot clipping trials,

III. Mob grazing experiments, forage response to grazing animals,

IV. Animal response, effect of forage on animal output,

V. Forage-livestock feeding systems.

Forage quality, in vitro organic matter digestibility (IVOMD) is taken into consideration in the first three phases, while phases IV and V involve animal response in nutrient digestibility, performance per animal and production per hectare.

Definitions for stocking rate, grazing pressure, and carrying capacity were given by Mott (1960). Stocking rate is defined as the number of animals per unit area of land, the term bearing no relationship to the amount of forage. Grazing pressure is the amount of forage dry matter on offer per animal per day and carrying capacity, also called grazing capacity, is defined as the stocking rate at the






17



optimum grazing pressure. McMeekan (1956) considered the stocking rate as the most powerful factor, influencing the efficiency of pasture conversion to animal products on a per unit area basis. Petersen et al. (1965) developed quantitative expressions which related grazing pressure and carrying capacity with output per animal. Their quantitative theory suggests that animal gain is constant as the stocking rate is increased up to a "critical point," where the grazeable forage is equal to the amount of forage consumed by grazing animals. If the stocking rate is increased beyond the critical point, then the gain per animal decreases, and the animal gain per unit area also decreases. Conway (1965), using three stocking rates, 1.0, 1.75, and 2.5 animals per acre, found that by increasing the stocking rate from 1.0 to 1.75 animals per acre, liveweight gain per animal decreased. However, with 2.5 animals per acre, the liveweight per animal was drastically reduced resulting also in a reduction of liveweight per acre. Jones (1979) in Australia studied the effect of five stocking rates, namely 0.8, 1.3, 1.8, 2.3, and 2.8 an ha1 in combination with three resting periods of 17, 39, and 50 days using a randomized complete block design with two replications. The area allocated to each treatment varied from 0.02 to 0.24 ha, using just one animal per experimental unit. Increasing the stocking rate had the greatest influence upon the persistence of the legume which in this case was Siratro. Also, the resting period became important to the legume productivity under heavy grazing pressure conditions. Results suggest that under heavy grazing pressure a longer rest period would allow the legume to increase or maintain its reserves of nutrients for subsequent growth. Echandi (1956) suggested that carrying capacity






18



should be determined taking into consideration the amount of litter left in the field as residue after each grazing period. Mott (1973) emphasized the use of variable stocking rates in grazing experiments by maintaining the number of animals in equilibrium with the available forage. He noted that the important advantage of the system was that it permits estimation of the carrying capacity of the pasture and the seasonal changes which occur. Evans and Bryan (1973) pointed out the need for more studies to evaluate different grazing pressures in order to measure the yield of the pasture and persistence of tropical legumes.

According to Mott (1973) the optimum grazing pressure must be

considered as an optimum range instead of a "critical point" and that such an optimum relates only to animal output and may or may not be the optima for plant species in the pasture. Mott and Lucas (1952), Mott (1960), and Matches (1970) described the put-and-take technique for grazing trials. They suggest that the stocking rate must be variable in which the grazing pressure is maintained at a constant level and the stocking rate adjusted as the availability of forage changes. These authors also distinguish the terms "testers" for animals which should remain in the pasture throughout the grazing experiment and "grazer," or put-and-take animals, as those used to maintain the grazing pressure at the optimum. Mott pointed out that "if the number of animals per unit area is to give an accurate appraisal of carrying capacity, then this unit of measure must not be fixed but be subject to adjustment, so that the number of animals per unit of forage is maintained at an equivalent level for







19



all treatments" (1960, p. 601-602). Serrao (1976) suggested that variable or fixed stocking rates can be used in continuous or rotational grazing systems. He concludes that the put-and-take system is more appropriate when plant and animal relationships are to be measured.


Grazing Systems


Continuous Grazing


Heady (1970) noted there is much confusion in the definition of grazing systems which are used for describing the day to day provision of livestock feed from a wide variety of sources such as conserved forage and by direct use of pasture by grazing animals. He also defined continuous grazing as a grazing system in which the animals have unrestricted access to any part of the pasture through a grazing period, which can be a season, a year or more. Spending (1965) suggested a pasture under a continuous grazing system could be called correctly grazed when the amount of removed forage by the grazing animals was equal to the amount of forage daily yield. Continuous grazing is the most commonly used system in the tropics, especially on vast areas, far away from the main consumer centers and in many cases areas without any suitable highway system (Chaverra, 1979). Rotational Grazing


A rotational grazing system is defined by Heady (1970) and Heath (1978) as a system in which the animals are allowed to graze the pasture for variable periods of time, normally with a heavy stocking rate, during which the pasture is grazed and ungrazed several times during a grazing season or year.







20


The terms grazing period and rest period are commonly used in

the rotational grazing system (Heady, 1970). According to Heady (1970) the grazing period is the portion of the grazing time during which grazing takes place and rest period is the time during which the pasture is not grazed. Comparing both continuous and rotational grazing systems on animal production, Stobbs (1969), in Africa, found that animal production was slightly higher for rotational grazing when it was carried out in three paddocks, but it was lower than continuous grazing when it was done in six paddocks. From Australia, Grof and Harding (1970) reported that a mixed Guineagrass-centro pasture with a carrying capacity of 3.5 an ha-l produced a liveweight gain of 934 kg ha-l for 2 years and 1075 kg ha- 1 for 2 years for continuous and rotational grazing, respectively. From Pichilingue, Ecuador, Paredes (1974) reported that during the rainy season there were no differences in stocking rate between continuous and rotational grazing, but during the dry season continuous grazing under variable stocking rate was superior to rotational grazing with fixed or variable stocking rate. He also found that continuous grazing with variable stocking rates gave the highest average daily gain of 0.457 kg an day


Estimates of Dry Matter Production and Yield


Estimation of production and yield are major problems in grazing experiments, because of the heterogeneity of the pasture sward and the amount of time required to obtain an adequate sample. One must decide the number of samples, the area to be sampled, and choose an adequate technique for sampling a highly variable population where cover, density,






21




height, weight, and other factors differ from one species to another (Kennedy, 1972). For many years, hand cutting and weighing of aboveground vegetative parts has been the most popular and useful method for estimating forage yield (Heady, 1970) and for estimating grazing pressure (Mott, 1960). Other methods, such as the simple disk meter, described by Bransby (1975), work on the principle of measuring the height of a disk supported by the resistance and compression of the vegetation. Santillan (1976) proved that the simple disk meter was very satisfactory for use in tropical species.

Mott (1974) suggested that estimation of total yield or yield of components is based upon the following relationship: yield per unit area = f (density, height). The total yield of an area of vegetation is related to the density and height of individual components. Ground cover and sward height have been used on different types of grassland to estimate dry matter yield.

Where pasture vegetation is utilized by grazing animals, the

amount of feed present at any one time may be only one of the factors associated with the intake by the grazing animals. Of interest to the pasture scientist is an estimate of that portion of the pasture which is consumed by the grazing animal since they are very selective of plant species and plant parts, which makes for a more complex situation (t'Mannetje, 1978).

A double sampling technique is probably one of the best ways to ensure a more precise yield determination. Eye-estimation in combination with a few harvested samples which act as a control on the observer's accuracy is one of the simplest forms of estimating total forage present or annual production potential of a pasture (t'Mannetje, 1978).






22




Measuring Botanical Composition


Botanical composition is a very essential measurement, especially in pastures subjected to grazing conditions, because the number of samples and yield of individual species may vary over a wide range depending upon environmental conditions and management factors. T'Mannetie et al. (1976) indicated that botanical composition can be measured in terms of the yield of component species, the number of plants covering the area and also the frequency of occurrence.

Determination of botanical composition and sampling techniques are difficult tasks in pasture research. Some methods have been developed to determine botanical composition; the most common are visual estimation and hand separation of harvested material into component species (Gardner, 1972). Visual estimation is a reliable method for studying pasture species, but in some cases it may be difficult to relate dry matter production of each of the component species, especially when growth habit and density differ widely. Tothill and Petersen (1962) indicated that the weight in situ, as well as the estimation of weight of each individual species, and visual estimation are the most useful methods for surveying vegetation of pasture species.


Effect of the Grazing Animal on Botanical Composition


There are at least three factors that are known to affect the balance of the grass-legume mixtures in the tropics once the pasture has been established. They are stocking rate or grazing pressure frequency of defoliation, and fertilization.







23



The grazing animal has a direct effect on pasture species due mainly to selectivity, deposition of feces and trampling, and an indirect effect on the soil due to the removal of nutrients by the removal of harvested forage. The animal alters the physical and chemical properties such as structure, texture, porosity, water retention, and in some cases, the accumulation of organic matter at certain points due to fecal deposits. All of these changes have a direct consequence, first on the botanical composition and subsequently on the final performance of that grazing animal (Alarcon and Lotero, 1970).

Research workers such as Davis (1967) and Wells (1967) agreed that grazing affects the soil cover and botanical composition which results in (1) a reduction of the basal cover, (2) a reduction of height of species, (3) a loss of soil cover and severe erosion, (4) a reduction of root systems, (5) a reduction in the emergence of new shoots, and (6) weed invasion.

Bryan and Evans (1973) found that the proportion of legumes Stylosanthes guianensis, Centrosema pubescens, and Pueraria phaseoboides on the dry matter basis was not affected by stocking rates during the first and second year. In the third year a marked reduction took place on the high stocking rate of 6 animals ha 1, in which the proportion of the legumes fell from 22 to 12%, while the companion grass (Panicum maximum) was more affected by the high stocking rate. In this case the reduction was significant since it went from 78 to 65% and to 38% in the first, second and third year, respectively. The other two stocking rates of 2 and 4 animals hal had little effect on






24



the grass and legume during the three experimental years. Similar results were reported from Africa by Stobbs (1969).

Cowan et al. (1975), in the Atherton Tableland, Queensland,

Australia, found a highly significant correlation coefficient between milk yield ha-I and the amount of the legume glycine present in the pasture. They studied four stocking rates, 1.3, 1.6, 1.9, and 2.5 cows ha -. On the other hand, while stocking rate was increased
-i
the milk production ha also was increased but the percentage of dry matter of legume present in the pasture decreased when the stocking rate increased. Cowan and O'Grady (1976) reported the same trend under similar conditions in another study carried out later at the same location.

Some investigations carried out in Cuba have shown that under heavy stocking rates the legumes fail to persist in the mixture. Febles and Padilla (1972), using a stocking rate of 4 cows ha-l, found that a mixture of guineagrass and the legumes glycine, Siratro, Stylosanthes guianensis, Desmodium intortum, and Desmodium uncinatum, did not persist more than 36 weeks. Funes and Perez (1976), using six animals ha -, also found that the three commercial cultivars of glycine (Tinaroo, Cooper, and Clarence) failed to persist under those conditions and that at 36 weeks the proportion of weedy species was drastically increased. Their final conclusion was that it is better to use light stocking rates and that the legumes are useful in those areas in which the grazing management is extensive. On the other hand, there are some results that show a positive effect of stocking rate upon the proportion of legume in the pasture. Vilela (1979), in Brazil, found that







25



the legume percentage was 8.7, 10.5, and 15.6 for the 0.5, 1.0, and 1.5 animals ha- 1, respectively, when it was measured with the point quadrat method.

Jensen and Schumacher (1970) observed that the percentage of

botanical species is not only affected by grazing animals, but also by environmental conditions such as season, precipitation, rain distribution, temperature, and solar radiation. Taking into consideration most of the above factors, Tothill (1978) noted that if the primary objective of the investigation is to obtain an estimate of botanical composition of pastures in assessing animal production, weight of species is the most suitable value to measure. If rainfall interception or photosynthesis are under study, then cover may be the more appropriate parameter. He also mentioned that the importance of this distinction is that number, weight and cover measurements are comparable in time and space, but are independent of the mode of sampling since they are measured directly and expressed in relation to a unit area. Shaw and Bryan (1976) mentioned that the proportion of species on a weight basis is generally the most useful where the main interest is in pasture production and where samples are cut for yield determination, and botanical composition can be determined by hand-separating the sample into component species. They also added that this is the most precise and satisfactory method for yield and botanical composition determinations.


Response Surface Methodology


Littell and Mott (1975) indicated that the purpose of response

surface methodology is to estimate the functional relati-onship-,-between a response variable such as yield and an experimental variable or







26


control variables, such as rates of P. They also suggested that the range of values determines the experimental region, and the functional relationship is called the response surface.

Factorial arrangements of treatments provide good information, but they require greater numbers of experimental units and more physical resources; with the same amount of resources using response surface methodology, it is possible to obtain results from a much greater number of variables and levels within each variable. These designs, such as the rotatable central composite, non-rotatable central composite and San Cristobal, have been used in grazing trials and have proved their usefulness in obtaining valuable data (Maraschin, 1975; Mott, 1977; Serrao, 1976; Villasmil et al., 1975).

Maraschin (1975) and Serrao (1976), both using a central composite design, studied the effect of three variables: grazing days (1, 3.5, 7, 10.5, and 14), resting periods (0, 14, 28, 42, and 56 days), and dry matter residue left after grazing (500, 1000, 1500, 2000, and 2500 kg DM ha-1 ) upon the botanical composition of Cynodon dactylonDesmodium intortum-Macroptilium atropurpureum-Lotononis bainesiiTrifolium repens mixture. The great advantage of this design is that it used only 24 treatments instead of 125 that the complete factorial would have required for a single replication, to obtain the coefficients needed to evaluate the grazing management systems. Serrao (1976) reported that the most important factors in determining dry matter yield from pasture and also legume percentage maintained in the mixture were grazing pressure and rest periods. He found that the legume content of the mixture was increased with an increase in length of rest period






27




and that heavy grazing pressure and short rest periods almost eliminated the legumes from the pasture.













CHAPTER III
MATERIALS AND METHODS


This research was conducted at Estacion Experimental Tropical Pichilingue, belonging to INIAP and located 7 km from Canton Quevedo, Procincia de los Rios, at 10 06' S Lat. and 790 29' W Long. Altitude at the site is 64 m above sea level. The average minimum and maximum temperatures are 17.3 and 35.6C, respectively, having a mean annual temperature of 24.3C. Annual precipitation is 2152 mm. About 82% of the yearly rainfall occurs during the warmer months from December through June. The months of February and March have the highest rainfall intensity. By contrast, October and November are the driest months, often registering no rainfall. Annual mean relative humidity is 84% and mean sunlight is 846.2 hours per year, 66% of which occurs during the December to May period (Servicio Nacional de Hidrologia y Meterologia del Ecuador, 1980). Holdridge (1967) located Pichilingue in the tropical moist forest zone (Fig. 1).

Figures 2, 3, and 4 present the monthly rainfall range, average temperature and average solar radiation for the years 1978, 1979, and 1980.

The soils are classified as Torripsamments. Hardy (1960) states that the chemical analysis of this Pichilingue loam reveals a marked deficiency in available P, but an abundance of available K. The soil N status is fair to medium in recently cleared land, but declines rapidly with cultivation. More recent soil analyses at the experimental site revealed that the amount of P is medium, while B, S, Zn and Mo are low (INIAP, 1979).


28






29

a:
2 u 0 0 Ln
w 0 0 co rn a) v- cD r*j N
I I
o 0 C) o o 00
E 0 0
oo 8 C) 0 C) to :r n CY C) 0
>
C) -H
(*Ulotlz-ozl) UJWOO09-0009) 4-j
IS36W IVOWCAU 13M A63A
,v
LO to
(*UlOlt7l-ool)(UJWOO92-OOGZ) >
JLS380J-'lV:)Idaiiens 13M UGA co
GAIIV )
SONVISSV8D 0100 13M
to
vnovanEWL
GAIIV
SONVISM9 0100 '80,
(S3SSVW) 031Mrino 1 0 0
r? "0 9 3AIJ.VN) S = 'VA MO ca
mxmzin, 2
SON'nSSVJ0 HDIH C)
0
m
< 0 Z
GA LWN) -,4 4 '
SONVISS"D 0'100 ;, o ca
4 W
co w
OZV8OeV4lHO $4 4-J
co C:
V8(]Nn-L CnO:) A0 0 ca
NOLIV1393A W -i
a) 14
ONVqSMB 0-100 13M-A80 4 CC
4-4
Culool-stl) LUWOOGZ-OOZI
JLS3hlOJ -VOidouens JOW S-- $4
4-1
0
4-4
0
W 0
(*UIOZI-09) WWO002-0091 0
.LS380J 'IVOWO U 13M 4-4 Cd
Cul 09- tpz) (WWOOSI 009)
IS380J IVOW081. A80
Cul i7Z ) WLU009 HVNNVAVS AJ
L)
NIV300 DIJIC)Vd









30

























I z



0 13 P4

ca










0

ca
4j
w
w 00
-W r-4 co I 00



0 4-4 u 0

0
r--f 4 cc (1) 44 Cl

-r-4
LL












(mm) TTUJUT'e'd








31



































4-1 04

















0
u 00 0) 10


ca
0
LL







C4 C4 C4 C4 44 C4



(o,) ainava9duia,










32















0
*-H


4-4









*i-4







-i




s~c o







0 c-,







0 M0

*o I

I

LL -4







o 4 04 Go a 0 in 0


Cl 0 60






33




Table 1 presents the soil analysis on the experimental site used in this research.


Legume-Grass Mixture


The legume-grass mixture selected for the experiment consisted of Malawi glycine [Neonotonia wightii (R. Grah x Wightii and Am.) Lackey], commercial centro (Centrosema pubescens Benth.), guineagrass (Panicum maximum Jacq.), and Hybrid 534 elephantgrass (Pennisetum purpureum Schumach.).

Guineagrass and elephantgrass were considered because both are very common and useful grass species in the littoral region of Ecuador, and also because of their high-yielding capacity, drought tolerance, and ability to grow in mixture with some tropical legumes. Centro is a native legume and may be the most widespread in the lowlands extending from dry tropical forest to wet tropical forest (600-2500 mm of rainfall). Glycine had been selected from previous experiments as a very persistent and productive legume.


Experimental Variables


The experimental variables were

(1) Days grazing (X1)--l, 7, 14, 21, 28;

(2) Days rest (X2)--0, 14, 28, 42, 56;

(3) Grazing pressure (X3)--dry matter on offer per 100 kg body

weight (BW); and

(4) Phosphorus (P2 0 5) levels (X4).







34






Table 1. Soil analysis of experimental site (1978).



Depth of sample: 0-25 cm

pH: 6.30

Percent organic matter: 4.96



Element ppm


Phosphorus 62.5

Potassium 598.8

Sulfur 18.4

Zinc 13.2

Manganese 258.5

Copper 20.9

Boron 12.6







35



For this experiment grazing pressure was expressed as the amount of dry matter on offer per 100 kg body weight (BW) and by the residual dry matter in kg ha-1 left after each grazing.

Each experimental variable was studied at five levels. Therefore, the treatments comprised a factorial type of experiment of four factors, each at five levels (5 4 factorial) (Table 2).


Experimental Design


Due to the large number of experimental units required to

conduct a 5 4 factorial (626 treatment combinations without replications), a response surface design, namely, a modified central composite non-rotatable design was used. The number of design points (treatment combinations) was determined from the following formula: (see Table 2)

No. of design points = 2 4 (l) + 2 4 (2) + (2 x 4) + 1 = 41

Certain treatments were replicated twice (central point was replicated thrice) and these are indicated in Table 2. The total number of experimental units was 51.


Field Plan of the Experiment


In order to estimate the size of experimental units for each treatment, the following formula was used: (see Table 3) S NdR
DG

where S = size of experimental unit in m 2

N = kg body weight/pasture/day (assumed 300 kg BW),

d = number of days pasture is grazed during cycle,








36











41 41 to U -I Ln ON M -4 Ln M M r-i Ln M M r-4 Ln M M 0 0 10 'J'
>1 > Cq M -.T .C C-4 M T .O M -T %D C'J Cn -T "D o 0 Ln 00 0 o Ln 00
ca u u u u u
cc






co (D 4 -4 -4 -4 r--i r-4 r-q -4 -4 -1 r-i 4 r-4 A C-4 r- 4 C4




$4
r-4

x N ca
-ri 4 -4 to
"-q -4 0 C) C) 0 CD CD CD CD CD CD C) CD CD 0 0 CD CD C) CD CD CD CD 0
z a) 0000 cc CDC 0 0 CD 0 CD CD CD 0
0 4 > r-4 r--4 r-I r-4 r-4 r- r-i r-I M M M M M M M M
4-4 w w --T

41
4-4
w
$4
bo go m cn M cn
-H a,* C C C C C r C 119 ll 1.9 1 9 '19 1
CA N m
Q) co CU m m m m m m m m ID 110 110 '.0 r-4 r-4 r-q r--i w w 0 0
$4 4 Cn
0 P4 X
41
-H ca
ca 0
0 0 0 4j
ca ',, M C14 T -Z C14 C4 T T CA t IT I:T C*ll C4 0 0 1,0 1,0 0 0 ',0 %D
0 m a) t -T -i r-4 T -:T r-4 r- T T r--4 r-4 T Ln Ln Ln Ln
0 w
-H ca

0 10 p
to

41 r- 1-4 r- -4 rl 4 -1 00 -4 00 r-I 00 r--i 00
-H > ca C4 CN C4 C4 cq CN C4 C14 C14 C\l
cn 0 W
0 co M 0

0 c
u u
co r-i r-I r-4 r--i r--i -4 -1 r-4 -4 -1 -4 r-i r- -4 CN C4 CN4 C 4 Cq
r-i w
m
w rn
" r-q
C: Q)
Q) > On r-4 r-4 -4 -1 r-I -4 -4 -1 r--l r--q -4 r--f -1 r-4 C4 C%4 Cq C4 C4 C14 C l C4
X I


-H > 0
t4-4 -r4 U C14 -4 r-A -4 r--i r-A -I r--i r--i r-4 -4 -1 -1 l r-4 -q -4 C14 C4 CIJ C4 CN C*4 C14
-H 4-4 "a
0 41
co
r--l r-4 r--4 r-4 r-i r--4 "i r-4 i 4 i rA r i i CN C 4 CN CN4 CN C%4 C 4

C
w
r-4
.M
ca -i cn T Ln o r, co o, o -i cA m .T Ln o r 00 m C) -1 CN M T
H -4 r-4 -4 -4 -1 -1 -4 -q -I C4 C11 CJ







37










-1 4J 4J 41 41
u 9: 0 %.o -T 0 0 %-c -T 00 Z 0 %o '.0 "o %0 o
>% >1 0 0 Ln 00 0 0 Lm 00 C14 Ln 0 r, Ln LM Ln Ln Ln
ca u C-)






04 r-4 -4 C14 C*4 r- r-i r--4 r-I r--j r--j r--j j Cn




w "4 w I
N cc
-H C
r-i
H -i to CD 0 0 0 CD 0 0 0 0 0 0 0 CD CD CD CD CD
41 a) Nd C) 0 0 CD 0 CD CD CD 0 c CD CD CD CD 0 CD
4 > eN -T It It T -,I --T T -T C4 C14 C4 Cq C 4 04 11 C14
w Q) IT
44 4 x 1.11


P41
bo w i.
0 CQ w C 1 1 C 9 1 C! C 1 r
a) r-f r--l P-4 00 00 00 00 Ul) Ln Ln Ln r-q 00 Ln Ln Ln
w cn

W


S w C 4 CD o %0 0 CD o %o %0 00 00 C) % c 00 00 co 00 00
-W ca w x Ln Ln LM Ln C14 C14 Ln C4 C'J C4 C14
co p P4



to
r.
W N 4 r-4 00 r--l 00 r-i 00 r--i 00 r--j 00 -:T T -1 ,1 T T
>% ca C4 C14 C14 C14 r--l r--l r-I -1 -4 -4
ca W




-,I C14 CN C14 C4 C14 C4 C4 C14 0 0 0 0 0 C> C-4 C,4 C:) X I



Cn CJ CN C%4 CN CN C4 C4 C 0 CD CD C*4 CN 0 C) C:) 10

0
C4 C4 CN C%4 C4 C%4 CN C4 CD C C4 4J
r.
0
u
I
I -i CN 04 04 . C-4 04 0 0 C 0 0 C) CD
C4
0i

Ln %.D r CO 0", 0 -4 C4 M It Ln %,D r CO M C) -4 E z C4 C4 CII C14 Cq M M M M M M M Cv) M M -T -T








38








w co
-H E-4
> m ca 0 Ln 0 0 Ln 0 0 a 0 Ln 0 0 Ln a 0 0 a 0 0 0 Ln 0 a 0 0
co 4-4 41 w v p r- Ln Ln rl- Ln 0 a Ln r- Ln Ln r- Ln 0 a Ln 0 0 Ln r- 0 0 0 0
-j-4 4J 0 w w w r--l r-4 C*4 --q r-q T--i r-q C14 r-I r-q C14 C14 -It 14, 4 04
E-4 -c4 124 92.
>

&4 41
$4 44
0
CC) Cooccoo occoo Coo CDC cocco 4.J Q) Ln C 0 Ln a 0 0 C Ln C C Ln C 0 0 C C 0 C Ln C a C 0
Aj U*l Ln r- Ln C C Ln r" Ln Ln r- Ln C 0 Ln C 0 Ln r- 0 C Ln C
r-i CIA r-4 r-4 q r-4 CN r-I r--i CN 04 -T --T C14
En
4




r--l r-I 4 4 "4 "1 "1 -4 "-1 r--A r-4 -4 r-q P--i -4 CIA C14 -1 r-4 C4 Cq

44 r-4
cl
r-I
-W r.
H Q) w
s 0 ca
.rq N z
4 -H w
a) r-4
-r4o. -Hr-q
w x WQ),-N
w a) $4 > -,T r-4 r--i r-i r-4 r-4 ?--i r-I r-4 M M M M Cn M M M
"a w 4) x
-4
Q) Ln

co
41 41 (U
ca $4 t3:
4J z =
0 0 -H w b lo
14 C N m m M cn M %D ID 1.0 cn M m m %0 110 %0 %.c 110 %D %D cln cn M cn

w cn m m m C4 C4 C4 c c c 0
0 w U P4


41
-H
m
0 -Zt C-4 C-4 It -T Cq C4 IT IT C4 C4 -,1 -1 C-4 C C "0 C 0 .O .O
-W CO -i -I -T zr r--l r--4 --T T r--4 -1 -,1 --T r-4 -4 T T VI Ln Ln Ln
co
0 cu
u u p
co F-4
r-I W
co 60
p m 0
-q -ri .-,
w w N r-i r- rH r r-i r r-4 r- r--i r r--l rl -4 r r-I r- r..i -4 00 r-i 00 r-i 00 -4 co
(U > > ca C-i 04 04 C14 C14 C14 04 C14
a) ca W
r--4
10
w
-r-I >
44 -r-4
-H 4.4 Q)
lc w
0 41 :$
%D tn --T -i lZ M u) C4 tn W w z r-4 C*4 IT cq Ln cq m M m r-4 --T C14 -4 m m m I-q m C*4 -IT 'T
ca
04 Ln C4 rC14 cn
Q)
r-q
4
ca m -T Ln I'D r- w (71% r-q m -,T Ln M O' C -4 C4 M ,T
E z -1 -1 r-4 -4 -4 -q -4 r-4 CN C4 CN CN










39






$4
-H E
ca ca z 0 0 0 Ln 0 0 0 0 0 0 0 Ln Ln 0 Ln Ln Ln
Aj w a, w 0 0 Ln r 0 0 0 0 Ln 0 0 r- r Ln r- r C14
0 w w w C*4 C14 T -4t r--l C 4 CN cn r-I 04
E-4 -:4 1:4 la.



41
4-4

(n CD c a a C) o CD CD 0 0 CD c C) p CD CD CD
CD CD CD tn C) 0 C) C) CD a 0 Ln Ln 0 Ln Ln Ln 0 0 Ln r- CD 0 Ln C) Ln CD 0 r- r- Ln r r r
-14 x C14 'T r4 cn -4
cn w













cd



CCD
14 N CN 04 (N

44 0-4





00 w pq C: :1 81*
w 1.0 ko %10 %D M M m m CD 0 C) CD c en 0 CD 0
M -N .
co (D cn r-q -1 r-4 r--4 co 00 00 co Ln Ln Ln Un r-q CO Ln Ln Ln





41
CD CN C) 0 %D %.o CD 0 ND %D co 00 0 1.0 00 co 00 00 co 0 ca a) x U') U") tn Ln C4 CN a) cq clq .






M N 1-1 m r-i w q m W r-I W -11, 'T --T -cr -zr --T C14
>, CO cq CN -i V-4 -1 -1 r-q r-i r=
CO 4
C




$4
0 Z
u -W 0 CD all r- CD IT a\ 110 r-q Ln 1-1 r- r--4 Cq r--l CYN cr cc
I w Z -a, IT r--q -T -T Ln 'T -1 -1 CY)
1 0 a7
a4 1: C l cl ca
C r-4 CN 1-4
1-4
cc
r-q 4J
.0 0
0 8 Ln \Z r 00 C\ CD -4 C14 Cn -T V) \.C r- 00 M 0 r-I E--q
E--4 z C4 C4 C4 CIJ C4 M M M M M M M M M M -ZT T +-







40



R = kg dry matter offered/kg body weight/day,

D = number of days in cycle, and

G = growth rate in kg/rn2 /day (assumed .04 kg/rn2 /day).

As an example, treatment 14 days of grazing, 28 days of rest, and

5.0 kg DM on offer/iQO kg BW/day is calculated as follows:

S=300 x 14 x .05 = 25_ 2
S 42 x0.004 -20

This figure was rounded to 1500 mrn to avoid the inconvenience of odd pasture sizes. All pastures were 500 m 2or some multiple (Table 3). In making this calculation it was assumed that one animal weighing approximately 300 kg was used to graze the pasture during the allotted grazing period and designated grazing pressure. It was also assumed that the growth rate was 4 g/m 2/day. In estimating the size (S) of the experimental unit the smallest was set at 500 m 2in order to avoid difficulty in handling the steers in units of smaller size. A total of 7.3 hectares were required for the 51 experimental units in this study (Table 3). These pastures were randomly distributed as shown in Fig. 5. The sizes of the experimental pastures varied from 500 m 2to 4000 m 2; the larger areas were for the continuously grazed treatments (Fig. 5).


Land Preparation and Pasture Establishment


Land preparation of the experimental area began in May 1977, after the existing vegetation (common guineagrass) was partially eliminated by an application of a grass killer herbicide (Glyfosateo). At the beginning of July the 6-year-old guineagrass pasture was plowed under.









41







































ca cc












44 CL "0


44







42



In August the complete area was cross-disced in order to eliminate some weeds and obtain a good seedbed. The area was disced again in late September to eliminate the new growth and some weeds. In the same month, 20 small soil samples were taken from the whole area and mixed together for soil analysis; the samples were taken at 25 cm depth (see Table 1).

The next step was to prepare the legume seeds. Glycine and

centro seeds were scarified using sulfuric acid for a time of 6 and

8 minutes, respectively. After the seeds were washed and shade dried, they were inoculated using an appropriate Rhizobium strain obtained from CIAT (Centro Internacional de Agricultura Tropical). In early October of 1977, a mixture of glycine and centro was sown using a small grain seeder with a four row capacity at the rates of
-l
3.0 and 6.0 kg ha respectively. The mixture was sown in rows spaced 1.4 m apart. A day after the sowing, a mixture of preemer-I -i
gence herbicides was applied (Linuron 0.75 kg ha + 2 L ha of Laso) to control both grasses and broadleaf weeds coming from seed. Two weeks after the legumes were sown, the grasses were planted vegetatively, using plant divisions for guineagrass and small pieces of stems with two or three buds for elephantgrass. Both grasses were planted alternatively between the rows of the legumes, so that the distance between grasses and legumes was 0.70 m. It was necessary to make one hand weeding using machetes. Four irrigations of

4 mm each were necessary at 12 day intervals. The first irrigation was immediately after the legumes were sown.






43




In early January of 1978, the P fertilizer was applied taking into consideration the respective levels of simple superphosphate kg ha- 1 (whose composition was 18.2% P 2 0 5 and 14% elemental S); at the same time a mixture of micronutrients was applied to the whole area. The composition of this mixture was 3 kg Zn chelate, 4 kg Fe chelate, 3 kg CuSO 49 3 kg Borax, and 0.8 kg Molybdenum nitrate ha- 1 .


Construction of Physical Facilities


In February, the total area was surveyed for the purpose of locating the fence lines, and four-strand wire fences were built. During the second half of March, the entire area was moderately grazed and the remaining vegetation was mowed at 15 cm height. Wooden mineral boxes were built which were used to supply the following formula: 50% of sodium chloride + 50% of mineralized salt containing 7% P as dicalcium phosphate, 0.48% Zinc sulfate, 0.12% manganese carbonate, 0.14% copper sulfate, 0.32% ferrous sulfate, and 0.006% cobalt chloride. Water tanks were provided in each experimental pasture. An area of approximately 10 ha adjacent to the experiment was available as reserve pasture for 50 animals used for adjusting the stocking rates in the experiment.


.Collection of Data in the Three Experimental Years.


On May 12, 1978, experimental grazing was initiated. Fifty-six Criollo and Holstein/Brahman steers were used to graze the experimental pastures. At the start of the experimental period, the






44



animals were 16 months old and their average weight was 268 kg. Every 28 days all animals were removed from the pastures and from the reserve pasture. They were taken to the main corral for tick treatments and every 56 days they were weighed in order to have a basis for estimating stocking rates for the next grazing cycle.

At the end of the first experimental year, the pool of animals

was removed from all the pastures and showed average weights of 428 kg. A new pool of younger animals replaced the first group for the second year, having an average weight of 302 kg. The animals for the second experimental year were from the same herds as those of the first year and were 18 months old.


Pasture Measurements


Dry Matter Determination Before and After Grazing


Dry matter ha- 1 estimates before (on offer) and after (residual) each grazing period were made in order to apply the required grazing pressure and to determine the net dry matter production.

Stocking rate was determined for grazing pressure-grazing period combination on the basis of the total dry matter available before each grazing plus the estimated growth rate during the grazing period. Growth rate during the grazing period was assumed to be the same as that of the previous rest period. The accuracy of this technique for determining stocking rate was checked, using the residual dry matter left on the field after each grazing period. Grazing pressure was also based on visual observations of the amount of dry matter during each grazing period, in order to add or remove animals from pastures





'A







45



depending on the specified dry matter on offer and days of grazing. In order to estimate the grazing pressure, the following formula was used:

G/P (A r + gd) S
RNd

where G/P grazing pressure in terms of 100 kg BW,

A available forage before grazing,

r residue for a specified grazing pressure,

g growth rate-preceding rest period,

d number of days for grazing period,
2
S size of the experimental unit in m

R kg dry matter offered/kg body weight/day, and

N kg body weight/pasture/day.

A double sampling technique was employed. As soon as possible, before and after each grazing (during the first 3 1/2 months only), 15 areas measuring 1.0 m 2 were randomly selected. From September 30, area measurements of the same size were taken, corresponding to the circular frame of a forage disk meter (one m 2 ) similar to the one described by Bransby (1975) and used in double-sampling measurements. In each sampling unit the percent dry matter yield of each component of the mixture was visually estimated, followed by an estimate of the dry matter yield in situ. The disk meter was lowered on the forage and after a settling time of approximately 5 to 10 seconds, the disk height (in centimeters) was read off a graduated scale mounted on the center shaft of the disk meter and recorded.

During the first three and one-half months, from the 15 sampling units, three were randomly selected and clipped at ground level for







46



dry matter yield and botanical composition. In September, five sampling units out of 30 disk readings were randomly selected and clipped for the above determinations. These samples were later hand separated into their components, placed inside cloth bags properly identified and dried for 20 hours at 72*C. The sum of the dry weights of the components yielded the total dry weight of the sample. Sickles were used for harvesting the forage samples. The clipped samples were used to adjust through regression analysis the 15 or 30 disk-meter readings of dry matter yield. The forage meter readings and visual estimates of dry weight were used as independent variables in regression equations to generate the regression coefficients needed for calibration of the disk meter and for adjustment of the visual sample estimation. The visual estimate of percent yield was made for the component grasses (guineagrass and elephantgrass), legumes (centro and glycine), and weeds. Although the above grasses and others such as Paspalum fasciculatum, P. paniculatum and Eleusine indica were present in scattered, small patches in some pastures, they were included in the weed components. Likewise, some native Desmodiums were accounted for in the legume component.

The response of the pasture mixture to the experimental variables was measured in terms of the following parameters:

(1) Aerial biomass (DM) kg ha = grass (DM) + legume (DM) +

weeds (DM),

(2) Available forage (DM) kg ha = grass (DM) + legume (DM),

(3) Grass yield (DMI),






47



(4) Legume yield (DM),

(5) Yield of weeds (DM),

(6) Grass percentage, and

(7) Legume percentage.

The above parameters were statistically analyzed for the partial wet season (May-June) of 1978, the dry season of 1978, wet and dry seasons of 1979, and wet season of 1980. The data were processed using the programs RSREG for response surface design, and 63D for plotting the three-dimensional graphs of the Statistical Analysis System of the Northeast Regional Data Center of the University of Florida.












CHAPTER IV
RESULTS AND DISCUSSION


Only two of the four experimental variables included in this experiment, namely, lengths of rest period and levels of grazing pressure are discussed in this section. The other two variables,

days grazing and fertilizer level had negligible effects upon the response of the pasture sward. Each of the response variables will be discussed in a separate section beginning in the wet season of 1978 and ending in the wet season of 1980.


Effect of Lengths of Rest Periods and Levels of Grazing
Pressure on Aerial Biomass (DM)



The effect of the lengths of rest period and levels of grazing pressure on the aerial biomass is presented for each of the five seasons in Table 4. The total biomass is the average amount of dry matter present before each grazing period for the rotational grazing treatment combinations, and for each grazing period of 56 days for continuous grazing.


Biomass Production (DM) for the Wet Season of 1978


During the first wet season only lengths of rest periods indicated an effect (P < 0.01) on the aerial biomass produced (Appendix Table 11) The linear components of the model accounted for 20% of the total variation, while the quadratic effects and interactions represented only

4 and 13% of the total variation, respectively (Appendix Table 11).



48








49





00 En r- %D -.1* 0 o w r- r- N M -11 tn r-4 -11 00 Cq ND r-I CN r- r, Ln CN
(7N -- 'T M 0% Ln r- ON r- a% CD 00 r- L'I -T %.0 a -T Ln 00 00 -It 110
r-4 :3: cj m m m t c,4 m Ln c,4 r m Ln c 4 r- m -i -i r, m Zo





00000000 coca 00000 0000000 r- M tn %Q %0 00 Ln 04 M M r-4 0 r-I r-4 CO 00 00 r- r- M ON %D Ln Ln M 00 Ln %D 00 -zt 00 0 r- Ln C4 r- tn 0 M CS r- -.T C-4 lZ r- 0 M 00 T
04 04 C14 r--l -,t C14 M -T CsJ M 04 Cq M C4 Ln C4 f-4 r-4 M C4 -,I M M -T





m 4+ Co OCCDC 0000 0000 coo 0000000 co
P, m 4 M -4 tn 0) M C4 -Zr M 04 --T -4 M Ln aN M -T -1 1,0 1,0 ',C -7 M T
M 0 C4 00 --T 00 r-i Cq 00 M -,T %0 T 00 ,O -4 r- CN r r- 0 ON M %0 ON
:3 60 M M M M %zr M -zr tn M T M T C4 04 *,C M -I Ln %T M M %.D Ln
.hd




w Co 1.+ 000000000000000000000000
rz r- En r-4 M ',D Cn ON Ln 04 %-T -4 C14 00 M C4 --T C) 0 -T C4 C14 O -Zr 00 0 Ln
4J T r-- r- Ln %D 0 0 %0 M M 00 C4 1.0 M*-Zr r-4 C4 -1 %0 Cq Ln Ln 00 rCo C*4 C%4 CN C4 M CY) -Z -T r-I C14 04 Cq C14 -,T M r--l -i M CN M M T Ln
w
$4
4-1


Co 0000000000
M m %0 rl -1 IT Ln Ln C) ON Crs V) r- Ln Ln M Ln M 00 0 00 1.0
m r- C7% %D 00 M r-4 r- M r--4 00 M U* C'J r-A 00 0 r- M r- T C*4 00 M T
r.-I C r-4 r-I CN r-4 Cq C4 C4 r-I M CN C14 r-I CIJ CY) -i Vn l 4 M 04 Cq -4 M M
00 CD 0 C, C:, 0 C:, a C:, C) 0 C) 0 C, C:,
0
w
Ca
Cu
w
w
CL
Ca Q) r-q r-4 Cq r-I -1 C*q
w 04
>1




Ca
11

hd
0
.r.1
41


0
w . . . . . .
M M M M M "o "o .o %C M M M m ,o r-f '.4 r4 O 0 0 0
X

Ca
E Cu
0
Co 04 Cq -T -zt C-4 CIJ -.T -,t C-4 C-4 -,T -Z C-4 CN -,T -T C14 CD 0 NC %.0 o 0 o "o
-4 r-4 -31 IT 1-4 "q -T -T 1-4 1-1 I:r It 1-4 r-4 IT IT Ln tn Ul) Ln
Co

Q)
r-4 r- -4 r-- r- r-A r- r-I r r-q r-I 00 -4 00 r-4 Co rq 00
Cq Cli CN (N C14 C14 Clq C*4 C14 Cq rq



r-q
Q
Ca MIT Ln"or-W Ma-icl4m-T Ln D rw mc -f M 'T
z -4 r-q r.-q r-4 ?--I r-f -1 r-4 r--( -4 CNJ N" Cq CN








50



0++
00 rn r-4 Cli Ln M w %o -IT %D ID C14 m m ol ID ON %0 a
ON 0 w Ln 0 %0 r--4 Ln r-I a,% -T w 0 r--l Ln r- Ln (3)
m C14 m IT m -It r- %D Clq M r--l It C*4 m m cq





ON ++ 00000000000000000 r4
W m ON Ln I'D tn Ln a m w 0 Ln N .o r-q M r-4 "
-- r-I 0 r-4 Ul) %.0 r-q o w r- CIN %0 m m C14 fl- w IT
r-q cm N r--4 N C14 m m Ln N r-i m m m N N N
$4
w
t4-4



M m (14 w ON %0 0 r- Ln M m Ln W %0 m C14 0
m N N 0 -It r--i 1.0 N -T 0
r-q 1--i
14




co ++ 00000000000000000
w 04 a% Ln M CN r- 00 -1 r- C14 % O P-1 0 Ln r--l r- "0
1 C4 r- 'T %0 %D Ul) 00 00 r- -T r- r a*, a% Ln r--l cn 0
?--I gm r--4 CIJ C4 T -,T C4 r-4 M r-4 C14 C4 CN CN a




co 44 CD CD CD CD C) 0 C) CD C) CD 0 CD CD 0 0 0 CD w
r Ln rn CD Ln CD M r rl- M r- r- CYN M -T cn %0 o %o $4
0., a a% m r- %0 IT -T r- Ln Ln %D r-4 00 Ln 00 0% cn 0
r-q C4 r-I Cq r--l 04 CN M CN -4 M -i r-I i Cq to
ca





r-4 r-4 C14 C14 r--i --I C14 04 q -q 1-4 "4 r-4 r-4 q q M N co



r-4
I a4
Ic CD C CD CD CD CD CD 0 CD CD CD CD CD CD CD C) C) u
rM4 CD CD 0 CD CD CD 0 0 CD 0 0 CD CD 0 CD C)
to -T -T -ZT %T T -,t T -,T Cq CIJ C4 CN C4 C4 -4T CN 4j
14 m
a) 0
co

-I-
P4 'D %.0 cn cn rn cn CD 0 CD 0 o M 0 0 CD "a >
. . w
cn -f --1 -4 -1 00 00 00 CO Ln Ln Ln U") -4 00 Ln Ln Ln


En


ca W. C,4 0 a %0 .o a C %-0 c CO 00 0 "D 00 00 00 00 00
Ln Ln Ln U*) C4 C4 Ln C4 C14 CIJ Cq C4 -H
N 0
co En

0
q r-4 W r-4 W r--l M -4 W r--f W -,T t -,t
C4 C14 C14 C14 --q -1 r-4 -1 -4 4 I-q CO w
CZ


cn
M Ln 1.0 r- 00 m CD .-I M --T Ln 1 0 r- 00 m 0 r--q :3
z cli cli cq C14 C4 M m M m Cn cn M cn CY) en 'T ++






51


The biomass production varied from 1570 kg DM ha- I for treatment 17 (1 day grazing, 0 days rest, 1.6 kg DM on offer/100 kg BW, and

0 kg ha-1 superphosphate) to 3750 kg DM ha- 1 for treatment 19 (1 day grazing, 56 days rest, 1.6 kg DM on offer/100 kg BW, and 0 kg ha-1 superphosphate).

The aerial biomass production response to lengths of rest periods

can only be explained by the management imposed during the first grazing season. All of the experimental pastures were grazed at a medium grazing pressure for 15 days during the latter part of March of 1978, after which they were mowed at about 15 cm above ground level. The wet season of 1978 was represented by a growth period of about 3 months from the first of April until June 30 when the wet season ended. The production data were obtained during this 3 month period beginning on May 12 after only about a month of regrowth. The pastures with short rest periods were included in the first sampling date and for some of these pastures the length of the grazing cycle was only 21 days which means that they had the opportunity to be sampled twice between May 12 and June 30. Pastures with a longer rest period accumulated more dry matter during the growth periods and, thus, showed a higher amount of dry matter produced. The analysis of variance for the wet season of 1978 is presented in Appendix Table 11. Biomass Production (DM) for the Dry Season of 1978


The biomass production during the dry season of 1978 is also presented in Table 4. The biomass production varied from 790 to 4880 kg DM ha- 1, corresponding to treatments 26 (28 days grazing,







52



0 days rest period, 1.6 kg DM on offer/100 kg BW, and 400 kg ha -1 of superphosphate) and 32, respectively (28 days grazing, 56 days rest,

8.3 kg DM on offer/100 kg BW, and 400 kg of simple superphosphate). In general, the lowest values of biomass production corresponded to treatment combinations of short rest periods and high levels of grazing pressure, while the highest levels of biomass production corresponded to the treatment combinations with long rest periods and low levels of grazing pressure.

The linear components of the model accounted for 74% of the total variation, while the quadratic effects and interactions represented only 1 and 2%, respectively (Appendix Table 12) The experimental variables, length of rest period (X 2 ) and level of grazing pressure (X 3 ) each showed a linear effect upon biomass production (P < 0.01). There was a suggestion that levels of superphosphate (X 4 ) might be having some effect (P < 0.10) but the effects of days grazing (X 1 ) was nil. In all cases individual quadratic effects or interactions were not significant. The lack of interactions between X 2 and X 3 indicated that both variables were behaving independently. The biomass production was increased as the lengths of the rest period were increased and as grazing pressure was reduced. The analysis of variance for the dry season of 1978 is presented in Appendix Table 12.


Biomass Production (DM) for the Wet Season of 1979


The biomass production during the wet season of 1979 for each

treatment combination is given in Table 4. Biomass production varied






53



from 1210 to 6630 kg DM ha1 corresponding to the treatments 17 [1 day grazing, 0 rest period (continuous grazing), 1.6 kg DM on of fer/loo kg BW, and 0 kg ha1 of superphosphate] and 31, respectively (1 day grazing, 56 days rest, 8.3 kg DM on offer/iQO kg BW, and 400 kg ha1 of superphosphate).

During the wet season of 1979 both length of rest period and

level of grazing pressure had an effect upon biomass production (P< 0.01), while no quadratic effects and interactions were noted. The linear components of the model accounted for 67% of the total variation, and only 2 and 1% of the total variation was represented by the quadratic and interactions. As is evident, the biomass production is increased as the length of the rest period is increased. The higher grazing pressures also decrease the biomass production and the greatest biomass production was reached when a long rest period and low grazing pressure were imposed. The analysis of variance for the wet season of 1979 is presented in Appendix Table 13.


Biomass Production (DM) for the Dry Season of 1979


Biomass production during the dry season of 1979 for each treatment combination is presented in Table 4. Biomass production varied from 1090 to 5980 kg DM ha 1, corresponding to treatments 26 [28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on of fer/l00 kg BW, and 400 kg ha -1of superphosphate] and 32 (28 days grazing, 56 days rest, 8.3 kg DM on offer/l00 kg BW, and 400 kg ha- of superphosphate). Again, the lowest values of biomass on offer ha- corresponded to treatment combinations of short rest periods or continuous grazing






54



with the highest grazing pressure, but both experimental variables acted independently from each other.

The linear components of the model accounted for 45% of the

total variation, while the quadratic effects and interactions represented only 1 and 6% of the total variation, respectively (Appendix Table 14).

An examination of the individual linear effects showed that only the experimental variables, days rest (X 2 ) and grazing pressure (X 3) were affecting biomass production (P < 0.01). The number of days grazing (X 1 ) and fertilizer (X 4 ) had no effect upon biomass production. For each 14 days increase in the number of days rest there was a biomass production increase of 230 kg (DM) ha- 1, while for each unit (1.6 kg DM) decrease in level of grazing pressure there was an increase of 480 kg ha-1 in biomass production. The analysis of variance for the dry season of 1979 is presented in Appendix Table 14. Biomass Production (DM) for the Wet Season of 1980


The lengths of rest period and levels of grazing pressure had a

direct effect upon biomass production for the wet season of 1980 (Table 4). There was no interaction between these two experimental variables during this wet season.

The linear components of the model accounted for 43% of the total variation, while the quadratic effects and interactions represented only 8 and 1% of the total variation, respectively (Appendix Table 15). The total biomass production varied from 1460 to 7540 kg DM ha- 1 for treatments 18 [28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on offer/100 kg BW, and 0 kg ha- 1 of superphosphate and 31






55



(1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 400 kg ha-1 of superphosphate). During this particular season, the amount of biomass production was again most affected by the length of rest period and by the levels of grazing pressure. The highest values of biomass production corresponded to the treatment combinations of long rest periods and low grazing pressures. For each 14 days increase in rest period, there was an increase of 550 kg DM ha- 1 of biomass production, while for each unit (1.6 kg DM) decrease in grazing pressure there was a 450 unit increase in biomass production. The analysis of variance for the wet season of 1980 is presented in Appendix Table 15.


Summary of Biomass Production (DM)


In comparison of aerial biomass production (DM) among seasons

and years, large differences, especially between wet and dry seasons, are observed. The principal environmental factors involved in these differences between seasons are precipitation, temperature, and solar radiation (see Fig. 2 of Materials and Methods). Every year the rainy season begins in the second half of December, initially with light showers and then increasing in amount and intensity reaching the highest peak of precipitation in February or March. After this time the amount and duration of the showers decreases until it reaches almost zero during the second half of June. About 90% of the total precipitation falls from December to June with the remaining months almost completely dry. The second important factor is temperature which is always higher during the rainy season reaching an average of 26*C during the wet season, while the average during the dry season






56



is 22* C. The third important factor is solar radiation which is also higher during the rainy season, reaching values of 500-600 hours of sunlight while from July to November this value averages 350-400 hours of sunlight. These three factors acting together are the main determinants of total aerial biomass production.

Another very important factor which must be taken into consideration is nitrogen. This element, at the beginning of the wet season, is rapidly mobilized from the soil organic matter, which has accumulated in the soil during the dry season. Sanchez (1977) has indicated that N in the tropics is rapidly utilized by actively growing plants with the occurrence of the first rains at the beginning of the wet season but that the remaining N is slowly released during the rest of the season. These three environmental factors plus the availability of N have a great impact upon the aerial biomass production.

The most important experimental variables influencing the aerial biomass production were days of rest (X 2 ) and grazing pressure (X 3). Each variable behaved independently and there was a direct relationship between the number of days rest which permitted the plants a better opportunity to accumulate reserves for more rapid regrowth and development after each grazing period.

A similar trend was also found for grazing pressure which was

directly related to the amount of forage removed by the grazing animals which in turn was related to the high removal and damage to the growing points and axillary buds. Grazing pressure was related to the rate of recovery and subsequently the final production. Hodgson and Ollerenshaw (1969) mentioned that if grazing pressure is increased while resting






57



periods are decreased, the frequency and severity of defoliation is increased affecting directly the subsequent regrowth and the total dry matter yields. Harris (1978) reported that although the function of reserves, availability of growing points, and uptake characteristics are influenced by the level of stubble biomass, the relationship between stubble biomass and growth rate relates to the amount of photosynthetic tissues.


Effect of Lengths of Rest Period and Levels of Grazing
Pressure Upon Available Forage (DM)


Available forage in this context represents the sum of the grass and legume component but is exclusive of the weeds.

The effect of the lengths of the rest period and grazing pressure upon available forage is presented in Table 5 for the five seasons of this experiment. The available forage is the average of the estimated amount of dry matter present before each grazing for the rotational grazing treatment combinations and for 56 days for the continuous treatment.


Available Forage for the Wet Season of 1978


During the first wet season, only the length of the rest period had an effect on available forage (P < 0.01). The linear components of the model accounted for only 20% of the total variation, while the quadratic effects and interactions represented 5 and 12%, respectively (Appendix Table 16).

The means of total available forage varied from 1420 kg DM ha- 1

to 3720 kg DM ha-l (Table 5) for treatments 34 (28 days grazing, 28 days








58




0++
00 (n Ln Ln -zr N r Ln r- N IT -Z m IT N N m w Ln %.o r-4 N N -11 Ln N
m -- Ln r-i r-i M 0 r- CN r- Ln Ln W M tn M N \0 %0 Ln Ln W P-4 W -,T 14
r--l :3: r--i Cn M M r C*4 M Ln M C4 Ln C14 Cq r- M r m 110 I-T Ln






Ln 0 r-I r- Ln eq m m r-4 r- 0%0 w %D r %0 -zr -11 N --T
ON m %0 w N m 0 r -.T 0 r- -IT 0 N ON r 1.0 r- I'D r- 1.0 a% w -T
r-4 C N C4 r--f -4' N M -,I' N M N N M N Ln N cn C14 -.? cn M T





Ch ++ ca
r M M CN Ln %.0 %10 r- -It r-4 0 M r-I r IT a en 00 M 00 Ln M Cj C)
m -, C*4 r-4 00 M 1.0 0 C4 00 r-4 C4 1.0 -T 00 %0 r-I %D 00 M r- M W 1.0 M
r-4 1: 00 CN M M M -,T Cn t Ln Cn t Cn -T (N CN %0 M 1-4 Ln I:r m m %.o Ln le




00-1-+
r- W rl- N %0
m N r- r- Ln %0 ON 0 %D 00 00 00 C4 1 0 CO M CD C) %0 Cq Ln Ln r- rr-i C4 N N N M N -3' -T r-I N N N N T M r-I r-4 M M M T Ln
0
u
41

00++
r, W -:r Ln M m m m Ln M 0 m o aN M Ln a% m w 0 w 0
ON 110 m tn w M r--4 .o m r--l r- m tn C14 m r- 0 I'D m fl- m N w m -It
r-4 r-q CN r-i Cq N C14 r-I M CN C4 r-I CN C4 r-I Cn r-4 r--l CY) C14 r-4 M Cn
4J

10



r-I r--l r-4 r--l -i r--4 r-4 4 r-4 r--i r-4 i r-I r-A "-i r-A r-4 4 C 4 C14 r--l -1 C14 CN
0
w
ca


r-4
$: 1
ca co
(D -I.- Ic 0 C CD 0 0 a 0 0 CD a CD 0 0 a CD CD CD 0 0 0 0 a a 0
w r-4 r--l r-i r-4 r-4 r-4 -4 r-I M M M M M M M M


:3:

"D %0 "0 %DD %,o "0 ,c m m
C C4 C C4 1 1 1 1 C4 C 14 C4 1 1 1 14 C 0 C
00
co
w m
0 41
44 0
Q)
r-4 -W -T 4' C4 -,T -,T CIJ -ZT C4 CN
4 ca r-A r-4 IT T r- r- Ln Ln Ln Ln
co W
q p
.r4
co

r-4 r r-4 r -4 r r-i r r--i r r-4 r r-4 r 4 P r-4 r-4 00 r-I M -4 W r-q M
C14 C*4 cli C14 C14 C14 CN 04 C4 cli





ca C4 M -,t Ln ,o r- W M CD -1 Cli T Ln ,o r 00 C C) cj cn IF-4 z -4 -A -4 -1 -4 C4 C14 C14 C14









59



C ++
00 to -4 -zr 1.0 M M T r-4 1,0 ON C4 M O r- Ln 0 0
()N M 0 tn 0 Ln -4 M 0 M -J- -4 Ln
m -It Ln 1.0 N 1-1 m m





++
W 0 .o r-q .o m IC Ln Ln 1.0 ra% -- --I Ln O r-4 1 0 W r- ,O ',0 W r-q r- W
r- 9z N M M M tn r--4 4 -4 M C4 C4 04



r-4
ON ++
r- En Ln M %0 as -.T M P- cn r- m P--l r- Ln M M Ln 0)
ON -- ON 0% cq a r-4 Ln 00 cn 0 r-q r -:T M Ln r-4 cn a% N
"-I :2c -,T -3* M M %C D M M r--l -,T -4 M M M C4 n-I
14



4.4
00 ++ 0 0 C) a 0 a C) 0 0 0 C) C) C) C) a C) 0
m cn M Ln M r-q -zT w -,o w 0 --q Ln M Ln --T 11
1-4 r --T %D o -IT w w 1.0 r- r- w r- Ln 0 m r-I r-i C 4 CN C 4 N -Z -,T CN CN r-4 M r-4 CNJ CN4 C 4 44


ca

00 ++ 00000000000000000 a;
r- En M r-I CN 0 CYN V) r-4 T %0 CIJ t M %0 %0 o Cq M 4
m -- m 00 en r- %,D en T Ln -zr -T 110 r- -It 00 aN C14 :)
r--l :3: r-4 r-i CN 04 -1 CIJ C14 C4 C*4 C4 M r-4 Cq -I C14 M




ca co

r-4 r-4 V4 4 r-I C14 C4 r--4 r-I r--i r--4 -4 -1 M N
cc
w



r-4

ca
+-.C 000000cDcoccoccoccD
44 00000000000000 CC:) 4.J
60 IT 4r -T --T T -T -Cr -I' C*4 C4 C14 04 C14 Cq -T w
w 0
cc
cn 0)
PQ >1 (n
cu
coo ,Ccn cc 0 "o >,
M r-q 14 14 0 0 0 O L U 14 O L L L
X



E IM
4j
w Ca 04 0 CD 110 0 CD ID 10 co 00 C) %C co co 00 00 00 C:
-- X Ln Ln Ln U-i C4 C14 U-) 04 C-4 Cq CII -r-4
tQ 0
.rq 0 w
w ca
0

1-4 00 r-I 00 r-q 00 r-4 00 r-.1 00 -IT -It -It -It 14, -11 -It 41
C4 CN C4 r-q r-I 4 -1 -4 -1 CC W
LA CZ "0 3:
if
CD cn
M Ln 1.0 r- 00 O' 0 -4 C14 M %:r Ln %D r, 00 (DN CD r--l 3:
z cq CIJ CIJ cq M M en M M M M M M en ++






60



rest, 5.0 kg DM on offer/lOG kg BW, and 200 kg ha -1of superphosphate) and 19 (1 day grazing, 56 days rest, 1.6 kg DM on offer/lOG kg BW, and 0 kg hal of superphosphate), respectively.

Considering the individual linear effects, days rest (X 2) was the only experimental variable affecting available forage (P < 0.01). The other three experimental variables did not affect the available forage. Individual quadratic effects or interactions were not significant. The lack of interaction indicates days rest is acting independently over available forage, which means that the amount of available forage increases as the length of the rest period increases.

For each 14 days of increase in days rest (X 2), there was 190 kg hal increase in the available forage.


Available Forage for the Dry Season of 1978 (DM)


Available forage varied from 790 kg DM ha to 4880 kg DM ha1 corresponding to treatment combinations 26 [28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on offer/lO0 kg BW, and 400 kg ha1 of superphosphatel and 32 (28 days grazing, 56 days rest, 8.3 kg DM on off er/lOG kg BW, and 400 kg ha- of superphosphate), respectively. During the first dry season, the lowest values of available forage were found in those treatment combinations of short rest periods or continuous grazing with the highest grazing pressure, while the highest values were the longest rest periods and the lowest grazing pressures.

The linear components of the model accounted for 74% of the total variation, while the quadratic effects and interactions represented only 1 and 1% of the total variation, respectively (Appendix Table 17).







61


An examination of the individual linear effects reveals that

only the experimental variables days rest (X 2 ) and grazing pressure (X 3 ) had any effect upon available forage (P < 0.01).

The length of rest period increased the available forage as was also the case with a decrease in grazing pressure. For each increase of 14 days in the rest period there was an increase of 440 kg ha- 1 of available forage and for each unit (1.6 kg DM) decrease in grazing pressure there was a 500 kg ha-1 increase in available forage. The analysis of variance for the dry season of 1978 is presented in Appendix Table 17.


Available Forage for the Wet Season of 1979 (DM)


During the wet season of 1979 the available forage (DM) varied from 830 kg DM ha- 1 to 6630 kg DM ha-1, corresponding to the treatment combinations 17 [1 day grazing, o days rest (continuous grazing), 1.6 kg DM on offer/100 kg BW, and 0 kg ha- 1 of superphosphate and 23 (1 day grazing, 56 days rest, 8.3 kg DM on offer/lOO kg BW, and 0 kg ha-1 of superphosphate), respectively. The treatment combinations with the shortest rest periods and lowest grazing pressure levels yielded the lowest values of available forage for this season; also it was noted that in this case both variables X 2 and X 3 were acting independently. There was no evidence of any interactions occurring.

The linear components of the model accounted for 69% of the total variation, while the quadratic effects and interactions represented only 2 and 1% of the total variation, respectively (Appendix Table 18). Only the linear effects of the experimental variables X 2 and X 3 had






62



an effect upon the available forage (P < 0.01). Increasing the days of rest and decreasing the grazing pressure increased the available forage during the wet season of 1979. For 14 days increase in rest period there was an increase of 750 kg ha-1 of available forage and for each unit (1.6 kg DM) decrease in grazing pressure there was a 500 kg ha-1 increase in available forage. The analysis of variance for the wet season of 1979 is presented in Appendix Table 18. Available Forage for the Dry Season of 1979 (DM)


For the second dry season, the available forage for each treatment combination is presented in Table 5. The available forage varied

-1 1
from 400 kg DM ha to 5980 kg DM ha corresponding to the following treatment combinations: 25 [1 day grazing, 0 days rest (continuous grazing), 1.6 kg DM on offer/lOO kg BW, and 400 kg ha- 1 of superphosphate] and 15 (7 days grazing, 42 days rest, 6.6 kg DM on offer/100 kg BW, and 300 kg ha- I of superphosphate), respectively. Only days rest (X 2 ) and grazing pressure (X 3 ) showed any effect upon available forage (P 0.01). Each of these two experimental variables was acting independently because no interactions were found.

The linear components of the model accounted for 54% of the

total variation, while the quadratic effects and interactions represented less than 1 and 6% of the total variation, respectively (Appendix Table 19). For each 14 day increase in the rest period, there was an increase of 320 kg ha- 1 of available forage and for each unit (1.6 kg DM) decrease in grazing pressure, an increase of 570 kg ha- I of available forage was realized. The analysis of variance for the dry season of 1979 is presented in Appendix Table 19.







63



Available Forage for the Wet Season of 1980 (DM)


The average amount of available forage (DM) for each treatment combination during the wet season of 1980 is presented in Table 5.

The available forage varied from 340 kg DM ha- 1 to 7510 kg DM ha 1, corresponding to treatments: 26 [28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on offer/100 kg BW, and 400 kg ha-1 of superphosphate], and 31 (1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 400 kg ha- of superphosphate), respectively. Only the two experimental variables days rest (X 2) and grazing pressure (X 3) were found to be affecting the available forage (P< <0.01). The other two variables and the interactions were not significant.

The linear components of the model accounted for 57% of the

total variation, while the quadratic effects and interactions represented only 3 and 4% of the total variation, respectively (Appendix Table 20). The other experimental variables, X 1and X 4'had no significant effect upon the available forage.

As the rest period was increased by 14 days there was an

increase of 760 kg ha1 of available forage. On the other hand, each

unit (1.6 kg DM) decrease in grazing pressure resulted in an increase of 650 kg ha1 of available forage. The analysis of variance for the wet season of 1980 is presented in Appendix Table 20. Summary of Available Forage (DM)


Available forage may be considered a more important response variable than aerial biomass since available forage takes into consideration only the components of the vegetation considered edible






64



by the grazing animal. The available forage was estimated by determining the botanical composition of the total aerial biomass and summing the grass and legume components. Over the five seasons during which this study was conducted the available forage was greatly influenced by the environmental factors which are discussed in the aerial biomass section. The experimental variables, days rest (X 2) and grazing pressure (X 3), each had a highly significant effect upon the available forage. The response of available forage to these two experimental variables was essentially linear over the range which was studied. The longer rest period permitted an accumulation of reserves for pasture regrowth and the lower grazing pressure retained more of the aerial parts of both the grasses and legumes affecting mainly the growing points, axillary buds, and traditional leaf area for the production of photosynthate which stimulated subsequent regrowth. It is significant that these two variables seem to act independently as no interactions between them were found within the range of levels in this experiment.

Similarities in responses to the same lengths of rest period

and with the same grazing pressures were observed by Maraschin (1975) and Serrao (1976). In each of these two cases interactions were found between these two variables where maximum yields were obtained at high grazing pressures in combination with long rest periods and low grazing pressures with short rest periods. The above studies were conducted with Cynodon dactylon and Desmodium intortum which have quite different growth habits than the grasses and legumes included in this study. Cynodon dactylon is a stoloniferous grass with the







65


growing points close to the soil surface as compared with guineagrass and elephantgrass, each of which has apical meristems and lateral buds located much higher from the crown. The two grasses included in the current study are more sensitive to levels of defoliation and short rest periods because these grasses do not have rhizomes or stolons to store reserves. The legumes included in the current study are climbing and twining tropical legumes which no doubt are more sensitive to close defoliation than many other species.


Effect of Lengths of Rest Period and Levels of Grazing Pressure on Grass Yield (DM)


The effects of lengths of rest period and levels of grazing pressure on grass yield are presented in Table 6. The grass yield is that available before each grazing cycle and each 56 days for the continuous grazing treatments. Grass Yield (DM) for the Wet Season of 1978


During this first experimental season no significant effects were shown by any of the four variables (Xi. X 29 X 32 X 4 ) (Table 6 and Appendix Table 21).


Grass Yield (DM) for the Dry Season of 1978


The grass yield for the dry season of 1978 for each treatment combination is given in Table 6. The grass yield varied from 400 kg DM ha -1 tO 4550 kg DM ha -1 which was found on treatments 26 [28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on









66



0++
00 cn lz w w a Ln -It r-4 a% Ln 0 a CN .o 00 Ln 0 Ln Ln M -.T rm 0 m r- Ln m r r-q Ln r- a% 0 r- r- %0 (n cn r-q 00 r- r- 14, M
r-4 1: r-4 N M M M tn CN C14 M Ln C14 r--i ,O Cn r- cn Ln IT Ln %0





M++
r rn %0 Ln M r r- Ln -It 0 N Ln m %0 %.0 N m cq r- r-.4 r- m
Ln 0 M -.t Ln a% M -** r- c,4 M C14 C14 %.C Ln Ln -T M %0 Ln 0 Ln r, -4 r--l C,4 r-A -i Cn r-i C14 T r--4 C14 C14 -4 Ln m C14 -It cn (n -It





m ++
r En r- 0 m 00 %0 -4 r- M m m fl- r--q -T m Ln m 110 r-I 1.0 r-q N m IT
m -- %0 1,0 Ln 0 r- -T 0 T U Cn C14 M -,I r- r- r Ln a% Ln C% M Cn Ln M
,-4 1: 00 r-i C4 M M M 04 Zr Ln M Cn C?) CIJ r--i Ln Cn tn cn cn cn %0





00 ++ 000000000000000000000000
r- En Ln cy) ,c c cn T -zr r- r- cn r, Ln r, c,4 oo Zo %c -T %o cl -:T cD r, Ln
m -- %0 r-4 P-4 r-4 a% M M r, Cq 4 -i r- M C) Cf) .T r ,0 0 r %.0 ,O M Ln
14 C j cq CN (N en cn r--l eq -4 C14 m C14 M 1-4 cq C4 m It



0
u
00 ++ 0 CD C C) 0 0 0 0 C 0 0 0 C) 0 a 0 0 0 0 0 0 0 0 0
-W r rn m -14, T -It r-I m 110 -It --T Ln ON a m w M 0 M %0 r- 0 M M r,
M Ln M W -4 IT C4 M M 04 M -4* ,O ON r- 0 M W -ct -T 0 0 0 4 r--l r-4 r-4 P-4 r--i -4 -4 r-4 r-I r-4 r-I 4 4 r--l r-4 r--l r--q r--l C4







r-i r-4 r- r-4 r-4 r-q r--q r-q r--q r--i r-4 r--l r--4 -4 q
ca


0
r-q
cc
44
to r-.4 r-4 r.-I m m m m m m m cn
ca -14

:Bc
>1 m
8N o cn m cn % o 0 "o cn cn cn cn "o '.0 "o %0 "D z o z cn cn cn cn
U .
Cn M en cn cn 10 ID %0 %10 cn cn cn cn %0 %0 110 o I-q 1-4 -4 1--1 co 00 co 00
X


"a cu

"-q co 0 I T C*4 -:r -zt CN Cq T -,T C*4 -T T C14 0 ,C %0 0 0 C .O
>1 -4 IT -T 1-1 q 1-t IT 1-4 IT IT -4 r-q IT --T Ln Ln V") Ln



14

r-4 r r-q r q r r-4 r% -4 r r-i r -1 -4 W V-1 m -4 CO r--q W
C14 C14 C14 C*4 C14 C4 cq C14 C14 C14 CN





cc r-4 C14 m IT Ln 110 r- 00 m C) -i C14 CY) -It Ln \.c r- co m C) 1-4 C14 M -T
z -4 -4 -j r-4 r-f -4 r-4 -1 r-4 r-4 C 4 CN CN4 C 4








67




++ 00000000000000000
00 En C14 m r- (ON 110 w %o r--4 Ln 0 --T w 0 m w C7%
C% N 1-4 m 0 w 0 -It r--l C4 --T Ln 0 fl- 110 %.o 0 C'4
r-I 3: m r m r- %.0 N r-I -zr CJ C4 C14 04




++ 00000000000000000
r--i m IT r-I r- ND 00 C14 r- 1.0 00 0 r- CY) co -11 Cq M C> T M -,T IA f- CN 00 0 M N -,t a% C14 %0 CN CN Cq M Ln C4 r-4 r-4 M r-4 r-4 -4



r-4 I
a% ++ ca a 0 0 a 0 0 C) CD 0 0 C) 0 0 0 CD C) 0 r rn 4 ON IT cli ON 00 04 cn 00 00 Ln 00 %.0 M Ln r--4 co rm -- r- -1 00 IT 110 Ln rl- Ln C14 cn r- cli r- %C rl cn
r-A 1: 00 IT m Clq cli 110 Ln C%4 C11 -4 CY) -4 CA C14 C14 C14
S4
W
N
-H
r-q
00 ++ 0 0 C) 0 0 0 0 C) 0 C) 0 0 0 0 0 a 0 .1-1
rl W r- 0 -T Ln -It r-4 % o 00 cc% -It r- %0 r-A C> C14 00 $4
M -T 0 C Ln C14 r--4 ON IT r-q 00 0 C) r- --T fl- w
CN4 4 -T -.T r--l r-4 r-4 CN r-4 C%4 -4 r--i r-4 t"
11
pr4

00 ++ 00000000000000000 a;
M 00 q C14 110 C) IT m en 110 0 cn -T ON 1-1 r" r- p
-4 m tn r-I en IT Ln -4 00 Ln M r--i M r- It C14 :3 q -1 r-I r-q r-4 r--q -4 -4 r-i r--i r- r-i r-4 r--i -4 "-1 0
w





r-i r-I C 4 r-q r--l N Cq r- r-i 4 r-4 i A r-4 r-4 CY) N ca
w





0 0 CD 0 C) 0 C C) 0 C 0 C) 0 0 0 CD 0 CD 0 CD 0 CD 0 0 0 0 0 a C> CD CD cc
T " CN C4 Cq -t w
w 0
cc

44 ca
cn en m 0 0 0 0 %D M 0 C) C) a >1
. . . . . . 4
Cf) -1 1-1 r--i r-4 w w w W Ln Ln Ln Ln r-A W Ln Ln Ln 10

41 0. it
0 !:) cn
(U IFi ^
Q) CO P4 C-4 0 0 110 1.0 0 0 o 110 w m 0 1.0 w w w 00 00 0
0 w -- >< U*) Ln Ln Ln CN C*4 Ln CN C4 C4 Cq C14 -H
lz $4 N 0
-H E-4 0 w
4.J 4 co
0 w w
0
I-i r-I 00 r-4 00 r-4 00 r- 00 -4 00 %T -Z %T -T -,T -1 4.J
C14 C14 C14 C4 P--f -1 -4 4 r-i -I M Q)
IV




co Ln \.O r m 0 1-4 C14 cn IT Ln .o r- m 0 -4 rn 3:
E N C14 m m m m m m m m m cn t -T +- ++







68


offer/100 kg BW, and 400 kg ha-1 of superphosphate] and 24 (28 days grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 0 kg ha-I of superphosphate), respectively. Both rest period (X 2 ) and grazing pressure (X 3 ) had an effect upon the grass yield (P < 0.01). The application of superphosphate (X 4 ) also had a significant effect at the 10% level of significance. Each of these variables acted independently for grass yield since no interactions occurred.

The linear components of the model accounted for 74% of the total variation, while the quadratic effects and interactions represented only 2 and 1% of the total variation, respectively (Appendix Table 22).

An examination of the individual linear effects shows that

the experimental variable, days grazing (X 1 ),did not have any effect on the amount of grass produced. The greatest effects were obtained from days rest (X 2 ) and grazing pressure (X 3 ). As the length of rest period increased from zero days (continuous grazing) the amount of grass yield tends to increase, whereas, when the grazing pressure is decreased from 1.6 to 8.3 kg DM on offer/100 kg BW, the grass yield was increased. For each 14 days increase in rest period, there was an increase of 440 kg ha- I of grass yield. For each increment of decrease in the grazing pressure, there was an increase of 430 kg ha- 1 of grass yield. The analysis of variance is presented in Appendix Table 22.






69



Grass Yield (DM) for the Wet Season of 1979


The grass yield for the wet season of 1979 for all treatment combinations is presented in Table 6. During this season the grass yield varied from 560 kg DM ha-1 to 6530 kg DM ha -1 for treatment combinations 17 [1 day grazing, 0 days rest (continuous grazing),

1.6 kg DM on offer/100 kg BW, and 0 kg ha- 1 of superphosphate and 23 (1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 0 kg ha- 1 of superphosphate), respectively. Grass yield was again influenced by days rest (X 2 ) and grazing pressure (X 3 ) (P < 0.01), while the experimental variables days grazing and fertilizer level were not significant. There were no interactions between experimental variables during this season.

The linear components of the model accounted for 72% of the total variation, while the quadratic and interaction effects represented only 2 and 3%, respectively (Appendix Table 23). The grass yield was positively related to increasing lengths of rest period and negatively related to increasing amounts of forage on offer to the grazing animals. For each 14 days increase in the rest period, there was an increase of 800 kg ha- I of grass yield; on the other hand, for each unit (1.6 kg DM) decrease in grazing pressure, there was an increase of 430 kg of grass yield. The analysis of variance is presented in Appendix Table 23.


Grass Yield (DM) for the Dry Season of 1979


The grass yield for the dry season of 1979 for each treatment combination is presented in Table 6. Grass yield varied from 210







70


kg DM ha1 to 572 kg DM ha- for the following treatment combinations: 25 [1 day grazing, 0 days rest (continuous grazing), 1.6 kg DM on of fer/iQO kg BW, and 400 kg ha- of superphosphate] and 32 (28 days grazing, 56 days rest, 8.3 kg DM on offer/lOG kg BW, and 400 kg ha1 of superphosphate), respectively. Again the grass yield was increased by increasing the days of rest and decreasing the grazing pressure (P < 0.01). The variables X1and X4were not significant and no interactions were found between the experimental variables. The linear components of the model accounted for 57% of the total variation, while the quadratic and interaction effects represented only 2 and 5% of the total variation, respectively (Appendix Table 24).

For each 14 days increase in rest period, there was a corres-1
ponding increase of 420 kg ha of grass yield, while each unit (1.6 kg DM) decrease in grazing pressure increased the grass yield by 540 kg ha 1. The analysis of variance is presented in Appendix Table 24.


Grass Yield (DM) for the Wet Season of 1980


The grass yield (DM) for the wet season of 1980 for each treatment combination is presented in Table 6. Grass yield varied from 120 kg DM ha- to 7150 kg DM ha- for treatments 26 [28 days grazing,

0 days rest (continuous grazing), 1.6 kg DM on of fer/l00 kg BW, and 400 kg ha- of superphosphate] and 19 (1 day grazing, 56 days rest,

1.6 kg DM on of fer/l00 kg BW, and 0 kg ha- of superphosphate), respectively. Grass dry matter yield was again affected by increasing the number of days rest and decreasing the grazing pressure (P < 0.01).







71



The other two experimental variables, X 1 and X V were not significant and no interactions were found between the experimental variables. The linear components of the model accounted for 59% of the total variation, while the quadratic and interaction effects represented only 5 and 3% of the total variation, respectively (Appendix Table 25). An increase of one unit (14 days) of rest increased the grass dry matter yield by 870 kg ha- 1, while a decrease in the grazing pressure increased the grass dry matter yield by 610 kg ha-l. The analysis of variance is presented in Appendix Table 25. Summary of Grass Yield (DM)


During the first rainy season (1978), with less than 2 months duration from May 12 to June 30, the experimental variables had no effect upon grass yield.

Beginning with the dry season of 1978 through the last wet

season of 1980, the grass yield was positively related to increasing lengths of rest period. The two grasses, guineagrass and elephantgrass, used in this experiment are tall growing species with a high capacity for dry matter production and each of them are favored by long rest periods. As the length of the rest period was increased the dominance of each of these grasses was evident, leaving very little space for other species, such as legumes and weeds. The grass yield also increased with time as these two species of grasses became securely established. The mean grass yield for the wet season of 1978 was 1530 kg DM ha-l' while for the last wet season of 1980, the grass yield was 3630 kg DM ha-l. It is evident that the amount






72



of grass increased considerably from 1978 to 1980 (Figs. 6, 7, and 8), due mainly to its rapid growth capacity and ability to eliminate the other companion species. Vicente-Chandler (1975) reported that elephantgrass and guineagrass. are the most productive tropical grasses reaching values of 78 and 45 tons DM ha- year 1. INIAP (1980) reported that under a cutting system, the rate of growth for elephantgrass and guineagrass is in the order of 166 kg and 149 kg DM haday ,' respectively. These values corresponded to those obtained in this experiment during the wet season of 1979 but as the year progressed, these values decreased and reached the lowest rate at the end of the dry season of that particular year, being in the order of 32 and 28 kg DM ha1 day1 for elephantgrass and guineagrass, respectively.

Grazing pressure expressed as the amount of forage offered per 100 kg BW increased the grass yield as the grazing pressure was decreased. The lowest yields were obtained under the highest grazing pressure and the highest yields were obtained in the treatments with the lowest grazing pressure. If the two experimental variables, rest period and grazing pressure are considered in combination, then the lowest grass yields were obtained under continuous grazing and the highest grazing pressure, while the highest yields were observed on treatment combinations having the longest rest period with the lowest grazing pressure. As the grazing pressure was increased, the amount of plant shoots that were removed by the grazing animal also increased and when this was accompanied by an increase in frequency of defoliation (short rest periods or continuous grazing), the










73











cu 0

cn p
w w 41
W 44 w
$4 44 0. 0

to C: CD CD 0 I= 0
0 0 a CD cn -H
-4 co ull C14 N $4
cli r-4 00 ca 0
4
bn




Ln



>1

to

co

to

0


W

r-i Cn
m

$4



00

cz

41


C14 cu
co a
> 0
ca -H
4


4J O
En r
C)
,a C) CD C) C)
r-i r-q 00 C4 4-4 4-4
WH C14 r-f 0 0
-r4 I
>1 ca
u 0 co (U w
cc bd 44 ca
4-4 W
w









74
























cn -W 4-4
M 0
4
4-4
0



C> to -W
Ln PO



.0 4-J




14, CU $4
4-4 4-4 0
0 44 (M
ca rn

C) 0 ri)
CY) C) cc
U") bo
w
N
co
C)
C
00 -,A 4-1
cn >1 0

U) M cn $4 En
0. ca
M 4 >
M to

4-4
0
cl
ca co C) 4
z
0 0
Lr) 4-1 -H
C: 4 00 0 (3) aN m C14 co Q. -q
Q) Lr) C
CN



co
cz









75





:3:
pq



w CD


$4
(D
4.4
Ia. 44
0

-H 0
0 N 41
CD co
0 p 4
C4 co 0
to




tn cn

en Q)












Q)
4


cu

12,

00 bo
C14
"-4
N




$4
cq co
to
>1
co 0
Im -r-4



4J
Ul) m 00
Q) 0* C C) 0 C) p r--A
0 C 0
,a 0 4 4-4
C14 0 0


>1 co u 0
W cn ca t" ca
co bo 44 0)
U)






76



intensity of plant removal was the major determinant of grass yield. This was true for both guineagrass and elephanatgrass and the effect of severe defoliation was to reduce the vigor and the regrowth capacity of these two species. At this point, some bare soil areas were observed and weed invasion occurred. This was very evident under short rest periods when combined with high grazing pressure.

The animals had a tendency to selectively graze the grass species during the wet seasons and to selectively graze the legumes during the dry seasons. Similar results have been reported by Humphreys (1978) who observed a selective consumption of Panicum maximum in preference to Stylosanthes guianensis while in the dry season, S. guianensis was well-consumed.

It is well known that tall-growing grasses are very susceptible to a higher degree of defoliation, especially if these species are subjected to low cutting heights or high grazing pressures. Most of the leaves, growing points, and axillary buds are located much higher than on short-growing species. The two tall-growing species in this study have long internodes with the axillary buds located far apart on the stem so they are very vulnerable to intense defoliation.


Effects of Lengths of Rest Period and Levels of Grazing Pressure Upon Legume Yield (DM)


Since one of the main objectives of this research was to

determine the optimum combinations of the components of grazing management which would favor the legume component of the pasture and permit a higher order of persistence, the legume yield and the changes which occurred during the five seasons are of primary interest.





77




Legume Yield (DM) for the Wet Season of 1978


Legume yields for the wet season of 1978 for each treatment combination are presented in Table 7. Legume yield varied from 410 kg DM ha- 1 to 1380 kg DM ha-1 for treatments 2 (21 days grazing, 14 days rest, 3.3 kg DM on offer/100 kg BW, and 100 kg ha-1 of superphosphate) and 14 (21 days grazing, 14 days rest, 6.6 kg DM on offer/ 100 kg BW, and 300 kg ha-1 superphosphate), respectively.

Length of rest period was the only factor which appeared to have much effect upon legume yield (P < 0.01). The other three experimental variables, X1. X 30 and X 41' and all of the interactions were not significant. The linear components of the model accounted for 16% of the total variation, while the quadratic and interaction components represented only 4 and 8% of the total variation, respectively (Appendix Table 26). For each unit (14 days) increase in the rest period, there was an increase of only 60 kg of legume yield. The analysis of variance is presented in Appendix Table 26.

This lack of response to the experimental variables was expected since the wet season of 1978 was very short, so the experimental variables did not have time to create differences in response. Legume Yield (DM) for the Dry Season of 1978


Legume yield for each treatment combination during the dry season of 1978 is presented in Table 7. The legume yields varied from 380 kg DM ha -1 to 1160 kg DM ha -1 for treatments 28 (28 days grazing, 56 days rest, 1.6 kg DM on offer/100 kg BW, and 400 kg ha- 1 of superphosphate) and 24 (28 days grazing, 56 days rest, 8.3 kg DM on offer/








78





00 r- r-4 C'4 "0
ON -,I --T -i r-4 Ln r- N lzr aN Ul 1.0 -T M C*4 cn r-4






(n r-4 m r-q Ln -1 r-4 1-4 r-i 0 r- r--4 r- Ln r- Ln r- r- --T a's m r- -IT 0
ON tn cn r- en %.D ON r-4 cn r- r- M N rl .o IT M4 N m r-4 % o cn cn





0
r- En %0 r- Ln 00 C) %lo C\l C14 a a -T 0 C) r-4 Ln m r- %0 as as CN a% 00 Ln
m Ln in CN N m 10 N m Ln m M Ln 'T w m r-4 'T f--i I.T D m




0
00
r W N m m r-I rl -It w r m W Ln r- -.T -IT ON en It CV) -It IT Ln Cl) C*4 %D
m 1 0 LA Ln -zr 1.0 r" 10 00 Ln ID -11 %0 00 a% Ln I-T -T Ul) Ln 00 m -11 -1





co 0 0 0 0 0 a 0 C) CD CD 0 CD a CD C) CD 0 0 0 0 0 0 C 0
f-% U3 NT 1-4 NT 00 cq aN Ln r- r- 00 r-4 0 M 00 r--l :T Ln r-4 110 Ln r r-q Ln Cli
aN IT C14 Ln Ln CYN r-q %0 C14 f-- 0 110 00 m 00 C*4 %0 %0 00 00 00 r- -4 cn
co V-4 r-i r-A r-q

C;
0
w
ca w
w
r-4 r--l r-4 r-4 r-4 r-4 r-I r--i r-4 r-I i r-4 "4 r-4 r-i r-4 C14 C14 r-4 -4 N






ca
C) 0 0 C) C) 0 C CD 0 CD CD CD 0 CD 0 0 CD 0 0 0 CD CD CD CD CD CD 0 CD CD CD CD CD CD CD 0 0 CD CD CD 0 w r-q r-4 P-4 r--l -1 -1 r--l cn m on m CY) m m m

r-4
4
0
4-4 04 M m cn cn cn m Cl) 1.0 % o M m cn cn
1 1; C4 1 1 14 4 14 C C C C
41

S

T .T N C,4 .T -,T N CN .T -,t t -Z C-4 0 0 1 0 1.0 0 0 o 1 0
-1 r-I -.T It -1 -4 It -T r-I "i -a- -T -4 r-4 -T -It Ln Ln Ln Ln
ca


4
u 1-4 r r-4 r r-I r r-4 r 4 r r-4 r r--f r r-4 r rq r-4 W r-4 W -4 W r-4 W
C14 eq C14 C14 C%4 C14 C14 C%4




ca -1 Cq M -ZI Ln \.C r- 00 M C) Cq M ,T Ln D r- 00 0'% C -4 N Cl) -,I
p r--i -4 -4 -4 -1 -4 -4 4 -I C4 N N








79





00 W4 00000000 cnco 0000 C400000r4 c (2' -.0 '.00 '.0I 0 Ln -TLd






at ++ 00000000000000000 ON'- -I 1 .0 ~-4 T r-t '.o'.or- L,


I


a++C00000000 000000000




1:4

-4
44 0n L o00 -r- 0000n000000(n00 C1 -4n-I







00 ++ 00000000000000000 f.- (n r- 0 U0) ) 0M'0 M 0 cq' TC
a% 0 r-C r4 Ln0 m m MLn 14 C m -I U
-1 r-4 r- -4 --l -4 -4



r-4 -4 -1 4 C 4r-I q r-4r-fr-4r--4CV) -YN







00000000000000000
00000000000 00
w ca

fn (
1.0 C",cn cn n 0 m
. . ..Ua)



C44 CC 4 0 C ir4-4r4-



U") 1.0-' r-0 D -IC C 4 )10 r 0 : 3
-X C-4 C-4 C- N C'N CN-v-.~-4 en U)mc nm m m Z' j-+







80



100 kg BW, and 0 kg ha-1 of superphosphate), respectively. The legume yield was increased as the result of reducing the grazing pressure (P < 0.01). There was an interaction between days grazing (X 1 ) and grazing pressure (X 3 ) but this interaction did not occur again during the remaining three seasons. There appeared to be no linear effect for days grazing (X 1 ), days rest (X 2 ), and fertilizer level (X 4). The linear components of the model accounted for only 31% of the total variation, while the quadratic effects represented only 1% and the interactions 12% (Appendix Table 27).

During the dry season of 1978, there was a tendency for the legume yield to increase as the grazing pressures decreased. Legume Yield (DM) for the Wet Season of 1979


The legume yield for the wet season of 1979 for each treatment combination is presented in Table 7. The legume yield during the second wet season varied from 80 kg DM ha-l to 970 kg DM ha- 1 for treatments 23 (1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and o kg ha-I of superphosphate) and 34 (28 days grazing, 28 days rest, 5.0 kg DM on offer/100 kg BW, and 200 kg ha- I of superphosphate), respectively.

Only the length of rest period (X 2 ) and grazing pressure (X 3) were found to have an effect upon legume yield (P < 0.01). There were no effects of X 1 and X 4 nor any of the interactions between the experimental variables.

The linear components of the model accounted for 34% of the

total variation, while the quadratic and interaction effects represented







81


only 6 and 7% of the total variation, respectively (Appendix Table 28). During the 1979 wet season, the yield of legumes began to show a pattern associated with the length of the rest period. For each 14 days increase in rest period, there was a decrease of 56 kg ha -1 0 legume yield. Also, during this season, grazing pressure began to show an effect upon legume yield. For each unit (1.6 kg DM) decrease in grazing pressure, there was an increase of 65 kg ha- 1 of legume yield. The analysis of variance is presented in Appendix Table 28. Legume Yield (DM) for the Dry Season of 1979


The legume yield for the dry season of 1979 for each treatment combination is presented in Table 7. Legume yield varied from 0 kg DM ha-1 to 910 kg DM ha-I for treatments 36 (14 days grazing, 56 days rest, 5.0 kg DM on offer/100 kg BW, and 200 kg ha-1 of superphosphate) and 6 (21 days grazing, 14 days rest, 6.6 kg DM on offer/100 kg BW, and 100 kg ha-1 of superphosphate), respectively.

Legume yield was affected by length of rest period (X 2 ) (P < 0.01) and by grazing pressure (X 3 ) (P < 0.05). Days grazing and fertilizer levels had no significant effect and there were no interactions among the experimental variables. For length of rest period (X 2 ) there was both a linear and quadratic effect upon legume yield (P < 0.01). The linear and quadratic components of this experimental variable accounted for 43 and 27% of the total variation, while the interactions among the experimental variables accounted for only 3%, respectively (Appendix Table 29).






82



The quadratic effect of length of rest period (X 2 ) strongly

suggests that the range of rest periods included in this experiment may have adequately covered the point of maximum legume yield. The highest legume yields were obtained in the region of 14 to 28 days rest and the yields decreased when the rest periods were reduced to zero or increased up to 56 days. The effect of grazing pressure (X 3) appears to be almost linear. For each unit (1.6 kg DM) that grazing pressure was reduced from 1.6 to 8.3 kg DM on offer/100 kg BW, the amount of legume yield increased by 31 kg DM ha-l. The analysis of variance is presented in Appendix Table 29.


Legume Yield (DM) for the Wet Season of 1980


The legume yield for the wet season of 1980 for each species

combination is given in Table 7. Legume yields varied from 0 kg DM ha-l for treatments with the longest rest period (56 days) to 980 kg DM ha-l for treatment 30 [28 days grazing, 0 rest period (continuous grazing), 3.3 kg DM on offer/100 kg BW, and 400 kg ha-l of superphosphate].

There was both a linear and quadratic effect of length of rest period (X 2 ) upon legume yield (P < 0.01). There was a linear effect of grazing pressure (P < 0.05). The effects of days grazing and fertilizer level were not significant and neither were any of the interactions of the experimental variables.

The linear and quadratic effects accounted for 45 and 24% of

the total variation, respectively, while the interactions represented only 5% (Appendix Table 30).







83


The quadratic effect of rest period (X 2 ) gives a better representation of the legume yield which maximizes within the range of 14 and 28 days. As the length of the rest period is either decreased or increased, the legume yield is decreased.

As the grazing pressure decreases, the amount of legume yield increased by 37 kg ha- 1. The analysis of variance is presented in Appendix Table 30.


Observations and Summary of Legume Yield (DM)


Some very significant changes occurred in the legume population of these experimental pastures, some of which were recorded and subjected to analysis, whereas others were observations made over the five seasons of this trial. Drastic changes occurred as a result of the treatments imposed from the end of the wet season of 1978 to the end of the wet season of 1980. Some of these observations are recorded here.

When the experiment was initiated on May 12 of 1978, the

average legume yield for the period until June 30 was about 928 kg ha-1. During this initial period, only rest period showed any effect (P < 0.01) upon legume yield. It is difficult to find any satisfactory explanation for the effect of any experimental variable over this short period of time, but differences in the growth rate of grasses and legumes or the short-term effects of insects like red spider on centre could be a partial explanation (Fig. 9).

During the first dry season (July to December, 1978), the legume yield declined from 640 kg ha- I which indicated a linear response to








84












00 On~i CDC)e
1-H 4 1


0 04

1





c~bc


0 '-0

Lf~ w

w












cn


ci)



4

0))

0)~~U 00 ."
r-4>, (') 0L4 Q




)0

C C 4 0



co







85


grazing pressure resulting in various degrees of defoliation, selectivity, and effect of trampling. Defoliation and selectivity were probably the main determinants on the amount of both legumes. A small difference is also recorded in the behavior of these two species in that centre appeared to be more tolerant of heavy grazing pressure and the area covered by glycine increased as the grazing pressure decreased. During the dry season of 1979, it was also observed that the grazing animals were selectively consuming more legumes than during the wet season, probably because of the lower quality and reduced quantity of grass for that period. Humphreys (1978) and Norman (1970) observed similar preferences of animals grazing tropical grass-legume pastures.

During the wet season of 1979 (January-June), the average legume yield was 473 kg ha-1 which was considerably lower when compared with the predecing dry season (1978) which can be explained by the much greater rate of growth by both of these companion grasses and by their stronger competitive ability which shaded the legumes, especially in the treatment with long rest periods. During the 1979 wet season, some environmental factors such as humidity, temperature, light, and N played a very important role in the high growth rates of the tropical grasses. The dominance of the C 4 grass species began to manifest itself over the C 3 tropical legumes. Mott stated that "physiological differences between tropical grasses and legumes have important implications for legume-grass associations. Since their optima for light, temperature, and moisture differ, it is much more difficult to select compatible grasses and legumes in the tropics than among temperate species where the responses to






86




environmental factors are similar." (1981, p. 35-41). He also added that the viney growth habit of several genera of tropical legumes (Calopogonium, Centrosema, some species of Desmodium, Neonotonia, Macroptilium, and Pueraria); they confer an advantage over C 4 tropical grasses in that they are able to climb to the top of the canopy. Devising defoliation strategies that will maintain the regrowth potential of viney legumes is very important.

The advantage of short rest periods during the wet seasons in terms of legume population was evident, giving these species more opportunity to survive, persist, bloom, and produce some seed, even at heavy grazing pressures. Under high levels of defoliation, the competitive and shading effects from both of the grasses was greatly limited. Each of the tropical legumes bloomed profusely at the beginning of the dry season (July-August).

During the dry season of 1979 (July-December), the mean legume yield was 397 kg ha 1. During this second dry season a few pastures with the longest rest periods were almost 100% grass with no legumes nor weeds being observed. The decrease in legume yield may be partly explained by the excessive compititive ability of the grasses and also by the selective pressure on the legumes during this season. Pastures subjected to continuous grazing or short rest periods showed a greater proportion of legume dry matter in the total amount of available forage. Both legume species, centro and glycine, survived the effect of frequency and severity of defoliation when short rest periods were combined with high grazing pressures.




Full Text
Table 3. Modified central composite non-rotatable design with four experimental (X) variables,
at five levels each, 41 design points, 51 experimental units with their respective area.
Treatments
Reps
Size of
Exp. Unit
Totalt
Area
Required
per Treat.
No.
Pasture
No.
Days
Grazing
(xx)
Days
Rest
(x2)
Grazing
Pressure
(X3) % BW
Fertilizer
Level .
(X^) kg ha
1
18
7
14
3.3
100
1
750
750
2
28
21
14
3.3
100
1
1500
1500
3
44
7
42
3.3
100
1
500
500
4
23
21
42
3.3
100
1
750
750
5
20
7
14
6.6
100
1
1500
1500
6
50
21
14
6.6
100
1
2000
2000
7
24
7
42
6.6
100
1
1000
1000
8
33
21
42
6.6
100
1
1500
1500
9
8
7
14
3.3
300
1
750
750
10
30
21
14
3.3
300
1
1500
1500
11
4
7
42
3.3
300
1
500
500
12
37
21
42
3.3
300
1
750
750
13
16
7
14
6.6
300
1
1500
1500
14
43
21
14
6.6
300
1
2000
2000
15
26
7
42
6.6
300
1
1000
1000
16
15
21
42
6.6
300
1
1500
1500
17
34
1
0
1.6
0
1
2000
2000
18
31
28
0
1.6
0
1
2000
2000
19
25,36
1
56
1.6
0
2
500
500
20
6,13
28
56
1.6
0
2
750
750
21
35
1
0
8.3
0
1
4000
4000
22
22
28
0
8.3
0
1
4000
4000
23
32,45
1
56
8.3
0
2
500
1000
24
27,48
28
56
8.3
0
2
2000
2000


Table 33.
Analysis of variance, regression coefficients and probabilities for yield
of weeds (g DM/m ) for the wet season of 1979.
RESPONSE MEAN
ROOT MSE
R-SQUARE
COEF OF VARIATION
9. >1007
12 301 7
0. S'979539
1. 3096
REGRESSION
Df
TYPE I SB
R-SQUARE
F-RATIO
PROB
LINEAR
4
4241. 0931
0. 3493
6. 98
0. 0003
QUADRATIC
4
302. 4015
0. 0249
0. 50
0. 7373
CROSSPRODUCT
&
2131. 2005
0. 1756
2. 34
0. 0520
TOTAL REGRESS
14
6674.7031
0. 5498
3. 14
0. 0020
RESIDUAL
nr
SS
MEAN SQUARE
F-RATIO
PR on
LACK OF FIT
26
5359.0769
206. 1491
19. 400
0. OOOJ
PURE ERROR
JO
105. 0250
10. 5026
TOTAL ERROR
36
5465. 7027
151. 0251
PARAMETER
1)1-
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
10.5303
3. 1 163
3. 38
0. 0018
XI
1
-O. 34005697
1. 1761
-0. 30
0. 7690
X2
]
-5. 0020
1 1778
-4. 32
0. 0001
X3
J
-4.3070
1. 1761
-3. 66
0 0008
X4
]
-1.3353
1. 1761
- 1. 14
0. 2637
XI X1
J
1. 0024
1.9073
0. 95
0. 3509
X1*X2
J
0.46330995
0.64418459
0. 72
0. 4766
X2X2
J
-1. 0009
1.90/3
-0 99
0. 3306
X1#X3
1
0.66226250
0.61600659
1 07
0 2095
X2*X3
1
2. 2206
0 64418459
3. 46
0. 0014
X3#X3
1
1. 2541
1. 9073
0. 66
0. 5150
X1 X4
1
-0. 08508750
0.61608659
-0. 14
0. 8909
X2*X4
1
0.36385477
0.64418459
0. 56
0. 5757
X3*X4
1
0.14858750
0.61608659
0. 24
0. 8108
X4#X4
1
-0.97466604
1 9073
-0. 51
0 6125
FACTOR
DF
SB
MEAN SQUARE
F-RATIO
PROB
XI
5
393.6798
70.7360
0. 52
0. 7605
X2
it
4963. 6565
992. 7313
6. 54
0.0002
X3
it
3234. 5947
646 9189
4. 26
0 0030
X4
5
256. 0955
51. 2191
0. 34
0. 8869
-P'
00


Table 4.continued.
Treatments
No.
D/Gf
(x1)
D/Rf
(x2)
G/Pf
(X ) % BW
Ft -i
kg ha
Reps
1978
w/st
1978
D/St
1979
w/st
1979
D/St
1980
w/st
25
1
0
1.6
400
1
2030
1220
kg ha
1390
2130
3010
26
28
0
1.6
400
1
1900
790
1220
1090
2820
27
1
56
1.6
400
2
2350
2450
4280
2150
3550
28
28
56
1.6
400
2
2700
2630
4090
2560
4030
29
1
0
8.3
400
1
1690
2620
3160
2650
3680
30
28
0
8.3
400
1
2440
2520
3600
3150
4160
31
1
56
8.3
400
2
2470
4870
6870
3690
7540
32
28
56
8.3
400
2
3730
4880
6350
5830
6160
33
1
28
5.0
200
1
2570
2710
3090
2780
2960
34
28
28
5.0
200
1
1570
2470
3490
2900
3420
35
14
0
5.0
200
1
2690
1720
1820
1650
1830
36
14
56
5.0
200
1
3130
3760
4450
3320
4030
37
14
28
1.6
200
1
1840
1910
2180
3360
2190
38
14
28
8.3
200
1
2530
2900
3660
3210
3560
39
14
28
5.0
0
1
1860
2550
3230
2730
3190
40
14
28
5.0
400
1
1960
2110
2420
2810
2560
41
14
28
5.0
200
3
2360
2370
3000
2420
2900
fD/G = days grazing, D/R = days rest, G/P = grazing pressure as % body weight, F = fertilizer.
$W/S = wet season, D/S = dry season.


Table 13. Analysis of variance, regression coefficients and probabilities for aerial
biomass (g DM/m ) for the wet season of 1979.
RESPONSE MEAN
403. 2542
ROUT MSE
104. 5726
R-SQUARE
0.70350047
COEF OF VARIATION
0.25932180
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
085040
0. 6664
20. 23
0. 0001
QUADRATIC
4
30205. 4997
0. 0227
0. 69
0. 6033
CROSSPRODUCT
6
19204. 3230
0. 0145
0. 29
0. 9365
TOTAL REGRESS
14
934450
0. 7036
6. 10
0. 0001
RESIDUAL
DF
SB
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
186479
7172.2821
0. 346
0. 9856
PURE ERROR
10
207196
20719. 6250
TOTAL ERROR
36
393676
10935. 4329
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
345. 3054
26. 4479
13. 06
0. 0001
XI
1
-6.0460
9. 9815
-0. 61
0. 5485
X2
1
69. 6999
9. 9955
6. 97
0. 0001
X3
1
46. 086B
9. 9815
4 62
0. 0001
X4
1
-5. 0009
9. 9815
-0. 50
0. 5642
X1*X1
1
6. 9550
16. 1868
0. 43
0. 6700
X1*X2
1
-5. 9221
5. 4671
-1. 08
0. 2859
X2*X2
1
2. 984 1
16. 1860
0. 18
0 8548
X1*X3
1
0.32452500
5. 2286
0. 06
0. 9509
X2#X3
1
0.06867165
5. 4671
0. 01
0. 9900
X3*X3
1
-2. 3971
16. 1868
-0. 15
0. 8831
XI *X4
1
3. 0579
5. 2286
0. 58
0. 5623
X2*X4
1
1. 1317
5. 4671
0. 21
0. 8372
X 3 X 4
1
2. 3022
5. 2286
0. 44
0. 6623
X4#X4
1
7. 7929
16. 1860
0. 48
0. 6331
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
28799. 4146
5759. BB29
0. 53
0. 7545
X2
5
546074
109375
10. 00
0. 0001
X3
5
257608
51537. 6265
4. 71
0. 0021
X4
5
12114. 9188
2422. 9838
0. 22
0. 9509
128


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
9. Visual estimation of dry matter grass percent for year, season, and treatment combination
Treatments
1978
1978
1979
1979
1980
D/Gf
(x1)
D/Rf G/Pf
(X2) (X3) % BW
Ft -i
kg ha
Reps
w/st
D/St
w/st
D/St
w/st
7,
7
14
3.3
100
1
58
69
53
53
45
21
14
3.3
100
1
80
78
78
81
77
7
42
3.3
100
1
50
77
90
70
93
21
42
3.3
100
1
70
80
87
82
95
7
14
6.6
100
1
76
78
76
79
75
21
14
6.6
100
1
50
68
72
68
69
7
42
6.6
100
1
51
81
91
94
98
21
42
6.6
100
1
66
79
88
91
90
7
14
3.3
300
1
63
64
72
65
67
21
14
3.3
300
1
65
68
73
67
61
7
42
3.3
300
1
60
76
86
84
95
21
42
3.3
300
1
50
80
87
89
98
7
14
6.6
300
1
63
72
80
69
74
21
14
6.6
300
1
50
63
63
66
69
7
42
6.6
300
1
50
68
91
89
88
21
42
6.6
300
1
55
75
92
90
100
1
0
1.6
0
1
56
59
46
27
16
28
0
1.6
0
1
67
57
49
30
26
1
56
1.6
0
2
69
82
95
97
100
28
56
1.6
0
2
63
77
91
90
100
1
0
8.3
0
1
63
69
81
82
88
28
0
8.3
0
1
60
66
81
88
96
1
56
8.3
0
2
70
89
96
98
99
28
56
8.3
0
2
54
75
78
90
98
104


54
with the highest grazing pressure, but both experimental variables
acted independently from each other.
The linear components of the model accounted for 45% of the
total variation, while the quadratic effects and interactions repre
sented only 1 and 6% of the total variation, respectively (Appendix
Table 14).
An examination of the individual linear effects showed that only
the experimental variables, days rest (X2) and grazing pressure (X^)
were affecting biomass production (P < 0.01). The number of days
grazing (X^) and fertilizer (X^) had no effect upon biomass production.
For each 14 days increase in the number of days rest there was a biomass
production increase of 230 kg (DM) ha \ while for each unit (1.6 kg DM)
decrease in level of grazing pressure there was an increase of 480 kg
ha ^ in biomass production. The analysis of variance for the dry season
of 1979 is presented in Appendix Table 14.
Biomass Production (DM) for the Wet Season of 1980
The lengths of rest period and levels of grazing pressure had a
direct effect upon biomass production for the wet season of 1980 (Table
4). There was no interaction between these two experimental variables
during this wet season.
The linear components of the model accounted for 43% of the total
variation, while the quadratic effects and interactions represented
only 8 and 1% of the total variation, respectively (Appendix Table 15).
The total biomass production varied from 1460 to 7540 kg DM ha ^ for
treatments 18 [28 days grazing, 0 days rest (continuous grazing), 1.6
kg DM on offer/100 kg BW, and 0 kg ha ^ of superphosphate] and 31


27
28
29
30
31
32
33
34
35
36
37
38
39
PAGE
142
143
144
145
146
147
148
149
150
151
152
153
154
Analysis of variance, regression coefficients and
probabilities for legume yield (g DM/m ) for the
dry season of 1978
Analysis of variance, regression coefficients and
probabilities for legume yield (g DM/m2) for the
wet season of 1979
Analysis of variance, regression coefficients and
probabilities for legume yield (g DM/m^) for the
dry season of 1979
Analysis of variance, regression coefficients and
probabilities for legume yield (g DM/m2) for the
wet season of 1980
Analysis of variance, regression coefficients and
probabilities for yield of weeds (g DM/m^) for the
wet season of 1978
Analysis of variance, regression coefficients and
probabilities for yield of weeds (g DM/m^) for the
dry season of 1978
Analysis of variance, regression coefficients and
probabilities for yield of weeds (g DM/m ) for the
wet season of 1979
Analysis of variance, regression coefficients and
probabilities for yield of weeds (g DM/m^) for the
dry season of 1979
Analysis of variance, regression coefficients and
probabilities for yield of weeds (g DM/m ) for the
wet season of 1980
Analysis of variance, regression coefficients and
probabilities for visual estimation grass (%) for
the wet season of 1978
Analysis of variance, regression coefficients and
probabilities for visual estimation grass (%) for
the dry season of 1978
Analysis of variance, regression coefficients and
probabilities for visual estimation grass (%) for
the wet season of 1979
Analysis of variance, regression coefficients and
probabilities for visual estimation grass (%) for
the dry season of 1979
ix


169
Vicente-Chandler, J. 1975. Intensive management of pastures and
forages in Puerto Rico. In E. Bornemisza and A. Alvarado (eds.)
Soil management in tropical America. North Carolina State Univ
ersity, Raleigh, North Carolina. P. 409-433.
Vilela, L. A. 1979. Efeito de taxa de lotacoo e de alimentacao
suplementar sobre a producao de leite durante o periodo de seca.
Revista de la Sociedade Brasileira de Zootechnia. 8:681-685.
Villasmil, I. J. J. Atencio, J. Barcenas, A. Casanova, M. Urdaneta,
and D. H. Timm. 1975. Response surface designs and linear
programming applied to cattle-forage-feeding systems in the
tropics. In E. C. Doll and G. 0. Mott (eds.) Tropical forages
in livestock production systems. ASA Special Publ. No. 24. P.
53-70.
Wells, T. C. 1967. Changes in the botanical composition of a sown
pasture on a chalk in Kent 1956-1964. J. Br. Grassld. Soc.
22:277-281.
Werner, J. C. 1979. Response of two species of Stylosanthes sp. to
levels of lime, phosphorus, potassium and boron on three mineral
soils. Ph.D. Dissertation, University of Florida, Gainesville,
Florida. 205 p.
, F. A. Monteiro, and H. B. Mattos. 1975. The use of trace
elements as FTE on forage tropical legumes. Br. Industr. Anim.
32:347-361.
Whiteman, P. C. 1969. Seasonal changes in growth and nodulation of
perennial tropical pasture legume in the field. II. Effects of
controlled defoliation levels on nodulation of Desmodium intortum
and Phaseolus atropurpureum. Aust. J. Agrie. Res. 21:207-214.
Zapata, C. 1981. Respuesta de cuatro pasturas tropicales en termino
de ganancia de peso vivo en tres grupos raciales de bovinos.
Tesis Ingeniero Agronomo. Universidad de Guayaquil, Guayaquil,
Guayas, Ecuador. 34 p.


51
The biomass production varied from 1570 kg DM ha for treatment
17 (1 day grazing, 0 days rest, 1.6 kg DM on offer/100 kg BW, and
0 kg ha ^ superphosphate) to 3750 kg DM ha ^ for treatment 19 (1 day
grazing, 56 days rest, 1.6 kg DM on offer/100 kg BW, and 0 kg ha ^
superphosphate).
The aerial biomass production response to lengths of rest periods
can only be explained by the management imposed during the first grazing
season. All of the experimental pastures were grazed at a medium gra
zing pressure for 15 days during the latter part of March of 1978, after
which they were mowed at about 15 cm above ground level. The wet
season of 1978 was represented by a growth period of about 3 months
from the first of April until June 30 when the wet season ended.
The production data were obtained during this 3 month period beginning
on May 12 after only about a month of regrowth. The pastures with
short rest periods were included in the first sampling date and for
some of these pastures the length of the grazing cycle was only 21
days which means that they had the opportunity to be sampled twice
between May 12 and June 30. Pastures with a longer rest period
accumulated more dry matter during the growth periods and, thus,
showed a higher amount of dry matter produced. The analysis of
variance for the wet season of 1978 is presented in Appendix Table 11.
Biomass Production (DM) for the Dry Season of 1978
The biomass production during the dry season of 1978 is also
presented in Table 4. The biomass production varied from 790 to
4880 kg DM ha corresponding to treatments 26 (28 days grazing,



PAGE 1

5(63216( 2) $ 7523,&$/ /(*80(*5$66 $662&,$7,21 72 6<67(06 2) *5$=,1* 0$1$*(0(17 $1' /(9(/6 2) 3+263+2586 )(57,/,=$7,21 %< 5$8/ $/2162 6$17,//$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

'HGLFDWHG WR P\ ZLIH 0DJJLH P\ GDXJKWHUV $OH[DQGUD DQG &DUROLQD P\ IDWKHU DQG WR WKH PHPRU\ RI P\ PRWKHU

PAGE 3

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nV ,QVWLWXWR 1DFLRQDO GH ,QYHVn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

PAGE 4

7KH DXWKRU LV LQGHEWHG WR JUDGXDWH VWXGHQWV 0U /XLV 5 5RGULJXH] DQG KLV ZLIH 7HUHVLQKD 5RGULJXH] IRU WKHLU IULHQGVKLS DQG KRVSLWDOLW\ 7KDQNV DUH DOVR GXH WR JUDGXDWH VWXGHQW -XDQ +HUEDV DQG KLV ZLIH IRU WKHLU IULHQGVKLS 7KH DXWKRU ZLVKHV WR WKDQN 0UV 3DW )UHQFK IRU W\SLQJ WKH ILQDO FRS\ RI WKLV PDQXVFULSW 6SHFLDO DFNQRZOHGJPHQW LV GXH WR WKH DXWKRUfV ZLIH 0DJJLH DQG WKHLU WZR GDXJKWHUV $OH[DQGUD DQG &DUROLQD IRU WKHLU LQYDOXDEOH KHOS ORYH DQG GHYRWLRQ DQG ILQDOO\ WR P\ SDUHQWV DQG GHDU VLVWHUV &HFLOLD DQG /XF\ IRU WKHLU ORYH DQG ILQDQFLDO KHOS LY

PAGE 5

7$%/( 2) &217(176 3$*( $&.12:/('*0(176 LLL /,67 2) 7$%/(6 YLL /,67 2) ),*85(6 [L $%675$&7 [LLL &+$37(5 ,1752'8&7,21 &+$37(5 ,, /,7(5$785( 5(9,(: 7KH 7URSLFDO )RUDJHV 3HUVLVWHQFH RI 7URSLFDO /HJXPH*UDVV $VVRFLDn WLRQV 5HVSRQVH WR 1XWULHQWV $QLPDO 3URGXFWLYLW\ 3DVWXUH (YDOXDWLRQ *UD]LQJ 6\VWHPV (VWLPDWHV RI 'U\ 0DWWHU 3URGXFWLRQ DQG
PAGE 6

3$*( (IIHFW RI /HQJWKV RI 5HVW 3HULRG DQG /HYHOV RI *UD]LQJ 3UHVVXUH 8SRQ /HJXPH
PAGE 7

/,67 2) 7$%/(6 3$*( 6RLO DQDO\VLV RI H[SHULPHQWDO VLWH f 0RGLILHG FHQWUDO FRPSRVLWH QRQURWDWDEOH GHVLJQ ZLWK IRXU H[SHULPHQWDO ;f YDULDEOHV DW ILYH OHYHOV HDFK DQG GHVLJQ SRLQWV 0RGLILHG FHQWUDO FRPSRVLWH QRQURWDWDEOH GHVLJQ ZLWK IRXU H[SHULPHQWDO ;f YDULDEOHV DW ILYH OHYHOV HDFK GHVLJQ SRLQWV H[SHULPHQWDO XQLWV ZLWK WKHLU UHVSHFWLYH DUHD $HULDO ELRPDVV SURGXFWLRQ '0f E\ \HDU VHDVRQ DQG WUHDWPHQW $YDLODEOH IRUDJH '0f E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ *UDVV \LHOGV '0f E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQn /HJXPH \LHOGV '0f E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ
PAGE 8

3$*( $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0Pf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0Q7f IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0PAf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0PAf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0P f IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0Pf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0PAf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0Pf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0PAf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0Pf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0PAf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0Pf IRU WKH ZHW VHDVRQ RI YLLL

PAGE 9

3$*( $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0P f IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0Pf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0PAf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0Pf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0PAf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0PAf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0P f IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0PAf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0P f IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV bf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV bf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV bf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV bf IRU WKH GU\ VHDVRQ RI L[

PAGE 10

7$%/( 3$*( $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV bf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH bf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH bf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH bf IRU WKH ZHW VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH bf IRU WKH GU\ VHDVRQ RI $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH bf IRU WKH ZHW VHDVRQ RI [

PAGE 11

/,67 2) ),*85(6 ),*85( 3$*( 3URILOH RI WKUHH QDWXUDO UHJLRQV RI (FXDGRU VKRZLQJ WKH PDLQ YHJHWDWLYH ]RQHV LQ UHODWLRQ WR UDLQIDOO DQG WHPSHUDWXUH 5DLQIDOO UHFRUGHG DW (VWDFLQ ([SHULPHQWDO 3LFKL OLQJXH GXULQJ WKH SHULRG 7HPSHUDWXUH UHFRUGHG DW (VWDFLQ ([SHULPHQWDO 3LFKL OLQJXH GXULQJ WKH SHULRG 6RODU UDGLDWLRQ UHFRUGHG DW (VWDFLQ ([SHULPHQWDO 3LFKLOLQJXH GXULQJ WKH SHULRG )LHOG SODQ RI WKH H[SHULPHQWDO SDVWXUHV (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV \LHOG '0f IRU WKH ZHW VHDVRQ RI &RQWRXUV RI JUDVV \LHOG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV \LHOG '0f IRU WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ OHJXPH \LHOG '0f IRU WKH ZHW VHDVRQ RI &RQWRXUV RI OHJXPH \LHOG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ OHJXPH \LHOG '0f IRU WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ \LHOG RI ZHHG '0f IRU WKH ZHW VHDVRQ RI &RQWRXUV RI \LHOG RI ZHHG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ \LHOG RI ZHHG '0f IRU WKH ZHW VHDVRQ RI (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV SHUFHQWDJH IRU WKH ZHW VHDVRQ RI [L

PAGE 12

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

PAGE 13

$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\ 5(63216( 2) $ 7523,&$/ /(*80(*5$66 $662&,$7,21 72 6<67(06 2) *5$=,1* 0$1$*(0(17 $1' /(9(/6 2) 3+263+2586 )(57,/,=$7,21 %\ 5$8/ $/2162 6$17,//$1 $SULO &KDLUPDQ 'U 0RWW 0DMRU 'HSDUWPHQW $JURQRP\ $ OHJXPHJUDVV SDVWXUH FRPSRVHG RI JO\FLQH >1HRQRWRQLD ZLJKWLL 5 *UDK H[ :LJKWLL DQG $Pf /DFNH\@ FHQWUR &HQWURVHPD SXEHVFHQV %HQWKf JXLQHDJUDVV 3DQLFXP PD[LPXP -DFTf DQG HOHSKDQWJUDVV 3HQQLVHWXP SXUSXUHXP 6FKXPDFKf ZDV HYDOXDWHG LQ D JUD]LQJ WULDO IURP 0D\ WR -XQH DW WKH (VWDFLQ ([SHULPHQWDO 7URSLFDO 3LFKLOLQJXH ,QVWLWXWR 1DFLRQDO GH ,QYHVWLJDFLRQHV $JURSHFXDULDV ,1,$3f 4XHYHGR (FXDGRU 7KH PDLQ REMHFWLYHV RI WKH VWXG\ ZHUH Df WR GHWHUPLQH WKH HIIHFWV RI OHQJWK RI JUD]LQJ SHULRG ;Af OHQJWK RI UHVW SHULRG ;Af JUD]LQJ SUHVVXUH ;Af DQG OHYHOV RI 3 IHUWLOL]DWLRQ ;Af XSRQ WKH SDVWXUH PL[WXUH Ef WR GHWHUPLQH WKH SURSHU JUD]LQJ PDQDJHPHQW WR DWWDLQ WKH RSWLPXP OHJXPH FRQWULEXWLRQ DQG Ff WR PHDVXUH WKH SDVWXUH UHVSRQVH LQ WHUPV RI GU\ PDWWHU SURGXFWLRQ DQG ERWDQLFDO FRPSRVLWLRQ *UD]LQJ SHULRGV VWXGLHG ZHUH DQG GD\V UHVW SHULRGV ZHUH DQG GD\V JUD]LQJ SUHVVXUHV ZHUH DQG NJ '0 RQ RIIHU NJ ERG\ ZHLJKW DQG [LLL

PAGE 14

OHYHOV RI IHUWLOL]HU ZHUH DQG NJ KD A RI VXSHUSKRVSKDWH 7R FRYHU WKH ILYH OHYHOV RI WKH FRPSOHWH IDFWRULDO f D PRGLILHG QRQURWDWDEOH FHQWUDO FRPSRVLWH GHVLJQ PDGH XS RI WUHDWPHQW FRPELQDWLRQV ZDV XVHG 7KH UHVSRQVH YDULDEOHV LQFOXGHG DHULDO ELRPDVV '0f DYDLODEOH IRUDJH '0f JUDVV \LHOG '0f OHJXPH \LHOG '0f \LHOG RI ZHHGV '0f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

PAGE 15

&+$37(5 ,1752'8&7,21 (FXDGRU ZLWK DQ DUHD RI NP LV ORFDWHG LQ WKH QRUWKn ZHVWHUQ SDUW RI WKH 6RXWK $PHULFDQ FRQWLQHQW 7KH GRPLQDQW WRSRn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f RU HOHSKDQWJUDVV 3HQQLVHWXP SXUSXUHXP 6FKXPDFKf ZLWK RWKHU VSHFLHV PDNLQJ XS D YHU\ VPDOO SRUWLRQ RI WKH WRWDO KHFWDUDJH $W SUHVHQW WKHUH LV D JURZLQJ LQWHUHVW LQ WKH HVWDEOLVKPHQW DQG XWLOL]DWLRQ RI WURSLFDO JUDVVOHJXPH PL[WXUHV IRU DQLPDO SURGXFn WLRQ 7KHUH DUH PDQ\ DGYDQWDJHV WR KDYLQJ OHJXPH FRPSRQHQWV LQ WKH

PAGE 16

SDVWXUH 7KH\ LQFUHDVH RU PDLQWDLQ VRLO IHUWLOLW\ GXH WR WKHLU DELOLW\ WR IL[ 1 LPSURYH WKH TXDOLW\ RI WKH GLHW JUD]HG E\ WKH DQLPDOV SURYLGH EHWWHU VHDVRQDO GLVWULEXWLRQ RI WKH IRUDJH WKURXJKn RXW WKH \HDU HVSHFLDOO\ ZKHUH DOWHUQDWH ZHWGU\ VHDVRQV RFFXU 7KHUHIRUH D V\VWHP ZKLFK FRPELQHV DGHTXDWH JUD]LQJ PDQDJHPHQW SUDFWLFHV ZLWK KLJK\LHOGLQJ JUDVVHV DQG OHJXPHV JURZLQJ LQ PL[WXUHV ZKLFK SURYLGH IHHG GXULQJ WKH ZKROH \HDU LV QHHGHG WR LQFUHDVH FDUU\LQJ FDSDFLW\ DQG WKH ILQDO RXWSXW RI WKH ODQG 7KH REMHFWLYHV RI WKLV UHVHDUFK ZHUH f 7R GHWHUPLQH WKH UHVSRQVH RI D WURSLFDO OHJXPHJUDVV PL[WXUH WR YDULRXV WUHDWPHQW FRPELQDWLRQV RI JUD]LQJ PDQDJHPHQW IDFWRUV OHQJWK RI JUD]LQJ SHULRG OHQJWK RI UHVW SHULRG JUD]LQJ SUHVVXUH DQG OHYHOV RI SKRVSKRUXV IHUWLOL]DWLRQ f 7R GHWHUPLQH WKH SURSHU PDQDJHPHQW VWUDWHJ\ IRU DGDSWHG IRUDJH VSHFLHV WR DWWDLQ WKH RSWLPXP OHJXPH FRQWULEXWLRQ DV D FRPSRQHQW LQ WKH PL[WXUH DQG f 7R PHDVXUH WKH UHVXOWV RI OHJXPHJUDVV PL[WXUH LQ WHUPV RI DHULDO ELRPDVV DYDLODEOH IRUDJH DQG ERWDQLFDO FRPSRn VLWLRQ RI WKH IRUDJH RQ RIIHU

PAGE 17

&+$37(5 ,, /,7(5$785( 5(9,(: ,Q WURSLFDO UHJLRQV DQLPDO SURGXFWLRQ RQ SDVWXUHV LV GHSHQGHQW XSRQ WKH TXDOLW\ DQG TXDQWLW\ RI IRUDJH DYDLODEOH WKURXJKRXW WKH \HDU ,Q WKH KXPLG WURSLFV WKH VHDVRQDO JURZWK RI SDVWXUHV LV JUHDWO\ LQIOXHQFHG E\ WKH ZHW DQG GU\ SHULRGV $W WKH EHJLQQLQJ RI WKH ZHW VHDVRQ YHU\ UDSLG JURZWK RFFXUV DQG IUHTXHQWO\ D ODUJH DPRXQW RI IRUDJH DFFXPXODWHV $V WKH VHDVRQ SURJUHVVHV WKH UDWH RI JURZWK GHFUHDVHV UDSLGO\ DQG DSSURDFKHV ]HUR GXULQJ WKH GU\ VHDVRQ 7KH FRQFHQWUDWLRQ RI QLWURJHQ DQG PLQHUDOV LV UHODWLYHO\ KLJK DW WKH EHJLQQLQJ RI WKH ZHW VHDVRQ DQG DOVR IDOOV UDSLGO\ DV WKH VHDVRQ SURJUHVVHV ,Q DGGLWLRQ WKHUH LV D SURJUHVVLYH LQFUHDVH LQ ILEHU DQG GHFUHDVH LQ GLJHVWLELOLW\ IURP WKH EHJLQQLQJ RI WKH ZHW VHDVRQ WKURXJK WKH UHPDLQGHU RI WKH \HDU 3DODGLQHV DQG GH $OED f 6RPH WURSLFDO SDVWXUH VSHFLHV VHHP WR EH ZHOO DGDSWHG WR H[WUHPH HQYLURQPHQWDO FRQGLWLRQV KRZHYHU WKHLU SRWHQWLDO WR VXSSO\ IHHG IRU FDWWOH SURGXFWLRQ PD\ EH OLPLWHG QRW RQO\ E\ VHDVRQDO FKDQJHV EXW DOVR E\ VRLO IHUWLOLW\ 7HUJDV f PHFKDQLVPV RI DGDSWDWLRQ *DUWQHU HW DO f DQG JUD]LQJ PDQDJHPHQW 0RWW (YDQV 6WREEV f 7KH 7URSLFDO )RUDJHV 3DQLFXP PD[LPXP -DFT KDV EHHQ GHVFULEHG E\ +XPSKUH\V f DQG %RJGDQ f DV D GHQVHO\ WXIWHG SHUHQQLDO JUDVV ZLWK UHODWLYH GURXJKW DQG SRRU VRLO WROHUDQFH ,W LV RI KLJK QXWULWLYH YDOXH ZKHQ \RXQJ DQG FRPELQHV ZHOO ZLWK RWKHU WURSLFDO SDVWXUH VSHFLHV +XGJHQV

PAGE 18

&KDYH] 5RODQGR f ,W LV LQGLJHQRXV WR WURSLFDO $IULFD ZKHUH LW LV GRPLQDQW RYHU ODUJH DUHDV SDUWLFXODUO\ XQGHU KXPLG DQG VXEKXPLG FRQGLWLRQV ,Q WKH FRDVWDO DUHD RI (FXDGRU LW LV NQRZQ DV JXLQHD FDXFD VDER\D FKLOHQD %HWDQFRXUW f LQGLFDWHG WKDW JXLQHDJUDVV ZDV WKH PRVW ZLGHVSUHDG SDVWXUH JUDVV LQ WKH ORZODQGV RI (FXDGRU $FRVWD6ROL] f UHSRUWHG WKDW WKLV JUDVV JURZV IURP VHD OHYHO XS WR P *XLQHDJUDVV LV YHU\ SRSXODU EHFDXVH RI LWV DGDSWDWLRQ WR WKH GU\ VHDVRQ UHVLVWDQFH WR ILUH KLJK SURGXFWLRQ RI IRUDJH FDSDELOLW\ RI HVWDEOLVKPHQW E\ VHHG RU E\ GLYLVLRQ RI SODQW FURZQV DQG ILQDOO\ GXH WR LWV DELOLW\ WR SHUVLVW XQGHU KHDY\ XVH DQG DEXVH ,1,$3 f 5HFHQWO\ WKH ,QVWLWXWR 1DFLRQDO GH ,QYHVWLJDFLRQHV $JURSHFXDULDV ,1,$3f RI (FXDGRU KDV WHVWHG LQWURGXFWLRQV RI JXLQHDJUDVV LQ VPDOO JUD]LQJ WULDOV PDLQO\ IRU DGDSWDWLRQ DQG SHUVLVWHQFH 1HZ FXOWLYDUV ZLOO EH UHOHDVHG WR WKH IDUPHUV DV VRRQ DV VXIILFLHQW VHHG LV DYDLODEOH IRU ODUJH VFDOH XVH ,1,$3 f 3HQQLVHWXP SXUSXUHXP 6FKXPDFK HOHSKDQWJUDVV LV D WDOOJURZLQJ VSHFLHV XS WR P LQ KHLJKW ,W KDV EHHQ GHVFULEHG E\ %RJGDQ f &RUUHD f DQG 0FOOUR\ f $FRVWD6ROL] f DQG ,1,$3 f VWDWHG WKDW HOHSKDQWJUDVV JURZV IURP VHD OHYHO XS WR P LQ WKH ZDUP YDOOH\V RI WKH 6LHUUD UHJLRQ ,Q (FXDGRU LW LV PDLQO\ XVHG DV D SDVWXUH JUDVV ZLWK VRPH IDUPHUV UHSRUWLQJ SDVWXUHV XS WR \HDUV ROG VWLOO XQGHU JUD]LQJ FRQGLWLRQV 8QGHU FXWWLQJ IUHTXHQFLHV RI GD\V LW SURGXFHV XS WR WRQV '0 KD A \U A ZKHQ UHFHLYLQJ NJ RI 1 DQG LUULJDWLRQ GXULQJ WKH GU\ VHDVRQ ,1,$3 f 7KHUH DUH WKUHH FXOWLYDUV ZKLFK DUH

PAGE 19

LQ ZLGHVSUHDG XVH LQ WKH ORZODQGV RI (FXDGRU 7KH\ DUH nFRPPRQn IRU JUD]LQJ n+\EULG n DQG n0H[LFDQn IRU FXWWLQJ &HQWURVHPD SXEHVFHQV %HQWK FHQWUR D WUXH WURSLFDO OHJXPH KDV EHHQ GHVFULEHG E\ +XPSKUH\V f DV D FUHHSLQJ WZLQLQJ SHUHQQLDO OHJXPH QDWLYH RI 6RXWK $PHULFD *URI f UHSRUWHG WKDW WKH JHQXV &HQWURVHPD FRQWDLQV DERXW VSHFLHV JURZLQJ QDWXUDOO\ LQ WKH WURSLFDO DUHDV RI &HQWUDO DQG 6RXWK $PHULFD 0RRUH f VWDWHG WKDW WKLV WURSLFDO OHJXPH JURZV ZHOO LQ KXPLG DUHDV DQG PXVW EH FRQVLGHUHG DV D EDVLF FRPSRQHQW RI SDVWXUHV XQGHU WKHVH FRQGLWLRQV 6WXGLHV FRQGXFWHG DW 3LFKLOLQJXH (FXDGRU E\ +XGJHQV f &KDYH] f 5RODQGR f DQG ,1,$3 f LQGLFDWHG WKDW WKLV OHJXPH SHUIRUPV ZHOO LQ DVVRFLDn WLRQ ZLWK JXLQHDJUDVV DQG LW LV DOVR KLJKO\ SHUVLVWHQW XQGHU JUD]LQJ DQG SURGXFHV ODUJH DPRXQWV RI VHHG )DUIDQ f ,Q WKH ODVW IHZ \HDUV QDWLYH HFRW\SHV RI &HQWURVHPD KDYH EHHQ VHOHFWHG DQG WHVWHG LQ (FXDGRU $ IHZ RI WKHVH KDYH EHHQ GLVWULEXWHG WR IDUPHUV DQG DUH VKRZLQJ VRPH DGYDQWDJHV RYHU WKH $XVWUDOLDQ FRPPHUFLDO FXOWLYDUV VXFK DV KLJKHU '0 \LHOG EHWWHU WROHUDQFH WR LQVHFWV DQG GLVHDVH DQG EHWWHU DGDSWDWLRQ WR (FXDGRU FRQGLWLRQV ,1,$3 f 1HRQRWRQLD ZLJKWLL 5 *UDK H[ :LJKW DQG $UQf /DFNH\ JO\FLQH DOVR NQRZQ DV SHUHQQLDO VR\EHDQV KDV EHHQ GHVFULEHG E\ +XPSKUH\V f DV D SHUHQQLDO SODQW VOHQGHU WZLQLQJ DQG ZLWK ORQJ VWHPV KDYLQJ VRPH FDSDFLW\ IRU URRWLQJ DW WKH QRGHV *O\FLQH ZDV ILUVW LQWURGXFHG LQWR (FXDGRU LQ EXW LW ZDV QRW XVHG LQ SDVWXUHV XQWLO ZKHQ LW ZDV VKRZQ WR EH RQH RI WKH EHVW OHJXPHV IRU KXPLG DQG VXEKXPLG DUHDV 7KUHH FXOWLYDUV KDYH EHHQ GLVWULEXWHG WR IDUPHUV DQG WKHVH ZHUH VHOHFWHG IRU SHUVLVWHQFH

PAGE 20

DGDSWDELOLW\ VHHG SURGXFWLRQ DQG GLVHDVH DQG LQVHFW UHVLVWDQFH 7KHVH FXOWLYDUV DUH n0DODZLn IRU ORZHU DOWLWXGHV n&RRSHUn IRU PHGLXP DOWLWXGHV XS WR P DQG n7LQDURRn WKH KLJKHVW IRUDJH \LHOGHU JURZV ZHOO IURP WR P RI DOWLWXGH SURGXFLQJ ODUJH DPRXQWV RI VHHG IURP P ,1,$3 f 3HUVLVWHQFH RI 7URSLFDO /HJXPH*UDVV $VVRFLDWLRQV 6HUUDR f VXJJHVWHG WKDW WKH ILUVW UHTXLUHPHQW IRU VXFFHVVIXO XVH RI KLJK\LHOGLQJ SDVWXUH OHJXPHV ZDV WKHLU DGDSWDWLRQ WR ORFDO FOLPDWLF FRQGLWLRQV 6HFRQGO\ QXWULHQW UHTXLUHPHQWV PXVW EH PHW WR LQVXUH KLJK \LHOG DQG PDLQWHQDQFH $QG ILQDOO\ WKH\ PXVW SHUVLVW XQGHU KHDY\ JUD]LQJ WR VHFXUH D ORQJ ODVWLQJ EHQHILFLDO FRQWULEXWLRQ WR WKH FRPSDQLRQ JUDVV WR WKH VRLO DQG WR WKH JUD]LQJ DQLPDO *RPH] f KDV VXJJHVWHG VRPH RWKHU IDFWRUV ZKLFK FDQ DIIHFW WKH SHUVLVWHQFH RI WURSLFDO OHJXPHV ZKHQ WKH\ DUH JURZLQJ LQ DVVRFLDWLRQ ZLWK JUDVVHV $PRQJ WKH PRVW LPSRUWDQW DUH f HQYLURQPHQWDO IDFWRUV VXFK DV OLJKW WHPSHUDWXUH DQG PRLVWXUH f JURZWK KDELWV RI HDFK VSHFLHV JURZLQJ LQ WKH PL[WXUH f QRGXODWLRQ DELOLW\ DQG FDSDFLW\ IRU QLWURJHQ IL[DWLRQ f HGDSKLF IDFWRUV VXFK DV S+ QXWULHQW DYDLODELOLW\ IRUP RI VXSSO\ f IUHTXHQF\ DQG LQWHQVLW\ RI GHIROLDn WLRQ E\ JUD]LQJ DQLPDOV f DELOLW\ WR VXUYLYH GXULQJ ORQJ GURXJKW SHULRGV f VHHG SURGXFWLRQ FDSDFLW\ DQG f SHVW DQG GLVHDVH WROHUn DQFH /XGORZ DQG :LOVRQ f UHSRUWHG WKDW WURSLFDO JUDVVHV DFKLHYH XS WR WKUHH WLPHV WKH SKRWRV\QWKHWLF UDWH ZKHQ FRPSDUHG ZLWK WURSLFDO OHJXPHV 7KLV FKDUDFWHULVWLF REYLRXVO\ JLYHV HFRORJLFDO DGYDQWDJHV WR & JUDVVHV DIIRUGLQJ WKHP WKH RSSRUWXQLW\ WR JURZ IDVWHU GRPLQDWH

PAGE 21

DQG HYHQ H[FOXGH WKH & OHJXPHV IURP WKH PL[WXUH 7RZ f VKRZHG WKDW JUHHQ SDQLF 3DQLFXP PD[LPXP YDU WULFKRJOXPHf ZDV PXFK PRUH SURGXFWLYH DW DOO OLJKW LQWHQVLWLHV DQG KLJKHU URRW WHPSHUDWXUHV WKDQ JO\FLQH ZKHQ ERWK VSHFLHV ZHUH WHVWHG XQGHU FRQWUROOHG HQYLURQn PHQWDO FRQGLWLRQV 5REHUWV f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f UHSRUWHG WKDW LQ JHQHUDO JUDVVHV KDYH D EHWWHU UDQJH RI DGDSWDWLRQ DQG DOVR D PRUH YLJRURXV JURZWK KDELW WKDW DOORZV WKHP WR FRPSHWH DOZD\V DW DQ DGYDQWDJH RYHU PRVW WURSLFDO IRUDJH OHJXPHV /DFN RI OHJXPH SHUVLVWHQFH LV DWWULEXWHG WR WKH XVH RI XQDGDSWHG VSHFLHV DQG FXOWLYDUV LPSURSHU RU QR IHUWLOL]DWLRQ LQFRUn UHFW UKL]RELXP DQG RYHUJUD]LQJ Wn0DQQHWMH f *URZWK KDELW DQG OHDI PRUSKRORJ\ RI VSHFLHV WKDW FRPSRVH D PL[WXUH DUH LPSRUWDQW FKDUDFWHULVWLFV ZKLFK KDYH GLUHFW HIIHFWV XSRQ FRPSDWLELOLW\ DQG SHUVLVWHQFH GXH HVSHFLDOO\ WR OLJKW LQWHUFHSWLRQ DELOLW\ RI HDFK LQGLYLGXDO VSHFLHV 6DQWKLUDVHJDUDP f LQ WKH KXPLG WURSLF RI 3HUX UHSRUWHG WKDW LQ D ZHOOPDQDJHG JXLQHD JUDVVFHQWUR SDVWXUH WKH SHUVLVWHQFH RI WKH OHJXPH ZDV GXH WR LWV YLQH\ JURZWK KDELW HQDEOLQJ LW WR FOLPE WKH VWHPV DQG OHDYHV RI WKLV WDOO DQG DJJUHVVLYH JUDVV 7KXV WKH OHJXPH FDQ LQWHUFHSW VXIILFLHQW

PAGE 22

VRODU UDGLDWLRQ 7KH\ DOVR VXJJHVWHG WKDW WKH LGHDO W\SH RI WZLQLQJ WURSLFDO OHJXPH VKRXOG KDYH D VWURQJ VWRORQLIHURXV RU UKL]RPDWRXV JURZWK KDELW LQ RUGHU WR ZLWKVWDQG IUHTXHQW GHIROLDWLRQ DQG KHDY\ GDPDJH E\ JUD]LQJ DQLPDOV :KLWHPDQ f SRLQWHG RXW WKDW IUHTXHQW GHIROLDWLRQ UHVXOWV LQ ORZ \LHOGV RI WZLQLQJ OHJXPHV VXFK DV JO\FLQH DQG 6LUDWUR ZKHUHDV VOLJKW GHIROLDWLRQ RU DEVHQFH RI GHIROLDWLRQ UHVXOWV LQ D KLJKHU FRQWULEXWLRQ RI WKH OHJXPHV WR WKH WRWDO SDVWXUH SURGXFWLRQ ,Q $IULFD 'UDROX DQG 1DEXVLQ1DSXOX f VWXGLHG WKH HIIHFW RI FXWWLQJ LQWHUYDOV RI DQG ZHHNV DQG FXWWLQJ KHLJKWV RI DQG FP LQ PL[HG VZDUGV RI JXLQHD JUDVV DQG 6W\ORVDQWKHV JXLDQHQVLV 7KH\ FRQFOXGHG WKDW '0 SURGXFWLRQ ZDV JUHDWO\ UHGXFHG XQGHU WKH ORZHVW DQG PRVW IUHTXHQW FXWWLQJ WUHDWn PHQW 7KH DPRXQW RI OHJXPH LQ WKH PL[WXUH DOVR GHFUHDVHG DV WKH FXWWLQJ LQWHUYDOV ZHUH LQFUHDVHG DQG WKH OHJXPH ZDV SDUWLFXODUO\ VHQVLWLYH WR FORVH GHIROLDWLRQ ,Q $XVWUDOLD 0FOYRU HW DO f IRXQG WKH VDPH OLQHDU WHQGHQF\ ZLWK 'HVPRGLXP LQWRUWXP DQG 6HWDULD VSKDFHODWD PL[WXUHV XQGHU GLIIHUHQW FXWWLQJ KHLJKWV RI DQG FP DQG FXWWLQJ LQWHUYDOV RI DQG ZHHNV 7KH JURZWK UHVSRQVH RI GHVPRGLXP ZDV PDUNHGO\ GHSUHVVHG DW WKH ORZHVW FXWWLQJ KHLJKW RI FP DQG WKH VKRUWHVW FXWWLQJ LQWHUYDO RI ZHHNV 7KH\ FRQFOXGHG WKDW FXWWLQJ DW WKH KHLJKW RI FP DW LQWHUYDOV RI WR ZHHNV ZDV QHFHVVDU\ IRU WKH SHUVLVWHQFH RI WKH GHVPRGLXP LQ WKH PL[WXUH %U\DQ HW DO f LQGLFDWHG WKDW WKH VKRUW JURZWK KDELW RI WKH WURSLFDO OHJXPHV 6W\ORVDQWKHV KXPLOLV DQG /RWRQRQLV EDLQHVLL ZKLFK PD\ EH VKDGHG E\ WDOOHU FRPSDQLRQ JUDVVHV EHQHILWV IURP KHDY\ JUD]LQJ SUHVVXUH ZKLFK DOORZV OLJKW SHQHWUDWLRQ LQWR WKH FDQRS\ 7KH\ DOVR

PAGE 23

FRQFOXGHG WKDW WKH GRPLQDQFH RI FOLPELQJ OHJXPHV VXFK DV JO\FLQH DQG 6LUDWUR LV JHQHUDOO\ HQKDQFHG E\ OLJKW JUD]LQJ SUHVVXUH DQG DOVR E\ ORQJ LQWHUYDOV EHWZHHQ JUD]LQJ SHULRGV *UD]LQJ H[SHULPHQWV FRQGXFWHG LQ WKH ZHW WURSLFV RI (FXDGRU E\ %HUUH]XHWD f &KDYH] f ,1,$3 f DQG =DSDWD f VKRZHG WKDW JXLQHDJUDVVFHQWUR SDVWXUHV DQG JXLQHDJUDVVJO\FLQH SDVn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fJO\FLQH DQG JXLQHDJUDVVFHQWUR PL[WXUHV ZHUH VKRZQ WR EH WKH PRVW FRPSDWLEOH DQG WKH EHVW DFFHSWHG E\ WKH JUD]LQJ DQLPDOV ZKLOH WURSLFDO NXG]X 3XHUDULD SKDVHRORLGHVf DQG &DORSRJRQLXP PXFXQRLGHV ZHUH XQSDODWDEOH VSHFLHV D FKDUDFWHULVWLF WKDW FRXOG GHWHUPLQH WKH GRPLQDQFH RI WKLV VSHFLHV RYHU WKH FRPSDQLRQ JUDVVHV ,1,$3 f

PAGE 24

5HVSRQVH WR 1XWULHQWV /HJXPHV DUH JHQHUDOO\ PRUH VHQVLWLYH WR VRLO IDFWRUV WKDQ RWKHU SDVWXUH SODQWV SDUWLFXODUO\ JUDVVHV 7KLV VHQVLWLYLW\ RI OHJXPHV HPSKDVL]HV WKH LPSRUWDQFH RI XQGHUVWDQGLQJ WKH HIIHFWV RI WKHVH OLPLn WDWLRQV LQ WURSLFDO FRQGLWLRQV 5XVVHOO f VXJJHVWHG WKDW LPn SURYLQJ WKH JURZWK RI OHJXPHV LQ ORZ IHUWLOLW\ VRLOV FRXOG EH DSSURDFKHG LQ WZR ZD\V f E\ DPHOLRUDWLRQ RI VRLO FRQGLWLRQV WKURXJK WKH XVH RI IHUWLOL]HUV RU DPHQGPHQWV DQG f E\ WKH VHOHFWLRQ RI OHJXPH FXOWLYDUV RU JHQHUD ZKLFK DUH PRUH WROHUDQW WR WKH OLPLWLQJ FRQGLWLRQV 7KHUH VHHPV WR EH D JHQHUDO FRQVHQVXV WKDW 1 DQG 3 LQ WKDW RUGHU DUH WKH SODQW QXWULHQWV WKDW DUH PRUH RIWHQ GHILFLHQW LQ WKH WURSLFV )R[ f 7KH DPRXQW RI 1 IL[DWLRQ QRGXODWLRQ SHUVLVWHQFH DQG \LHOG RI WURSLFDO IRUDJH OHJXPHV PD\ EH DIIHFWHG E\ WKH VRLO S+ DQG DOVR E\ WKH DYDLODELOLW\ RI SODQW QXWULHQWV 0DQKDHV DQG 'REHUHLQHU fDQG )R[ HW DO f UHSRUWHG WKDW IRU JRRG OHJXPH HVWDEOLVKPHQW DGHn TXDWH DPRXQWV RI DYDLODEOH 3 ZHUH UHTXLUHG HVSHFLDOO\ GXULQJ WKH QRGXODWLRQ VWDJHV 7KHVH DXWKRUV GHWHUPLQHG WKDW JO\FLQH UHTXLUHG SSP RI 3A2 LQ WKH VRLO VROXWLRQ GXULQJ WKH HVWDEOLVKPHQW SKDVH 7KLV UHTXLUHPHQW GHFUHDVHG DIWHU WKH VHFRQG FXWWLQJ ,Q $XVWUDOLD $QGUHZ DQG 5RELQV f GHWHUPLQHG WKH FULWLFDO 3 FRQFHQWUDWLRQ LQ WKH WRSV DVVRFLDWHG ZLWK PD[LPXP SODQW JURZWK DV EHLQJ WKH FULWLFDO OHYHOV 7KHVH ZHUH DQG b IRU FHQWUR DQG JO\FLQH UHVSHFn WLYHO\ )RU DGHTXDWH SODQW XSWDNH KA32A LRQV LQ WKH VRLO VROXWLRQ VKRXOG EH EHWZHHQ WR SSP $FFRUGLQJ WR 6DQFKH] f VRPH

PAGE 25

WURSLFDO OHJXPHV WROHUDQW WR ORZ DYDLODEOH VRLO 3 HLWKHU DEVRUE 3 DW D IDVWHU UDWH RU DUH DEOH WR WUDQORFDWH LW WR WKH WRSV PRUH UDSLGO\ WKDQ GR VSHFLHV QRW WROHUDQW WR ORZ 3 DYDLODELOLW\ 6DOLQDV DQG 6DQFKH] $QGUHZ f &RQVLGHUDEOH \LHOG UHVSRQVH WR 3 IHUWLOL]HUV ZDV UHSRUWHG E\ -RQHV DQG )UHLWDV f ZLWK IRXU WURSLFDO OHJXPHV 6W\ORVDQWKHV JXLDQHQVLV FHQWUR JO\FLQH DQG 6LUDWURf 6LPLODU UHVXOWV ZHUH REWDLQHG E\ )UDQFD DQG &DUYDOKR f LQ JUHHQKRXVH VWXGLHV XVLQJ ILYH WURSLFDO OHJXPHV JO\FLQH YDU n&RPPRQn JO\FLQH YDU n7LQDURRn 6LUDWUR FHQWUR DQG 3XHUDULD SKDVHRORLGHV YDU n-DYDQLFDn %HQWKf ,Q ERWK FDVHV WKH 3 GHILFLHQF\ ZDV UHIOHFWHG LQ GHFUHDVHG QRGXOH ZHLJKW DQG 1 IL[DWLRQ FDSDELOLW\ 6Q\GHU DQG .UHWVFKPHU f REWDLQHG VPDOO OLQHDU LQFUHDVHV LQ GU\ PDWWHU \LHOGV RI 6LUDWUR n&RRNn VW\OR FHQWUR DQG 'HVPRGLXP KHWHURn FDUS RQ /f '& ZKHQ OLPH ZDV DSSOLHG LQ NJ KD A LQFUHPHQWV XS WR NJ KD A ZLWKRXW 3 IHUWLOL]DWLRQ :KHQ WKH VDPH OHYHOV RI OLPH ZHUH XVHG WRJHWKHU ZLWK NJ KD A RI 3 WKH UHVSRQVH LQ \LHOG ZDV OLQHDU XS WR NJ KD A RI OLPH DQG FXUYLOLQHDU WKHUHDIWHU (VWLPDWLRQ RI 3 UHTXLUHPHQWV IRU SODQW JURZWK VKRXOG EH EDVHG RQ WKH DPRXQW RI 3 QHHGHG WR JLYH DW OHDVW b RI PD[LPXP JURZWK 2]DQQH DQG 6KDZ f 1HPH DQG /RYDGLQL f ZRUNLQJ ZLWK JO\FLQH IRXQG WKDW D FRPELQDWLRQ RI NJ RI 3AXV PHWULF WRQV RI OLPH KD A JDYH D ODUJH LQFUHDVH LQ \LHOG :HUQHU f UHSRUWHG VXEVWDQWLDOO\ LQFUHDVHG \LHOGV RI FHQWUR WR 3 DQG IHUWLOL]DWLRQ 3DODFLRV f LQ (FXDGRU IRXQG WKDW FHQWUR UHVSRQGHG SRVLWLYHO\ WR 3 IHUWLOL]DWLRQ GXULQJ WKH HVWDEOLVKPHQW SHULRG EXW \LHOG UHVSRQVH ZDV QRW UHODWHG WR WKH DPRXQW RI 3 DSSOLHG WR WKH VRLO GXULQJ WKH VRZLQJ WLPH )DODGH

PAGE 26

f FRPSDULQJ VL[ OHYHOV RI 3 DQG PJ NJ RI VRLOSRWf IRXQG WKDW 3 FRQFHQWUDWLRQ LQ JXLQHDJUDVV DQG HOHSKDQWJUDVV ZDV LQFUHDVHG ZLWK WKH DGGLWLRQ RI 3 7KH VDPH UHVSRQVH ZDV UHSRUWHG E\ 9LFHQWH&KDQGOHU f IURP 3XHUWR 5LFR ZKHUH b RI WKH PD[LPXP GU\ PDWWHU SURGXFWLRQ RI SHDUO PLOOHW >3HQQLVHWXP DPHUL FDQXP /f 6FKXP@ ZDV REWDLQHG ZKHQ VRLO S+ ZDV UDLVHG WR DQG SSP RI 3 ZHUH DGGHG $W D S+ RI WZLFH DV PXFK 3 ZDV QHHGHG WR SURGXFH WKH VDPH IRUDJH \LHOG )R[ f UHSRUWHG WKDW WKH VWDQGDUG 3 UHTXLUHPHQW ZDV D UHODWLYH VRLO UHTXLUHPHQW QRW DQ DEVROXWH SODQW UHTXLUHPHQW )R[ HW DO f IURP +DZDLL UHSRUWHG ODUJH LQFUHDVHV LQ 3 XSWDNH E\ VHYHUDO SDVWXUH VSHFLHV RQFH VRLOV ZKLFK KDG KLJK 3IL[LQJ FDSDFLWLHV ZHUH OLPHG WR S+ DQG 3KRVSKRUXV UHTXLUHn PHQWV RI VRLOV FDQ UDQJH IURP ]HUR WR PRUH WKDQ NJ 3 KD ? $IWHU 1 DQG 3 6 LV FRQVLGHUHG E\ PDQ\ DV WKH QH[W PRVW LPSRUWDQW HOHPHQW QHHGHG IRU WURSLFDO OHJXPH JURZWK 7HUJDV f QRWHG WKH VLJQLILFDQFH RI 6 RQ WKH JURZWK DQG QRGXODWLRQ RI VHYHUDO GLIIHUHQW WURSLFDO IRUDJH OHJXPHV 6LUDWUR DQG FHQWUR GU\ PDWWHU DQG QRGXOH ZHLJKW LQFUHDVHG DV 6 ZDV LQFUHDVHG 6DQFKH] f PHQWLRQHG WKDW 6 GHILFLHQFLHV DUH ZLGHVSUHDG WKURXJKRXW WKH WURSLFV DQG WKDW VRPH SDVWXUH OHJXPHV DUH PRUH VXVFHSWLEOH WR 6 GHILFLHQF\ WKDQ PRVW JUDVVHV 0HGLQD f UHSRUWHG 6 GHILFLHQFLHV LQ VRPH FURSV JURZLQJ LQ WKH OLWWRUDO UHJLRQ RI (FXDGRU $OVR D VWURQJ UHVSRQVH IURP JXLQHD JUDVV SDUDJUDVV JO\FLQH DQG FHQWUR ZDV REVHUYHG ZKHQ 6 ZDV DSSOLHG DORQH RU VXSSOLHG E\ WKH RUGLQDU\ VXSHUSKRVSKDWH RU E\ DPPRQLXP VXOSKDWH ,1,$3 f

PAGE 27

,Q WKH WURSLFV ZKHUH WKHUH DUH ZHOO GHILQHG ZHW DQG GU\ VHDVRQV UDLQIDOO SOD\V DQ LPSRUWDQW UROH LQ WKH XSWDNH DQG QXWULHQW FRQWHQW RI WURSLFDO IRUDJH VSHFLHV %OXH DQG 7HUJDV f UHSRUWHG D GURS LQ 1 3 DQG FRQWHQWV GXULQJ WKH GU\ VHDVRQ OLNHZLVH D GHFUHDVH RI QXWULHQW FRQWHQW GXULQJ WKH ZHW VHDVRQ ZDV IRXQG DQG LW ZDV SRVWXODWHG WR EH GXH WR WUDQVORFDWLRQ RI QXWULHQWV WR WKH URRWV 5DSLG JURZWK GXULQJ ZHW VHDVRQV PD\ UHVXOW LQ WUDFH PLHQUDOV EHLQJ WUDQVORFDWHG WR SODQW WRSV ZKHUH WKH\ DUH UDSLGO\ GLOXWHG ZLWK DHULDO WLVVXH FDXVLQJ GHILFLHQF\ V\PSWRPV LQ ROGHU WLVVXHV 5HXWHU f 0LFURQXWULHQWV FDQ SOD\ DQ LPSRUWDQW UROH LQ WURSLFDO OHJXPH SDVWXUHV JURZWK PDLQO\ EHFDXVH RI WKHLU IXQFWLRQ LQ VHYHUDO HQ]\PH V\VWHPV DQG LQ 1IL[DWLRQ E\ UKL]RELXPOHJXPH DVVRFLDWLRQV :HUQHU HW DO f VWXGLHG WURSLFDO OHJXPH UHVSRQVH WR WKH PLFURQXWULHQWV 0R &X =Q % 0Q DQG &R LQ WKH IRUP RI )7( %5 DQG DOVR LQ WKH VDOW IRUP 8VLQJ WKUHH WURSLFDO OHJXPHV SODQWHG LQ SRWV WKH\ REVHUYHG YLVLEOH V\PSWRPV RI 0Q WR[LFLW\ RQ JO\FLQH DQG % WR[LFLW\ RQ VW\OR 7KLV ZRUN HPSKDVL]HV WKDW WKHUH LV D QDUURZ UDQJH EHWZHHQ GHILFLHQW DQG WR[LF OHYHOV RI VRPH PLFURQXWULHQWV 0HGLQD f IRXQG VRPH PLFURQXWULHQW GHILFLHQF\ V\PSWRPV LQ VRPH WURSLFDO FURSV JURZLQJ LQ WKH 4XHYHGR DUHD +H UHSRUWHG WKDW % =Q DQG )H ZHUH WKH PRVW GHILFLHQW HOHPHQWV ,1,$3 f IRXQG WKDW DW WKH EHJLQQLQJ RI WKH UDLQ\ VHDVRQ WKH SHULRG RI PRVW UDSLG JURZWK IRU JUDVVHV VXFK DV SDUDJUDVV =Q GHILFLHQF\ V\PSWRPV ZHUH HYLGHQW EXW WKHVH GLVDSSHDU LQ WZR RU WKUHH ZHHNV 0RO\EGHQXP DOVR KDV EHHQ UHFRJQL]HG DV DQ HVVHQWLDO HOHPHQW IRU OHJXPH JURZWK HVSHFLDOO\ IRU HVWDEOLVKPHQW DQG PDLQWHQDQFH 6RPH DXWKRUV KDYH VXJJHVWHG WKDW WKLV

PAGE 28

HOHPHQW LV QHFHVVDU\ IRU GHYHORSPHQW RI HQ]\PHV UHODWHG WR 1 IL[DWLRQ QLWUDWH UHGXFWLRQ DQG OHJXPH QRGXODWLRQ $QGUHZ (SVWHLQ f $QLPDO 3URGXFWLYLW\ ,Q 3LFKLOLQJXH (FXDGRU =DSDWD f REVHUYHG WKDW OLYHZHLJKW JDLQ RI VWHHUV JUD]LQJ RQ FRPPRQ JXLQHDJUDVV LPSURYHG *XLQHDJUDVV %UDFKLDULD KXPLGLFROD DQG LPSURYHG JXLQHDJUDVVJO\FLQH SDVWXUHV ZHUH LQ WKH RUGHU JLYHQ DQG NJ DQLPDO A GD\ DQG DQ DQQXDO OLYHZHLJKW JDLQ RI DQG NJ KD UHVSHFWLYHO\ 7KH VDPH DXWKRU DOVR QRWHG D GLIIHUHQFH LQ OLYHZHLJKW JDLQ GXH WR WKH EUHHG RI DQLPDO EHLQJ DQG NJ DQ A GD\ A IRU UHG FULROOR EUDKDP DQG EUDKDP [ KROVWHLQ FURVVHV UHVSHFn WLYHO\ 6LPLODU UHVXOWV ZHUH IRXQG DW P DOWLWXGH E\ &RZDQ HW DO f EXW LQ WKLV FDVH PLONLQJ FRZV ZHUH JUD]HG RQ JXLQHDJUDVV JO\FLQH DQG NLNX\JUDVV 3HQQLVHWXP FODQGHVWLQXPf SDVWXUHV ZLWKRXW DQ\ VXSSOHPHQWDWLRQ 0LON SURGXFWLRQ DYHUDJHG DQG NJ FRZ A GD\ IRU MHUVH\ DQG KROVWHLQ FRZV UHVSHFWLYHO\ ,Q $XVWUDOLD *URI DQG +DUGLQJ f IRXQG WKDW D JXLQHDJUDVVFHQWUR SDVWXUH \LHOGHG b PRUH OLYHZHLJKW WKDQ *XLQHDJUDVV DORQH RYHU D WZR \HDU SHULRG +DOO f UHSRUWHG WKDW DQLPDO SURGXFWLRQ RQ XQLPSURYHG QDWLYH JUDVV ZDV OHVV WKDQ NJ KD A ZKHQ FRPSDUHG WR JXLQHDJUDVV6LUDWUR PL[WXUH ZKLFK \LHOGHG NJ OLYHZHLJKW KD ? ,Q &RVWD 5LFD .UHWVFKPHU f UHSRUWHG WKDW FHQWUR LQ PL[WXUHV ZLWK JXLQHDJUDVV LQFUHDVHG IRUDJH \LHOG E\ b GXULQJ WKH GU\ VHDVRQ DQG E\ b GXULQJ WKH ZHW VHDVRQ ZKHQ FRPSDUHG ZLWK JUDVV DORQH 7HUJDV f UHSRUWHG WKDW FHQWUR LQFUHDVHG WRWDO IRUDJH SURGXFWLRQ

PAGE 29

E\ b ZKHQ LW ZDV JURZQ LQ PL[WXUHV ZLWK JXLQHDJUDVV DQG WKH DPRXQW RI 1 IL[HG E\ WKH OHJXPH ZDV HVWLPDWHG WR EH NJ 1 KD A \U ? ,Q QRUWKHUQ $XVWUDOLD 1RUPDQ f VKRZHG WKDW JDLQ SHU DQLPDO JUD]LQJ OHJXPHJUDVV PL[WXUHV ZDV OLQHDUO\ UHODWHG WR WKH SURSRUWLRQ RI OHJXPH LQ WKH SDVWXUH $OVR DQLPDO OLYHZHLJKW JDLQ GXULQJ WKH GU\ VHDVRQ ZDV UHODWHG WR WKH QXPEHU RI GD\V GXULQJ ZKLFK WKH DQLPDOV KDG EHHQ JUD]LQJ WKH OHJXPHV +H DOVR UHSRUWHG WKDW DQLPDOV JUD]LQJ RQ JUDVV SDVWXUHV JDLQHG NJ KHDG ? ZKLOH WKRVH DOORZHG WR JUD]H JUDVVOHJXPH SDVWXUHV JDLQHG NJ KHDG ? 'XULQJ WKH SHULRG RI GD\V RI WKH GU\ VHDVRQ WKH ILUVW JURXS RQ JUDVV SDVWXUHV ORVW DOPRVW NJ DQLPDO ? ZKLOH WKH VHFRQG JDLQHG NJ DQLPDO ? GXH WR WKH FRPSDQLRQ OHJXPH 6W\ORVDQWKHV KXPLOLV ,Q (FXDGRU &KDYH] f UHSRUWHG WKDW D JXLQHD JUDVVFHQWUR PL[WXUH SURGXFHG NJ OLYHZHLJKW JDLQ KD A \HDU A ZLWK WKH EHQHILFLDO HIIHFW RI WKH OHJXPH EHLQJ PRUH DSSDUHQW GXULQJ WKH GU\ VHDVRQ ZKHQ FUXGH SURWHLQ FRQWHQW RI WKH JXLQHDJUDVV DORQH ZDV EHORZ b DQG FUXGH SURWHLQ FRQWHQW IRU WKH JUDVVOHJXPH SDVWXUHV RYHU b 6LPLODU UHVXOWV ZHUH UHSRUWHG E\ ,1,$3 f RQ JXLQHDJUDVVJO\FLQH YDU n7LQDURRn PL[WXUH ZKLFK SURGXFHG NJ RI OLYHZHLJKW KD A \HDU ? ,Q ERWK FDVHV WKH UHVXOWV ZHUH REWDLQHG XQGHU URWDWLRQDO JUD]LQJ XVLQJ WKH SXWDQGWDNH WHFKQLTXH GHYHORSHG E\ 0RWW f $FFRUGLQJ WR 0LQVRQ DQG 0LOIRUG f WURSLFDO OHJXPHV PDLQWDLQ DGHTXDWH QXWULWLYH YDOXHV IRU D ORQJHU SHULRG RI WLPH WKDQ PRVW WURSLFDO JUDVVHV ZKHQ HDFK ZHUH XQGHU WKH VDPH PDQDJHPHQW V\VWHP 7KH FULWLFDO OHYHO RI FUXGH SURWHLQ UHTXLUHG LQ D SDVWXUH EHIRUH LQWDNH LV UHGXFHG E\ SURWHLQ GHILFLHQF\ LV HVWLPDWHG WR OLH EHWZHHQ DQG b (YHQ KLJKO\ 1IHUWLOL]HG WURSLFDO JUDVVHV DW ODWH JURZWK VWDJHV PD\ KDYH

PAGE 30

YDOXHV EHORZ WKHVH SRLQWV 9HQWXUD HW DO f 7URSLFDO IRUDJH OHJXPHV RQ WKH RWKHU KDQG UHWDLQ KLJKHU FUXGH SURWHLQ OHYHOV GXULQJ WKH GU\ VHDVRQ HYHQ LQ DGYDQFHG PDWXULW\ VWDJHV 0LOIRUG DQG +D\GRFN f 6HOHFWLYLW\ E\ JUD]LQJ DQLPDOV LV D SRVVLEOH H[SODQDWLRQ IRU KLJKHU SHUIRUPDQFH RI JUD]LQJ FDWWOH GXULQJ WKH GU\VHDVRQ SHULRGV LQ WURSLFDO UHJLRQV HVSHFLDOO\ ZKHQ DYDLODEOH IRUDJH PHHWV RU H[FHHGV DQLPDO UHTXLUHPHQWV Wn0DQQHWMH f 3DVWXUH (YDOXDWLRQ 0RWW DQG 0RRUH f GHYHORSHG D ILYHSKDVH VFKHPH IRU IRUDJH HYDOXDWLRQ 6XFK D VFKHPH LQYROYHV TXDQWLW\ DQG TXDOLW\ GHWHUPLQDWLRQV 3KDVH ,QWURGXFWLRQ DQG EUHHGHUnV OLQHV ,,6PDOO SORW FOLSSLQJ WULDOV ,,,0RE JUD]LQJ H[SHULPHQWV IRUDJH UHVSRQVH WR JUD]LQJ DQLPDOV ,9$QLPDO UHVSRQVH HIIHFW RI IRUDJH RQ DQLPDO RXWSXW 9)RUDJHOLYHVWRFN IHHGLQJ V\VWHPV )RUDJH TXDOLW\ LQ YLWUR RUJDQLF PDWWHU GLJHVWLELOLW\ ,920'f LV WDNHQ LQWR FRQVLGHUDWLRQ LQ WKH ILUVW WKUHH SKDVHV ZKLOH SKDVHV ,9 DQG 9 LQYROYH DQLPDO UHVSRQVH LQ QXWULHQW GLJHVWLELOLW\ SHUIRUPDQFH SHU DQLPDO DQG SURGXFWLRQ SHU KHFWDUH 'HILQLWLRQV IRU VWRFNLQJ UDWH JUD]LQJ SUHVVXUH DQG FDUU\LQJ FDSDFLW\ ZHUH JLYHQ E\ 0RWW f 6WRFNLQJ UDWH LV GHILQHG DV WKH QXPEHU RI DQLPDOV SHU XQLW DUHD RI ODQG WKH WHUP EHDULQJ QR UHODWLRQn VKLS WR WKH DPRXQW RI IRUDJH *UD]LQJ SUHVVXUH LV WKH DPRXQW RI IRUDJH GU\ PDWWHU RQ RIIHU SHU DQLPDO SHU GD\ DQG FDUU\LQJ FDSDFLW\ DOVR FDOOHG JUD]LQJ FDSDFLW\ LV GHILQHG DV WKH VWRFNLQJ UDWH DW WKH

PAGE 31

RSWLPXP JUD]LQJ SUHVVXUH 0F0HHNDQ f FRQVLGHUHG WKH VWRFNLQJ UDWH DV WKH PRVW SRZHUIXO IDFWRU LQIOXHQFLQJ WKH HIILFLHQF\ RI SDVWXUH FRQYHUVLRQ WR DQLPDO SURGXFWV RQ D SHU XQLW DUHD EDVLV 3HWHUVHQ HW DO f GHYHORSHG TXDQWLWDWLYH H[SUHVVLRQV ZKLFK UHODWHG JUD]LQJ SUHVVXUH DQG FDUU\LQJ FDSDFLW\ ZLWK RXWSXW SHU DQLPDO 7KHLU TXDQWLWDWLYH WKHRU\ VXJJHVWV WKDW DQLPDO JDLQ LV FRQVWDQW DV WKH VWRFNLQJ UDWH LV LQFUHDVHG XS WR D FULWLFDO SRLQW ZKHUH WKH JUD]HDEOH IRUDJH LV HTXDO WR WKH DPRXQW RI IRUDJH FRQVXPHG E\ JUD]LQJ DQLPDOV ,I WKH VWRFNLQJ UDWH LV LQFUHDVHG EH\RQG WKH FULWLFDO SRLQW WKHQ WKH JDLQ SHU DQLPDO GHFUHDVHV DQG WKH DQLPDO JDLQ SHU XQLW DUHD DOVR GHFUHDVHV &RQZD\ f XVLQJ WKUHH VWRFNLQJ UDWHV DQG DQLPDOV SHU DFUH IRXQG WKDW E\ LQFUHDVLQJ WKH VWRFNLQJ UDWH IURP WR DQLPDOV SHU DFUH OLYHZHLJKW JDLQ SHU DQLPDO GHFUHDVHG +RZHYHU ZLWK DQLPDOV SHU DFUH WKH OLYHZHLJKW SHU DQLPDO ZDV GUDVWLFDOO\ UHGXFHG UHVXOWLQJ DOVR LQ D UHGXFWLRQ RI OLYHZHLJKW SHU DFUH -RQHV f LQ $XVWUDOLD VWXGLHG WKH HIIHFW RI ILYH VWRFNLQJ UDWHV QDPHO\ DQG DQ KD A LQ FRPELQDWLRQ ZLWK WKUHH UHVWLQJ SHULRGV RI DQG GD\V XVLQJ D UDQGRPL]HG FRPSOHWH EORFN GHVLJQ ZLWK WZR UHSOLFDn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f VXJJHVWHG WKDW FDUU\LQJ FDSDFLW\

PAGE 32

VKRXOG EH GHWHUPLQHG WDNLQJ LQWR FRQVLGHUDWLRQ WKH DPRXQW RI OLWWHU OHIW LQ WKH ILHOG DV UHVLGXH DIWHU HDFK JUD]LQJ SHULRG 0RWW f HPSKDVL]HG WKH XVH RI YDULDEOH VWRFNLQJ UDWHV LQ JUD]LQJ H[SHULPHQWV E\ PDLQWDLQLQJ WKH QXPEHU RI DQLPDOV LQ HTXLOLEULXP ZLWK WKH DYDLODEOH IRUDJH +H QRWHG WKDW WKH LPSRUWDQW DGYDQWDJH RI WKH V\VWHP ZDV WKDW LW SHUPLWV HVWLPDWLRQ RI WKH FDUU\LQJ FDSDFLW\ RI WKH SDVWXUH DQG WKH VHDVRQDO FKDQJHV ZKLFK RFFXU (YDQV DQG %U\DQ f SRLQWHG RXW WKH QHHG IRU PRUH VWXGLHV WR HYDOXDWH GLIIHUHQW JUD]LQJ SUHVVXUHV LQ RUGHU WR PHDVXUH WKH \LHOG RI WKH SDVWXUH DQG SHUVLVWHQFH RI WURSLFDO OHJXPHV $FFRUGLQJ WR 0RWW f WKH RSWLPXP JUD]LQJ SUHVVXUH PXVW EH FRQVLGHUHG DV DQ RSWLPXP UDQJH LQVWHDG RI D FULWLFDO SRLQW DQG WKDW VXFK DQ RSWLPXP UHODWHV RQO\ WR DQLPDO RXWSXW DQG PD\ RU PD\ QRW EH WKH RSWLPD IRU SODQW VSHFLHV LQ WKH SDVWXUH 0RWW DQG /XFDV f 0RWW f DQG 0DWFKHV f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

PAGE 33

DOO WUHDWPHQWV S f 6HUUDR f VXJJHVWHG WKDW YDULDEOH RU IL[HG VWRFNLQJ UDWHV FDQ EH XVHG LQ FRQWLQXRXV RU URWDn WLRQDO JUD]LQJ V\VWHPV +H FRQFOXGHV WKDW WKH SXWDQGWDNH V\VWHP LV PRUH DSSURSULDWH ZKHQ SODQW DQG DQLPDO UHODWLRQVKLSV DUH WR EH PHDVXUHG *UD]LQJ 6\VWHPV &RQWLQXRXV *UD]LQJ +HDG\ f QRWHG WKHUH LV PXFK FRQIXVLRQ LQ WKH GHILQLWLRQ RI JUD]LQJ V\VWHPV ZKLFK DUH XVHG IRU GHVFULELQJ WKH GD\ WR GD\ SURYLVLRQ RI OLYHVWRFN IHHG IURP D ZLGH YDULHW\ RI VRXUFHV VXFK DV FRQVHUYHG IRUDJH DQG E\ GLUHFW XVH RI SDVWXUH E\ JUD]LQJ DQLPDOV +H DOVR GHILQHG FRQWLQXRXV JUD]LQJ DV D JUD]LQJ V\VWHP LQ ZKLFK WKH DQLPDOV KDYH XQUHVWULFWHG DFFHVV WR DQ\ SDUW RI WKH SDVWXUH WKURXJK D JUD]LQJ SHULRG ZKLFK FDQ EH D VHDVRQ D \HDU RU PRUH 6SHGGLQJ f VXJn JHVWHG D SDVWXUH XQGHU D FRQWLQXRXV JUD]LQJ V\VWHP FRXOG EH FDOOHG FRUUHFWO\ JUD]HG ZKHQ WKH DPRXQW RI UHPRYHG IRUDJH E\ WKH JUD]LQJ DQLPDOV ZDV HTXDO WR WKH DPRXQW RI IRUDJH GDLO\ \LHOG &RQWLQXRXV JUD]LQJ LV WKH PRVW FRPPRQO\ XVHG V\VWHP LQ WKH WURSLFV HVSHFLDOO\ RQ YDVW DUHDV IDU DZD\ IURP WKH PDLQ FRQVXPHU FHQWHUV DQG LQ PDQ\ FDVHV DUHDV ZLWKRXW DQ\ VXLWDEOH KLJKZD\ V\VWHP &KDYHUUD f 5RWDWLRQDO *UD]LQJ $ URWDWLRQDO JUD]LQJ V\VWHP LV GHILQHG E\ +HDG\ f DQG +HDWK f DV D V\VWHP LQ ZKLFK WKH DQLPDOV DUH DOORZHG WR JUD]H WKH SDVWXUH IRU YDULDEOH SHULRGV RI WLPH QRUPDOO\ ZLWK D KHDY\ VWRFNLQJ UDWH GXULQJ ZKLFK WKH SDVWXUH LV JUD]HG DQG XQJUD]HG VHYHUDO WLPHV GXULQJ D JUD]LQJ VHDVRQ RU \HDU

PAGE 34

7KH WHUPV JUD]LQJ SHULRG DQG UHVW SHULRG DUH FRPPRQO\ XVHG LQ WKH URWDWLRQDO JUD]LQJ V\VWHP +HDG\ f $FFRUGLQJ WR +HDG\ f WKH JUD]LQJ SHULRG LV WKH SRUWLRQ RI WKH JUD]LQJ WLPH GXULQJ ZKLFK JUD]LQJ WDNHV SODFH DQG UHVW SHULRG LV WKH WLPH GXULQJ ZKLFK WKH SDVWXUH LV QRW JUD]HG &RPSDULQJ ERWK FRQWLQXRXV DQG URWDWLRQDO JUD]LQJ V\VWHPV RQ DQLPDO SURGXFWLRQ 6WREEV f LQ $IULFD IRXQG WKDW DQLPDO SURGXFWLRQ ZDV VOLJKWO\ KLJKHU IRU URWDWLRQDO JUD]LQJ ZKHQ LW ZDV FDUULHG RXW LQ WKUHH SDGGRFNV EXW LW ZDV ORZHU WKDQ FRQWLQXRXV JUD]LQJ ZKHQ LW ZDV GRQH LQ VL[ SDGGRFNV )URP $XVWUDOLD *URI DQG +DUGLQJ f UHSRUWHG WKDW D PL[HG *XLQHDJUDVVFHQWUR SDVWXUH ZLWK D FDUU\LQJ FDSDFLW\ RI DQ KD A SURGXFHG D OLYHZHLJKW JDLQ RI NJ KD IRU \HDUV DQG NJ KD A IRU \HDUV IRU FRQWLQXRXV DQG URWDWLRQDO JUD]LQJ UHVSHFWLYHO\ )URP 3LFKLOLQJXH (FXDGRU 3DUHGHV f UHSRUWHG WKDW GXULQJ WKH UDLQ\ VHDVRQ WKHUH ZHUH QR GLIIHUHQFHV LQ VWRFNLQJ UDWH EHWZHHQ FRQWLQXRXV DQG URWDWLRQDO JUD]LQJ EXW GXULQJ WKH GU\ VHDVRQ FRQWLQXRXV JUD]LQJ XQGHU YDULDEOH VWRFNLQJ UDWH ZDV VXSHULRU WR URWDWLRQDO JUD]LQJ ZLWK IL[HG RU YDULDEOH VWRFNLQJ UDWH +H DOVR IRXQG WKDW FRQWLQXRXV JUD]LQJ ZLWK YDULDEOH VWRFNLQJ UDWHV JDYH WKH KLJKHVW DYHUDJH GDLO\ JDLQ RI NJ DQ A GD\ ? (VWLPDWHV RI 'U\ 0DWWHU 3URGXFWLRQ DQG
PAGE 35

KHLJKW ZHLJKW DQG RWKHU IDFWRUV GLIIHU IURP RQH VSHFLHV WR DQRWKHU .HQQHG\ f )RU PDQ\ \HDUV KDQG FXWWLQJ DQG ZHLJKLQJ RI DERYHn JURXQG YHJHWDWLYH SDUWV KDV EHHQ WKH PRVW SRSXODU DQG XVHIXO PHWKRG IRU HVWLPDWLQJ IRUDJH \LHOG +HDG\ f DQG IRU HVWLPDWLQJ JUD]LQJ SUHVVXUH 0RWW f 2WKHU PHWKRGV VXFK DV WKH VLPSOH GLVN PHWHU GHVFULEHG E\ %UDQVE\ f ZRUN RQ WKH SULQFLSOH RI PHDVXULQJ WKH KHLJKW RI D GLVN VXSSRUWHG E\ WKH UHVLVWDQFH DQG FRPSUHVVLRQ RI WKH YHJHWDWLRQ 6DQWLOODQ f SURYHG WKDW WKH VLPSOH GLVN PHWHU ZDV YHU\ VDWLVIDFWRU\ IRU XVH LQ WURSLFDO VSHFLHV 0RWW f VXJJHVWHG WKDW HVWLPDWLRQ RI WRWDO \LHOG RU \LHOG RI FRPSRQHQWV LV EDVHG XSRQ WKH IROORZLQJ UHODWLRQVKLS \LHOG SHU XQLW DUHD I GHQVLW\ KHLJKWf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n0DQQHWMH f $ GRXEOH VDPSOLQJ WHFKQLTXH LV SUREDEO\ RQH RI WKH EHVW ZD\V WR HQVXUH D PRUH SUHFLVH \LHOG GHWHUPLQDWLRQ (\HHVWLPDWLRQ LQ FRPELQDn WLRQ ZLWK D IHZ KDUYHVWHG VDPSOHV ZKLFK DFW DV D FRQWURO RQ WKH REVHUn YHUnV DFFXUDF\ LV RQH RI WKH VLPSOHVW IRUPV RI HVWLPDWLQJ WRWDO IRUDJH SUHVHQW RU DQQXDO SURGXFWLRQ SRWHQWLDO RI D SDVWXUH Wn0DQQHWMH f

PAGE 36

0HDVXULQJ %RWDQLFDO &RPSRVLWLRQ %RWDQLFDO FRPSRVLWLRQ LV D YHU\ HVVHQWLDO PHDVXUHPHQW HVSHFLDOO\ LQ SDVWXUHV VXEMHFWHG WR JUD]LQJ FRQGLWLRQV EHFDXVH WKH QXPEHU RI VDPSOHV DQG \LHOG RI LQGLYLGXDO VSHFLHV PD\ YDU\ RYHU D ZLGH UDQJH GHSHQGLQJ XSRQ HQYLURQPHQWDO FRQGLWLRQV DQG PDQDJHPHQW IDFWRUV 7n0DQQHWMH HW DO f LQGLFDWHG WKDW ERWDQLFDO FRPSRVLWLRQ FDQ EH PHDVXUHG LQ WHUPV RI WKH \LHOG RI FRPSRQHQW VSHFLHV WKH QXPEHU RI SODQWV FRYHULQJ WKH DUHD DQG DOVR WKH IUHTXHQF\ RI RFFXUUHQFH 'HWHUPLQDWLRQ RI ERWDQLFDO FRPSRVLWLRQ DQG VDPSOLQJ WHFKQLTXHV DUH GLIILFXOW WDVNV LQ SDVWXUH UHVHDUFK 6RPH PHWKRGV KDYH EHHQ GHYHORSHG WR GHWHUPLQH ERWDQLFDO FRPSRVLWLRQ WKH PRVW FRPPRQ DUH YLVXDO HVWLPDWLRQ DQG KDQG VHSDUDWLRQ RI KDUYHVWHG PDWHULDO LQWR FRPSRn QHQW VSHFLHV *DUGQHU f 9LVXDO HVWLPDWLRQ LV D UHOLDEOH PHWKRG IRU VWXG\LQJ SDVWXUH VSHFLHV EXW LQ VRPH FDVHV LW PD\ EH GLIILFXOW WR UHODWH GU\ PDWWHU SURGXFWLRQ RI HDFK RI WKH FRPSRQHQW VSHFLHV HVSHFLn DOO\ ZKHQ JURZWK KDELW DQG GHQVLW\ GLIIHU ZLGHO\ 7RWKLOO DQG 3HWHUVHQ f LQGLFDWHG WKDW WKH ZHLJKW MLQ VLWX DV ZHOO DV WKH HVWLPDWLRQ RI ZHLJKW RI HDFK LQGLYLGXDO VSHFLHV DQG YLVXDO HVWLPDWLRQ DUH WKH PRVW XVHIXO PHWKRGV IRU VXUYH\LQJ YHJHWDWLRQ RI SDVWXUH VSHFLHV (IIHFW RI WKH *UD]LQJ $QLPDO RQ %RWDQLFDO &RPSRVLWLRQ 7KHUH DUH DW OHDVW WKUHH IDFWRUV WKDW DUH NQRZQ WR DIIHFW WKH EDODQFH RI WKH JUDVVOHJXPH PL[WXUHV LQ WKH WURSLFV RQFH WKH SDVWXUH KDV EHHQ HVWDEOLVKHG 7KH\ DUH VWRFNLQJ UDWH RU JUD]LQJ SUHVVXUH IUHTXHQF\ RI GHIROLDWLRQ DQG IHUWLOL]DWLRQ

PAGE 37

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f 5HVHDUFK ZRUNHUV VXFK DV 'DYLV f DQG :HOOV f DJUHHG WKDW JUD]LQJ DIIHFWV WKH VRLO FRYHU DQG ERWDQLFDO FRPSRVLWLRQ ZKLFK UHVXOWV LQ f D UHGXFWLRQ RI WKH EDVDO FRYHU f D UHGXFWLRQ RI KHLJKW RI VSHFLHV f D ORVV RI VRLO FRYHU DQG VHYHUH HURVLRQ f D UHGXFWLRQ RI URRW V\VWHPV f D UHGXFWLRQ LQ WKH HPHUJHQFH RI QHZ VKRRWV DQG f ZHHG LQYDVLRQ %U\DQ DQG (YDQV f IRXQG WKDW WKH SURSRUWLRQ RI OHJXPHV 6W\ORVDQWKHV JXLDQHQVLV &HQWURVHPD SXEHVFHQV DQG 3XHUDULD SKDVHR ORLGHV RQ WKH GU\ PDWWHU EDVLV ZDV QRW DIIHFWHG E\ VWRFNLQJ UDWHV GXULQJ WKH ILUVW DQG VHFRQG \HDU ,Q WKH WKLUG \HDU D PDUNHG UHGXFWLRQ WRRN SODFH RQ WKH KLJK VWRFNLQJ UDWH RI DQLPDOV KD ? LQ ZKLFK WKH SURSRUWLRQ RI WKH OHJXPHV IHOO IURP WR b ZKLOH WKH FRPSDQLRQ JUDVV 3DQLFXP PD[LPXPf ZDV PRUH DIIHFWHG E\ WKH KLJK VWRFNLQJ UDWH ,Q WKLV FDVH WKH UHGXFWLRQ ZDV VLJQLILFDQW VLQFH LW ZHQW IURP WR b DQG WR b LQ WKH ILUVW VHFRQG DQG WKLUG \HDU UHVSHFWLYHO\ 7KH RWKHU WZR VWRFNLQJ UDWHV RI DQG DQLPDOV KD A KDG OLWWOH HIIHFW RQ

PAGE 38

WKH JUDVV DQG OHJXPH GXULQJ WKH WKUHH H[SHULPHQWDO \HDUV 6LPLODU UHVXOWV ZHUH UHSRUWHG IURP $IULFD E\ 6WREEV f &RZDQ HW DO f LQ WKH $WKHUWRQ 7DEOHODQG 4XHHQVODQG $XVWUDOLD IRXQG D KLJKO\ VLJQLILFDQW FRUUHODWLRQ FRHIILFLHQW EHWZHHQ PLON \LHOG KD A DQG WKH DPRXQW RI WKH OHJXPH JO\FLQH SUHVHQW LQ WKH SDVWXUH 7KH\ VWXGLHG IRXU VWRFNLQJ UDWHV DQG FRZV KD ‘r‘ 2Q WKH RWKHU KDQG ZKLOH VWRFNLQJ UDWH ZDV LQFUHDVHG WKH PLON SURGXFWLRQ KD A DOVR ZDV LQFUHDVHG EXW WKH SHUFHQWDJH RI GU\ PDWWHU RI OHJXPH SUHVHQW LQ WKH SDVWXUH GHFUHDVHG ZKHQ WKH VWRFNLQJ UDWH LQFUHDVHG &RZDQ DQG 2n*UDG\ f UHSRUWHG WKH VDPH WUHQG XQGHU VLPLODU FRQGLWLRQV LQ DQRWKHU VWXG\ FDUULHG RXW ODWHU DW WKH VDPH ORFDWLRQ 6RPH LQYHVWLJDWLRQV FDUULHG RXW LQ &XED KDYH VKRZQ WKDW XQGHU KHDY\ VWRFNLQJ UDWHV WKH OHJXPHV IDLO WR SHUVLVW LQ WKH PL[WXUH )HEOHV DQG 3DGLOOD f XVLQJ D VWRFNLQJ UDWH RI FRZV KD ? IRXQG WKDW D PL[WXUH RI JXLQHDJUDVV DQG WKH OHJXPHV JO\FLQH 6LUDWUR 6W\OR VDQWKHV JXLDQHQVLV 'HVPRGLXP LQWRUWXP DQG 'HVPRGLXP XQFLQDWXP GLG QRW SHUVLVW PRUH WKDQ ZHHNV )XQHV DQG 3HUH] f XVLQJ VL[ DQLn PDOV KD ? DOVR IRXQG WKDW WKH WKUHH FRPPHUFLDO FXOWLYDUV RI JO\FLQH 7LQDURR &RRSHU DQG &ODUHQFHf IDLOHG WR SHUVLVW XQGHU WKRVH FRQGLWLRQV DQG WKDW DW ZHHNV WKH SURSRUWLRQ RI ZHHG\ VSHFLHV ZDV GUDVWLFDOO\ LQFUHDVHG 7KHLU ILQDO FRQFOXVLRQ ZDV WKDW LW LV EHWWHU WR XVH OLJKW VWRFNLQJ UDWHV DQG WKDW WKH OHJXPHV DUH XVHIXO LQ WKRVH DUHDV LQ ZKLFK WKH JUD]LQJ PDQDJHPHQW LV H[WHQVLYH 2Q WKH RWKHU KDQG WKHUH DUH VRPH UHVXOWV WKDW VKRZ D SRVLWLYH HIIHFW RI VWRFNLQJ UDWH XSRQ WKH SURSRUn WLRQ RI OHJXPH LQ WKH SDVWXUH 9LOHOD f LQ %UD]LO IRXQG WKDW

PAGE 39

WKH OHJXPH SHUFHQWDJH ZDV DQG IRU WKH DQG DQLPDOV KD ? UHVSHFWLYHO\ ZKHQ LW ZDV PHDVXUHG ZLWK WKH SRLQW TXDGUDW PHWKRG -HQVHQ DQG 6FKXPDFKHU f REVHUYHG WKDW WKH SHUFHQWDJH RI ERWDQLFDO VSHFLHV LV QRW RQO\ DIIHFWHG E\ JUD]LQJ DQLPDOV EXW DOVR E\ HQYLURQPHQWDO FRQGLWLRQV VXFK DV VHDVRQ SUHFLSLWDWLRQ UDLQ GLVWULn EXWLRQ WHPSHUDWXUH DQG VRODU UDGLDWLRQ 7DNLQJ LQWR FRQVLGHUDWLRQ PRVW RI WKH DERYH IDFWRUV 7RWKLOO f QRWHG WKDW LI WKH SULPDU\ REMHFWLYH RI WKH LQYHVWLJDWLRQ LV WR REWDLQ DQ HVWLPDWH RI ERWDQLFDO FRPSRVLWLRQ RI SDVWXUHV LQ DVVHVVLQJ DQLPDO SURGXFWLRQ ZHLJKW RI VSHFLHV LV WKH PRVW VXLWDEOH YDOXH WR PHDVXUH ,I UDLQIDOO LQWHUFHSWLRQ RU SKRWRV\QWKHVLV DUH XQGHU VWXG\ WKHQ FRYHU PD\ EH WKH PRUH DSSURSULn DWH SDUDPHWHU +H DOVR PHQWLRQHG WKDW WKH LPSRUWDQFH RI WKLV GLVWLQFn WLRQ LV WKDW QXPEHU ZHLJKW DQG FRYHU PHDVXUHPHQWV DUH FRPSDUDEOH LQ WLPH DQG VSDFH EXW DUH LQGHSHQGHQW RI WKH PRGH RI VDPSOLQJ VLQFH WKH\ DUH PHDVXUHG GLUHFWO\ DQG H[SUHVVHG LQ UHODWLRQ WR D XQLW DUHD 6KDZ DQG %U\DQ f PHQWLRQHG WKDW WKH SURSRUWLRQ RI VSHFLHV RQ D ZHLJKW EDVLV LV JHQHUDOO\ WKH PRVW XVHIXO ZKHUH WKH PDLQ LQWHUHVW LV LQ SDVWXUH SURGXFWLRQ DQG ZKHUH VDPSOHV DUH FXW IRU \LHOG GHWHUPLQDWLRQ DQG ERWDQn LFDO FRPSRVLWLRQ FDQ EH GHWHUPLQHG E\ KDQGVHSDUDWLQJ WKH VDPSOH LQWR FRPSRQHQW VSHFLHV 7KH\ DOVR DGGHG WKDW WKLV LV WKH PRVW SUHFLVH DQG VDWLVIDFWRU\ PHWKRG IRU \LHOG DQG ERWDQLFDO FRPSRVLWLRQ GHWHUPLQDWLRQV 5HVSRQVH 6XUIDFH 0HWKRGRORJ\ /LWWHOO DQG 0RWW f LQGLFDWHG WKDW WKH SXUSRVH RI UHVSRQVH VXUIDFH PHWKRGRORJ\ LV WR HVWLPDWH WKH IXQFWLRQDO UHODWLRQVKLSEHWZHHQ D UHVSRQVH YDULDEOH VXFK DV \LHOG DQG DQ H[SHULPHQWDO YDULDEOH RU

PAGE 40

FRQWURO YDULDEOHV VXFK DV UDWHV RI 3 7KH\ DOVR VXJJHVWHG WKDW WKH UDQJH RI YDOXHV GHWHUPLQHV WKH H[SHULPHQWDO UHJLRQ DQG WKH IXQFn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f 0DUDVFKLQ f DQG 6HUUDR f ERWK XVLQJ D FHQWUDO FRPSRVLWH GHVLJQ VWXGLHG WKH HIIHFW RI WKUHH YDULDEOHV JUD]LQJ GD\V DQG f UHVWLQJ SHULRGV DQG GD\Vf DQG GU\ PDWWHU UHVLGXH OHIW DIWHU JUD]LQJ DQG NJ '0 KD Af XSRQ WKH ERWDQLFDO FRPSRVLWLRQ RI &\QRGRQ GDFW\ORQ 'HVPRGLXP LQWRUWXP0DFURSWLOLXP DWURSXUSXUHXP/RWRQRQLV EDLQHVLL 7ULIROLXP UHSHQV PL[WXUH 7KH JUHDW DGYDQWDJH RI WKLV GHVLJQ LV WKDW LW XVHG RQO\ WUHDWPHQWV LQVWHDG RI WKDW WKH FRPSOHWH IDFWRULDO ZRXOG KDYH UHTXLUHG IRU D VLQJOH UHSOLFDWLRQ WR REWDLQ WKH FRHIILFLHQWV QHHGHG WR HYDOXDWH WKH JUD]LQJ PDQDJHPHQW V\VWHPV 6HUUDR f UHn SRUWHG WKDW WKH PRVW LPSRUWDQW IDFWRUV LQ GHWHUPLQLQJ GU\ PDWWHU \LHOG IURP SDVWXUH DQG DOVR OHJXPH SHUFHQWDJH PDLQWDLQHG LQ WKH PL[WXUH ZHUH JUD]LQJ SUHVVXUH DQG UHVW SHULRGV +H IRXQG WKDW WKH OHJXPH FRQWHQW RI WKH PL[WXUH ZDV LQFUHDVHG ZLWK DQ LQFUHDVH LQ OHQJWK RI UHVW SHULRG

PAGE 41

DQG WKDW KHDY\ JUD]LQJ SUHVVXUH DQG VKRUW UHVW SHULRGV DOPRVW HOLPLn QDWHG WKH OHJXPHV IURP WKH SDVWXUH

PAGE 42

&+$37(5 ,,, 0$7(5,$/6 $1' 0(7+2'6 7KLV UHVHDUFK ZDV FRQGXFWHG DW (VWDFLQ ([SHULPHQWDO 7URSLFDO 3LFKLOLQJXH EHORQJLQJ WR ,1,$3 DQG ORFDWHG NP IURP &DQWRQ 4XHYHGR 3URFLQFLD GH ORV 5LRV DW ,R n 6 /DW DQG r n : /RQJ $OWLWXGH DW WKH VLWH LV P DERYH VHD OHYHO 7KH DYHUDJH PLQLPXP DQG PD[LPXP WHPSHUDWXUHV DUH DQG r& UHVSHFWLYHO\ KDYLQJ D PHDQ DQQXDO WHPSHUDWXUH RI r& $QQXDO SUHFLSLWDWLRQ LV PP $ERXW b RI WKH \HDUO\ UDLQIDOO RFFXUV GXULQJ WKH ZDUPHU PRQWKV IURP 'HFHPEHU WKURXJK -XQH 7KH PRQWKV RI )HEUXDU\ DQG 0DUFK KDYH WKH KLJKHVW UDLQIDOO LQWHQVLW\ %\ FRQWUDVW 2FWREHU DQG 1RYHPEHU DUH WKH GULHVW PRQWKV RIWHQ UHJLVWHULQJ QR UDLQIDOO $QQXDO PHDQ UHODn WLYH KXPLGLW\ LV b DQG PHDQ VXQOLJKW LV KRXUV SHU \HDU b RI ZKLFK RFFXUV GXULQJ WKH 'HFHPEHU WR 0D\ SHULRG 6HUYLFLR 1DFLRQDO GH +LGURORJD \ 0HWHURORJLD GHO (FXDGRU f +ROGULGJH f ORFDWHG 3LFKLOLQJXH LQ WKH WURSLFDO PRLVW IRUHVW ]RQH )LJ f )LJXUHV DQG SUHVHQW WKH PRQWKO\ UDLQIDOO UDQJH DYHUDJH WHPSHUDWXUH DQG DYHUDJH VRODU UDGLDWLRQ IRU WKH \HDUV DQG 7KH VRLOV DUH FODVVLILHG DV 7RUULSVDPPHQWV +DUG\ f VWDWHV WKDW WKH FKHPLFDO DQDO\VLV RI WKLV 3LFKLOLQJXH ORDP UHYHDOV D PDUNHG GHILFLHQF\ LQ DYDLODEOH 3 EXW DQ DEXQGDQFH RI DYDLODEOH 7KH VRLO 1 VWDWXV LV IDLU WR PHGLXP LQ UHFHQWO\ FOHDUHG ODQG EXW GHFOLQHV UDSLGO\ ZLWK FXOWLYDWLRQ 0RUH UHFHQW VRLO DQDO\VHV DW WKH H[SHULn PHQWDO VLWH UHYHDOHG WKDW WKH DPRXQW RI 3 LV PHGLXP ZKLOH % 6 =Q DQG 0R DUH ORZ ,1,$3 f

PAGE 43

RQ +r 72 U_ 7 L +r 2 f +L 3$&,),& 2&($1 f5< 6$9$11$+ R Q! LL U + UI P IL ,f§_ &' 7! & UU & UU UL Wf§ 7 UM 72 UU ,f§r & 2 U &2 0 2 F &/ 2 LL FQ UU R 6O W/ 72 72 1 2 &IO +r '5< 7523,&$/ )25(67 PPf LQf :(7 7523,&$/ )25(67 OPP LQf :(7 68%7523,&$/ )25(67 $ PP LQf '5<:(7 &2/' *5$66/$1' 9(*(7$7,21 9(5< &2/' 781'5$ &+,0%25$=2 '5< &2/' *5$66/$1'6 L 1$7,9(f PP LQf 5<:(7 +,*+ *5$66/$1'6 n&8/7,9$7('f '5< 9$//(<6 1$7,9( 6 &8/7,9$7(' *5$66(6f '5< &2/' *5$66/$1'6 1$7,9(f 781*85$*8$ :(7 &2/' *5$66/$1'6 1$7,9(f 9(5< :(7 68%7523,&$/ )25(67 PPf LQf 9(5< :(7 7523,&$/ )25(67 PP LQf RM A XL R R R 2 R R A 2 R R R 2&2 R $ R EERERQ_ =

PAGE 44

5DLQIDOO PPf )LJ 5DLQIDOO UHFRUGHG DW (VWDFLQ ([SHULPHQWDO 3LFKLOLQJXH GXULQJ WKH SHULRG RI X! R

PAGE 45

7HPSHUDWXUH r&f )LJ 7HPSHUDWXUH UHFRUGHG DW (VWDFLQ ([SHULPHQWDO 3LFKLOLQJXH GXULQJ WKH SHULRG X!

PAGE 46

+RXUV )LJ 6RODU UDGLDWLRQ DW (VWDFLQ ([SHULPHQWDO 3LFKLOLQJXH GXULQJ WKH SHULRG RI /2 1

PAGE 47

7DEOH SUHVHQWV WKH VRLO DQDO\VLV RQ WKH H[SHULPHQWDO VLWH XVHG LQ WKLV UHVHDUFK /HJXPH*UDVV 0L[WXUH 7KH OHJXPHJUDVV PL[WXUH VHOHFWHG IRU WKH H[SHULPHQW FRQVLVWHG RI 0DODZL JO\FLQH >1HRQRWRQLD ZLJKWLL 5 *UDK [ :LJKWLL DQG $UQf /DFNH\@ FRPPHUFLDO FHQWUR &HQWURVHPD SXEHVFHQV %HQWKf JXLQHDJUDVV 3DQLFXP PD[LPXP -DFTf DQG +\EULG HOHSKDQWJUDVV 3HQQLVHWXP SXUSXUHXP 6FKXPDFKf *XLQHDJUDVV DQG HOHSKDQWJUDVV ZHUH FRQVLGHUHG EHFDXVH ERWK DUH YHU\ FRPPRQ DQG XVHIXO JUDVV VSHFLHV LQ WKH OLWWRUDO UHJLRQ RI (FXDn GRU DQG DOVR EHFDXVH RI WKHLU KLJK\LHOGLQJ FDSDFLW\ GURXJKW WROHUDQFH DQG DELOLW\ WR JURZ LQ PL[WXUH ZLWK VRPH WURSLFDO OHJXPHV &HQWUR LV D QDWLYH OHJXPH DQG PD\ EH WKH PRVW ZLGHVSUHDG LQ WKH ORZODQGV H[n WHQGLQJ IURP GU\ WURSLFDO IRUHVW WR ZHW WURSLFDO IRUHVW PP RI UDLQIDOOf *O\FLQH KDG EHHQ VHOHFWHG IURP SUHYLRXV H[SHULPHQWV DV D YHU\ SHUVLVWHQW DQG SURGXFWLYH OHJXPH ([SHULPHQWDO 9DULDEOHV 7KH H[SHULPHQWDO YDULDEOHV ZHUH f 'D\V JUD]LQJ ;Af§ f 'D\V UHVW ;Aff§ f *UD]LQJ SUHVVXUH ;Aff§GU\ PDWWHU RQ RIIHU SHU NJ ERG\ ZHLJKW %:f DQG f 3KRVSKRUXV 3RBf OHYHOV ;f

PAGE 48

7DEOH 6RLO DQDO\VLV RI H[SHULPHQWDO VLWH f 'HSWK RI VDPSOH FP S+ 3HUFHQW RUJDQLF PDWWHU (OHPHQW SSP 3KRVSKRUXV 3RWDVVLXP 6XOIXU =LQF 0DQJDQHVH &RSSHU %RURQ

PAGE 49

)RU WKLV H[SHULPHQW JUD]LQJ SUHVVXUH ZDV H[SUHVVHG DV WKH DPRXQW RI GU\ PDWWHU RQ RIIHU SHU NJ ERG\ ZHLJKW %:f DQG E\ WKH UHVLGXDO GU\ PDWWHU LQ NJ KD OHIW DIWHU HDFK JUD]LQJ (DFK H[SHULPHQWDO YDULDEOH ZDV VWXGLHG DW ILYH OHYHOV 7KHUHn IRUH WKH WUHDWPHQWV FRPSULVHG D IDFWRULDO W\SH RI H[SHULPHQW RI IRXU IDFWRUV HDFK DW ILYH OHYHOV IDFWRULDOf 7DEOH f ([SHULPHQWDO 'HVLJQ 'XH WR WKH ODUJH QXPEHU RI H[SHULPHQWDO XQLWV UHTXLUHG WR FRQGXFW D IDFWRULDO WUHDWPHQW FRPELQDWLRQV ZLWKRXW UHSOLFDn WLRQVf D UHVSRQVH VXUIDFH GHVLJQ QDPHO\ D PRGLILHG FHQWUDO FRPSRn VLWH QRQURWDWDEOH GHVLJQ ZDV XVHG 7KH QXPEHU RI GHVLJQ SRLQWV WUHDWPHQW FRPELQDWLRQVf ZDV GHWHUPLQHG IURP WKH IROORZLQJ IRUPXOD VHH 7DEOH f 1R RI GHVLJQ SRLQWV sf sf [ f &HUWDLQ WUHDWPHQWV ZHUH UHSOLFDWHG WZLFH FHQWUDO SRLQW ZDV UHSOLFDWHG WKULFHf DQG WKHVH DUH LQGLFDWHG LQ 7DEOH 7KH WRWDO QXPEHU RI H[SHULPHQWDO XQLWV ZDV )LHOG 3ODQ RI WKH ([SHULPHQW ,Q RUGHU WR HVWLPDWH WKH VL]H RI H[SHULPHQWDO XQLWV IRU HDFK WUHDWPHQW WKH IROORZLQJ IRUPXOD ZDV XVHG VHH 7DEOH f 1G5 E '* ZKHUH 6 VL]H RI H[SHULPHQWDO XQLW LQ P 1 NJ ERG\ ZHLJKWSDVWXUHGD\ DVVXPHG NJ %:f G QXPEHU RI GD\V SDVWXUH LV JUD]HG GXULQJ F\FOH

PAGE 50

7DEOH 0RGLILHG FHQWUDO FRPSRVLWH QRQURWDWDEOH GHVLJQ ZLWK IRXU H[SHULPHQWDO ;f YDULDEOHV DW ILYH OHYHOV HDFK DQG GHVLJQ SRLQWV 7UHDWPHQWV 5HSV 'D\V LQ F\FOH 1R ; &RGHG ; ; ; 'D\V *UD]LQJ [f 'D\V 5HVW [f *UD]LQJ 3UHVVXUH ;f b %:5 )HUWLOL]HU /HYHO ;Af NJ KD &RQW &RQW &RQW &RQW X! 21

PAGE 51

7DEOH f§FRQWLQXHG 7UHDWPHQWV 5HSV 'D\V LQ F\FOH 1R ; &RGHG ; ; ; 'D\V *UD]LQJ [f 'D\V 5HVW [f *UD]LQJ 3UHVVXUH ;f b %:5 )HUWLOL]HU /HYHO ;$f NJ KD &RQW &RQW &RQW &RQW &RQW /2

PAGE 52

7DEOH 0RGLILHG FHQWUDO FRPSRVLWH QRQURWDWDEOH GHVLJQ ZLWK IRXU H[SHULPHQWDO ;f YDULDEOHV DW ILYH OHYHOV HDFK GHVLJQ SRLQWV H[SHULPHQWDO XQLWV ZLWK WKHLU UHVSHFWLYH DUHD 7UHDWPHQWV 5HSV 6L]H RI ([S 8QLW 7RWDOW $UHD 5HTXLUHG SHU 7UHDW 1R 3DVWXUH 1R 'D\V *UD]LQJ [[f 'D\V 5HVW [f *UD]LQJ 3UHVVXUH ;f b %: )HUWLOL]HU /HYHO ;Af NJ KD

PAGE 53

7DEOH f§FRQWLQXHG 7UHDWPHQWV 5HSV 6L]H RI ([S 8QLW 6 7RWDOI $UHD 5HTXLUHG SHU 7UHDW 1R 3DVWXUH 1R 'D\V *UD]LQJ [[f 'D\V 5HVW [f *UD]LQJ 3UHVVXUH ; f b %: )HUWLOL]HU /HYHO ;Af NJ KD I7RWDO DUHD P

PAGE 54

5 NJ GU\ PDWWHU RIIHUHGNJ ERG\ ZHLJKWGD\ QXPEHU RI GD\V LQ F\FOH DQG JURZWK UDWH LQ NJP GD\ DVVXPHG NJP GD\f $V DQ H[DPSOH WUHDWPHQW GD\V RI JUD]LQJ GD\V RI UHVW DQG NJ '0 RQ RIIHU NJ %:GD\ LV FDOFXODWHG DV IROORZV f [ [ RUUL 6 f7R $aQZ P [ 7KLV ILJXUH ZDV URXQGHG WR P WR DYRLG WKH LQFRQYHQLHQFH RI RGG SDVWXUH VL]HV $OO SDVWXUHV ZHUH P RU VRPH PXOWLSOH 7DEOH f ,Q PDNLQJ WKLV FDOFXODWLRQ LW ZDV DVVXPHG WKDW RQH DQLPDO ZHLJKLQJ DSSUR[LPDWHO\ NJ ZDV XVHG WR JUD]H WKH SDVWXUH GXULQJ WKH DOORWWHG JUD]LQJ SHULRG DQG GHVLJQDWHG JUD]LQJ SUHVVXUH ,W ZDV DOVR DVVXPHG WKDW WKH JURZWK UDWH ZDV JUD GD\ ,Q HVWLPDWLQJ WKH VL]H 6f RI WKH H[SHULPHQWDO XQLW WKH VPDOOHVW ZDV VHW DW P LQ RUGHU WR DYRLG GLIILFXOW\ LQ KDQGOLQJ WKH VWHHUV LQ XQLWV RI VPDOOHU VL]H $ WRWDO RI KHFWDUHV ZHUH UHTXLUHG IRU WKH H[SHULPHQWDO XQLWV LQ WKLV VWXG\ 7DEOH f 7KHVH SDVWXUHV ZHUH UDQGRPO\ GLVWULEXWHG DV VKRZQ LQ )LJ 7KH VL]HV RI WKH H[SHULPHQWDO SDVWXUHV YDULHG IURP P WR P WKH ODUJHU DUHDV ZHUH IRU WKH FRQWLQXRXVO\ JUD]HG WUHDWPHQWV )LJ f /DQG 3UHSDUDWLRQ DQG 3DVWXUH (VWDEOLVKPHQW /DQG SUHSDUDWLRQ RI WKH H[SHULPHQWDO DUHD EHJDQ LQ 0D\ DIWHU WKH H[LVWLQJ YHJHWDWLRQ FRPPRQ JXLQHDJUDVVf ZDV SDUWLDOO\ HOLPLQDWHG E\ DQ DSSOLFDWLRQ RI D JUDVV NLOOHU KHUELFLGH *O\IRVDWHpf $W WKH EHJLQQLQJ RI -XO\ WKH \HDUROG JXLQHDJUDVV SDVWXUH ZDV SORZHG XQGHU

PAGE 55

)LJ )LHOG SODQ RI WKH H[SHULPHQWDO SDVWXUHV

PAGE 56

,Q $XJXVW WKH FRPSOHWH DUHD ZDV FURVVGLVFHG LQ RUGHU WR HOLPLQDWH VRPH ZHHGV DQG REWDLQ D JRRG VHHGEHG 7KH DUHD ZDV GLVFHG DJDLQ LQ ODWH 6HSWHPEHU WR HOLPLQDWH WKH QHZ JURZWK DQG VRPH ZHHGV ,Q WKH VDPH PRQWK VPDOO VRLO VDPSOHV ZHUH WDNHQ IURP WKH ZKROH DUHD DQG PL[HG WRJHWKHU IRU VRLO DQDO\VLV WKH VDPSOHV ZHUH WDNHQ DW FP GHSWK VHH 7DEOH f 7KH QH[W VWHS ZDV WR SUHSDUH WKH OHJXPH VHHGV *O\FLQH DQG FHQWUR VHHGV ZHUH VFDULILHG XVLQJ VXOIXULF DFLG IRU D WLPH RI DQG PLQXWHV UHVSHFWLYHO\ $IWHU WKH VHHGV ZHUH ZDVKHG DQG VKDGH GULHG WKH\ ZHUH LQRFXODWHG XVLQJ DQ DSSURSULDWH 5KL]RELXP VWUDLQ REWDLQHG IURP &,$7 &HQWUR ,QWHUQDFLRQDO GH $JULFXOWXUD 7URSLFDOf ,Q HDUO\ 2FWREHU RI D PL[WXUH RI JO\FLQH DQG FHQWUR ZDV VRZQ XVLQJ D VPDOO JUDLQ VHHGHU ZLWK D IRXU URZ FDSDFLW\ DW WKH UDWHV RI DQG NJ KD UHVSHFWLYHO\ 7KH PL[WXUH ZDV VRZQ LQ URZV VSDFHG P DSDUW $ GD\ DIWHU WKH VRZLQJ D PL[WXUH RI SUHHPHUn JHQFH KHUELFLGHV ZDV DSSOLHG /LQXURQp NJ KD A / KD A RI /DVRpf WR FRQWURO ERWK JUDVVHV DQG EURDGOHDI ZHHGV FRPLQJ IURP VHHG 7ZR ZHHNV DIWHU WKH OHJXPHV ZHUH VRZQ WKH JUDVVHV ZHUH SODQWHG YHJHWDWLYHO\ XVLQJ SODQW GLYLVLRQV IRU JXLQHDJUDVV DQG VPDOO SLHFHV RI VWHPV ZLWK WZR RU WKUHH EXGV IRU HOHSKDQWJUDVV %RWK JUDVVHV ZHUH SODQWHG DOWHUQDWLYHO\ EHWZHHQ WKH URZV RI WKH OHJXPHV VR WKDW WKH GLVWDQFH EHWZHHQ JUDVVHV DQG OHJXPHV ZDV P ,W ZDV QHFHVn VDU\ WR PDNH RQH KDQG ZHHGLQJ XVLQJ PDFKHWHV )RXU LUULJDWLRQV RI PP HDFK ZHUH QHFHVVDU\ DW GD\ LQWHUYDOV 7KH ILUVW LUULJDWLRQ ZDV LPPHGLDWHO\ DIWHU WKH OHJXPHV ZHUH VRZQ

PAGE 57

,Q HDUO\ -DQXDU\ RI WKH 3 IHUWLOL]HU ZDV DSSOLHG WDNLQJ LQWR FRQVLGHUDWLRQ WKH UHVSHFWLYH OHYHOV RI VLPSOH VXSHUSKRVSKDWH NJ KD A ZKRVH FRPSRVLWLRQ ZDV b AA DQFr b HOHPHQWDO 6f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n ORZLQJ IRUPXOD b RI VRGLXP FKORULGH b RI PLQHUDOL]HG VDOW FRQWDLQLQJ b 3 DV GLFDOFLXP SKRVSKDWH b =LQF VXOIDWH b PDQJDQHVH FDUERQDWH b FRSSHU VXOIDWH b IHUURXV VXOIDWH DQG b FREDOW FKORULGH :DWHU WDQNV ZHUH SURYLGHG LQ HDFK H[SHULn PHQWDO SDVWXUH $Q DUHD RI DSSUR[LPDWHO\ KD DGMDFHQW WR WKH H[SHULn PHQW ZDV DYDLODEOH DV UHVHUYH SDVWXUH IRU DQLPDOV XVHG IRU DGMXVWLQJ WKH VWRFNLQJ UDWHV LQ WKH H[SHULPHQW &ROOHFWLRQ RI 'DWD LQ WKH 7KUHH ([SHULPHQWDO
PAGE 58

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f DQG DIWHU UHVLGXDOf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

PAGE 59

GHSHQGLQJ RQ WKH VSHFLILHG GU\ PDWWHU RQ RIIHU DQG GD\V RI JUD]LQJ ,Q RUGHU WR HVWLPDWH WKH JUD]LQJ SUHVVXUH WKH IROORZLQJ IRUPXOD ZDV XVHG *3 $ U JGf 6 51G ZKHUH *3 JUD]LQJ SUHVVXUH LQ WHUPV RI NJ %: $ DYDLODEOH IRUDJH EHIRUH JUD]LQJ U UHVLGXH IRU D VSHFLILHG JUD]LQJ SUHVVXUH J JURZWK UDWHSUHFHGLQJ UHVW SHULRG G QXPEHU RI GD\V IRU JUD]LQJ SHULRG 6 VL]H RI WKH H[SHULPHQWDO XQLW LQ P 5 NJ GU\ PDWWHU RIIHUHGNJ ERG\ ZHLJKWGD\ DQG 1 NJ ERG\ ZHLJKWSDVWXUHGD\ $ GRXEOH VDPSOLQJ WHFKQLTXH ZDV HPSOR\HG $V VRRQ DV SRVVLEOH EHIRUH DQG DIWHU HDFK JUD]LQJ GXULQJ WKH ILUVW PRQWKV RQO\f DUHDV PHDVXULQJ P ZHUH UDQGRPO\ VHOHFWHG )URP 6HSWHPEHU DUHD PHDVXUHPHQWV RI WKH VDPH VL]H ZHUH WDNHQ FRUUHVSRQGLQJ WR WKH FLUFXODU IUDPH RI D IRUDJH GLVN PHWHU RQH P f VLPLODU WR WKH RQH GHVFULEHG E\ %UDQVE\ f DQG XVHG LQ GRXEOHVDPSOLQJ PHDVXUHPHQWV ,Q HDFK VDPSOLQJ XQLW WKH SHUFHQW GU\ PDWWHU \LHOG RI HDFK FRPSRQHQW RI WKH PL[WXUH ZDV YLVXDOO\ HVWLPDWHG IROORZHG E\ DQ HVWLPDWH RI WKH GU\ PDWWHU \LHOG LQ VLWX 7KH GLVN PHWHU ZDV ORZHUHG RQ WKH IRUDJH DQG DIWHU D VHWWOLQJ WLPH RI DSSUR[LPDWHO\ WR VHFRQGV WKH GLVN KHLJKW LQ FHQWLPHWHUVf ZDV UHDG RII D JUDGXDWHG VFDOH PRXQWHG RQ WKH FHQWHU VKDIW RI WKH GLVN PHWHU DQG UHFRUGHG 'XULQJ WKH ILUVW WKUHH DQG RQHKDOI PRQWKV IURP WKH VDPSOLQJ XQLWV WKUHH ZHUH UDQGRPO\ VHOHFWHG DQG FOLSSHG DW JURXQG OHYHO IRU

PAGE 60

GU\ PDWWHU \LHOG DQG ERWDQLFDO FRPSRVLWLRQ ,Q 6HSWHPEHU ILYH VDPSOLQJ XQLWV RXW RI GLVN UHDGLQJV ZHUH UDQGRPO\ VHOHFWHG DQG FOLSSHG IRU WKH DERYH GHWHUPLQDWLRQV 7KHVH VDPSOHV ZHUH ODWHU KDQG VHSDUDWHG LQWR WKHLU FRPSRQHQWV SODFHG LQVLGH FORWK EDJV SURSHUO\ LGHQWLILHG DQG GULHG IRU KRXUV DW r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f OHJXPHV FHQWUR DQG JO\FLQHf DQG ZHHGV $OWKRXJK WKH DERYH JUDVVHV DQG RWKHUV VXFK DV 3DVSDOXP IDVFLFXODWXP 3 SDQLFXODWXP DQG (OHXVLQH LQGLFD ZHUH SUHVHQW LQ VFDWWHUHG VPDOO SDWFKHV LQ VRPH SDVWXUHV WKH\ ZHUH LQFOXGHG LQ WKH ZHHG FRPSRQHQWV /LNHZLVH VRPH QDWLYH 'HVPRGLXPV ZHUH DFFRXQWHG IRU LQ WKH OHJXPH FRPSRQHQW 7KH UHVSRQVH RI WKH SDVWXUH PL[WXUH WR WKH H[SHULPHQWDO YDULDEOHV ZDV PHDVXUHG LQ WHUPV RI WKH IROORZLQJ SDUDPHWHUV f $HULDO ELRPDVV '0f NJ KD A JUDVV '0f OHJXPH '0f ZHHGV '0f f $YDLODEOH IRUDJH '0f NJ KD A JUDVV '0f OHJXPH '0f f *UDVV \LHOG '0f

PAGE 61

f /HJXPH \LHOG '0f f
PAGE 62

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f 7KH HIIHFW RI WKH OHQJWKV RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH RQ WKH DHULDO ELRPDVV LV SUHVHQWHG IRU HDFK RI WKH ILYH VHDVRQV LQ 7DEOH 7KH WRWDO ELRPDVV LV WKH DYHUDJH DPRXQW RI GU\ PDWWHU SUHVHQW EHIRUH HDFK JUD]LQJ SHULRG IRU WKH URWDWLRQDO JUD]LQJ WUHDWPHQW FRPELQDWLRQV DQG IRU HDFK JUD]LQJ SHULRG RI GD\V IRU FRQWLQXRXV JUD]LQJ %LRPDVV 3URGXFWLRQ '0f IRU WKH :HW 6HDVRQ RI 'XULQJ WKH ILUVW ZHW VHDVRQ RQO\ OHQJWKV RI UHVW SHULRGV LQGLFDWHG DQ HIIHFW 3 f RQ WKH DHULDO ELRPDVV SURGXFHG $SSHQGL[ 7DEOH f 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDUn LDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f

PAGE 63

7DEOH $HULDO ELRPDVV SURGXFWLRQ '0f E\ \HDU VHDVRQ DQG WUHDWPHQW 7UHDWPHQWV '*W :LW :LW I77 1R ;f ;f ;f b %: NJ KD 5HSV :6 '6 :6 '6 :VW NJ KD

PAGE 64

7DEOH f§FRQWLQXHG 7UHDWPHQWV 1R '*I [f '5I [f *3I ; f b %: )W L NJ KD 5HSV ZVW '6W ZVW '6W ZVW NJ KD I'* GD\V JUD]LQJ '5 GD\V UHVW *3 JUD]LQJ SUHVVXUH DV b ERG\ ZHLJKW ) IHUWLOL]HU :6 ZHW VHDVRQ '6 GU\ VHDVRQ

PAGE 65

7KH ELRPDVV SURGXFWLRQ YDULHG IURP NJ '0 KD IRU WUHDWPHQW GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A VXSHUSKRVSKDWHf WR NJ '0 KD A IRU WUHDWPHQW GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A VXSHUSKRVSKDWHf 7KH DHULDO ELRPDVV SURGXFWLRQ UHVSRQVH WR OHQJWKV RI UHVW SHULRGV FDQ RQO\ EH H[SODLQHG E\ WKH PDQDJHPHQW LPSRVHG GXULQJ WKH ILUVW JUD]LQJ VHDVRQ $OO RI WKH H[SHULPHQWDO SDVWXUHV ZHUH JUD]HG DW D PHGLXP JUDn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f IRU WKH 'U\ 6HDVRQ RI 7KH ELRPDVV SURGXFWLRQ GXULQJ WKH GU\ VHDVRQ RI LV DOVR SUHVHQWHG LQ 7DEOH 7KH ELRPDVV SURGXFWLRQ YDULHG IURP WR NJ '0 KD FRUUHVSRQGLQJ WR WUHDWPHQWV GD\V JUD]LQJ

PAGE 66

GD\V UHVW SHULRG NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf DQG UHVSHFWLYHO\ GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ RI VLPSOH VXSHUSKRVSKDWHf ,Q JHQHUDO WKH ORZHVW YDOXHV RI ELRPDVV SURGXFWLRQ FRUUHVSRQGHG WR WUHDWPHQW FRPELQDWLRQV RI VKRUW UHVW SHULRGV DQG KLJK OHYHOV RI JUD]LQJ SUHVVXUH ZKLOH WKH KLJKHVW OHYHOV RI ELRPDVV SURGXFWLRQ FRUUHVSRQGHG WR WKH WUHDWPHQW FRPELQDWLRQV ZLWK ORQJ UHVW SHULRGV DQG ORZ OHYHOV RI JUD]LQJ SUHVVXUH 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH H[SHULPHQWDO YDULDEOHV OHQJWK RI UHVW SHULRG ;ef DQG OHYHO RI JUD]LQJ SUHVVXUH ;Af HDFK VKRZHG D OLQHDU HIIHFW XSRQ ELRPDVV SURGXFWLRQ 3 f 7KHUH ZDV D VXJJHVWLRQ WKDW OHYHOV RI VXSHUSKRVSKDWH ;Af PLJKW EH KDYLQJ VRPH HIIHFW 3 f EXW WKH HIIHFWV RI GD\V JUD]LQJ ;Af ZDV QLO ,Q DOO FDVHV LQGLYLGXDO TXDGUDWLF HIIHFWV RU LQWHUDFWLRQV ZHUH QRW VLJQLILFDQW 7KH ODFN RI LQWHUDFWLRQV EHWZHHQ ;A DQG ;A LQGLFDWHG WKDW ERWK YDULDEOHV ZHUH EHKDYLQJ LQGHSHQGHQWO\ 7KH ELRPDVV SURGXFWLRQ ZDV LQFUHDVHG DV WKH OHQJWKV RI WKH UHVW SHULRG ZHUH LQFUHDVHG DQG DV JUD]LQJ SUHVVXUH ZDV UHGXFHG 7KH DQDO\VLV RI YDULDQFH IRU WKH GU\ VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH %LRPDVV 3URGXFWLRQ '0f IRU WKH :HW 6HDVRQ RI 7KH ELRPDVV SURGXFWLRQ GXULQJ WKH ZHW VHDVRQ RI IRU HDFK WUHDWPHQW FRPELQDWLRQ LV JLYHQ LQ 7DEOH %LRPDVV SURGXFWLRQ YDULHG

PAGE 67

IURP WR NJ '0 KD ? FRUUHVSRQGLQJ WR WKH WUHDWPHQWV > GD\ JUD]LQJ UHVW SHULRG FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHUORR NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG UHVSHFWLYHO\ GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWHf 'XULQJ WKH ZHW VHDVRQ RI ERWK OHQJWK RI UHVW SHULRG DQG OHYHO RI JUD]LQJ SUHVVXUH KDG DQ HIIHFW XSRQ ELRPDVV SURGXFWLRQ 3 f ZKLOH QR TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV ZHUH QRWHG 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ DQG RQO\ DQG b RI WKH WRWDO YDULDWLRQ ZDV UHSUHVHQWHG E\ WKH TXDGUDWLF DQG LQWHUDFWLRQV $V LV HYLGHQW WKH ELRPDVV SURGXFWLRQ LV LQFUHDVHG DV WKH OHQJWK RI WKH UHVW SHULRG LV LQFUHDVHG 7KH KLJKHU JUD]LQJ SUHVVXUHV DOVR GHFUHDVH WKH ELRPDVV SURGXFWLRQ DQG WKH JUHDWHVW ELRPDVV SURGXFWLRQ ZDV UHDFKHG ZKHQ D ORQJ UHVW SHULRG DQG ORZ JUD]LQJ SUHVVXUH ZHUH LPSRVHG 7KH DQDO\VLV RI YDULDQFH IRU WKH ZHW VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH %LRPDVV 3URGXFWLRQ '0f IRU WKH 'U\ 6HDVRQ RI %LRPDVV SURGXFWLRQ GXULQJ WKH GU\ VHDVRQ RI IRU HDFK WUHDWn PHQW FRPELQDWLRQ LV SUHVHQWHG LQ 7DEOH %LRPDVV SURGXFWLRQ YDULHG IURP WR NJ '0 KD ? FRUUHVSRQGLQJ WR WUHDWPHQWV > GD\V JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf $JDLQ WKH ORZHVW YDOXHV RI ELRPDVV RQ RIIHU KD A FRUUHVSRQGHG WR WUHDWPHQW FRPELQDWLRQV RI VKRUW UHVW SHULRGV RU FRQWLQXRXV JUD]LQJ

PAGE 68

ZLWK WKH KLJKHVW JUD]LQJ SUHVVXUH EXW ERWK H[SHULPHQWDO YDULDEOHV DFWHG LQGHSHQGHQWO\ IURP HDFK RWKHU 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHn VHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $Q H[DPLQDWLRQ RI WKH LQGLYLGXDO OLQHDU HIIHFWV VKRZHG WKDW RQO\ WKH H[SHULPHQWDO YDULDEOHV GD\V UHVW ;f DQG JUD]LQJ SUHVVXUH ;Af ZHUH DIIHFWLQJ ELRPDVV SURGXFWLRQ 3 f 7KH QXPEHU RI GD\V JUD]LQJ ;Af DQG IHUWLOL]HU ;Af KDG QR HIIHFW XSRQ ELRPDVV SURGXFWLRQ )RU HDFK GD\V LQFUHDVH LQ WKH QXPEHU RI GD\V UHVW WKHUH ZDV D ELRPDVV SURGXFWLRQ LQFUHDVH RI NJ '0f KD ? ZKLOH IRU HDFK XQLW NJ '0f GHFUHDVH LQ OHYHO RI JUD]LQJ SUHVVXUH WKHUH ZDV DQ LQFUHDVH RI NJ KD A LQ ELRPDVV SURGXFWLRQ 7KH DQDO\VLV RI YDULDQFH IRU WKH GU\ VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH %LRPDVV 3URGXFWLRQ '0f IRU WKH :HW 6HDVRQ RI 7KH OHQJWKV RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH KDG D GLUHFW HIIHFW XSRQ ELRPDVV SURGXFWLRQ IRU WKH ZHW VHDVRQ RI 7DEOH f 7KHUH ZDV QR LQWHUDFWLRQ EHWZHHQ WKHVH WZR H[SHULPHQWDO YDULDEOHV GXULQJ WKLV ZHW VHDVRQ 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH WRWDO ELRPDVV SURGXFWLRQ YDULHG IURP WR NJ '0 KD A IRU WUHDWPHQWV > GD\V JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG

PAGE 69

GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWHf 'XULQJ WKLV SDUWLFXODU VHDVRQ WKH DPRXQW RI ELRPDVV SURGXFWLRQ ZDV DJDLQ PRVW DIIHFWHG E\ WKH OHQJWK RI UHVW SHULRG DQG E\ WKH OHYHOV RI JUD]LQJ SUHVVXUH 7KH KLJKHVW YDOXHV RI ELRPDVV SURGXFWLRQ FRUUHVSRQGHG WR WKH WUHDWPHQW FRPELQDWLRQV RI ORQJ UHVW SHULRGV DQG ORZ JUD]LQJ SUHVVXUHV )RU HDFK GD\V LQFUHDVH LQ UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ '0 KD A RI ELRPDVV SURGXFn WLRQ ZKLOH IRU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH WKHUH ZDV D XQLW LQFUHDVH LQ ELRPDVV SURGXFWLRQ 7KH DQDO\VLV RI YDULDQFH IRU WKH ZHW VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH 6XPPDU\ RI %LRPDVV 3URGXFWLRQ '0f ,Q FRPSDULVRQ RI DHULDO ELRPDVV SURGXFWLRQ '0f DPRQJ VHDVRQV DQG \HDUV ODUJH GLIIHUHQFHV HVSHFLDOO\ EHWZHHQ ZHW DQG GU\ VHDVRQV DUH REVHUYHG 7KH SULQFLSDO HQYLURQPHQWDO IDFWRUV LQYROYHG LQ WKHVH GLIIHUHQFHV EHWZHHQ VHDVRQV DUH SUHFLSLWDWLRQ WHPSHUDWXUH DQG VRODU UDGLDWLRQ VHH )LJ RI 0DWHULDOV DQG 0HWKRGVf (YHU\ \HDU WKH UDLQ\ VHDVRQ EHJLQV LQ WKH VHFRQG KDOI RI 'HFHPEHU LQLWLDOO\ ZLWK OLJKW VKRZHUV DQG WKHQ LQFUHDVLQJ LQ DPRXQW DQG LQWHQVLW\ UHDFKLQJ WKH KLJKHVW SHDN RI SUHFLSLWDWLRQ LQ )HEUXDU\ RU 0DUFK $IWHU WKLV WLPH WKH DPRXQW DQG GXUDWLRQ RI WKH VKRZHUV GHFUHDVHV XQWLO LW UHDFKHV DOPRVW ]HUR GXULQJ WKH VHFRQG KDOI RI -XQH $ERXW b RI WKH WRWDO SUHFLSLWDWLRQ IDOOV IURP 'HFHPEHU WR -XQH ZLWK WKH UHPDLQLQJ PRQWKV DOPRVW FRPSOHWHO\ GU\ 7KH VHFRQG LPSRUWDQW IDFWRU LV WHPSHUDWXUH ZKLFK LV DOZD\V KLJKHU GXULQJ WKH UDLQ\ VHDVRQ UHDFKLQJ DQ DYHUDJH RI r& GXULQJ WKH ZHW VHDVRQ ZKLOH WKH DYHUDJH GXULQJ WKH GU\ VHDVRQ

PAGE 70

LV r & 7KH WKLUG LPSRUWDQW IDFWRU LV VRODU UDGLDWLRQ ZKLFK LV DOVR KLJKHU GXULQJ WKH UDLQ\ VHDVRQ UHDFKLQJ YDOXHV RI KRXUV RI VXQOLJKW ZKLOH IURP -XO\ WR 1RYHPEHU WKLV YDOXH DYHUDJHV KRXUV RI VXQOLJKW 7KHVH WKUHH IDFWRUV DFWLQJ WRJHWKHU DUH WKH PDLQ GHWHUPLQDQWV RI WRWDO DHULDO ELRPDVV SURGXFWLRQ $QRWKHU YHU\ LPSRUWDQW IDFWRU ZKLFK PXVW EH WDNHQ LQWR FRQVLGHUDn WLRQ LV QLWURJHQ 7KLV HOHPHQW DW WKH EHJLQQLQJ RI WKH ZHW VHDVRQ LV UDSLEO\ PRELOL]HG IURP WKH VRLO RUJDQLF PDWWHU ZKLFK KDV DFFXPXn ODWHG LQ WKH VRLO GXULQJ WKH GU\ VHDVRQ 6DQFKH] f KDV LQGLFDWHG WKDW 1 LQ WKH WURSLFV LV UDSLGO\ XWLOL]HG E\ DFWLYHO\ JURZLQJ SODQWV ZLWK WKH RFFXUUHQFH RI WKH ILUVW UDLQV DW WKH EHJLQQLQJ RI WKH ZHW VHDVRQ EXW WKDW WKH UHPDLQLQJ 1 LV VORZO\ UHOHDVHG GXULQJ WKH UHVW RI WKH VHDVRQ 7KHVH WKUHH HQYLURQPHQWDO IDFWRUV SOXV WKH DYDLODELOLW\ RI 1 KDYH D JUHDW LPSDFW XSRQ WKH DHULDO ELRPDVV SURGXFWLRQ 7KH PRVW LPSRUWDQW H[SHULPHQWDO YDULDEOHV LQIOXHQFLQJ WKH DHULDO ELRPDVV SURGXFWLRQ ZHUH GD\V RI UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af (DFK YDULDEOH EHKDYHG LQGHSHQGHQWO\ DQG WKHUH ZDV D GLUHFW UHODWLRQVKLS EHWZHHQ WKH QXPEHU RI GD\V UHVW ZKLFK SHUPLWWHG WKH SODQWV D EHWWHU RSSRUWXQLW\ WR DFFXPXODWH UHVHUYHV IRU PRUH UDSLG UHJURZWK DQG GHYHOn RSPHQW DIWHU HDFK JUD]LQJ SHULRG $ VLPLODU WUHQG ZDV DOVR IRXQG IRU JUD]LQJ SUHVVXUH ZKLFK ZDV GLUHFWO\ UHODWHG WR WKH DPRXQW RI IRUDJH UHPRYHG E\ WKH JUD]LQJ DQLPDOV ZKLFK LQ WXUQ ZDV UHODWHG WR WKH KLJK UHPRYDO DQG GDPDJH WR WKH JURZLQJ SRLQWV DQG D[LOODU\ EXGV *UD]LQJ SUHVVXUH ZDV UHODWHG WR WKH UDWH RI UHFRYHU\ DQG VXEVHTXHQWO\ WKH ILQDO SURGXFWLRQ +RGJVRQ DQG 2OOHUHQVKDZ f PHQWLRQHG WKDW LI JUD]LQJ SUHVVXUH LV LQFUHDVHG ZKLOH UHVWLQJ

PAGE 71

SHULRGV DUH GHFUHDVHG WKH IUHTXHQF\ DQG VHYHULW\ RI GHIROLDWLRQ LV LQFUHDVHG DIIHFWLQJ GLUHFWO\ WKH VXEVHTXHQW UHJURZWK DQG WKH WRWDO GU\ PDWWHU \LHOGV +DUULV f UHSRUWHG WKDW DOWKRXJK WKH IXQFWLRQ RI UHVHUYHV DYDLODELOLW\ RI JURZLQJ SRLQWV DQG XSWDNH FKDUDFWHULVWLFV DUH LQIOXHQFHG E\ WKH OHYHO RI VWXEEOH ELRPDVV WKH UHODWLRQVKLS EHWZHHQ VWXEEOH ELRPDVV DQG JURZWK UDWH UHODWHV WR WKH DPRXQW RI SKRWRV\QWKHWLF WLVVXHV (IIHFW RI /HQJWKV RI 5HVW 3HULRG DQG /HYHOV RI *UD]LQJ 3UHVVXUH 8SRQ $YDLODEOH )RUDJH '0f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f 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU RQO\ b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH PHDQV RI WRWDO DYDLODEOH IRUDJH YDULHG IURP NJ '0 KD A WR NJ '0 KD 7DEOH f IRU WUHDWPHQWV GD\V JUD]LQJ GD\V

PAGE 72

7DEOH $YDLODEOH IRUDJH '0f E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ 7UHDWPHQWV '*W '5I *3I )W L NJ KD 1R [f [f ; f b %: 5HSV :6 '6 :6 '6 :6 NJ KD

PAGE 73

7DEOH f§FRQWLQXHG 7UHDWPHQWV '*I [[f '5I [f *3I ; f b %: Q L NJ KD 1R 5HSV ZVW '6 ZVW '6W ZVW NJ KD W'* GD\V JUD]LQJ '5 GD\V UHVW *3 JUD]LQJ SUHVVXUH DQG ) IHUWLOL]HU :6 ZHW VHDVRQ '6 GU\ VHDVRQ

PAGE 74

UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf DQG GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ &RQVLGHULQJ WKH LQGLYLGXDO OLQHDU HIIHFWV GD\V UHVW ;Af ZDV WKH RQO\ H[SHULPHQWDO YDULDEOH DIIHFWLQJ DYDLODEOH IRUDJH 3 f 7KH RWKHU WKUHH H[SHULPHQWDO YDULDEOHV GLG QRW DIIHFW WKH DYDLODEOH IRUDJH ,QGLYLGXDO TXDGUDWLF HIIHFWV RU LQWHUDFWLRQV ZHUH QRW VLJQLILFDQW 7KH ODFN RI LQWHUDFWLRQ LQGLFDWHV GD\V UHVW LV DFWLQJ LQGHSHQGHQWO\ RYHU DYDLODEOH IRUDJH ZKLFK PHDQV WKDW WKH DPRXQW RI DYDLODEOH IRUDJH LQFUHDVHV DV WKH OHQJWK RI WKH UHVW SHULRG LQFUHDVHV )RU HDFK GD\V RI LQFUHDVH LQ GD\V UHVW ;Af WKHUH ZDV NJ KD A LQFUHDVH LQ WKH DYDLODEOH IRUDJH $YDLODEOH )RUDJH IRU WKH 'U\ 6HDVRQ RI '0f $YDLODEOH IRUDJH YDULHG IURP NJ '0 KD A WR NJ '0 KD ? FRUUHVSRQGLQJ WR WUHDWPHQW FRPELQDWLRQV > GD\V JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWH@ DQG GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ 'XULQJ WKH ILUVW GU\ VHDVRQ WKH ORZHVW YDOXHV RI DYDLODEOH IRUDJH ZHUH IRXQG LQ WKRVH WUHDWPHQW FRPELQDWLRQV RI VKRUW UHVW SHULRGV RU FRQWLQXRXV JUD]LQJ ZLWK WKH KLJKHVW JUD]LQJ SUHVVXUH ZKLOH WKH KLJKHVW YDOXHV ZHUH WKH ORQJHVW UHVW SHULRGV DQG WKH ORZHVW JUD]LQJ SUHVVXUHV 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f

PAGE 75

$Q H[DPLQDWLRQ RI WKH LQGLYLGXDO OLQHDU HIIHFWV UHYHDOV WKDW RQO\ WKH H[SHULPHQWDO YDULDEOHV GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af KDG DQ\ HIIHFW XSRQ DYDLODEOH IRUDJH 3 f 7KH OHQJWK RI UHVW SHULRG LQFUHDVHG WKH DYDLODEOH IRUDJH DV ZDV DOVR WKH FDVH ZLWK D GHFUHDVH LQ JUD]LQJ SUHVVXUH )RU HDFK LQFUHDVH RI GD\V LQ WKH UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI DYDLODEOH IRUDJH DQG IRU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH WKHUH ZDV D NJ KD A LQFUHDVH LQ DYDLODEOH IRUDJH 7KH DQDO\VLV RI YDULDQFH IRU WKH GU\ VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH $YDLODEOH )RUDJH IRU WKH :HW 6HDVRQ RI '0f 'XULQJ WKH ZHW VHDVRQ RI WKH DYDLODEOH IRUDJH '0f YDULHG IURP NJ '0 KD A WR NJ '0 KD ? FRUUHVSRQGLQJ WR WKH WUHDWn PHQW FRPELQDWLRQV > GD\ JUD]LQJ R GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ 7KH WUHDWPHQW FRPELQDWLRQV ZLWK WKH VKRUWHVW UHVW SHULRGV DQG ORZHVW JUD]LQJ SUHVVXUH OHYHOV \LHOGHG WKH ORZHVW YDOXHV RI DYDLODEOH IRUDJH IRU WKLV VHDVRQ DOVR LW ZDV QRWHG WKDW LQ WKLV FDVH ERWK YDULDEOHV ;A DQG ;A ZHUH DFWLQJ LQGHSHQGHQWO\ 7KHUH ZDV QR HYLGHQFH RI DQ\ LQWHUDFWLRQV RFFXUULQJ 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 2QO\ WKH OLQHDU HIIHFWV RI WKH H[SHULPHQWDO YDULDEOHV ;A DQG ;A KDG

PAGE 76

DQ HIIHFW XSRQ WKH DYDLODEOH IRUDJH 3 f ,QFUHDVLQJ WKH GD\V RI UHVW DQG GHFUHDVLQJ WKH JUD]LQJ SUHVVXUH LQFUHDVHG WKH DYDLODEOH IRUDJH GXULQJ WKH ZHW VHDVRQ RI )RU GD\V LQFUHDVH LQ UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI DYDLODEOH IRUDJH DQG IRU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH WKHUH ZDV D NJ KD A LQFUHDVH LQ DYDLODEOH IRUDJH 7KH DQDO\VLV RI YDULDQFH IRU WKH ZHW VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH $YDLODEOH )RUDJH IRU WKH 'U\ 6HDVRQ RI '0f )RU WKH VHFRQG GU\ VHDVRQ WKH DYDLODEOH IRUDJH IRU HDFK WUHDWn PHQW FRPELQDWLRQ LV SUHVHQWHG LQ 7DEOH 7KH DYDLODEOH IRUDJH YDULHG IURP NJ '0 KD A WR NJ '0 KD ? FRUUHVSRQGLQJ WR WKH IROORZLQJ WUHDWPHQW FRPELQDWLRQV > GD\ JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVn SKDWH@ DQG GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ 2QO\ GD\V UHVW ;f DQG JUD]LQJ SUHVVXUH ;Af VKRZHG DQ\ HIIHFW XSRQ DYDLODEOH IRUDJH 3 f (DFK RI WKHVH WZR H[SHULPHQWDO YDULDEOHV ZDV DFWLQJ LQGHSHQGHQWO\ EHFDXVH QR LQWHUDFWLRQV ZHUH IRXQG 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHn VHQWHG OHVV WKDQ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQn GL[ 7DEOH f )RU HDFK GD\ LQFUHDVH LQ WKH UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI DYDLODEOH IRUDJH DQG IRU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH DQ LQFUHDVH RI NJ KD A RI DYDLOn DEOH IRUDJH ZDV UHDOL]HG 7KH DQDO\VLV RI YDULDQFH IRU WKH GU\ VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH

PAGE 77

$YDLODEOH )RUDJH IRU WKH :HW 6HDVRQ RI '0f 7KH DYHUDJH DPRXQW RI DYDLODEOH IRUDJH '0f IRU HDFK WUHDWPHQW FRPELQDWLRQ GXULQJ WKH ZHW VHDVRQ RI LV SUHVHQWHG LQ 7DEOH 7KH DYDLODEOH IRUDJH YDULHG IURP NJ '0 KD A WR NJ '0 KD ? FRUUHVSRQGLQJ WR WUHDWPHQWV > GD\V JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ 2QO\ WKH WZR H[SHULPHQWDO YDULDEOHV GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af ZHUH IRXQG WR EH DIIHFWLQJ WKH DYDLODEOH IRUDJH 3 f 7KH RWKHU WZR YDULDEOHV DQG WKH LQWHUDFWLRQV ZHUH QRW VLJQLILFDQW 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHn VHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH RWKHU H[SHULPHQWDO YDULDEOHV ;A DQG ;A KDG QR VLJQLILFDQW HIIHFW XSRQ WKH DYDLODEOH IRUDJH $V WKH UHVW SHULRG ZDV LQFUHDVHG E\ GD\V WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI DYDLODEOH IRUDJH 2Q WKH RWKHU KDQG HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH UHVXOWHG LQ DQ LQFUHDVH RI NJ KD A RI DYDLODEOH IRUDJH 7KH DQDO\VLV RI YDULDQFH IRU WKH ZHW VHDVRQ RI LV SUHVHQWHG LQ $SSHQGL[ 7DEOH 6XPPDU\ RI $YDLODEOH )RUDJH '0f $YDLODEOH IRUDJH PD\ EH FRQVLGHUHG D PRUH LPSRUWDQW UHVSRQVH YDULDEOH WKDQ DHULDO ELRPDVV VLQFH DYDLODEOH IRUDJH WDNHV LQWR FRQn VLGHUDWLRQ RQO\ WKH FRPSRQHQWV RI WKH YHJHWDWLRQ FRQVLGHUHG HGLEOH

PAGE 78

E\ WKH JUD]LQJ DQLPDO 7KH DYDLODEOH IRUDJH ZDV HVWLPDWHG E\ GHWHUPLQLQJ WKH ERWDQLFDO FRPSRVLWLRQ RI WKH WRWDO DHULDO ELRPDVV DQG VXPPLQJ WKH JUDVV DQG OHJXPH FRPSRQHQWV 2YHU WKH ILYH VHDVRQV GXULQJ ZKLFK WKLV VWXG\ ZDV FRQGXFWHG WKH DYDLODEOH IRUDJH ZDV JUHDWO\ LQIOXHQFHG E\ WKH HQYLURQPHQWDO IDFWRUV ZKLFK DUH GLVFXVVHG LQ WKH DHULDO ELRPDVV VHFWLRQ 7KH H[SHULPHQWDO YDULDEOHV GD\V UHVW ;f DQG JUD]LQJ SUHVVXUH ;Af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n JURZWK ,W LV VLJQLILFDQW WKDW WKHVH WZR YDULDEOHV VHHP WR DFW LQGHn SHQGHQWO\ DV QR LQWHUDFWLRQV EHWZHHQ WKHP ZHUH IRXQG ZLWKLQ WKH UDQJH RI OHYHOV LQ WKLV H[SHULPHQW 6LPLODULWLHV LQ UHVSRQVHV WR WKH VDPH OHQJWKV RI UHVW SHULRG DQG ZLWK WKH VDPH JUD]LQJ SUHVVXUHV ZHUH REVHUYHG E\ 0DUDVFKLQ f DQG 6HUUDR f ,Q HDFK RI WKHVH WZR FDVHV LQWHUDFWLRQV ZHUH IRXQG EHWZHHQ WKHVH WZR YDULDEOHV ZKHUH PD[LPXP \LHOGV ZHUH REWDLQHG DW KLJK JUD]LQJ SUHVVXUHV LQ FRPELQDWLRQ ZLWK ORQJ UHVW SHULRGV DQG ORZ JUD]LQJ SUHVVXUHV ZLWK VKRUW UHVW SHULRGV 7KH DERYH VWXGLHV ZHUH FRQGXFWHG ZLWK &\QRGRQ GDFW\ORQ DQG 'HVPRGLXP LQWRUWXP ZKLFK KDYH TXLWH GLIIHUHQW JURZWK KDELWV WKDQ WKH JUDVVHV DQG OHJXPHV LQFOXGHG LQ WKLV VWXG\ &\QRGRQ GDFW\ORQ LV D VWRORQLIHURXV JUDVV ZLWK WKH

PAGE 79

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f NJ '0 RQ

PAGE 80

7DEOH *UDVV \LHOGV '0f E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ 7UHDWPHQWV '*I [A '5I [f *3I ; f b %: )W L NJ KD 1R 5HSV ZVW '6 ZVW '6M ZVW NJ KD

PAGE 81

7DEOH f§FRQWLQXHG 7UHDWPHQWV '*I [f '5I [f *3I ;f b %: ) L NJ KD 1R 5HSV ZVW '6 ZVW '6L ZVW NJ KD I'* GD\V JUD]LQJ '5 GD\V UHVW*3 JUD]LQJ SUHVVXUH ) IHUWLOL]HU :6 ZHW VHDVRQ '6 GU\ VHDVRQ

PAGE 82

RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ %RWK UHVW SHULRG ;Af DQG JUD]LQJ SUHVVXUH ;Af KDG DQ HIIHFW XSRQ WKH JUDVV \LHOG 3 f 7KH DSSOLFDWLRQ RI VXSHUSKRVSKDWH ;Af DOVR KDG D VLJQLILFDQW HIIHFW DW WKH b OHYHO RI VLJQLILFDQFH (DFK RI WKHVH YDULDEOHV DFWHG LQGHn SHQGHQWO\ IRU JUDVV \LHOG VLQFH QR LQWHUDFWLRQV RFFXUUHG 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV DQG LQWHUDFWLRQV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $Q H[DPLQDWLRQ RI WKH LQGLYLGXDO OLQHDU HIIHFWV VKRZV WKDW WKH H[SHULPHQWDO YDULDEOH GD\V JUD]LQJ ;Af GLG QRW KDYH DQ\ HIIHFW RQ WKH DPRXQW RI JUDVV SURGXFHG 7KH JUHDWHVW HIIHFWV ZHUH REWDLQHG IURP GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Aff $V WKH OHQJWK RI UHVW SHULRG LQFUHDVHG IURP ]HUR GD\V FRQWLQXRXV JUD]LQJf WKH DPRXQW RI JUDVV \LHOG WHQGV WR LQFUHDVH ZKHUHDV ZKHQ WKH JUD]LQJ SUHVVXUH LV GHFUHDVHG IURP WR NJ '0 RQ RIIHU NJ %: WKH JUDVV \LHOG ZDV LQFUHDVHG )RU HDFK GD\V LQFUHDVH LQ UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI JUDVV \LHOG )RU HDFK LQFUHPHQW RI GHFUHDVH LQ WKH JUD]LQJ SUHVVXUH WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI JUDVV \LHOG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH

PAGE 83

*UDVV GD\ JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ *UDVV \LHOG ZDV DJDLQ LQIOXHQFHG E\ GD\V UHVW DQG JUD]LQJ SUHVVXUH ;Af 3 f ZKLOH WKH H[SHULPHQWDO YDULDEOHV GD\V JUD]LQJ DQG IHUWLOL]HU OHYHO ZHUH QRW VLJQLILFDQW 7KHUH ZHUH QR LQWHUDFWLRQV EHWZHHQ H[SHULPHQn WDO YDULDEOHV GXULQJ WKLV VHDVRQ 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG RQO\ DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH JUDVV \LHOG ZDV SRVLWLYHO\ UHODWHG WR LQFUHDVLQJ OHQJWKV RI UHVW SHULRG DQG QHJDn WLYHO\ UHODWHG WR LQFUHDVLQJ DPRXQWV RI IRUDJH RQ RIIHU WR WKH JUD]LQJ DQLPDOV )RU HDFK GD\V LQFUHDVH LQ WKH UHVW SHULRG WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI JUDVV \LHOG RQ WKH RWKHU KDQG IRU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH WKHUH ZDV DQ LQFUHDVH RI NJ RI JUDVV \LHOG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH *UDVV
PAGE 84

NJ '0 KD WR NJ '0 KD IRU WKH IROORZLQJ WUHDWPHQW FRPELQDWLRQV > GD\ JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\V JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ $JDLQ WKH JUDVV \LHOG ZDV LQFUHDVHG E\ LQFUHDVLQJ WKH GD\V RI UHVW DQG GHFUHDVLQJ WKH JUD]LQJ SUHVVXUH 3 f 7KH YDULDEOHV ; DQG ZHUH QRW VLJQLILFDQW DQG QR LQWHUDFWLRQV ZHUH IRXQG EHWZHHQ WKH H[SHULPHQWDO YDULDEOHV 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDn WLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f )RU HDFK GD\V LQFUHDVH LQ UHVW SHULRG WKHUH ZDV D FRUUHVn SRQGLQJ LQFUHDVH RI NJ KD A RI JUDVV \LHOG ZKLOH HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH LQFUHDVHG WKH JUDVV \LHOG E\ NJ KD ? 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH *UDVV GD\V JUD]LQJ GD\V UHVW FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWH@ DQG GD\ JUD]LQJ GD\V UHVW NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ *UDVV GU\ PDWWHU \LHOG ZDV DJDLQ DIIHFWHG E\ LQFUHDVLQJ WKH QXPEHU RI GD\V UHVW DQG GHFUHDVLQJ WKH JUD]LQJ SUHVVXUH 3 f

PAGE 85

7KH RWKHU WZR H[SHULPHQWDO YDULDEOHV DQG ;A ZHUH QRW VLJQLILFDQW DQG QR LQWHUDFWLRQV ZHUH IRXQG EHWZHHQ WKH H[SHULPHQWDO YDULDEOHV 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $Q LQFUHDVH RI RQH XQLW GD\Vf RI UHVW LQFUHDVHG WKH JUDVV GU\ PDWWHU \LHOG E\ NJ KD ? ZKLOH D GHFUHDVH LQ WKH JUD]LQJ SUHVVXUH LQFUHDVHG WKH JUDVV GU\ PDWWHU \LHOG E\ NJ KD 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH 6XPPDU\ RI *UDVV
PAGE 86

RI JUDVV LQFUHDVHG FRQVLGHUDEO\ IURP WR )LJV DQG f GXH PDLQO\ WR LWV UDSLG JURZWK FDSDFLW\ DQG DELOLW\ WR HOLPLQDWH WKH RWKHU FRPSDQLRQ VSHFLHV 9LFHQWH&KDQGOHU f UHSRUWHG WKDW HOHSKDQWJUDVV DQG JXLQHDJUDVV DUH WKH PRVW SURGXFWLYH WURSLFDO JUDVVHV UHDFKLQJ YDOXHV RI DQG WRQV '0 KD A \HDU ? ,1,$3 f UHn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f WKH

PAGE 87

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV \LHOG '0f IRU WKH ZHW VHDVRQ RI fYM /2

PAGE 88

'D\V UHVW )LJ &RQWRXUV RI JUDVV \LHOG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI !

PAGE 89

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV \LHOG '0f IRU WKH ZHW VHDVRQ RI /Q

PAGE 90

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f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
PAGE 91

/HJXPH
PAGE 92

7DEOH 'U\ PDWWHU PHDQV IRU OHJXPH E\ \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ 7UHDWPHQWV '* ?f '5 [f *3 ; f b %: ) 1R NJ KD A 5HSV :6 '6 :6 '6 :6 NJ KD

PAGE 93

7DEOH f§FRQWLQXHG 7UHDWPHQWV '*I [[f '5I [f *3I ;f b %: ) L NJ KD 1R 5HSV ZVW '6W ZVW '6W ZVW NJ KD W'* MZV GD\V JUD]HG '5 ZHW VHDVRQ '6 GD\V UHVW *3 GU\ VHDVRQ JUD]LQJ SUHVVXUH ) IHUWLOL]HU

PAGE 94

NJ %: DQG NJ KD A RI VXSHUSKRVSKDWHf UHVSHFWLYHO\ 7KH OHJXPH \LHOG ZDV LQFUHDVHG DV WKH UHVXOW RI UHGXFLQJ WKH JUD]LQJ SUHVVXUH 3 f 7KHUH ZDV DQ LQWHUDFWLRQ EHWZHHQ GD\V JUD]LQJ ;Af DQG JUD]LQJ SUHVVXUH ;Af EXW WKLV LQWHUDFWLRQ GLG QRW RFFXU DJDLQ GXULQJ WKH UHPDLQLQJ WKUHH VHDVRQV 7KHUH DSSHDUHG WR EH QR OLQHDU HIIHFW IRU GD\V JUD]LQJ ;Af GD\V UHVW ;Af DQG IHUWLOL]HU OHYHO ;Af 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU RQO\ b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF HIIHFWV UHSUHVHQWHG RQO\ b DQG WKH LQWHUDFWLRQV b $SSHQGL[ 7DEOH f 'XULQJ WKH GU\ VHDVRQ RI WKHUH ZDV D WHQGHQF\ IRU WKH OHJXPH \LHOG WR LQFUHDVH DV WKH JUD]LQJ SUHVVXUHV GHFUHDVHG /HJXPH
PAGE 95

RQO\ DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 'XULQJ WKH ZHW VHDVRQ WKH \LHOG RI OHJXPHV EHJDQ WR VKRZ D SDWWHUQ DVVRFLDWHG ZLWK WKH OHQJWK RI WKH UHVW SHULRG )RU HDFK GD\V LQFUHDVH LQ UHVW SHULRG WKHUH ZDV D GHFUHDVH RI NJ KD A RI OHJXPH \LHOG $OVR GXULQJ WKLV VHDVRQ JUD]LQJ SUHVVXUH EHJDQ WR VKRZ DQ HIIHFW XSRQ OHJXPH \LHOG )RU HDFK XQLW NJ '0f GHFUHDVH LQ JUD]LQJ SUHVVXUH WKHUH ZDV DQ LQFUHDVH RI NJ KD A RI OHJXPH \LHOG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH /HJXPH
PAGE 96

7KH TXDGUDWLF HIIHFW RI OHQJWK RI UHVW SHULRG ;f VWURQJO\ VXJJHVWV WKDW WKH UDQJH RI UHVW SHULRGV LQFOXGHG LQ WKLV H[SHULPHQW PD\ KDYH DGHTXDWHO\ FRYHUHG WKH SRLQW RI PD[LPXP OHJXPH \LHOG 7KH KLJKHVW OHJXPH \LHOGV ZHUH REWDLQHG LQ WKH UHJLRQ RI WR GD\V UHVW DQG WKH \LHOGV GHFUHDVHG ZKHQ WKH UHVW SHULRGV ZHUH UHGXFHG WR ]HUR RU LQFUHDVHG XS WR GD\V 7KH HIIHFW RI JUD]LQJ SUHVVXUH ;Af DSSHDUV WR EH DOPRVW OLQHDU )RU HDFK XQLW NJ '0f WKDW JUD]LQJ SUHVVXUH ZDV UHGXFHG IURP WR NJ '0 RQ RIIHU NJ %: WKH DPRXQW RI OHJXPH \LHOG LQFUHDVHG E\ NJ '0 KD ? 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH /HJXPH GD\V JUD]LQJ UHVW SHULRG FRQWLQXRXV JUD]LQJf NJ '0 RQ RIIHU NJ %: DQG NJ KD A RI VXSHUSKRVn SKDWH @ 7KHUH ZDV ERWK D OLQHDU DQG TXDGUDWLF HIIHFW RI OHQJWK RI UHVW SHULRG ;A XSRQ OHJXPH \LHOG 3 f 7KHUH ZDV D OLQHDU HIIHFW RI JUD]LQJ SUHVVXUH 3 f 7KH HIIHFWV RI GD\V JUD]LQJ DQG IHUWLOL]HU OHYHO ZHUH QRW VLJQLILFDQW DQG QHLWKHU ZHUH DQ\ RI WKH LQWHUDFWLRQV RI WKH H[SHULPHQWDO YDULDEOHV 7KH OLQHDU DQG TXDGUDWLF HIIHFWV DFFRXQWHG IRU DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ ZKLOH WKH LQWHUDFWLRQV UHSUHVHQWHG RQO\ b $SSHQGL[ 7DEOH f

PAGE 97

7KH TXDGUDWLF HIIHFW RI UHVW SHULRG ;JLYHV D EHWWHU UHSn UHVHQWDWLRQ RI WKH OHJXPH \LHOG ZKLFK PD[LPL]HV ZLWKLQ WKH UDQJH RI DQG GD\V $V WKH OHQJWK RI WKH UHVW SHULRG LV HLWKHU GHFUHDVHG RU LQFUHDVHG WKH OHJXPH \LHOG LV GHFUHDVHG $V WKH JUD]LQJ SUHVVXUH GHFUHDVHV WKH DPRXQW RI OHJXPH \LHOG LQFUHDVHG E\ NJ KD A 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH 2EVHUYDWLRQV DQG 6XPPDU\ RI /HJXPH
PAGE 98

)LJ /HJXPH \LHOG NJ KDaO *UD]LQJ SUHVVXUH NJ '0 RQ RIIHU NJ %:f 'D\V UHVW f (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ OHJXPH \LHOG '0f IRU WKH ZHW VHDVRQ RI

PAGE 99

JUD]LQJ SUHVVXUH UHVXOWLQJ LQ YDULRXV GHJUHHV RI GHIROLDWLRQ VHOHFWLYLW\ DQG HIIHFW RI WUDPSOLQJ 'HIROLDWLRQ DQG VHOHFWLYLW\ ZHUH SUREDEO\ WKH PDLQ GHWHUPLQDQWV RQ WKH DPRXQW RI ERWK OHJXPHV $ VPDOO GLIIHUHQFH LV DOVR UHFRUGHG LQ WKH EHKDYLRU RI WKHVH WZR VSHFLHV LQ WKDW FHQWUR DSSHDUHG WR EH PRUH WROHUDQW RI KHDY\ JUD]LQJ SUHVVXUH DQG WKH DUHD FRYHUHG E\ JO\FLQH LQFUHDVHG DV WKH JUD]LQJ SUHVVXUH GHFUHDVHG 'XULQJ WKH GU\ VHDVRQ RI LW ZDV DOVR REVHUYHG WKDW WKH JUD]LQJ DQLPDOV ZHUH VHOHFWLYHO\ FRQVXPLQJ PRUH OHJXPHV WKDQ GXULQJ WKH ZHW VHDVRQ SUREDEO\ EHFDXVH RI WKH ORZHU TXDOLW\ DQG UHGXFHG TXDQWLW\ RI JUDVV IRU WKDW SHULRG +XPSKUH\V f DQG 1RUPDQ f REVHUYHG VLPLODU SUHIHUHQFHV RI DQLPDOV JUD]LQJ WURSLFDO JUDVVOHJXPH SDVWXUHV 'XULQJ WKH ZHW VHDVRQ RI -DQXDU\-XQHf WKH DYHUDJH OHJXPH \LHOG ZDV NJ KD A ZKLFK ZDV FRQVLGHUDEO\ ORZHU ZKHQ FRPSDUHG ZLWK WKH SUHGHFLQJ GU\ VHDVRQ f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

PAGE 100

HQYLURQPHQWDO IDFWRUV DUH VLPLODU S f +H DOVR DGGHG WKDW WKH YLQH\ JURZWK KDELW RI VHYHUDO JHQHUD RI WURSLFDO OHJXPHV &DORSRJRQLXP &HQWURVHPD VRPH VSHFLHV RI 'HVPRGLXP 1HRQRWRQLD 0DFURSWLOLXP DQG 3XHUDULDf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f 'XULQJ WKH GU\ VHDVRQ RI -XO\'HFHPEHUf WKH PHDQ OHJXPH \LHOG ZDV NJ KD 'XULQJ WKLV VHFRQG GU\ VHDVRQ D IHZ SDVWXUHV ZLWK WKH ORQJHVW UHVW SHULRGV ZHUH DOPRVW b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

PAGE 101

'XULQJ WKH PRQWK RI -XQH DQG SDUW RI -XO\ RI D VHYHUH DWWDFN RI UHG VSLGHU ZDV REVHUYHG HVSHFLDOO\ RQ FHQWUR ZKLOH JO\FLQH DSSHDUHG WR EH PRUH WROHUDQW DQG ZDV GDPDJHG OHVV WKDQ FHQWUR ZKLFK OHG WR WKH GRPLQDQFH RI JO\FLQH GXULQJ WKLV ZKROH VHDVRQ /LQHDU DQG TXDGUDWLF HIIHFWV RI UHVW SHULRG ;f ZHUH REVHUYHG 3 f ZKLOH WKH OLQHDU HIIHFW RI JUD]LQJ SUHVVXUH ;Af ZDV QRWHG 3 f )LHOG REVHUYDWLRQV VXJJHVWHG WKDW UHVW SHULRG ZDV D JUHDWHU GHWHUPLQDQW RI OHJXPH VXUYLYDO SURGXFWLYLW\ DQG SHUVLVWHQFH WKDQ ZDV JUD]LQJ SUHVVXUH )RU WKH ZHW VHDVRQ RI -DQXDU\-XQHf WKHUH ZDV DOVR D GHFUHDVH LQ WKH DYHUDJH OHJXPH \LHOG DW NJ KD ,Q WKH ODVW ZHW VHDVRQ VRPH SDVWXUHV VKRZHG D FRPSOHWH GRPLQDQFH RI JUDVV FRPSOHWHO\ H[FOXGLQJ ERWK OHJXPHV DQG ZHHGV 7KHVH FKDQJHV ZHUH REVHUYHG RQ WUHDWPHQWV ZLWK WKH ORQJHVW UHVW SHULRG ZKLFK ZDV FRQn ILUPHG E\ WKH DQDO\VLV ZKLFK VKRZHG UHVW SHULRG ;Af ZLWK ERWK OLQHDU DQG TXDGUDWLF HIIHFWV 3 f 7KH HIIHFW RI JUD]LQJ SUHVVXUH ;Af ZDV OLQHDU 3 f ZKLFK VXJJHVWHG WKDW UHVW SHULRG ZDV WKH PDMRU GHWHUPLQDQW IRU WKH VXUYLYDO DQG SURGXFWLYLW\ RI WKH OHJXPHV )LJV DQG f 7KH WUHDWPHQWV ZLWK FRQWLQXRXV JUD]LQJ RU VKRUW UHVW SHULRGV LQ FRPELQDWLRQ ZLWK WKH KLJKHVW JUD]LQJ SUHVVXUHV GUDVWLFDOO\ DIIHFWHG WKH \LHOGV RI OHJXPHV JUDVVHV DQG ZHHGV 7KLV ZDV HVSHFLDOO\ WUXH IRU WKH SODQWHG JUDVVHV DQG OHJXPHV ZKLFK ZHUH LQ VRPH FDVHV DOPRVW FRPSOHWHO\ HOLPLQDWHG E\ WKH FORVH DQG IUHTXHQW GHIROLDWLRQ ZKLFK UHPRYHG WKH PDMRU SRUWLRQ RI WKH \RXQJ DFWLYH OHDI PDWHULDO DQG DSLFDO PHULVWHPV OHDGLQJ WR D UHGXFWLRQ LQ UDWH RI UHFRYHU\ DQG DELOLW\ WR

PAGE 102

*UD]LQJ SUHVVXUH NJ '0 RQ RIIHU NJ %:f )LJ &RQWRXUV RI OHJXPH \LHOG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI

PAGE 103

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ OHJXPH \LHOG '0f IRU WKH ZHW VHDVRQ RI 92

PAGE 104

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

PAGE 105

(IIHFW RI /HQJWKV RI 5HVW 3HULRG DQG /HYHOV RI *UD]LQJ 3UHVVXUH RQ WKH
PAGE 106

1! K2 UR UR UR 0 +r +r K} 0 Kr +} +} K 6 /2 UR +r R 92 21 /Q A /2 UR +r R 92 2n /Q A /2 UR 0 R R UR UR UR UR UR UR UR UR UR UR UR UR ; +} 0 &2 + +r ,f§ n2 +r ,r ?O K +r n2 Kr 0 Kr R f§E + /Q /Q /Q /Q 1 Kr ,f§r 6 1 ,r A A Kr Kr ‘! A ,f§r +r ; R 21 21 R 2 2Q 21 R R UR UR A UR UR 1 A UR UR A 1 UR UR A f3 UR SD t f§K UW I' 3 UW &2 RR +r 0 +} +r 21 21 21 21 /2 //2 /2 21 21 21 21 //2 /2 /2 ; /2 R /X! /2 21 21 21 21 21 21 21 21 /2 /2 /2 /2 21 21 21 21 /2 //2 /6n" f§W : 37 / /2 /2 /2 /2 /2 /2 /2 0 0 +r K 0 0 0 0 72 R R R R R R R R R R R R R R R R 7O R R R R R R R R R R R R R R R R R R R R R R R R ;7 f§W + 1 UR +r UR UR K K +r +r 0 Kr +} 0 +r 0 If§r Kr +r 0 0 SD r &2 V r K} +} r 92 2Q 21 UR Y2 r /2 92 /2 21 ‘! 0 /Q & n2 R R R R R R R R R R R R R R R R R R R R R R R R RR R 0 0 0 Y2 X! 1 /Q 3! 92 UR 21 0 /Q /FQ nR R R R R R R 2 R R R R R R R R R R R R R R R R 72 V r /2 /2 UR UR ,f§r Kr 92 6 Kr R /Q UR R UR Ir /2 92 21 ;7 R R R R R R R R R R R R R R R R R R R R R R R R 3! f:r 92 R 0 /Q UR r UR +r 2 n‘r!f§ 92 +r K UR /2 +} 92 21 1 If§r /2 R R /Q 21 92 fR R R R R R R R R R R R R R R R R R R R R R R R R 92 0 V f§r +r UR A !r UR 92 92 UR &n 92 21 +r 92 /2 UR ‘3 KL K! UR UR Kr UR RR R R R R R R R R R R R R R R R R R R 2 R R R R R R = 7DEOH 'U\ PDWWHU PHDQV IRU ZHHGV IRU \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ

PAGE 107

7DEOH f§FRQWLQXHG 7UHDWPHQWV 5HSV 1R '*I [[f '5I ;f *3I ; f b %: )W L NJ KD ZVW '6W ZVW '6W ZVW f§ NJ KD A f§ I'* GD\V JUD]LQJ '5 GD\V UHVW *3 JUD]LQJ SUHVVXUH ) IHUWLOL]HU W:6 ZHW VHDVRQ '6 GU\ VHDVRQ

PAGE 108

'XULQJ WKH ILUVW VHDVRQ D VLJQLILFDQW OLQHDU TXDGUDWLF DQG LQWHUn DFWLRQ HIIHFWV ZHUH IRXQG ZKLFK DFFRXQWHG IRU DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f
PAGE 109

JUD]LQJ SHULRG DQG IHUWLOL]HU OHYHO KDG QR HIIHFW XSRQ WKH \LHOG RI ZHHGV 7KH OLQHDU DQG LQWHUDFWLRQV FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH \LHOG RI ZHHGV WHQGHG WR GHFUHDVH DV WKH OHQJWK RI WKH UHVW SHULRG LQFUHDVHG IURP FRQWLQXRXV JUD]LQJ WR GD\V RI UHVW DQG D VLPLODU WUHQG ZDV REVHUYHG DV WKH JUD]LQJ SUHVVXUH LQFUHDVHG IURP WR NJ '0 RQ RIIHU NJ %: (DFK XQLW RI UHVW GD\Vf GHFUHDVHG WKH \LHOG RI ZHHGV E\ NJ KD A DQG IRU HDFK XQLW LQFUHDVH LQ JUD]LQJ SUHVVXUH WKH \LHOG RI ZHHGV ZDV GHFUHDVHG E\ NJ KD ? 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH
PAGE 110


PAGE 111

WKH \LHOGV RI ZHHGV ZHUH ORZ GXULQJ WKH ILUVW VHDVRQ WKHUH SUHVHQFH LQ UHVSRQVH ZDV QRWHZRUWK\ 'XULQJ WKH GU\ VHDVRQ RI WKHUH ZDV D TXDGUDWLF UHVSRQVH WR 3 IHUWLOL]HU DQG LW DSSHDUHG WKDW VRPH VSHFLHV RI ZHHGV ZHUH WKH RQO\ VSHFLHV ZKLFK UHVSRQGHG SRVLWLYHO\ WR 3 7KHUH ZDV DOVR VRPH HYLGHQFH WKDW WKH \LHOG RI ZHHGV ZDV UHVSRQGLQJ GXULQJ WKH ILUVW GU\ VHDVRQ WR WKH OHQJWK RI UHVW SHULRG ;f VLQFH WKHUH ZDV D GHFOLQH LQ WKH ZHHG SRSXODWLRQ DV WKH OHQJWK RI UHVW SHULRG LQFUHDVHG 7KLV PD\ KDYH EHHQ GXH WR WKH FRPSHWLWLRQ IURP WKH VPDOOJURZLQJ WURSLFDO JUDVVHV )LJ f 'XULQJ WKH ZHW VHDVRQ RI WKH HIIHFW RI GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af UHVXOWHG LQ D GHFOLQH RI WKH \LHOG RI ZHHGV 3 f 7KHUH ZDV DOVR D VWURQJ LQWHUDFWLRQ EHWZHHQ WKHVH WZR YDULDEOHV 3 f 7KLV LQWHUDFWLRQ VXJJHVWV WKDW WKH \LHOG RI ZHHGV LV UHGXFHG E\ LQFUHDVLQJ WKH OHQJWK RI WKH UHVW SHULRG LQ DVVRn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f GD\V UHVW ;f DQG JUD]LQJ SUHVVXUH ;Af UHVXOWHG LQ D GHFOLQH LQ WKH \LHOG RI ZHHGV 3 f 7KH H[SHULPHQWDO YDULDEOHV ; DQG ;A

PAGE 112

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ \LHOG RI ZHHG '0f IRU WKH ZHW VHDVRQ RI

PAGE 113

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f JUD]LQJ SUHVVXUH ;Af DQG 3 IHUWLOL]HU ;Af ZHUH YHU\ HYLGHQW 3 f 7KHUH ZDV DOVR D TXDGUDWLF HIIHFW RI ;A 3 f 7KH LQWHUDFWLRQV EHWZHHQ ; [ ;A ; [ ;A DQG ;A [ ;A ZHUH DOVR VLJQLILFDQW 3 f )URP WKH UHVXOWV RI WKHVH DQDO\VHV LW DSSHDUHG WKDW GD\V UHVW DQG JUD]LQJ SUHVVXUH ZHUH WKH SULQFLSDO GHWHUPLQDQWV RI WKH \LHOG RI ZHHGV VHH )LJV DQG f 7KH LQWHUn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

PAGE 114

'D\V UHVW ? ? ? Y ? ? ? ? ? ? *UD]LQJ SUHVVXUH NJ '0 RQ RIIHU NJ %:f )LJ &RQWRXUV RI \LHOG ZHHG '0f DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI

PAGE 115

)LJ (IIHFWV RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ \LHOG RI ZHHG '0f IRU WKH ZHW VHDVRQ RI R

PAGE 116

ZLWK ORQJ UHVW SHULRGV DQGRU ORZ JUD]LQJ SUHVVXUHV ZLWK ORZ 3 OHYHOV ,W LV ZHOO NQRZQ WKDW WDOOJURZLQJ JUDVVHV DQG FOLPELQJ OHJXPHV KDYH WKH SRWHQWLDO WR H[SORLW WKH VRLO DQG DHULDO HQYLURQPHQWV ZKLFK PD\ OLPLW WKH JURZWK RI FRPSDQLRQ RU LQYDGLQJ VSHFLHV HVSHFLDOO\ ZKHQ WKH PDQDJHPHQW SUDFWLFHV IDYRU WKHLU PD[LPXP SURGXFWLRQ )UHTXHQW DQG LQWHQVLYH GHIROLDWLRQ RI WKH WZR VSHFLHV PD\ EH KLJKO\ GHWULn PHQWDO WR WKHLU YLJRU DQG UHJURZWK FDSDFLW\ VR WKDW WKH VZDUG FDQ HDVLO\ EH GRPLQDWHG E\ XQGHVLUDEOH VSHFLHV +DUDVFKLQ f UHSRUWHG VLPLODU UHVXOWV ZLWK ZHHGV ZKLFK ZHUH DIIHFWHG E\ WKH UHVW SHULRG DQG JUD]LQJ SUHVVXUH DQG KH FRQFOXGHG WKDW VKRUW UHVW SHULRGV DQG KLJK JUD]LQJ SUHVVXUH FUHDWH WKH PRVW IDYRUDEOH FRQGLWLRQV IRU ZHHGV LQYDVLRQ -HQVHQ DQG 6FKXPDFKHU f UHSRUWHG WKDW WKH ERWDQLFDO FRPSRn VLWLRQ LV QRW RQO\ DIIHFWHG E\ WKH JUD]LQJ DQLPDO EXW DOVR E\ VRPH HQYLURQPHQWDO IDFWRUV VXFK DV WKH VHDVRQ UDLQIDOO WHPSHUDWXUH DQG VRLO QXWULHQWV ZKLFK PD\ DOORZ FHUWDLQ ZHHG\ VSHFLHV WR LQYDGH WKH SDVWXUH VZDUG 7KH SUHGRPLQDQW ZHHG VSHFLHV LQ WKLV VWXG\ ZHUH 6LGD DFXWD 6RODQXP FDUROLQHQVLV $PDUDQWKXV VS &DSVLFXP VS $HVFSOHSLDV VS 0LPRVD SXELFD &DVVLD WRUD & RFFLGHQWDOLV EURDGOHDI ZHHGVf 3DVSDOXP IDVFLFXODWXP 3 SDQLFXODWXP 3 FRQMXJDWXP 'LJLWDULD VDQJXLQDOLV (OHXVLQH LQGLFD *UDVVHVf DQG &\SHUXV URWXQGXV 6HGJHf 6RPH RI WKHVH ZHHGV ZHUH DOVR UHSRUWHG E\ 6DQWLOODQ f JURZLQJ LQ -DUDJXDJUDVV +\SDUUKHQLD UXIDf

PAGE 117

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b ZKLFK UHSUHVHQWHG WKH YDULDWLRQ WKDW RFFXUUHG GXULQJ WKH ILUVW WZR DQG RQHKDOI PRQWKV RI WKH H[SHULPHQW 7KHUH ZHUH QR VLJQLILFDQW HIIHFWV RI WUHDWPHQW GXULQJ WKLV ILUVW VHDVRQ 7KH DQDn O\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH ,Q WKH GU\ VHDVRQ RI WKH YLVXDO HVWLPDWHV DUH DOVR SUHVHQWHG LQ 7DEOH DQG WKH SHUFHQWDJH JUDVV YDULHG IURP WR b 7KH OLQHDU HIIHFWV RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH 3 f EHFDPH HYLGHQW GXULQJ WKH ILUVW GU\ VHDVRQ EXW WKH HIIHFWV RI GD\V JUD]LQJ DQG IHUWLOL]HU OHYHO ZHUH QRW VLJQLILFDQW $OVR WKHUH ZHUH QR LQWHUn DFWLRQV EHWZHHQ WKH H[SHULPHQWDO YDULDEOHV 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 'XULQJ WKH VHDVRQ

PAGE 118

9LVXDO HVWLPDWLRQ RI GU\ PDWWHU JUDVV SHUFHQW IRU \HDU VHDVRQ DQG WUHDWPHQW FRPELQDWLRQ 7UHDWPHQWV '*I [f '5I *3I ;f ;f b %: )W L NJ KD 5HSV ZVW '6W ZVW '6W ZVW

PAGE 119

7DEOH f§FRQWLQXHG 7UHDWPHQWV 5HSV 1R '*I [[f '5I [f *3I ;f b %: )W L NJ KD ZVW '6 :6'6W ZVW b I'* GD\V JUD]LQJ '5 GD\V UHVW*3 JUD]LQJ SUHVVXUH DQG ) IHUWLOL]HU UDWH :6 ZHW VHDVRQ '6 GU\ VHDVRQ

PAGE 120

WKH SHUFHQWDJH JUDVV WHQGHG WR LQFUHDVH DV WKH OHQJWK RI UHVW SHULRG LQFUHDVHG DQG WR GHFUHDVH DV WKH JUD]LQJ SUHVVXUH LQFUHDVHG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH )RU WKH ZHW VHDVRQ RI WKH SHUFHQWDJH JUDVV YDULHG IURP WR b $JDLQ WKH OHQJWK RI WKH UHVW SHULRG ;f DQG JUD]LQJ SUHVVXUH ;Af DIIHFWHG WKH SHUFHQWDJH JUDVV 3 f DQG WKHUH ZHUH DOVR VLJQLILFDQW LQWHUDFWLRQV IRU ;A [ ;A DQG ; [ ;A 3 f 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $JDLQ WKH SHUFHQWDJH JUDVV WHQGHG WR LQFUHDVH DV WKH OHQJWK RI WKH UHVW SHULRG LQFUHDVHG ZKLOH LQFUHDVHV LQ JUD]LQJ SUHVVXUH WHQGHG WR UHGXFH WKH SHUFHQWDJH JUDVV 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH 'XULQJ WKH GU\ VHDVRQ RI WKH VSUHDG LQ WKH SHUFHQWDJH JUDVV KDV QRZ LQFUHDVHG IURP WR b 7KH GD\V UHVW ;f DQG JUD]LQJ SUHVVXUH ;Af ZHUH HYHQ PRUH HYLGHQW GXULQJ WKH IRXUWK VHDVRQ DQG WKHUH ZHUH DOVR VLJQLILFDQW LQWHUDFWLRQV EHWZHHQ ; [ ;A 3 f 7KH HIIHFW XSRQ SHUFHQWDJH JUDVV RI GD\V JUD]LQJ DQG IHUWLOL]HU OHYHO ZHUH QLO 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO LQ WKH GU\ VHDVRQ RI DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $JDLQ WKH SHUFHQWDJH JUDVV WHQGHG WR LQFUHDVH DV WKH OHQJWK RI UHVW SHULRG LQFUHDVHG DQG WR GHFUHDVH DV WKH JUD]LQJ SUHVVXUH ZDV LQFUHDVHG

PAGE 121

'XULQJ WKH ZHW VHDVRQ RI ZKLFK ZDV WKH ILQDO VHDVRQ RI WKH H[SHULPHQW WKH SHUFHQWDJH JUDVV UDQJHG IURP WR b 7KH OLQHDU DQG TXDGUDWLF HIIHFWV RI GD\V UHVW DQG JUD]LQJ SUHVVXUH ;Af ZHUH YHU\ HYLGHQW 3 f DQG WKHUH ZDV DOVR D VWURQJ LQWHUDFWLRQ EHWZHHQ [ ;A $JDLQ GD\V JUD]LQJ ;Af DQG OHYHO RI IHUWLOL]HU ;Af KDG QR HIIHFW 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 9LVXDO HVWLPDWLRQ RI SHUFHQWDJH JUDVV ZDV UHFRUGHG RQ WKH VDPH GDWHV WKDW JUDVV \LHOGV ZHUH WDNHQ 6RPH UDWKHU GUDVWLF FKDQJHV RFFXUUHG RYHU WLPH DQG WKH HIIHFWV RI WUHDWPHQWV IURP WKH EHJLQQLQJ RI WKH H[SHULPHQW WKURXJK WKH ODVW JUD]LQJ VHDVRQ DUH VXPPDUL]HG EHORZ 'XULQJ WKH ILUVW ZHW VHDVRQ 0D\-XQHf RI QR HIIHFWV RI WUHDWPHQWV ZHUH UHFRUGHG ZKHUHDV LQ WKH ODVW ZHW VHDVRQ RI ERWK OLQHDU DQG TXDGUDWLF HIIHFWV 3 f RI GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af DV ZHOO DV D VLJQLILFDQW LQWHUDFWLRQ EHWZHHQ ;e [ ;A ZHUH YHU\ HYLGHQW :KLOH LW LV QRW FHUWDLQ WKDW WZR \HDUV DUH VXIILFLHQW WR SURGXFH VWDEOH DVVRFLDWLRQV LW LV FOHDU WKDW ERWK JUDVVHV HOHSKDQWJUDVV DQG *XLQHDJUDVVf DUH YHU\ UHVSRQVLYH WR GD\V UHVW DQG WR JUD]LQJ SUHVVXUH 7KH UHODWLRQVKLS RI WKHVH WZR H[SHULn PHQWDO YDULDEOHV LV REYLRXVO\ FXUYLOLQHDU VHH )LJV DQG f /RZ UHVW SHULRGV UHVXOWHG LQ VZDUGV ZLWK b JUDVV ZKHUHDV WKHVH WDOO JURZLQJ JUDVVHV DOPRVW HOLPLQDWHG XQGHU FRQWLQXRXV LQWHQVLYH JUD]LQJ (DFK RI WKHVH VSHFLHV KDV D KLJK JURZWK FDSDFLW\ DQG WKH DELOLW\ WR

PAGE 122

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ JUDVV SHUFHQWDJH IRU WKH ZHW VHDVRQ RI

PAGE 123

'D\V UHVW *UD]LQJ SUHVVXUH NJ '0 RQ RIIHU NJ %:f )LJ &RQWRXUV RI JUDVV SHUFHQWDJH DV DIIHFWHG E\ OHYHOV RI UHVW SHULRGV DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI

PAGE 124

R 9} R Y 2

PAGE 125

,OO XVH PRUH HIILFLHQWO\ WKH HQYLURQPHQWDO IDFWRUV VXFK DV OLJKW WHPSHUDWXUH PRLVWXUH QXWULHQWV DQG VSDFH SURGXFLQJ D KLJKO\ FRPSHWLWLYH VLWXDWLRQ DQG ILQDOO\ HOLPLQDWLQJ WKH OHVV DJUHVVLYH RU XQDGDSWHG FRPSDQLRQ VSHFLHV 7KH GHJUHH RI GHIROLDWLRQ DV UHSUHVHQWHG LQ WKLV H[SHULPHQW E\ WKH LQWHQVLW\ RI JUD]LQJ SUHVVXUH DOVR KDG D PDUNHG HIIHFW XSRQ WKH SHUFHQWDJH JUDVV $JDLQ WKH UHODWLRQVKLS LV TXDGUDWLF DV VKRZQ E\ )LJV DQG +DUULV f PHQWLRQHG WKDW FORVH FRQWLQXRXV GHIROLDWLRQ OHDGV WR D PRUH VSHFLHVULFK DVVRFLDWLRQ GRPLQDWHG E\ VSHFLHV ZLWK SURVWUDWH UKL]RPDWRXV VWRORQLIHURXV RU EDVDO URVHWWH KDELW ZKLOH WDOO VSHFLHV WHQG WR GLVDSSHDU 7KH GRPLQDQFH RU VXSSUHVVLRQ RI WDOOJURZLQJ VSHFLHV FRXOG ODUJHO\ EH FRQWUROOHG E\ WKH GHJUHH RI GHIROLDWLRQ SHUPLWWLQJ RU QRW SHUPLWWLQJ OLJKW SHQHWUDWLRQ WR OHYHOV ZKHUH SURVWUDWH VSHFLHV GLVSRVH WKHLU OHDI FDQRSLHV 7KH SHUFHQWDJH JUDVV DW WKH EHJLQQLQJ RI WKH H[SHULPHQW ZDV DERXW b ZKLOH GXULQJ WKH ODVW ZHW VHDVRQ f WKH PHDQ SHUFHQWDJH JUDVV ZDV b ZKLFK LQFOXGHG D VSUHDG RI IURP WR b JUDVV GHSHQGLQJ XSRQ WKH WUHDWPHQW FRPELQDWLRQ /RZ JUD]LQJ SUHVn VXUH LQ FRPELQDWLRQ ZLWK ORQJ UHVW SHULRGV SURGXFHG VZDUGV ZLWK DOPRVW b JUDVV ZKHUHDV FRPELQDWLRQV RI KLJK JUD]LQJ SUHVVXUH DQG VKRUW UHVW SHULRGV RU FRQWLQXRXV JUD]LQJ UHVXOWHG LQ DOPRVW D FRPSOHWH HOLPLQDWLRQ RI WKH JUDVV FRPSRQHQW 7KHVH UHVXOWV ZHUH REWDLQHG QRWZLWKVWDQGLQJ WKDW LW LV ZHOO NQRZQ WKDW WKHVH WZR VSHFLHV DUH YHU\ ZHOO DGDSWHG WR WKH HQYLURQPHQWDO FRQGLWLRQV RI WKH UHJLRQ

PAGE 126

,W LV DOVR RI LQWHUHVW WR QRWH WKDW WKH H[SHULPHQWDO YDULDEOHV LQFOXGHG LQ WKH PRGHO DFFRXQWHG IRU RYHU b RI WKH WRWDO YDULDELn OLW\ LQ WKH SHUFHQWDJH RI JUDVV LQ WKH PL[WXUH 9LVXDO (VWLPDWLRQ RI 3HUFHQWDJH /HJXPH 7KH YLVXDO HVWLPDWH RI SHUFHQWDJH RI OHJXPHV LQ WKH PL[WXUH LV IRXQG LQ 7DEOH IRU HDFK WUHDWPHQW FRPELQDWLRQ DQG IRU WKH ILYH VHDVRQV GXULQJ ZKLFK WKLV H[SHULPHQW ZDV FRQGXFWHG )RU WKH ILUVW H[SHULPHQWDO VHDVRQ 0D\-XQH RI f QR VLJQLn ILFDQW HIIHFWV GXH WR WKH WUHDWPHQWV ZHUH IRXQG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH $V H[SHFWHG WKH VHFRQG RUGHU PRGHO DFFRXQWHG IRU RQO\ DERXW b RI WKH WRWDO YDULDELOLW\ LQ SHUFHQWDJH OHJXPH 'XULQJ WKH GU\ VHDVRQ RI WKH SHUFHQWDJH OHJXPH YDULHG IURP WR b /LQHDU HIIHFWV RI GD\V JUD]LQJ ;Af 3 f DQG GD\V UHVW ;Af DQG JUD]LQJ SUHVVXUH ;Af 3 f EHJDQ WR DSSHDU LQ WKH SHUFHQWDJH OHJXPH 7KH HIIHFW RI ;A ZDV QLO DQG QR LQWHUn DFWLRQV DPRQJ WKH H[SHULPHQWDO YDULDEOHV ZHUH IRXQG 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b UHVSHFWLYHO\ $SSHQGL[ 7DEOH f 7KH SHUFHQWDJH OHJXPH WHQGHG WR LQFUHDVH ZLWK LQFUHDVLQJ OHQJWK RI JUD]LQJ SHULRG ZKLOH D QHJDWLYH UHODWLRQVKLS ZDV REVHUYHG DV WKH OHQJWK RI WKH UHVW SHULRG LQFUHDVHG 7KH SHUFHQWDJH OHJXPH DOVR GHFUHDVHG DV WKH JUD]LQJ SUHVVXUH GHFUHDVHG 7KH DQDO\VLV RI YDULDQFH LV SUHVHQWHG LQ $SSHQGL[ 7DEOH

PAGE 127

9LVXDO HVWLPDWLRQ RI GU\ PDWWHU OHJXPH SHUFHQW IRU \HDU VHDVRQ DQG WUHDWPHQW 7UHDWPHQWV '*I '5I *3I )I ;[f ;f ;f b %: NJ KD 5HSV :6W '6W :6W '6W b

PAGE 128

7DEOH f§FRQWLQXHG 7UHDWPHQWV '*I [[f '5I [f *3I ;f b %: )W NJ KD 1R 5HSV ZVI '6L ZVW '6_ ZVL Y 9R '* WZV GD\V JUD]LQJ '5 GD\V UHVW *3 ZHW VHDVRQ '6 GU\ VHDVRQ JUD]LQJ SUHVVXUH DQG ) IHUWLOL]HU UDWH

PAGE 129

'XULQJ WKH ZHW VHDVRQ RI WKH SHUFHQWDJH RI OHJXPH YDULHG IURP WR b 7KH OLQHDU HIIHFW RI GD\V UHVW ;f DQG IRU WKH LQWHUDFWLRQ RI [ ;A ZDV VLJQLILFDQW 7KHUH ZHUH QR GLUHFW HIIHFWV RI GD\V JUD]LQJ ;Af JUD]LQJ SUHVVXUH ;Af DQG IHUWLOL]HU OHYHO ;Af 7KH OLQHDU FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHn VHQWHG DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $JDLQ WKH SHUFHQWDJH RI OHJXPH WHQGHG WR GHFUHDVH ZLWK LQn FUHDVLQJ OHQJWKV RI UHVW SHULRG 'XULQJ WKH GU\ VHDVRQ RI WKH SHUFHQWDJH OHJXPH YDULHG IURP WR b 7KH QXPEHU RI GD\V JUD]LQJ ;Af KDG D OLQHDU HIIHFW 3 f XSRQ WKH SHUFHQWDJH OHJXPH DQG WKH OHQJWK RI WKH UHVW SHULRG ;DOVR KDG DQ HIIHFW XSRQ SHUFHQWDJH OHJXPH 3 f 7KH H[SHULPHQWDO YDULDEOHV ;A DQG ;A KDG QR HIIHFW DQG WKHUH ZHUH QR LQWHUDFWLRQV DPRQJ WKH H[SHULPHQWDO YDULDEOHV 7KH OLQHDU FRPn SRQHQWV RI WKH PRGHO DFFRXQWHG IRU b RI WKH WRWDO YDULDWLRQ ZKLOH WKH TXDGUDWLF DQG LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG DQG b RI WKH WRWDO YDULDWLRQ UHVSHFWLYHO\ $SSHQGL[ 7DEOH f $JDLQ WKH OHJXPH SHUFHQWDJH WHQGHG WR LQFUHDVH ZLWK LQFUHDVLQJ OHQJWKV RI JUD]LQJ SHULRG ZKLOH WKH UHYHUVH ZDV WUXH IRU LQFUHDVLQJ OHQJWKV RI UHVW SHULRG 'XULQJ WKH ZHW VHDVRQ RI WKH SHUFHQWDJH OHJXPH KDG EHJXQ WR VWDELOL]H DQG YDULHG IURP WR b 7KH OHQJWK RI UHVW SHULRG ;f VKRZHG ERWK D VWURQJ OLQHDU DQG TXDGUDWLF HIIHFW XSRQ WKH SHUn FHQWDJH OHJXPH LQ WKH PL[WXUH 7KHUH DSSHDUHG WR EH OLWWOH HIIHFW RI GD\V JUD]LQJ ;A JUD]LQJ SUHVVXUH ;f DQG IHUWLOLW\ OHYHO ;Af

PAGE 130

XSRQ SHUFHQWDJH OHJXPH 1R LQWHUDFWLRQV DPRQJ WKH H[SHULPHQWDO YDULDEOHV ZHUH REVHUYHG 7KH OLQHDU DQG TXDGUDWLF FRPSRQHQWV RI WKH PRGHO DFFRXQWHG IRU DQG b RI WKH WRWDO YDULDWLRQ UHVSHFn WLYHO\ ZKLOH WKH LQWHUDFWLRQ HIIHFWV UHSUHVHQWHG RQO\ b $SSHQGL[ 7DEOH f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f LI ZH H[DPLQH WKH WUHQGV IURP WKH ZHW VHDVRQ RI WKURXJK WKH ZHW VHDVRQ RI 'XULQJ WKH ZHW VHDVRQ RI LW LV FOHDUO\ HYLGHQW WKDW WKH OHQJWK RI UHVW SHULRG LV WKH IDFWRU PRVW LQIOXHQWLDO LQ GHWHUPLQLQJ WKH SHUFHQWDJH OHJXPH 7KH KLJKHVW SHUFHQWDJH OHJXPH DSSHDUV WR RFFXU GXULQJ UHVW SHULRGV RI WR GD\V DQG XQGHU PRGHUDWH FRQGLn WLRQV RI JUD]LQJ SUHVVXUH %RWK H[WUHPHO\ VKRUW RU ORQJ UHVW SHULRGV ZHUH KLJKO\ GHWULPHQWDO WR WKH OHJXPH SRSXODWLRQ ZLWK WKH JUHDWHVW GHVWUXFWLRQ RFFXUULQJ IRU UHVW SHULRGV RI GD\V RU PRUH 7KLV VLWXDWLRQ FDQ SUREDEO\ EHVW EH H[SODLQHG E\ WKH DJJUHVVLYHQHVV RI ERWK RI WKH FRPSDQLRQ JUDVVHV HVSHFLDOO\ GXULQJ WKH ZHW VHDVRQ ZKHQ WKH HQYLURQPHQWDO FRQGLWLRQV ZHUH PRUH IDYRUDEOH IRU PD[LPXP JUDVV

PAGE 131

JURZWK 5HVW SHULRGV RI GD\V DSSHDUHG WR EH PRVW DSSURSULDWH IRU JUDVVOHJXPH \LHOG DQG EDODQFH EHFDXVH WKH KLJKHU YDOXHV IRU OHJXPH SHUFHQWDJH DW WKH HQG RI WKH ZHW VHDVRQ RI ZHUH UHFRUGHG RQ SDVWXUHV ZLWK GD\V UHVW IROORZHG E\ WKRVH ZKLFK KDG D UHVW SHULRG RI GD\V RU FRQWLQXRXV JUD]LQJ 3DVWXUHV ZLWK RU GD\V UHVW VKRZHG ]HUR RU YHU\ ORZ OHJXPH SHUFHQWDJHV 7KHVH UHVXOWV DUH VRPHZKDW FRQWUDU\ WR WKRVH IRXQG E\ 0DUDVFKLQ f DQG 6HUUDR f ZKR UHSRUWHG WKDW 'HVPRGLXP LQWRUWXP LQn FUHDVHG ZLWK WKH OHQJWK RI UHVW SHULRG ZKHQ JURZQ LQ DVVRFLDWLRQ ZLWK &\QRGRQ GDFW\ORQ 7KH UHVXOWV REWDLQHG E\ WKHVH DXWKRUV FDQ SUREDEO\ EH H[SODLQHG E\ WKH IDFW WKDW WKH FRPSHWLWLYH JUDVV LV D ORZJURZLQJ VSHFLHV ZKLFK GLG QRW KDYH QHDUO\ WKH GHVWUXFWLYH HIIHFW XSRQ WKH DVVRFLDWHG OHJXPH =DSDWD f DQG ,1,$3 f UHSRUWHG WKDW JXLQHDJUDVVJO\FLQH SDVWXUHV ZLWK PRUH WKDQ VL[ \HDUV PDLQn WDLQHG DQ DFFHSWDEOH EDODQFH RI ERWK VSHFLHV ZLWK b RU PRUH RI OHJXPH ZKHQ WKHVH SDVWXUHV ZHUH VXEMHFWHG WR D URWDWLRQDO JUD]LQJ V\VWHP RI GD\V JUD]LQJ DQG GD\V UHVW ,Q WKLV H[SHULPHQW WKH SHUFHQWDJH OHJXPH ZDV DOVR UHVSRQVLYH WR OHQJWK RI JUD]LQJ 3 f IRU WKH GU\ VHDVRQ RI DQG WKH ZHW DQG GU\ VHDVRQ RI EXW WKLV HIIHFW GLVDSSHDUHG GXULQJ WKH ODVW VHDVRQ RI %RWK RI WKH OHJXPHV LQ WKLV VWXG\ FHQWUR DQG JO\FLQH WHQGHG WR LQFUHDVH DV WKH OHQJWK RI JUD]LQJ ZDV LQFUHDVHG HVSHFLDOO\ XQGHU WKH FRQWLQXRXV JUD]LQJ ZKHUH GXH WR IUHTXHQW GHIROLDWLRQ WKH JUDVVOHJXPH SHUFHQWDJH ZDV PRUH VWDEOH EXW GHFOLQHG GXULQJ WKH HQWLUH FRXUVH RI WKH H[SHULPHQW

PAGE 132

*UD]LQJ SUHVVXUH DSSHDUHG DV D VLJQLILFDQW IDFWRU 3 f RQO\ GXULQJ WKH GU\ VHDVRQ RI ZKHUH OHJXPH SHUFHQWDJH ZDV PRUH UHVSRQVLYH WR KLJKHU OHYHOV RI JUD]LQJ SUHVVXUH :KLOH WKH JUD]LQJ SUHVVXUH GHFUHDVHG WKH OHJXPH SHUFHQWDJH DOVR VKRZHG D GHFOLQLQJ WUHQG 'XULQJ WKH UHPDLQLQJ IRXU VHDVRQV RI WKH H[SHULPHQW QR VLJQLILFDQW HIIHFWV RI WKLV YDULDEOH ZHUH REVHUYHG 7KH HIIHFWV RI UHVW SHULRG ;Af DQG JUD]LQJ SUHVVXUH ;Af RQ OHJXPH SHUFHQWDJH LV SUHVHQWHG LQ )LJV DQG IRU WKH ZHW VHDVRQ RI DQG WKH ZHW VHDVRQ RI UHVSHFWLYHO\ 7KH FRQWRXUV RI WKH UHVSRQVH VXUIDFH RI SHUFHQWDJH OHJXPH DSSHDUV LQ )LJ IRU WKH ZHW VHDVRQ RI 3HUFHQWDJH OHJXPH DSSHDUV WR EH DOZD\V KLJKHU WKDQ WKH DFWXDO OHJXPH \LHOG 7KLV PD\ EH H[SODLQHG E\ WKH SODQWV JURZWK KDELW GLVWULEXWLRQ RQ WKH VRLO VXUIDFH DQG XSRQ WKH DVVRFLDWHG JUDVHV DQG ILQDOO\ E\ WKHLU PRUSKRORJLFDO FKDUDFWHULVWLFV EURDGOHDYHVf ZKLFK OHDGV WKH REVHUYHU WR RYHUHVWLPDWH WKH DFWXDO SHUFHQWDJH RI WURSLFDO OHJXPHV 7KLV LV HYLGHQW LI ZH FRPSDUH WKH UHVSRQVH VXUIDFHV IRU OHJXPH \LHOG DQG SHUFHQWDJH OHJXPH LQ )LJV DQG

PAGE 133

WHe XmYH HWFHW9 WD6H *WDAY RWWHW mf§} rr HFHW9 WD6H 7fD9V G D79 G e WD6 LR6 WH6V XrH HW A LV (IIHFW RI UHV/ IRU WKH ZHW VHDVRQ R b L}

PAGE 134

'D\V UHVW )LJ &RQWRXUV RI OHJXPH SHUFHQWDJH DV DIIHFWHG E\ OHQJWK RI UHVW SHULRG DQG OHYHOV RI JUD]LQJ SUHVVXUH LQ WKH ZHW VHDVRQ RI

PAGE 135

)LJ (IIHFW RI UHVW SHULRG DQG JUD]LQJ SUHVVXUH XSRQ OHJXPH SHUFHQWDJH IRU WKH ZHW VHDVRQ RI

PAGE 136

&+$37(5 9 6800$5< $1' &21&/86,216 $ JUD]LQJ WULDO ZDV FRQGXFWHG DW WKH (VWDFLQ ([SHULPHQWDO 7URSLFDO 3LFKLOLQJXH EHORQJLQJ WR ,QVWLWXWR 1DFLRQDO GH ,QYHVWLn JDFLRQHV $JURSHFXDULDV ,1,$3f DQG ORFDWHG NP IURP &DQWRQ 4XHYHGR DW ,R n 6 /DW DQG r n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n ]DWLRQ DQG NJ KD A RI VXSHUSKRVSKDWH $ PRGLILHG QRQURWDWDEOH FHQWUDO FRPSRVLWH UHVSRQVH VXUIDFH GHVLJQ ZDV XVHG ZKLFK LQFOXGHG VHOHFWHG WUHDWPHQW FRPELQDWLRQV &HUWDLQ WUHDWPHQWV ZHUH UHSOLFDWHG ZKLFK DFFRXQWHG IRU WKH H[SHULn PHQWDO XQLWV

PAGE 137

7KH FROOHFWLRQ RI GDWD VWDUWHG LQ 0D\ DQG HQGHG LQ -XQH $HULDO ELRPDVV '0f DYDLODEOH IRUDJH '0f JUDVV \LHOG '0f OHJXPH \LHOG '0f \LHOG RI ZHHGV '0f JUDVV SHUFHQWDJH DQG OHJXPH SHUFHQWDJH ZHUH HVWLPDWHG IRU HDFK JUD]LQJ F\FOH E\ D GRXEOHVDPSOLQJ SURFHGXUH 'XULQJ WKH ILUVW PRQWKV UDQGRP REVHUYDWLRQV RI RQH P HDFK ZHUH WDNHQ ZLWK D IRUDJH GLVN PHWHU )URP WKHVH VDPSOLQJ XQLWV WKUHH ZHUH UDQGRPO\ VHOHFWHG DQG FOLSSHG DW JURXQG OHYHO IRU DFWXDO \LHOG GHWHUPLQDWLRQV '0f DQG SHUFHQW FRPSRVLWLRQ )URP 6HSWHPEHU WKH QXPEHU RI VDPSOHV WDNHQ ZDV LQFUHDVHG WR XQLWV RI WKH VDPH VL]H DQG ILYH VDPSOLQJ XQLWV RXW RI WKH ZHUH UDQGRPO\ VHOHFWHG DQG FOLSSHG IRU WKH DERYH GHWHUPLQDWLRQV 7KHVH VDPSOHV ZHUH ODWHU KDQG VHSDUDWHG LQWR WKHLU FRPSRQHQWV DQG GULHG IRU KRXUV 7KH VXP RI WKH GU\ ZHLJKWV RI WKH FRPSRQHQWV \LHOGHG WKH WRWDO GU\ ZHLJKW RI WKH VDPSOH 7KHVH YDOXHV ZHUH XVHG IRU DHULDO ELRPDVV '0f DYDLODEOH IRUDJH '0f DQG IRU WKH LQGLYLGXDO FRPSRQHQW \LHOGV '0f (VWLPDWLRQV IRU JURZWK GXULQJ JUD]LQJ ZHUH PDGH WR DFKLHYH WKH WRWDO DYDLODEOH IRUDJH '0f 9LVXDO HVWLPDWHV RI ERWDQLFDO FRPSRVLWLRQ ZHUH DOVR WDNHQ LQ RUGHU WR GHWHUPLQH WKH DPRXQW RI WKH LQGLYLGXDO FRPSRQHQWV LQ WHUPV RI SHUFHQWDJH 7KH YLVXDO HVWLPDWH RI SHUFHQW \LHOG ZDV PDGH IRU WKH FRPSRQHQWV JUDVVHV OHJXPHV DQG ZHHGV )URP WKH UHVXOWV RI WKLV H[SHULPHQW EDVHG RQ WKH LQIRUPDWLRQ REWDLQHG IURP WKH UHVSRQVHV ZKLFK ZHUH PHDVXUHG WKH IROORZLQJ FRQn FOXVLRQV DSSHDU WR EH MXVWLILHG $HULDO ELRPDVV '0f DQG DYDLODEOH IRUDJH '0f ZHUH LQFUHDVHG DV WKH UHVW SHULRGV LQFUHDVHG DQG JUD]LQJ SUHVVXUH GHFUHDVHG IRUDJH

PAGE 138

GU\ PDWWHU RQ RIIHU ZDV LQFUHDVHGf 1R LQWHUDFWLRQV EHWZHHQ WKHVH WZR IDFWRUV ZHUH IRXQG VXJJHVWLQJ WKDW ORQJHU UHVW SHULRGV DQG ORZHU JUD]LQJ SUHVVXUHV LQGHSHQGHQWO\ ZHUH UHTXLUHG WR PDLQWDLQ KLJK GU\ PDWWHU SURGXFWLRQ *UDVV '0f \LHOG DQG JUDVV SHUFHQWDJH DSSHDUHG WR EH KLJKO\ VHQVLWLYH WR VKRUW UHVW SHULRGV DQG KLJK JUD]LQJ SUHVVXUHV 7KH QHJDWLYH LQIOXHQFH RI LQFUHDVHG JUD]LQJ SUHVVXUH XSRQ WKH JUDVV FRPSRQHQW ZDV SDUWLDOO\ RIIVHW E\ LQFUHDVLQJ WKH UHVW SHULRGV /HJXPH '0f \LHOGV DQG OHJXPH SHUFHQWDJHV ZHUH KLJKO\ VHQVLWLYH WR OHQJWKV RI UHVW SHULRG DQG OHVV VHQVLWLYH WR KLJK JUD]LQJ SUHVVXUHV 6KRUW UHVW SHULRGV IDYRUHG OHJXPH FRQWHQW EXW LWV JUHDWHVW FRQWULEXn WLRQ DSSHDUHG WR EH QHDU WKH PLGGOH UDQJH RI UHVW SHULRGV 7KLV UHVSRQVH DSSDUHQWO\ LV LQGHSHQGHQW RI OHQJWK RI JUD]LQJ SHULRG /HJXPH '0f \LHOG DQG OHJXPH SHUFHQWDJH VKRZHG D VOLJKW GHFOLQH GXULQJ WKH ILYH H[SHULPHQWDO VHDVRQV ZKLOH JUDVV \LHOG '0f DQG JUDVV SHUFHQWDJH LQFUHDVHG GXULQJ WKH VDPH SHULRG RI WLPH 7KH ZHHG FRPSRQHQW ZDV UHGXFHG E\ ORQJ UHVW SHULRGV DQG E\ ORZ JUD]LQJ SUHVVXUHV ,QWHUDFWLRQV RFFXUUHG EHWZHHQ WKHVH WZR YDULDEOHV %RWK KLJK JUD]LQJ SUHVVXUHV DQG VKRUW UHVW SHULRGV ZHUH UHVSRQVLEOH IRU LQFUHDVLQJ WKH \LHOG RI ZHHGV '0f 7KH UHVXOWV RI WKLV VWXG\ VXJJHVW WKDW D JUD]LQJ PDQDJHPHQW V\VWHP ZKLFK FRPELQHV D PRGHUDWH UHVW SHULRG DQG D PRGHUDWH OHYHO RI JUD]LQJ SUHVVXUH ZRXOG EH DQ RSWLPXP PDQDJHPHQW VWUDWHJ\ WR PDLQWDLQ D ORZ DPRXQW RI ZHHGV D ODUJH DPRXQW RI OHJXPHV DQG KLJK \LHOGV RI DYDLODEOH IRUDJH 7KH RWKHU WZR YDULDEOHV GD\V JUD]LQJ DQG OHYHOV RI SKRVSKRUXV IHUWLOL]Dn WLRQ KDG QHJOLJLEOH HIIHFWV XSRQ WKH UHVSRQVH RI WKH SDVWXUH VZDUG

PAGE 139

$33(1',;

PAGE 140

7DEOH ELRPDVALIF0A7f UHVVORQ fHIILFLHQWV DQG SUREDELOLWLHV ORPDVV J '0P f IRU WKH ZHW VHDVRQ RI IRU DHULDO 5(63216( 0($1 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352' /,1($5 7 7 7 48$'5$7,& 7 &5266352'8&7 7 727$/ 5(*5(66 ,7 5(6,'8$/ ') 6% 0($1 648$5( )5$7,2 352% /$&. 2) ),7 7% 385( (5525 -2 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, f7 ; ; ;7 ; ; ;, r; ;r; ,7 ;r; 7 ;r; 7 ;r; ;, }; 7 ; r ; 7 ;t; 7 ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, 6 ; ; 7 7 ;7 7

PAGE 141

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV ELRPDVV J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &*() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 6% 0($1 648$5( )5$7,2 /$&. 2) ),7 385( (5525 WR 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 ,17(5&(37 ;, ; ; ; ;r; ;r; ;r; ;; ;; ;; ;r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 ;, ; ; ; IRU DHULDO 35 2% 352% 352% 2 352%

PAGE 142

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 6% 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, ; ; % ; ;r; ;r; ;r; ;r; ;; ;r; ;, r; ;r; ; ; ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, %% ; ; ;

PAGE 143

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352% /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$ 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 % ;, ; ; ; ;r;, ; r; ;; ; ; ;r; ;; ;; %% ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; % ; ;

PAGE 144

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DHULDO ELRPDVV J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 56*8$5( )5$7 352% /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7 352% ,17(5&(37 ;, ; ; ; ;, r; ;}; ;r; ;; ;}; ;r; ;r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 145

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( ,17(5&(37 ;, ; ; ; ;, r; ;, r; ;r; ;r; ;; ;}; ;r; ;r; ;r; ;; )$&725 ') 66 ;, ; ; ; 5648$5( )5$7 35 2% 0($1 648$5( )5$7,2 352% 67' '(9 75$7,2 352% 0($1 648$5( )5$7,2 352%

PAGE 146

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5642$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 6% 5648$5( )5$7,2 352' /,1($5 % 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;, r; ;:; ;t; ;+I; ;m; ;r; % ;, r; ;r; ;:; ;; )$&725 ') 6% 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 147

7DEOH $QDO\VLV RI YDULDQFH IRUDJH J '0P f IRU 5(63216( 0($1 5227 06( 56*8$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') ,17(5&(37 ;, ; ; ; ;r; ;; ;r; ;; ;r; ;}; [Lr[ ;r; ;r; ;r; )$&725 ') ;, ; ; ; UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH WKH ZHW VHDVRQ RI 7<3( 66 5648$5( )5$7,2 352% 66 0($1 648$5( )5$7, 352% (67,0$7( 67' '(9 75$7,2 352% 66 0($1 648$5( )5$7,2 352%

PAGE 148

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 52'7 06( 56*8$5( &2() 2) 9$5,$7,21 2 5(*5(66,21 /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ,17(5&(37 ;, ; ;r; ;r; ;r; ;; ;r; ;r; ;; ;r; ;r; ;r; )$&725 ;, ; ; ; '3 7<3) 66 56*8$5( )5$7,2 352' % % ') 66 0($1 648$5( )5$7,2 352% ') (67,0$7( 67' '(9 75$7,2 352% %22 % ') 66 0($1 648$5( )5$7,2 352%

PAGE 149

7DEOH $QDO\VLA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU DYDLODEOH IRUDJH J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &52%6352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 3523 /$&. 2) ),7 385( (5525 % 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 35 2% ,17(5&(37 2% ;, ; ; ; ;, r; ;r; ;}; ;r; ;}; ;r; ; ; ;r; ;; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; % ; ;

PAGE 150

7DEOH $QDO\VLA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 m 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ SL 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 'I (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; L ;r; ;; ;r; ;; ;r; ;r; L ;, r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ; X! 21

PAGE 151

7DEOH $QDO\VLVA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ,0 7<3( 66 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ QU 66 0($1 648$5( )5$7,2 3523 /$&. 2) ),7 385( (5525 WR 727$/ (5525 3$5$0(7(5 f (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, ; ; ; ;, r; ;r; ;r; ;r; ;r; ;r; a ;, r; ;r; ;r; ;m; )$&725 ') 66 0($1 648$5( )5$7,2 35 2% ;, ; ; ; }

PAGE 152

7DEOH $QDO\VLVA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( 2 &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 26 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ '3n 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, ; ; ; ;, r; ;, r; ;r; ;r; ;r; ;r; ;r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 153

7DEOH $QDO\VLA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 '3 7<3( %6 5648$5( )5$7,2 352' /,1($5 48$'5$7,& % &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ '3 66 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 f (67,0$7( 6,' '(9 75$7,2 352' ,17(5&(37 ;, ; ; ; ;, r; ; r; ;r; ;r; ;r; ;r; ;r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ; % ; ; ;

PAGE 154

7DEOH f $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU JUDVV \LHOG J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5287 06( 56*8$5( f &2() 2) 9$5,$7,21 5(*5(66,21 '3 7<3( 66 56*8$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 $ % 727$/ 5(*5(66 5(6,'8$/ O!) 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 $ 385( (5525 727$/ (5525 3$5$0(7(5 P (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; f§ ;, r; ;r; ; } ; ;, ; ;r; ;r; ; ; ;r; ;r; ;r; )$&725 'Ln 66 0($1 648$5( )5$7,2 352% ;, OL ; ; ;

PAGE 155

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV \LHOG J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 QU 7<3( 66 5648$5( )5$7,2 /,1($5 48$'5$7,& &5266352'8&7 8 727$/ 5(*5(66 5(6,'8$/ QU 66 0)$1 648$5( )5$7,2 /$&. 2) ),7 2 385( (5525 727$/ (5525 3$5$0(7(5 ', (67,0$7( 67' '(9 75$7,2 ,17(5&(37 ;, 2 ; ; ; ;r; ; ; ;r; ;, r; ;r; ;r; ;, r; % ;r; ;}; % ;r; )$&725 ') 66 0($1 648$5( )5$7,2 ;, ; ; ; IRU OHJXPH 35 2% 2 35 2% 2 352' 352% 2 2 2

PAGE 156

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( '% 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352% /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 -2 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 % ;, ; ; ; ;, r; ;; 2 ;r; ;r; ;r; L ;r; ;, r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, 2n ; ; ;

PAGE 157

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 'I 7<3( 66 5648$5( )5$7,2 3525 /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ QU 66 0($1 648$5( )5$7-2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,* 352' ,17(5&(37 ;, ; ; ; ;, r; 2 ;, ; ;r; ;r; ;r; ;; % ; r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352' ;, ; ; ;

PAGE 158

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOG J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 QU 7<3( 66 5 648$5( )5$7,2 35 2% /,1($5 48$'5$7,& &5266352'8&7 $ 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 35 f /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, ; 222O ; @ ; ;, r; ;-8; ;r; ;; ;; ;; ; ; ;r; ;r; ;; )$&7'5 ') 66 0($1 648$5( )5$7,2 352' ;, ; V ; f! ;

PAGE 159

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU OHJXPH \LHOGV J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 7<3( 66 5648$5( )5$7,2 352' /,1($5 WLOO 222O 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 -2 727$/ (5525 3$5$0(7(5 QL (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; 222O ; @ ; ;r; ;; ;r; ;; ;; ;r; ;, }; ;r; ;r; ;r; )$&725 QU 6% 0($1 648$5( )5$7,2 352% ;, ; 222O ; ; 1 /Q

PAGE 160

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( G 56*8$5( &2() 2) 9$5,$7,21 5(*5(66,21 'O I <3( 36 5 648$5( )f§5$7 352' /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 ,G 5(6,'8$/ QU 66 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 QU (67,0$7( 67' '(9 75$7 352% ,17(5&(37 L ;, @ ; L ; L ; L ;, r; L ;; L ;; L ;r; L ;r; L ;}; L ;r; M ;r; L ;r; L ;; L )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ; U!

PAGE 161

7DEOH $QDO\VLA RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV RI ZHHGV J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 'O 7<3( 26 5 648$5( )5$7,2 35 2% /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ OfI 26 0($1 648$5( )5$7,2 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 QU (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;r; ;-8; ;r; ;r; ;r; ;r; ; }; ;r; ;; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ; ,%

PAGE 162

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0P f IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 6n} 5(*5(66,21 'I 7<3( 6% 5648$5( )5$7,2 352% /,1($5 48$'5$7,& &5266352'8&7 t 727$/ 5(*5(66 5(6,'8$/ QU 66 0($1 648$5( )5$7,2 35 RQ /$&. 2) ),7 222385( (5525 -2 727$/ (5525 3$5$0(7(5 f (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, 2 ; @ ; ; @ ;, }; ;r; ;}; ;; ;r; ;; ; }; ;r; ;r; ;; )$&725 ') 6% 0($1 648$5( )5$7,2 352% ;, ; LW ; LW ; 3n

PAGE 163

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG RI ZHHGV J '0P f IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 2) 7<3( 66 5f§648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, ; ; ; ;r; ; r; ;r; ; ; ;r; ;r; ;, }; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352' ;, ; ; (L ;

PAGE 164

7DEOH $QDO\VLV RI YDULDQFH RI ZHHG NJ'0PAf IRU 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 P /,1($5 ‘ 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ QU /$&. 2) ),7 385( (55'5 727$/ (5525 3$5$0(7(5 f ,17(5&(37 ;, ; ; ; ;r; ;r; ;}; ;m; ;r; ;r; ;r; ;r; ;; ;r; )$&725 ') ;, ; ; ; U} UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU \LHOG WKH ZHW VHDVRQ RI } 2 7<3( 66 5 648$5( )5$7 352' 66 0($1 648$5( )5$7,2 352% (67,0$7( 67' '(9 75$7,2 352% " 66 0($1 648$5( )5$7,2 352'

PAGE 165

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 -26 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 f 7<3( 66 5648$5( )5$7,2 352 /,1($5 % 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, a ; ; ; ;r; a ;r; ;r; ;; ;r; ;; ;r; ;r; ;r; ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 166

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( 66 5 648$5( )5$7,2 352% /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ U!U 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 'O (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, '% ; ; ; f§ ;, }; ;r; ;r; ;r; ;r; ;r; % ;r; ;r; ;t; ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 167

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 25227 06( 8 5648$5( &2() 2) 9$5,$7,21 f 5(*5(66,21 ') 7<3( 66 5648$5( )5$7,2 352% /,1($5 48$'5$7,& &52%6352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7, 352% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;, r; 2 ;r; ;r; ;r; % ;r; ;r; ; ; ;r; ;; 2 ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 168

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 ') 7<3( QV 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ QU 6% 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 '3 (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, 2 ; ; ; ;, r; ;; ;r; ;; ;r; ;; ;r; ;r; ;; a ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352' ;, ; ; ;

PAGE 169

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ JUDVV IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5287 06( 5648$5( &2() 2) 9$5,$7,21 5(*5(66,21 5 7<3( 65 5648$5( )5$7,2 352' /,1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ,f) 66 0($1 648$5( )5$7-2 35 2% /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 QU (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 O 222O ;, L 2 ; L ; O ; L ;r; L 2 ;; L ; t ; O ;; L ;; L ;r; O ;r; L ;r; L ;; L ;r; L )$&725 )f) 66 0($1 648$5( )5$7,2 352% ; ; ; ;

PAGE 170

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( &2() 2) 9$5,$7,21 f 5(*5(66,21 ,}O 7<3( 06 5 648$5( )5$7 35 RQ /,1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ '. 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ') (67,0$7( 67' '(9 75$7,2 352' ,17(5&(37 ;, 2 ; ; ; ;r; ;; ;r; ;r; ;r; ;r; ;, r; ;r; ;r; ;; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ; Ur

PAGE 171

$QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 5648$5( f &2() 2) 9$5,$7,21 5(*5(66,21 QU 7<3( 66 5648$5( )5$7,2 352' /,1($5 48$'5$7,& f &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 352' /$&. 2) ),7 385( (5525 727$/ (5525 3$5$0(7(5 ', (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;r; ;r; ;r; ;; ;r; ;r; 2 ;r; ;r; ;r; ;r; )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 172

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 -! 5227 06( 56*8$5( &2() 2) 9$5,$7,21 5(*5(66,21 LQ 7<3( 66 5 648$5( )5$7,2 35 2% /,1($5 48$'5$7,& &526%352'8&7 $ 727$/ 5(*5(66 5(6,'8$/ f 62 0($1 648$5( )5$7,2 35 2% /$&. 2) ),7 +2 385( (5525 727$/ (5525 3$5$0(7(5 QU (67,0$7( 67' '(9 75$7 35 2% ,17(5&(37 L ;, L 2 ; L ; O ; L ;, r; L ;, r; L ;r; L ;; L ;r; L ;}; ;, }; O ;r; O ;r; L a ;; L )$&725 Of) 66 0($1 648$5( )5$7,2 352' ;, ; ; ;

PAGE 173

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH IRU WKH GU\ VHDVRQ RI 5(63216( 0($1 5227 06( 56*8$5( &2() 2) 9$5,$7,21 5(*5(66,21 2I 7<3( 26 56*8$5( )5$7,2 352' /, 1($5 48$'5$7,& &526%352'8&7 727$/ 5(*5(66 5(6,'8$/ 3L 62 0($1 648$5( )5$7,2 35/ /$&. 2) ),7 385( (5525 LR 727$/ (5525 3$5$0(7(5 ' (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;; ;, r; ;r; ; ; ;r; ;r; ;, r; ;r; ;; ;; )$&725 ') 66 0($1 648$5( )5$7,2 352' ;, ; ; } ;

PAGE 174

7DEOH $QDO\VLV RI YDULDQFH UHJUHVVLRQ FRHIILFLHQWV DQG SUREDELOLWLHV IRU YLVXDO HVWLPDWLRQ OHJXPH IRU WKH ZHW VHDVRQ RI 5(63216( 0($1 5287 06( 56*8$5( &2() 2) 9$5,$7,21 5(*5(66,21 !, 7<3( 66 56*8$5( )5$7,2 3523 /, 1($5 48$'5$7,& &5266352'8&7 727$/ 5(*5(66 5(6,'8$/ ') 66 0($1 648$5( )5$7,2 35 2% /$&. 2) ),7 6 385( (5525 727$/ (5525 3$5$0(7(5 RU (67,0$7( 67' '(9 75$7,2 352% ,17(5&(37 ;, ; ; ; ;r; ;r; @ ;}; ;r; ;r; 2 ;r; ; r; ;r; ;Z; ;r; @ )$&725 ') 66 0($1 648$5( )5$7,2 352% ;, ; ; ;

PAGE 175

/,7(5$785( &,7(' $FRVWD6ROL] 0 'LYLVLRQV ILWRJHRJUDILFDV \ IRUPDFLRQHV JHRERW£QLFDV GHO (FXDGRU ,QVWLWXWR (FXDWRULDQR GH &LHQFLDV 1DWXUDOHV 4XLWR (FXDGRU S $ODUFRQ ( DQG /RWHUR (VWDEOHFLPLHQWR IHUWLOL]DFLQ \ PDQHMR GH ODV SULQFLSDOHV JUDPLQHDV \ OHJXPLQRVDV IRUUDMHUDV HQ GRV SLVRV WUPLFRV GH &RORPELD ,QVWLWXWR &RORPELDQR $JURSHFXDULR ,&$f &RORPELD S PLPHRf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f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

PAGE 176

3 6KDUSH DQG 3 +D\GRFN 6RPH IDFWRUV DIIHFWLQJ WKH JURZWK RI /RWRQRQLV /RWRQRQLV EDLQHVLLf $XVW ([S $JULH $QLP +XVK &KDYHUUD + 0DQHMR GH SRWUHURV 3DVWRV \ JDQDGR SDUD OD FRVWD $WO£QWLFD ,QVWLWXWR &RORPELDQR $JURSHFXDULR ,&$f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
PAGE 177

DQG : : %U\DQ (IIHFWV RI VRLOV IHUWLOL]HUV DQG VWRFNLQJ UDWHV RQ SDVWXUHV DQG EHHI SURGXFWLRQ $XVW ([S $JULH $QLP +XVE )DODGH $ 7KH HIIHFW RI SKRVSKRUXV RQ JURZWK DQG PLQHUDO FRPSRVLWLRQ RI ILYH WURSLFDO JUDVVHV (DVW $IU $JULH )RU )DUIDQ & (IHFWR GH SUDFWLFDV FXOWXUDOHV HQ OD SURGXFFLQ GH VHPLOODV GH SODQWDV IRUUDMHUHV WURSLFDOHV 7HVLV ,QJHQLHUR $JURQRPR 8QLYHUVLGDG 7HFKQLFD GH 0DQDEL 3RUWRYLHMR 0DQDEL (FXDGRU S )HEOHV DQG & 3DGLOOD (IHFWR GHO SDVWRUHR HQ DVRFLDn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f 0RQWHYLGHR 8UXJXD\ S *DUWQHU $ ( )HUJXVRQ 5 : :DONHU DQG ( $ *RZDUG (YDOXDWLQJ SHUHQQLDO JUDVVOHJXPH RQ WKH $WKHUWRQ WDEOHODQG RI 1RUWK 4XHHQVODQG 4XHHQVO $JULH $QLP 6FL *RPH] 7 (VWDEOLVKPHQW PHWKRGV DQG FRPSDUDWLYH SHUVLVn WHQFH RI ILYH WURSLFDO OHJXPHV LQ JUDVV VRGV 3K' 'LVVHUWDWLRQ 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )ORULGD S *URI %
PAGE 178

+DUG\ ) 5HSRUW RQ D YLVLW WR WKH 5LYHULQH EHOW RI (FXDGRU ,QWHU$PHULFDQ ,QVWLWXWH RI $JULFXOWXUDO 6FLHQFH 7XUULDOED &RVWD 5LFD 5SW 1R S +DUULV : 'HIROLDWLRQ DV D GHWHUPLQDQW RI WKH JURZWK SHUVLVWHQFH DQG FRPSRVLWLRQ RI SDVWXUH ,UL 5 :LOVRQ HGf 3ODQW UHODWLRQV LQ SDVWXUHV &6,52 %ULVEDQH $XVWUDOLD 3 +HDG\ + ) *UD]LQJ V\VWHPV 7KHUPV DQG GHILQLWLRQV 5DQJH 0DQDJH +HDWK 0 ( *UDVVODQG DJULFXOWXUH ,Q 0 ( +HDWK 6 0HWFDOIH DQG 5 ) %DUQHV HGVf )RUDJHV ,RZD 6WDWH 8QLYHUn VLW\ 3UHVV $PHV ,RZD 3 +RGJVRQ DQG + 2OOHUHQVKDZ 7KH IUHTXHQF\ DQG VHYHULW\ RI GHIROLDWLRQ RI LQGLYLGXDO WLOOHUV LQ VHUVWRFNHW VZDUGV %U *UDVVOG 6RF +ROGULGJH / 5 /LIH ]RQH HFRORJ\ 7URSLFDO 6FLHQFH &HQWHU 6DQ -RVH &RVWD 5LFD S +XGJHQV 5 ( 7KH FRPSDWLELOLW\ SHUVLVWHQFH DQG QXWULWLYH YDOXH RI JUDVVOHJXPH DVVRFLDWLRQV LQ WKH ZHWGU\ WURSLFV RI FRDVWDO (FXDGRU 0 6 7KHVLV 8QLYHUVLW\ RI )ORULGD *DLQHVn YLOOH )ORULGD S +XPSKUH\V / 5 7URSLFDO SDVWXUH DQG IRGGHU FURSV ,Q : $ 3D\QH HGf ,QWHUPHGLDWH WURSLFDO DJULFXOWXUH VHULHV /RQJPDQ *URXS /LPLWHG /RQGRQ 3 $ JXLGH WR EHWWHU SDVWXUH IRU WKH WURSLFV DQG VXEWURSLFV :ULJKW 6WHSKHQVRQ DQG &R $XVWUDOLD S ,1,$3 ,QVWLWXWR 1DFLRQDO GH ,QYHVWLJDFLRQHV $JURSHFXDULDVf ,QIRUPHV WFQLFRV DQXDOHV GHO SURJUDPD GH SDVWRV \ JDQDn GHUD GH OD (VWDFLQ ([SHULPHQWDO 7URSLFDO 3LFKLOLQJXH 4XLWR (FXDGRU -HQVHQ 3 1 DQG & 0 6FKXPDFKHU &KDQJHV LQ SUDLULH SODQW FRPSRVLWLRQ 5DQJH 0DQDJH -RQHV 5 0 (IIHFW RI VWRFNLQJ UDWH DQG JUD]LQJ IUHTXHQF\ RQ D 6LUDWUR 0DFURSWLOLXP DWURSXUSXUHXPf6HWDULD DQFHSV FY 1DQG\ SDVWXUH $XVW ([S $JULH $QLP +XVE -RQHV 0 % DQG / 0 0 GH )UHLWDV 5HVSRVWDV GH TXDWUR OHJXPLQRVDV WURSLFDLV D IRVWRUR SRWDVVLR H FDOFDULR QXP /DWRV VROR 9HUPHOKR$PDUHOR GH FDPSR FHUUDGR 3HVTXLVD $JURSHFXDULD %UDVLOHLUD 6HULH $JU .HQQHG\ 5 7KH VLFNOHGUDW D FLUFXODU TXDGUDW PRGLILFDn WLRQ XVHIXO LQ JUDVVODQG VWXGLHV 5DQJH 0DQDJH

PAGE 179

.UHWVFKPHU $ ( 1HZ OHJXPHV IRU WKH /DWLQ $PHULFDQ WURSLFV $JULH 5HV &HQWHU )RUW 3LHUFH )ORULGD 5SW 3 'LVWULEXWLRQ LQWURGXFWLRQ DQG HYDOXDWLRQ RI WURSLn FDO SDVWXUH VSHFLHV $JULH 5HV &HQWHU 5SW )/ )RUW 3LHUFH )ORULGD 3 /LWWHOO 5 & DQG 0RWW &RPSXWHU DVVLVWHG GHVLJQ DQG DQDO\VLV RI UHVSRQVH VXUIDFH H[SHULPHQWV LQ DJURQRP\ 6RLO &URS 6FL 6RF )OD 3URF /XGORZ 0 0 DQG / :LOVRQ 6WXGLHV RQ WKH SURGXFWLYLW\ RI WURSLFDO SDVWXUH SODQWV ,, *URZWK DQDO\VLV SKRWRV\QWKHVLV DQG UHVSLUDWLRQ RI VSHFLHV RI JUDVVHV DQG OHJXPHV LQ D FRQWUROOHG HQYLURQPHQW $XVW $JULH 5HV 0DQKDHV 6 6 DQG 'REHUHLQHU (IIHLWR GR IRVIRUR WHPSn HUDWXUD D XQLGDGH GR VROR QD QRGXODFDR H QR GHVPYROYLPLHQWR GH GXD YDULHGDGH GH VR\D SHUHQH *O\FLQH ZLJKWLLf 3HVTXLVD $JURn SHFXDULD %UDVLOHLUD 0DUDVFKLQ ( 5HVSRQVH RI D FRPSOH[ WURSLFDO SDVWXUH PL[WXUH WR GLIIHUHQW JUD]LQJ PDQDJHPHQW V\VWHPV 3K' 'LVVHUWDWLRQ 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )ORULGD S 0DWFKHV $ 3DVWXUH UHVHDUFK PHWKRGV 3URF 1DW &RQI )RUDJH 4XDOLW\ (YDOXDWLRQ DQG 8WLOL]DWLRQ 8QLYHUVLW\ RI 1HEUDVND /LQFROQ 1HEUDVND 0FOOUR\ 5 ,QWURGXFWLRQ DO FXOWLYR GH ORV SDVWRV WURSLn FDOHV 7UDG GH DO D (G ,QJOHVDf 87(+$ 0H[LFR S 0FOYRU 5 -RQHV & *DUGLQHU DQG : + :LQWHU 'HYHORSPHQW RI OHJXPH EDVHG SDVWXUHV IRU EHHI SURGXFWLRQ LQ GU\ WURSLFDO DUHDV RI 1RUWKHUQ $XVWUDOLD ;,9 ,QW *UDVVOG &RQJ 3URF 8QLYHUVLW\ RI .HQWXFN\ /H[LQJWRQ .HQWXFN\ S 0F0HHFNDQ & 3 *UD]LQJ PDQDJHPHQW DQG DQLPDO SURGXFWLRQ 9,, ,QW *UDVVOG &RQJ 3URF 1HZ =HDODQG 0HGLQD (VWXGLR GH OD GLVSRQLELOLGDG GH KLHUUR PDQn JDQHVR \ ]LQF GH OD ]RQD GH 4XHYHGR (FXDGRU 7HVLV ,QJHQLHUR $JURQRP\ 8QLYHUVLGDG GH *XD\DTXLO *XD\DTXLO *XD\RV (FXDGRU S 0LOIRUG 5 DQG 3 +D\GRFN 7KH QXWULWLYH YDOXH RI SURWHLQ LQ VXEWURSLFDO SDVWXUH VSHFLHV JURZQ LQ VRXWKHDVW 4XHHQVODQG $XVW ([S $JULH $QLP +XVE 0LQVRQ DQG 5 0LOIRUG 7KH YROXQWDU\ LQWDNH DQG GLJHVWLELOLW\ RI GLHWV FRQWDLQLQJ GLIIHUHQW SURSRUWLRQV RI OHn JXPHV DQG PDWXUH SDQJRODJUDVV 'LJLWDULD GHFXPEHQVf $XVW ([S $JULH $QLP +XVE

PAGE 180

0RRUH $ : 6\PELRWLF QLWURJHQ IL[DWLRQ LQ D JUD]HG WURSLFDO JUDVVOHJXPH SDVWXUH 1DWXUH 0RWW *UD]LQJ SUHVVXUH DQG WKH PDQDJHPHQW RI SDVWXUH SURGXFWLRQ 9,,, ,QW *UDVVOG &RQJ 3URF 8QLYHUVLW\ RI 5HDGLQJ (QJODQG 3 (YDOXDWLQJ IRUDJH SURGXFWLRQ M>Q 0 ( +HDWK ( 0HWFDOIH DQG 5 ) %DUQHV HGVf )RUDJHV ,RZD 6WDWH 8QLYHUn VLW\ 3UHVV $PHV ,RZD 3 1XWULHQW UHF\FOLQJ LQ SDVWXUHV ,Q $ 0D\V HGf )RUDJH IHUWLOL]DWLRQ 7KH $PHULFDQ 6RFLHW\ RI $JURQRP\ 0DGLVRQ :LVFRQVLQ 3 *UD]LQJ PDQDJHPHQW RI WURSLFDO OHJXPHJUDVV DVVRFLDn WLRQV WK $QQ &RQI RQ /LYHVWRFN DQG 3RXOWU\ LQ /DWLQ $PHULFD 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )ORULGD $ 3RWHQWLDO SURGXFWLYLW\ RI WHPSHUDWH DQG WURSLFDO JUDVVODQG V\VWHPV ;,9 ,QW *UDVVOG &RQJ 3URF 8QLYHUVLW\ RI .HQWXFN\ /H[LQJWRQ .HQWXFN\ 3 DQG + / /XFDV 7KH GHVLJQ FRQGXFW DQG LQWHUSUHn WDWLRQ RI JUD]LQJ WULDOV RQ FXOWLYDWHG DQG LPSURYHG SDVWXUHV 9, ,QW *UDVVOG &RQJ 3URF 3HQQV\OYDQLD 6WDWH 8QLYHUVLW\ 6WDWH &ROOHJH 3HQQV\OYDQLD 3 DQG ( 0RRUH )RUDJH HYDOXDWLRQ LQ SHUVSHFWLYH 3URF 1DW &RQJ )RUDJH 4XDOLW\ (YDOXDWLRQ DQG 8WLOL]DWLRQ 8QLYHUVLW\ RI 1HEUDVND /LQFROQ 1HEUDVND 1HPH 1 $ DQG / $ & /RYDGLQL (IHLWR GH DGXERV IRVIDWDGRV H FDOFDULR QD SURGXFDR GH IRUUDJHP GH VRMD SHUHQH *O\FLQH MDYDQLFD /f HP WHUUD GH &HUUDGR %UDJDQWLD 1RUPDQ 0 7 5HODWLRQVKLSV EHWZHHQ OLYHZHLJKW JDLQ RI JUD]LQJ EHHI VWHHUV DQG DYDLODELOLW\ RI 7RZQVYLOOH OXFHUQH ;, ,QW *UDVVOG &RQJ 3URF 6XUIHUnV 3DUDGLVH $XVWUDOLD 3 2]DQQH 3 DQG 7 & 6KDZ 3KRVSKDWH VRUSWLRQ E\ VRLOV DV D PHDVXUH RI WKH SKRVSKDWH UHTXLUHPHQW IRU SDVWXUH JURZWK $XVW $JULH 5HV 3DODFLRV & (IHFWRV GH OD QRGXODFLRQ SHSHWL]DFLRQ \ QXWULn FLQ PLQHUDO HQ OD ILIDFLRQ GH QLWUJHQR SRU &HQWURVHPD SXEHVFHQV %HQWK D QHYHO GH LQYHUQDGHUR 7HVLV ,QJHQLHUR $JURQRPR 8QLn YHUVLGDG GH *XD\DTXLO *XD\DTXLO *XD\DV (FXDGRU S 3DODGLQHV / DQG GH $OED $FHSWDFLQ GH IRUUDMHV WURSLFDOHV SRU HO JDQDGR 7XUULDOED 3DUHGHV (YDOXFDLRQ GH GLIHUHQWHV VLVWHPDV GH SDVWRUHR FRQ ERYLQRV HQ SDVWR JXLQHD 3DQLFXP PD[LPXP -DFTf 7HVLV ,QJHQLHUR $JURQRPR 8QLYHUVLGDG GH *XD\DTXLO *XD\DTXLO *XD\DV (FXDGRU S

PAGE 181

3HWHUVHQ 5 + / /XFDV DQG 0RWW 5HODWLRQVKLS EHWZHHQ UDWH RI VWRFNLQJ DQG SHU DQLPDO DQG SHU DFUH SHUIRUPDQFH RQ SDVWXUH $JURQ 5HXWHU 7KH UHFRJQLWLRQ DQG FRUUHODWLRQ RI WUDFH HOHPHQW GHILFLHQFLHV ,Q ' 1LFKODV DQG $ 5 (JDQ HGVf 7UDFH HOHPHQWV LQ VRLOSODQWDQLPDO V\VWHPV $FDGHPLF 3UHVV 1HZ +\SDUUKHQLD UXID 1HHVf 6WDSI@ VRPHWLGR D SDVWRUHR FRQWLQXR 7HVLV ,QJHQLHUR $JURQRPR 8QLYHUVLGDG GH *XD\DTXLO *XD\DTXLO *XD\DV (FXDGRU S (VWLPDWLQJ IRUDJH \LHOG ZLWK D GLVN PHWHU 06 7KHVLV 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )ORULGD S 6HUUDR ( $ 7KH XVH RI UHVSRQVH VXUIDFH GHVLJQ LQ WKH DJURQRPLF HYDOXDWLRQ RI D JUDVVOHJXPH PL[WXUH XQGHU JUD]LQJ 3K' 'LVVHUWDn WLRQ 8QLYHUVLW\ RI )ORULGD *DLQHVYLOOH )ORULGD S 6HUYLFLR 1DFLRQDO GH +LGURORJLD \ 0HWHURORJLD GHO (FXDGRU %ROHWLQHV FOLPDWROJLFRV $QR 1R 4XLWR (FXDGRU 6KDZ 1 + DQG : : %U\DQ 7URSLFDO SDVWXUH UHVHDUFK 3ULQFLSOHV DQG PHWKRGV &RPPRQZHDOWK $JULH %XUHD[ $OGHQ 3UHVV 2[IRUG (QJODQG

PAGE 182

6Q\GHU + DQG $ ( .UHWVFKPHU 7URSLFDO OHJXPH UHVSRQVH WR OLPH DQG VXSHUSKRVSKDWH LQ ROGVPDU ILQH VDQG 6RLO &URS 6FL 6RF )OD 3URF 6SHGGLQJ & 5 : 7KH SK\VLRORJLFDO EDVLV RI JUD]LQJ PDQDJHn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n0DQQHWMH / 5HODWLRQV EHWZHHQ SDVWXUHV DWWULEXWHV DQG OLYH ZHLJKW JDLQV RQ D VXEWURSLFDO SDVWXUH ;,, ,QW *UDVVOG &RQJ 3URF 0RVFRZ 5XVVLD 3 $Q LQWURGXFWLRQ WR JUDVVODQG YHJHWDWLRQ DQG LWV PHDVXUHPHQW ,Q / Wn0DQQHWMH HGf 0HDVXUHPHQW RI JUDVVODQG YHJHWDWLRQ DQG DQLPDO SURGXFWLRQ &RPPRQZHDOWK $JULH %XUHDX[ )DPKDQ 5R\DO %XFNV (QJODQG 3 7KH UROH RI D OHJXPH EDVHG SDVWXUH IRU EHHI SURGXFn WLRQ LQ WKH WURSLFV ,9 :RUOG &RQI $QLP 3URG %XHQRV $LUHV $UJHQWLQD 7RWKLOO & &RPSDUDWLYH DVSHFWV RI WKH HFRORJ\ RI SDVWXUHV ,Q 5 :LOVRQ HGf 3ODQW UHODWLRQV LQ SDVWXUHV &RPPRQZHDOWK 6FLHQWLILF DQG ,QGXVWULDO 5HVHDUFK 2UJDQL]DWLRQ %ULVEDQH $XVWUDn OLD 3 DQG 0 / 3HWHUVHQ %RWDQLFDO DQDO\VLV DQG VDPSOLQJ 7DPH SDVWXUHV ,Q 3DVWXUH DQG UDQJH UHVHDUFK WHFKQLTXHV ,WKDFD 1HZ
PAGE 183

9LFHQWH&KDQGOHU ,QWHQVLYH PDQDJHPHQW RI SDVWXUHV DQG IRUDJHV LQ 3XHUWR 5LFR ,Q ( %RUQHPLV]D DQG $ $OYDUDGR HGVf 6RLO PDQDJHPHQW LQ WURSLFDO $PHULFD 1RUWK &DUROLQD 6WDWH 8QLYn HUVLW\ 5DOHLJK 1RUWK &DUROLQD 3 9LOHOD / $ (IHLWR GH WD[D GH ORWDFRR H GH DOLPHQWDFDR VXSOHPHQWDU VREUH D SURGXFDR GH OHLWH GXUDQWH R SHULRGR GH VHFD 5HYLVWD GH OD 6RFLHGDGH %UDVLOHLUD GH =RRWHFKQLD 9LOODVPLO $WHQFLR %DUFHQDV $ &DVDQRYD 0 8UGDQHWD DQG + 7LPP 5HVSRQVH VXUIDFH GHVLJQV DQG OLQHDU SURJUDPPLQJ DSSOLHG WR FDWWOHIRUDJHIHHGLQJ V\VWHPV LQ WKH WURSLFV ,Q ( & 'ROO DQG 0RWW HGVf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

PAGE 184

%,2*5$3+,&$/ 6.(7&+ 5DXO $ 6DQWLOODQ ZDV ERUQ 'HFHPEHU LQ 5LREDPED &KLPERUD]R SURYLQFH (FXDGRU WR 5LJREHUWR DQG ,WDOD 6DQWLOODQ )URP 0D\ WR -DQXDU\ KH DWWHQGHG WKH 8QLYHUVLGDG GH *XD\DTXLO (FXDGRU DQG UHFHLYHG WKH ,QJHQLHUR $JURQRPR GHJUHH )RU VL[ PRQWKV KH ZRUNHG LQ WKH 3URJUDPD 1DFLRQDO GHO %DQDQR DQG LQ $XJXVW RI KH MRLQHG WKH ,QVWLWXWR 1DFLRQDO GH ,QYHVWLJDn FLRQHV $JURSHFXDULDV ,1,$3f ZRUNLQJ LQ WURSLFDO SDVWXUHV DW WKH ([WDFLRQ ([SHULPHQWDO 3LFKLOLQJXH ,Q KH ZDV DZDUGHG D JUDQW WR UHFHLYH D \HDUnV WUDLQLQJ LQ FDWWOH SURGXFWLRQ DW WKH &HQWUR ,QWHUQDFLRQDO GH $JULFXOWXUDO 7URSLFDO &,$7f LQ &DOL &RORPELD ,Q KH UHFHLYHG WKH 0DVWHU RI 6FLHQFH GHJUHH LQ DJULFXOWXUH ,Q KH FRQWLQXHG KLV VWXGLHV WRZDUG WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ LQ DJURQRP\ 5DXO $ 6DQWLOODQ LV PDUULHG WR WKH IRUPHU 0DJJLH 0RUHQR DQG WKH\ KDYH WZR GDXJKWHUV $OH[DQGUD DQG &DUROLQD 7KH DXWKRU LV D PHPEHU RI WKH $VRFLDFLQ (FXDWRULDQD GH 3URGXFFLQ $QLPDO DQG $VRFLDFLQ /DWLQRDPHULFDQD GH 3URGXFFLQ $QLPDO

PAGE 185

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

PAGE 186

, 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\ W "‘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} FML 'HDQ &ROOHJH RI $JULFXOWXUH I: FX\WL 'HDQ IRU *UDGXDWH 6WXGLHV DQG 5HVHDUFK


Days rest
Fig. 19. Contours of legume percentage as affected by length of rest period
and levels of grazing pressure in the wet season of 1980.
120


23
The grazing animal has a direct effect on pasture species due
mainly to selectivity, deposition of feces and trampling, and an
indirect effect on the soil due to the removal of nutrients by the
removal of harvested forage. The animal alters the physical and
chemical properties such as structure, texture, porosity, water
retention, and in some cases, the accumulation of organic matter at
certain points due to fecal deposits. All of these changes have a
direct consequence, first on the botanical composition and subsequently
on the final performance of that grazing animal (Alarcon and Lotero,
1970).
Research workers such as Davis (1967) and Wells (1967) agreed
that grazing affects the soil cover and botanical composition which
results in (1) a reduction of the basal cover, (2) a reduction of
height of species, (3) a loss of soil cover and severe erosion, (4) a
reduction of root systems, (5) a reduction in the emergence of new
shoots, and (6) weed invasion.
Bryan and Evans (1973) found that the proportion of legumes
Stylosanthes guianensis, Centrosema pubescens, and Pueraria phaseo-
loides on the dry matter basis was not affected by stocking rates
during the first and second year. In the third year a marked reduction
took place on the high stocking rate of 6 animals ha \ in which the
proportion of the legumes fell from 22 to 12%, while the companion
grass (Panicum maximum) was more affected by the high stocking rate.
In this case the reduction was significant since it went from 78 to
65% and to 38% in the first, second and third year, respectively. The
other two stocking rates of 2 and 4 animals ha ^ had little effect on


52
0 days rest period, 1.6 kg DM on offer/100 kg BW, and 400 kg ha ^ of
superphosphate) and 32, respectively (28 days grazing, 56 days rest,
8.3 kg DM on offer/100 kg BW, and 400 kg of simple superphosphate).
In general, the lowest values of biomass production corresponded to
treatment combinations of short rest periods and high levels of grazing
pressure, while the highest levels of biomass production corresponded
to the treatment combinations with long rest periods and low levels
of grazing pressure.
The linear components of the model accounted for 74% of the total
variation, while the quadratic effects and interactions represented
only 1 and 2%, respectively (Appendix Table 12). The experimental
variables, length of rest period (X£) and level of grazing pressure
(X^) each showed a linear effect upon biomass production (P < 0.01).
There was a suggestion that levels of superphosphate (X^) might be
having some effect (P < 0.10) but the effects of days grazing (X^)
was nil. In all cases individual quadratic effects or interactions
were not significant. The lack of interactions between X^ and X^
indicated that both variables were behaving independently. The
biomass production was increased as the lengths of the rest period
were increased and as grazing pressure was reduced. The analysis
of variance for the dry season of 1978 is presented in Appendix
Table 12.
Biomass Production (DM) for the Wet Season of 1979
The biomass production during the wet season of 1979 for each
treatment combination is given in Table 4. Biomass production varied


18
should be determined taking into consideration the amount of litter
left in the field as residue after each grazing period. Mott (1973)
emphasized the use of variable stocking rates in grazing experiments
by maintaining the number of animals in equilibrium with the available
forage. He noted that the important advantage of the system was that
it permits estimation of the carrying capacity of the pasture and
the seasonal changes which occur. Evans and Bryan (1973) pointed out
the need for more studies to evaluate different grazing pressures in
order to measure the yield of the pasture and persistence of tropical
legumes.
According to Mott (1973) the optimum grazing pressure must be
considered as an optimum range instead of a "critical point" and that
such an optimum relates only to animal output and may or may not be
the optima for plant species in the pasture. Mott and Lucas (1952),
Mott (1960), and Matches (1970) described the put-and-take technique
for grazing trials. They suggest that the stocking rate must be
variable in which the grazing pressure is maintained at a constant
level and the stocking rate adjusted as the availability of forage
changes. These authors also distinguish the terms "testers" for
animals which should remain in the pasture throughout the grazing
experiment and "grazer," or put-and-take animals, as those used to
maintain the grazing pressure at the optimum. Mott pointed
out that "if the number of animals per unit area is to give an
accurate appraisal of carrying capacity, then this unit of measure
must not be fixed but be subject to adjustment, so that the number of
animals per unit of forage is maintained at an equivalent level for


115
During the wet season of 1979, the percentage of legume varied
from 3 to 34%. The linear effect of days rest (X2) and for the
interaction of x X^ was significant. There were no direct effects
of days grazing (X^), grazing pressure (X^), and fertilizer level
(X^). The linear components of the model accounted for 61% of the
total variation, while the quadratic and interaction effects repre
sented 4 and 9% of the total variation, respectively (Appendix Table
43). Again the percentage of legume tended to decrease with in
creasing lengths of rest period.
During the dry season of 1979, the percentage legume varied
from 2 to 43%. The number of days grazing (X^) had a linear effect
(P < 0.05) upon the percentage legume and the length of the rest
period (Xalso had an effect upon percentage legume (P < 0.01).
The experimental variables X^ and X^ had no effect and there were
no interactions among the experimental variables. The linear com
ponents of the model accounted for 48% of the total variation, while
the quadratic and interaction effects represented 16 and 3% of the
total variation, respectively (Appendix Table 44). Again the legume
percentage tended to increase with increasing lengths of grazing
period while the reverse was true for increasing lengths of rest
period.
During the wet season of 1980, the percentage legume had begun
to stabilize and varied from 0 to 32%. The length of rest period
(X2) showed both a strong linear and quadratic effect upon the per
centage legume in the mixture. There appeared to be little effect of
days grazing (X^, grazing pressure (X3), and fertility level (X^)


22
Measuring Botanical Composition
Botanical composition is a very essential measurement, especially
in pastures subjected to grazing conditions, because the number of
samples and yield of individual species may vary over a wide range
depending upon environmental conditions and management factors.
T'Mannetje et al. (1976) indicated that botanical composition can be
measured in terms of the yield of component species, the number of
plants covering the area and also the frequency of occurrence.
Determination of botanical composition and sampling techniques
are difficult tasks in pasture research. Some methods have been
developed to determine botanical composition; the most common are
visual estimation and hand separation of harvested material into compo
nent species (Gardner, 1972). Visual estimation is a reliable method
for studying pasture species, but in some cases it may be difficult to
relate dry matter production of each of the component species, especi
ally when growth habit and density differ widely. Tothill and Petersen
(1962) indicated that the weight jin situ, as well as the estimation of
weight of each individual species, and visual estimation are the most
useful methods for surveying vegetation of pasture species.
Effect of the Grazing Animal on Botanical Composition
There are at least three factors that are known to affect the
balance of the grass-legume mixtures in the tropics once the pasture
has been established. They are stocking rate or grazing pressure,
frequency of defoliation, and fertilization.


90
compete with the invading weed species. At the end of the wet season
of 1980, some pasture which were subjected to the most intensive
systems of grazing were almost completely invaded by weed species.
Whether the lower survival under heavy grazing regimes was due to
selective pressure by the animals, trampling, or the competitve
advantage of the weeds, the final result was a steady decline in the
density of sown species. Pastures which were subjected to moderate
levels of rest period and grazing pressures had a higher survival
rate, greater vigor, density, and a better legume-grass balance.
On the other hand, long rest periods and light grazing pressure
permitted the grasses to increase in dominance eliminating other
species with a lower growth rate and ability to compete for space,
environmental factors, and pressure of the grazing animals. The
main concern of tropical legumes is whether they can persist and
produce efficiently under the unfavorable effects of the tropical
environnent. Some native legumes that appeared on the pastures were
included as part of the available forage and of the legume yield.
The most important of these were: Desmodium triflorum, D. canum,
D. barbatum, and lesser amounts of Calopogonium mucunoides, Phaseolus
sp. and Vigna luteola. The following legumes were recorded as part
of the weed components since they were almost completely rejected by
the grazing animals: Mimosa pdica, Mucuna pruriens, Cassia tora,
and C. occidentalis.


85
grazing pressure resulting in various degrees of defoliation,
selectivity, and effect of trampling. Defoliation and selectivity
were probably the main determinants on the amount of both legumes.
A small difference is also recorded in the behavior of these two
species in that centro appeared to be more tolerant of heavy grazing
pressure and the area covered by glycine increased as the grazing
pressure decreased. During the dry season of 1979, it was also
observed that the grazing animals were selectively consuming more
legumes than during the wet season, probably because of the lower
quality and reduced quantity of grass for that period. Humphreys
(1978) and Norman (1970) observed similar preferences of animals
grazing tropical grass-legume pastures.
During the wet season of 1979 (January-June), the average
legume yield was 473 kg ha ^ which was considerably lower when
compared with the predecing dry season (1978) which can be explained
by the much greater rate of growth by both of these companion grasses
and by their stronger competitive ability which shaded the legumes,
especially in the treatment with long rest periods. During the 1979
wet season, some environmental factors such as humidity, temperature,
light, and N played a very important role in the high growth rates
of the tropical grasses. The dominance of the grass species
began to manifest itself over the tropical legumes. Mott
stated that "physiological differences between tropical grasses and
legumes have important implications for legume-grass associations.
Since their optima for light, temperature, and moisture differ, it
is much more difficult to select compatible grasses and legumes in
the tropics than among temperate species where the responses to


82
The quadratic effect of length of rest period (X2) strongly
suggests that the range of rest periods included in this experiment
may have adequately covered the point of maximum legume yield. The
highest legume yields were obtained in the region of 14 to 28 days
rest and the yields decreased when the rest periods were reduced to
zero or increased up to 56 days. The effect of grazing pressure (X^)
appears to be almost linear. For each unit (1.6 kg DM) that grazing
pressure was reduced from 1.6 to 8.3 kg DM on offer/100 kg BW, the
amount of legume yield increased by 31 kg DM ha \ The analysis of
variance is presented in Appendix Table 29.
Legume Yield (DM) for the Wet Season of 1980
The legume yield for the wet season of 1980 for each species
combination is given in Table 7. Legume yields varied from 0 kg DM
ha for treatments with the longest rest period (56 days) to 980 kg
DM ha ^ for treatment 30 [28 days grazing, 0 rest period (continuous
grazing), 3.3 kg DM on offer/100 kg BW, and 400 kg ha ^ of superphos
phate ].
There was both a linear and quadratic effect of length of rest
period (X^ upon legume yield (P < 0.01). There was a linear
effect of grazing pressure (P < 0.05). The effects of days grazing
and fertilizer level were not significant and neither were any of
the interactions of the experimental variables.
The linear and quadratic effects accounted for 45 and 24% of
the total variation, respectively, while the interactions represented
only 5% (Appendix Table 30).


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
RESPONSE OF A TROPICAL LEGUME-GRASS ASSOCIATION
TO SYSTEMS OF GRAZING MANAGEMENT AND LEVELS OF
PHOSPHORUS FERTILIZATION
By
RAUL ALONSO SANTILLAN
April 1983
Chairman: Dr. G. 0. Mott
Major Department: Agronomy
A legume-grass pasture composed of glycine [Neonotonia wightii
(R. Grah ex Wightii and Am.) Lackey], centro (Centrosema pubescens
Benth.), guineagrass (Panicum maximum Jacq.), and elephantgrass
(Pennisetum purpureum Schumach.) was evaluated in a grazing trial
from May 1978 to June 1980 at the Estacin Experimental Tropical
Pichilingue, Instituto Nacional de Investigaciones Agropecuarias
(INIAP), Quevedo, Ecuador.
The main objectives of the study were (a) to determine the
effects of length of grazing period (X^), length of rest period
(X^), grazing pressure (X^), and levels of P fertilization (X^)
upon the pasture mixture; (b) to determine the proper grazing
management to attain the optimum legume contribution; and (c) to
measure the pasture response in terms of dry matter production and
botanical composition.
Grazing periods studied were 1, 7, 14, 21, and 28 days; rest
periods were 0, 14, 28, 42, and 56 days; grazing pressures were
1.6, 3.3, 5.0, 6.6, and 8.3 kg DM on offer/100 kg body weight; and
xiii


TABLE PAGE
40 Analysis of variance, regression coefficients and
probabilities for visual estimation grass (%) for
the wet season of 1980 155
41 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1978 156
42 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the dry season of 1978 157
43 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1979 158
44 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the dry season of 1979 159
45 Analysis of variance, regression coefficients and
probabilities for visual estimation legume (%) for
the wet season of 1980 160
x


102
with long rest periods and/or low grazing pressures with low P levels.
It is well known that tall-growing grasses and climbing legumes have
the potential to exploit the soil and aerial environments which may
limit the growth of companion or invading species, especially when
the management practices favor their maximum production. Frequent
and intensive defoliation of the two species may be highly detri
mental to their vigor and regrowth capacity, so that the sward can
easily be dominated by undesirable species.
Haraschin (1976) reported similar results with weeds which were
affected by the rest period and grazing pressure and he concluded
that short rest periods and high grazing pressure create the most
favorable conditions for weeds invasion.
Jensen and Schumacher (1970) reported that the botanical compo
sition is not only affected by the grazing animal, but also by
some environmental factors such as the season, rainfall, temperature,
and soil nutrients which may allow certain weedy species to invade
the pasture sward.
The predominant weed species in this study were: Sida acuta,
Solanum carolinensis, Amaranthus sp., Capsicum sp., Aescplepias sp.,
Mimosa pubica, Cassia tora, C. occidentalis (broadleaf weeds);
Paspalum fasciculatum, P. paniculatum, P. conjugatum, Digitaria
sanguinalis, Eleusine indica (Grasses); and Cyperus rotundus (Sedge).
Some of these weeds were also reported by Santillan (1971) growing
in Jaraguagrass (Hyparrhenia rufa).


Table 38. Analysis of variance, regression coefficients and probabilities for visual
estimation grass for the wet season of 1979.
RESPONSE MEAN
79. OJ 26
ROOT MSE
6. U715
R-SQUARE
0.03650645
COEF OF VARIATION
(). 00634 547
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROB
LINEAR
4
6424. 3477
0. 6140
33. 02
0. 0001
QUADRATIC
4
214 1237
0 0205
1. 13
0. 3571
CROBSPRODUCT
6
2114.3441
0. 202 J
7. 42
0. 0001
TOTAL REGRESS
14
0752. 0157
0 0366
13. 16
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATI
PROB
LACK OF FIT
26
1559. 7237
59. 9094
3. 999
0. 0131
PURE ERROR
10
147.9963
14.7776
TOTAL ERROR
36
1709. 7201
47. 4722
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
77. 4709
1. 7429
44. 45
0. 0001
XI
1
-0.37284437
0. 65779293
-0. 57
0. 5744
X2
1
7. 1006
0.65071424
10. 71
0. 0001
X3
1
2. 9444
0.65779293
4. 48
0. 0001
X4
1
0.00771147
0.65777293
0. 01
0. 7803
XI *X1
1
-O.57175009
1. 0667
-0. 54
0. 5953
X1*X2
1
-0. 63943240
0. 36020003
-1 77
0. 0044
X2*X2
1
1. 5741
1. 0667
1. 48
0. 1407
X1*X3
1
- 0. 8710B750
0. 344 57301
-2. 53
0. 0160
X2*X3
1
-2. 0750
0. 36028003
-5. 76
0. 0001
X3*X3
1
-1. 7097
1.0667
-1. 60
0. 1021
X1 #X4
1
0.40212717
0.34457301
1. 17
0. 2509
X2*X4
1
0.24119199
0.36020003
0. 67
0. 5075
X3#X4
1
0.00449503
O.34457301
0. 01
0. 9097
X4#X4
1
0.02400324
1. 0667
0. 77
0. 4448
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
592. 5914
110. 5103
2. 50
0. 0486
X2
5
7561. 2152
1512. 2430
31. 84
0 0001
X3
5
2417. 7502
403 5516
10. 18
0. 0001
X4
0
116.6154
23.3231
0. 49
0. 7807


10
Response to Nutrients
Legumes are generally more sensitive to soil factors than other
pasture plants, particularly grasses. This sensitivity of legumes
emphasizes the importance of understanding the effects of these limi
tations in tropical conditions. Russell (1978) suggested that im
proving the growth of legumes in low fertility soils could be approached
in two ways: (1) by amelioration of soil conditions through the use
of fertilizers or amendments, and (2) by the selection of legume
cultivars or genera which are more tolerant to the limiting conditions.
There seems to be a general consensus that N and P, in that order, are
the plant nutrients that are more often deficient in the tropics (Fox,
1979).
The amount of N fixation, nodulation, persistence, and yield of
tropical forage legumes may be affected by the soil pH and also by
the availability of plant nutrients. Manhaes and Dobereiner (1968)and
Fox et al. (1974) reported that for good legume establishment, ade
quate amounts of available P were required, especially during the
nodulation stages. These authors determined that glycine required
60 ppm of P^O in the soil solution during the establishment phase.
This requirement decreased after the second cutting. In Australia,
Andrew and Robins (1969) determined the critical P concentration in
the tops associated with maximum plant growth as being the critical
levels. These were 0.16 and 0.23% for centro and glycine, respec
tively. For adequate plant uptake, h^PO^ ions in the soil solution
should be between 0.07 to 0.2 ppm. According to Sanchez (1977) some


86
environmental factors are similar." (1981, p. 35-41). He also added
that the viney growth habit of several genera of tropical legumes
(Calopogonium, Centrosema, some species of Desmodium, Neonotonia,
Macroptilium, and Pueraria); they confer an advantage over
tropical grasses in that they are able to climb to the top of the
canopy. Devising defoliation strategies that will maintain the
regrowth potential of viney legumes is very important.
The advantage of short rest periods during the wet seasons in
terms of legume population was evident, giving these species more
opportunity to survive, persist, bloom, and produce some seed, even
at heavy grazing pressures. Under high levels of defoliation, the
competitive and shading effects from both of the grasses was greatly
limited. Each of the tropical legumes bloomed profusely at the
beginning of the dry season (July-August).
During the dry season of 1979 (July-December), the mean legume
yield was 397 kg ha During this second dry season a few pastures
with the longest rest periods were almost 100% grass with no legumes
nor weeds being observed. The decrease in legume yield may be partly
explained by the excessive compititive ability of the grasses and
also by the selective pressure on the legumes during this season.
Pastures subjected to continuous grazing or short rest periods showed
a greater proportion of legume dry matter in the total amount of
available forage. Both legume species, centro and glycine, survived
the effect of frequency and severity of defoliation when short rest
periods were combined with high grazing pressures.


Table 17. Analysis of variance, regression coefficients and probabilities for available
forage (g DM/m ) for the dry season of 1978.
RESPONSE MEAN
301.0755
ROOT MSE
72. 2359
R-SQOARE
0. 76545926
COEF OF VARIATION
0.23992627
REGRESSION
DF
TYPE I SB
R-SQUARE
F-RATIO
PROD
LINEAR
4
5B9779
0. 7364
20. 26
0. 0001
QUADRATIC
4
9460.3109
0. 0118
0. 45
0. 7691
CROSBPRODUCT
6
13826. 0157
0. 0173
0. 44
0. 8460
TOTAL REGRESS
14
613074
0. 7655
8. 39
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
83859. 3783
3225. 3607
0. 310
0. 9921
PURE ERROR
10
103990
10390. 9630
TOTAL ERROR
36
187049
5210. 0200
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
268. 8809
18. 2695
14. 72
0. 0001
XI
1
-2. 8808
6. 8950
-0. 42
0. 6786
X2
1
44. 1255
6. 9046
6. 39
0. 0001
X3
1
50.4124
6. 8950
7. 31
0. 0001
X4
1
-13. 1554
6. 8950
-1. 91
0. 0644
XI *X1
1
2. 5463
11. 1814
0. 23
0. 8211
X1WX2
1
0.04082591
3. 7765
0. 01
0. 9914
X2&X2
1
8. 4080
11. 1814
0. 76
0. 4527
XHfX3
1
4. 1572
3. 6118
1. 15
0. 2573
X2X3
1
3. 4631
3. 7765
0. 92
0. 3652
X3*X3
1
-1. 8853
11. 1814
-0. 17
0. B670
XI *X4
1
0. 59127037
3. 6118
0. 16
0. 8709
X2*X4
1
1. 2596
3. 7765
0. 33
0. 7407
X3WX4
1
-2. 124 3
3. 6118
-0. 59
0. 5601
X4#X4
1
-1. 3530
11. 1814
-0. 12
0. 9044
FACTOR
DF
SB
MEAN SQUARE
F-RATIO
PROB
XI
5
8304. 9503
1660. 9917
0. 32
0. 8987
X2
5
222956
44591. 1075
0. 55
0. 0001
X3
5
340980
68195. 9001
13. 07
0. 0001
X4
5
21302. 6361
4260. 5272
0. 82
0. 5459


61
An examination of the individual linear effects reveals that
only the experimental variables days rest (X^) and grazing pressure
(X^) had any effect upon available forage (P < 0.01).
The length of rest period increased the available forage as was
also the case with a decrease in grazing pressure. For each increase
of 14 days in the rest period there was an increase of 440 kg ha ^ of
available forage and for each unit (1.6 kg DM) decrease in grazing
pressure there was a 500 kg ha ^ increase in available forage. The
analysis of variance for the dry season of 1978 is presented in Appendix
Table 17.
Available Forage for the Wet Season of 1979 (DM)
During the wet season of 1979 the available forage (DM) varied
from 830 kg DM ha ^ to 6630 kg DM ha \ corresponding to the treat
ment combinations 17 [1 day grazing, o days rest (continuous grazing),
1.6 kg DM on offer/100 kg BW, and 0 kg ha ^ of superphosphate] and 23
(1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 0 kg
ha of superphosphate), respectively. The treatment combinations
with the shortest rest periods and lowest grazing pressure levels
yielded the lowest values of available forage for this season; also
it was noted that in this case both variables X^ and X^ were acting
independently. There was no evidence of any interactions occurring.
The linear components of the model accounted for 69% of the total
variation, while the quadratic effects and interactions represented
only 2 and 1% of the total variation, respectively (Appendix Table 18).
Only the linear effects of the experimental variables X^ and X^ had


62
an effect upon the available forage (P < 0.01). Increasing the days
of rest and decreasing the grazing pressure increased the available
forage during the wet season of 1979. For 14 days increase in rest
period there was an increase of 750 kg ha ^ of available forage and
for each unit (1.6 kg DM) decrease in grazing pressure there was a
500 kg ha ^ increase in available forage. The analysis of variance
for the wet season of 1979 is presented in Appendix Table 18.
Available Forage for the Dry Season of 1979 (DM)
For the second dry season, the available forage for each treat
ment combination is presented in Table 5. The available forage varied
from 400 kg DM ha ^ to 5980 kg DM ha \ corresponding to the following
treatment combinations: 25 [1 day grazing, 0 days rest (continuous
grazing), 1.6 kg DM on offer/100 kg BW, and 400 kg ha ^ of superphos
phate] and 15 (7 days grazing, 42 days rest, 6.6 kg DM on offer/100
kg BW, and 300 kg ha of superphosphate) respectively. Only days
rest (X2) and grazing pressure (X^) showed any effect upon available
forage (P 0.01). Each of these two experimental variables was
acting independently because no interactions were found.
The linear components of the model accounted for 54% of the
total variation, while the quadratic effects and interactions repre
sented less than 1 and 6% of the total variation, respectively (Appen
dix Table 19). For each 14 day increase in the rest period, there was
an increase of 320 kg ha ^ of available forage and for each unit (1.6
kg DM) decrease in grazing pressure, an increase of 570 kg ha ^ of avail
able forage was realized. The analysis of variance for the dry season
of 1979 is presented in Appendix Table 19.


Table 35. Analysis of variance,
of weed (kgDM/m^) for
RESPONSE MEAN
ROOT MSE
R-SQUARE
COEF OF VARIATION
REGRESSION
m
LINEAR
1
QUADRATIC
4
CROSSPRODUCT
6
TOTAL REGRESS
1 4
RESIDUAL
nr-
LACK OF FIT
26
PURE ERRDR
10
TOTAL ERROR
36
PARAMETER
1)1
INTERCEPT
1
XI
1
X2
1
X3
1
X4
I
X1*X1
1
X1*X2
1
X2X2
J
X1X3
1
X2*X3
1
X3*X3
1
X1*X4
1
X2*X4
1
X3#X4
J
X4*X4
1
FACTOR
DF
XI
5
X2
0
X3
0
X4
r
regression coefficients and probabilities for yield
the wet season of 1980.
24 2712
22. 74 22
O. 06301666
0. 77724J 56
TYPE I SS
R SQUARE
F-RAT 10
PROD
674 40. 21 40
0 4 552
29 91
0 0001
14622. 7170
0. 0707
6 49
0 0005
4 0706. 0677
0 3091
13. 54
0. 0001
127847
0. 0630
16 20
0. 0001
SS
MEAN SQUARE
F-RATIO
PROB
17040. 7432
751. 5670
9. 991
0. 0003
7 52. 2400
75. 2240
20292. 7032
563.6940
ESTIMATE
STD DEV
T-RATIO
PROB
7. 1000
6. 0040
1. 18
0. 2443
-2. 0540
2. 2662
-1. 26
0. 215?
-20. 6103
2. 2694
-9. 08
0. 0001
-19.7345
2. 2662
-0. 71
0. 0001
7. 1496
2. 2662
3. 15
0. 0032
1. 9100
3. 6751
0. 52
0. 6064
0.90740192
1. 2413
0. 00
0. 4315
-1. 4600
3. 6751
-0. 40
0. 6934
0.80295040
1 1071
0. 60
0. 5031
9. 9357
1. 2413
0. 00
0. 0001
9. 9307
3. 6751
2. 70
0. 0104
0.22705119
1. 1071
0. 19
0. 040?
-3. 64 53
1.2413
-2. 94
0 0050
-3. 2292
1. 1071
-2. 72
0. 0100
0.32748273
3. 6751
-0. 09
0. 9295
SS
MEAN SQUARE
F-RATIO
PROD
1430. 1045
207 6209
0. 51
0. 7666
00100 2609
17620. 0538
31. 26
0. 0001
69660. 8126
13933. 7625
24 72
0. 0001
12309. 5691
2461.9138
4. 37
0. 0033
150


1.6 3.3 5.0 6.6 8.3
Grazing pressure (kg DM on offer/100 kg BW)
Fig. 10. Contours of legume yield (DM) as affected by length of rest period
and levels of grazing pressure in the wet season of 1980.
00
00


Table 8.continued.
Treatments
Reps
1978
1978
1979
1979
1980
No.
D/Gf
(xx)
D/Rf
(X2)
G/Pf
(X ) % BW
Ft -i
kg ha
w/st
D/St
w/st
D/St
w/st
kg ha ^
25
1
0
1.6
400
1
40
80
440
1730
2600
26
28
0
1.6
400
1
90
0
220
240
2470
27
1
56
1.6
400
2
30
0
10
60
190
28
28
56
1.6
400
2
0
0
0
0
0
29
1
0
8.3
400
1
0
10
20
30
150
30
28
0
8.3
400
1
90
100
60
40
110
31
1
56
8.3
400
2
120
30
0
20
0
32
28
56
8.3
400
2
130
0
10
0
0
33
1
28
5.0
200
1
100
40
20
20
70
34
28
28
5.0
200
1
150
190
320
240
200
35
14
0
5.0
200
1
50
10
20
0
10
36
14
56
5.0
200
1
0
40
30
20
0
37
14
28
1.6
200
1
70
50
200
500
830
38
14
28
8.3
200
1
210
160
100
30
80
39
14
28
5.0
0
1
0
0
100
10
40
40
14
28
5.0
400
1
40
30
20
0
60
41
14
28
5.0
200
3
70
30
50
30
100
fD/G = days grazing, D/R = days rest, G/P = grazing pressure, F = fertilizer.
tW/S = wet season, D/S = dry season.


Table 9.continued
Treatments
Reps
1978
1978
1979
1979
1980
No.
D/Gf
(xx)
D/Rf
(x2)
G/Pf
(X3) % BW
Ft -i
kg ha
w/st
D/S
W/SJ
D/St
w/st
25
1
0
1.6
400
1
60
52
- %
34
15
7
26
28
0
1.6
400
1
53
54
61
35
10
27
1
56
1.6
400
2
57
83
94
95
91
28
28
56
1.6
400
2
60
82
94
91
99
29
1
0
8.3
400
1
73
77
78
79
76
30
28
0
8.3
400
1
53
58
71
77
70
31
1
56
8.3
400
2
56
85
92
96
100
32
28
56
8.3
400
2
56
83
90
94
100
33
1
28
5.0
200
1
40
69
81
80
71
34
28
28
5.0
200
1
56
64
60
61
68
35
14
0
5.0
200
1
53
63
74
62
81
36
14
56
5.0
200
1
63
77
84
99
100
37
14
28
1.6
200
1
60
55
58
52
29
38
14
28
8.3
200
1
50
67
73
75
63
39
14
28
5.0
0
1
56
62
74
66
76
40
14
28
5.0
400
1
76
67
79
81
80
41
14
28
5.0
200
3
55
72
74
62
73
fD/G = days grazing, D/R = days rest,G/P= grazing pressure, and F = fertilizer rate.
W/S = wet season, D/S = dry season.


CHAPTER I
INTRODUCTION
Ecuador, with an area of 273,670 km is located in the north
western part of the South American continent. The dominant topo
graphical features are two parallel ranges of the lofty Andes
mountains that separate the fertile littoral lowland on the west
and the more extensive and less fertile lowland of the Amazon Basin
on the east.
The diversity of natural features of the littoral region is
very great due to its multiple climatic conditions, soils, forms of
vegetation, and settlement patterns which set it apart from the more
homogeneous Sierra and Oriente regions.
In the littoral, about 2,500,000 ha are considered as pasture
land, supporting 2,875,000 head of cattle giving a carrying capacity
of 1.12 animals ha ^.
The seasonal pattern of rainfall distribution and the low soil
N levels are factors restricting forage production and quality
during the wet-dry seasons. It is well known that cattle production
is limited by the feed supply during the dry season, while in the
wet season, there is abundance of forage.
Pastures are mainly planted to guineagrass (Panicum maximum Jacq.)
or elephantgrass (Pennisetum purpureum Schumach.), with other species
making up a very small portion of the total hectarage.
At present, there is a growing interest in the establishment
and utilization of tropical grass-legume mixtures for animal produc
tion. There are many advantages to having legume components in the
1


CHAPTER III
MATERIALS AND METHODS
This research was conducted at Estacin Experimental Tropical
Pichilingue, belonging to INIAP and located 7 km from Canton
Quevedo, Procincia de los Rios, at Io 06' S Lat. and 79 29' W Long.
Altitude at the site is 64 m above sea level. The average minimum
and maximum temperatures are 17.3 and 35.6C, respectively, having
a mean annual temperature of 24.3C. Annual precipitation is 2152 mm.
About 82% of the yearly rainfall occurs during the warmer months from
December through June. The months of February and March have the
highest rainfall intensity. By contrast, October and November are
the driest months, often registering no rainfall. Annual mean rela
tive humidity is 84% and mean sunlight is 846.2 hours per year, 66%
of which occurs during the December to May period (Servicio Nacional
de Hidrologa y Meterologia del Ecuador, 1980). Holdridge (1967)
located Pichilingue in the tropical moist forest zone (Fig. 1).
Figures 2, 3, and 4 present the monthly rainfall range, average
temperature and average solar radiation for the years 1978, 1979,
and 1980.
The soils are classified as Torripsamments. Hardy (1960) states
that the chemical analysis of this Pichilingue loam reveals a marked
deficiency in available P, but an abundance of available K. The
soil N status is fair to medium in recently cleared land, but declines
rapidly with cultivation. More recent soil analyses at the experi
mental site revealed that the amount of P is medium, while B, S, Zn
and Mo are low (INIAP, 1979).
28


levels of fertilizer were 0, 100, 200, 300, and 400 kg ha ^ of
superphosphate. To cover the five levels of the complete factorial
4
(5 ), a modified non-rotatable central composite design made up of
41 treatment combinations was used.
The response variables included aerial biomass (DM), available
forage (DM), grass yield (DM), legume yield (DM), yield of weeds
(DM), percentage grass, and percentage legume. A double-sampling
procedure was used for estimating pasture production and botanical
composition.
Rest periods and grazing pressures had the greatest effects
on all response variables. Aerial biomass, available forage, grass
yield and grass percentage were increased by longer rest periods
and by lower grazing pressure. Legume yield and legume percentage
were decreased by long rest periods and by low grazing pressure.
Short rest periods and high grazing pressures resulted in high
yields of weeds. Medium levels of both rest periods and grazing
pressure were required for high forage dry matter production and
for high legume yield.
The other two variables, days grazing and levels of phosphorus
fertilization, had negligible effects upon the response of the pasture
sward.
xiv


64
by the grazing animal. The available forage was estimated by
determining the botanical composition of the total aerial biomass
and summing the grass and legume components. Over the five seasons
during which this study was conducted the available forage was greatly
influenced by the environmental factors which are discussed in the
aerial biomass section. The experimental variables, days rest (X2)
and grazing pressure (X^), each had a highly significant effect upon
the available forage. The response of available forage to these two
experimental variables was essentially linear over the range which
was studied. The longer rest period permitted an accumulation of
reserves for pasture regrowth and the lower grazing pressure retained
more of the aerial parts of both the grasses and legumes affecting
mainly the growing points, axillary buds, and traditional leaf area
for the production of photosynthate which stimulated subsequent re
growth. It is significant that these two variables seem to act inde
pendently as no interactions between them were found within the range
of levels in this experiment.
Similarities in responses to the same lengths of rest period
and with the same grazing pressures were observed by Maraschin (1975)
and Serrao (1976). In each of these two cases interactions were found
between these two variables where maximum yields were obtained at
high grazing pressures in combination with long rest periods and low
grazing pressures with short rest periods. The above studies were
conducted with Cynodon dactylon and Desmodium intortum which have
quite different growth habits than the grasses and legumes included
in this study. Cynodon dactylon is a stoloniferous grass with the


Table 26. Analysis of variance, regression coefficients and probabilities
yield (g DM/m ) for the wet season of 1978.
RESPONSE MEAN
92. 7395
ROOT MSE
20. 2070
R-SQUARE
0.29723332
COEF OF VARIATION
0.30400793
REGRESSION
nr
TYPE I SS
R-SQUARE
F-RATIO
LINEAR
4
6690. 6737
0. 1642
2. 10
QUADRATIC
4
1945 7639
0 0477
0. 61
CROSSPRODUCT
/.
U
3478 0272
0 0053
0. 73
TOTAL REGRESS
1 4
121 14. 4649
0. 2972
1. 09
RESIDUAL
nr
SS
MFAN SQUARE
F-RATIO
LACK OF FIT
26
20127 0462
774.1172
O. 707
PURE ERROR
10
8515. 8074
851 5887
TOTAL ERROR
36
20642. 9336
795.6370
PARAMETER
DI
ESTIMATE
STD DEV
T-RATIO
INTERCEPT
1
91.7044
7. 1340
12. 07
XI
1
-O.11058066
2. 6924
-0. 04
X2
1
6. 6558
2. 6961
2. 47
X3
1
3. 0137
2. 6924
1. 42
X4
1
0.85625607
2. 6924
0. 32
X1*X1
1
0. 50674037
4. 3662
0. 12
X1 #X2
1
0.61410957
1. 4747
0. 42
X2*X2
1
5. 7651
4. 3662
1. 37
XI *X3
1
1. 4719
1.4104
1. 04
X2*X3
1
0 00961492
1. 4747
-0. 01
X3*X3
1
-3. 6579
4. 3662
-0. 84
XI *X4
1
1. 2019
1. 4104
0. B5
X2*X4
1
-0. 73300431
1. 4747
-0. 63
X3X4
1
-1.7B36
1. 4104
-1. 41
X4*X4
1
-3. 2047
4.3662
-0. 75
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
XI
5
1598. 8015
319.7603
0. 40
X2
5
6970. 8820
1394. 1766
1. 75
X3
5
4743. 3062
940 6772
1. 19
X4
3
2739 1746
507. 0309
0. 74
for legume
PR OB
0. 1007
O. 6571
0. 6296
0. 4000
PR OB
O.6019
PROD
0. 0001
0.9675
0. 0104
0. 1652
0. 7523
0.9002
0. 6796
0 1003
0 3036
0. 9940
0. 4074
0. 3997
0. 5309
0. 1677
0. 4567
PROB
0. 0443
O 1470
O. 3323
O 5994


13
In the tropics, where there are well defined wet and dry seasons,
rainfall plays an important role in the uptake and nutrient content
of tropical forage species. Blue and Tergas (1969) reported a drop
in N, P, and K contents during the dry season; likewise a decrease of
nutrient content during the wet season was found and it was postulated
to be due to translocation of nutrients to the roots. Rapid growth
during wet seasons may result in trace mienrals being translocated to
plant tops where they are rapidly diluted with aerial tissue causing
deficiency symptoms in older tissues (Reuter, 1975).
Micronutrients can play an important role in tropical legume
pastures growth, mainly because of their function in several enzyme
systems and in N-fixation by rhizobium-legume associations. Werner
et al. (1975) studied tropical legume response to the micronutrients
Mo, Cu, Zn, B, Mn, and Co in the form of FTE BR-10 and also in the
salt form. Using three tropical legumes planted in pots, they observed
visible symptoms of Mn toxicity on glycine and B toxicity on stylo.
This work emphasizes that there is a narrow range between deficient
and toxic levels of some micronutrients.
Medina (1969) found some micronutrient deficiency symptoms in
some tropical crops growing in the Quevedo area. He reported that B,
Zn, and Fe were the most deficient elements. INIAP (1978) found that
at the beginning of the rainy season, the period of most rapid growth
for grasses, such as paragrass, Zn deficiency symptoms were evident
but these disappear in two or three weeks. Molybdenum also has been
recognized as an essential element for legume growth especially for
establishment and maintenance. Some authors have suggested that this


72
of grass increased considerably from 1978 to 1980 (Figs. 6, 7, and 8),
due mainly to its rapid growth capacity and ability to eliminate the
other companion species. Vicente-Chandler (1975) reported that
elephantgrass and guineagrass are the most productive tropical grasses
reaching values of 78 and 45 tons DM ha ^ year \ INIAP (1980) re
ported that under a cutting system, the rate of growth for elephant-
grass and guineagrass is in the order of 166 kg and 149 kg DM ha ^
day \ respectively. These values corresponded to those obtained
in this experiment during the wet season of 1979 but as the year
progressed, these values decreased and reached the lowest rate at
the end of the dry season of that particular year, being in the order
of 32 and 28 kg DM ha ^ day ^ for elephantgrass and guineagrass,
respectively.
Grazing pressure expressed as the amount of forage offered per
100 kg BW increased the grass yield as the grazing pressure was
decreased. The lowest yields were obtained under the highest grazing
pressure and the highest yields were obtained in the treatments with
the lowest grazing pressure. If the two experimental variables,
rest period and grazing pressure are considered in combination, then
the lowest grass yields were obtained under continuous grazing and
the highest grazing pressure, while the highest yields were observed
on treatment combinations having the longest rest period with the
lowest grazing pressure. As the grazing pressure was increased, the
amount of plant shoots that were removed by the grazing animal also
increased and when this was accompanied by an increase in frequency
of defoliation (short rest periods or continuous grazing), the


Table 44. Analysis of variance, regression coefficients and probabilities for visual
estimation legume for the dry season of 1979.
RESPONSE MEAN
16. 2602
ROOT MSE
7. 7953
R-SGUARE
0. 67757 J 72
COEF OF VARIATION
0 491470J 0
REGRESSION
Of
TYPE I OS
R-SGUARE
F-RATIO
PROD
LI NEAR
4
3462.0520
0. 4021
13. 54
0. 0001
QUADRATIC
4
1160 7140
0. 1616
4 54
0. 004 5
CROSBPRODUCT
6
257. 4202
0. 0358
0. 67
0 6735
TOTAL REGRESS
J 4
4 000. 6741
0. 6776
5. 45
0. 0001
RESIDUAL
Pi
SO
MEAN SQUARE
F-RATIO
PR0L1
LACK OF FIT
26
2127. 0230
81. 8086
4. 674
0. 0071
PURE ERROR
io
174.2756
17. 4296
TOTAL ERROR
36
2301.3185
63. 7255
PARAMETER
D-
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
24. 81 18
2. 0221
12. 27
0.0001
XI
1
1. 5100
0.76315706
1. 77
0. 0547
X2
J
-4. 544 5
0.76422/75
-5. 95
0.0001
X3
J
-1. 0371
0.76315706
-1. 36
0. 1818
X 4
J
0.40612713
0.76315706
0. 53
0. 5777
X1#X1
1
- 0. 52510726
1. 2376
-0. 42
0. 6738
XI *X2
J
-0. 37588526
0.41777743
-0. 90
0. 374 5
X2*X2
1
-1. 7035
1.2376
-1. 60
0. 1177
X1 # X 3
1
-0.44431250
0.39776716
-1. 11
0. 2730
X2*X3
I
0.25327036
0.41777743
0. 61
0. 5404
X3*X3
1
-0. 33505310
1. 2376
-0. 27
0.7881
XI *X4
1
0.17687083
0.37776716
0. 44
0. 6608
X2*X4
I
-0.47797649
0.41777943
-1. 19
0. 24 13
X3#X4
1
0.01204503
0.37776716
0. 03
0. 7761
X4#X4
1
-0 10140726
1. 2376
-0. 15
0. 8843
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROD
XI
5
362.2770
72 4570
1. 13
0 3605
X2
5
2632. 6178
526. 5236
8. 24
0. 0001
X3
1
205. 251 1
41.0502
0. 64
0. 6671
X4
5
107. 0427
21.4006
0. 33
0. 0004
159


68
offer/100 kg BW, and 400 kg ha ^ of superphosphate] and 24 (28 days
grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 0 kg ha ^
of superphosphate), respectively. Both rest period (X^) and grazing
pressure (X^) had an effect upon the grass yield (P < 0.01). The
application of superphosphate (X^) also had a significant effect at
the 10% level of significance. Each of these variables acted inde
pendently for grass yield since no interactions occurred.
The linear components of the model accounted for 74% of the
total variation, while the quadratic effects and interactions
represented only 2 and 1% of the total variation, respectively
(Appendix Table 22).
An examination of the individual linear effects shows that
the experimental variable, days grazing (X^), did not have any effect
on the amount of grass produced. The greatest effects were obtained
from days rest (X^) and grazing pressure (X^) As the length of
rest period increased from zero days (continuous grazing) the amount
of grass yield tends to increase, whereas, when the grazing pressure
is decreased from 1.6 to 8.3 kg DM on offer/100 kg BW, the grass
yield was increased. For each 14 days increase in rest period,
there was an increase of 440 kg ha ^ of grass yield. For each
increment of decrease in the grazing pressure, there was an increase
of 430 kg ha ^ of grass yield. The analysis of variance is presented
in Appendix Table 22.


2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
20
15
7
4
13
29
1
8
20
26
5
2
24
25
11
0
16
18
0
0
8
3
1
2
10. Visual estimation of dry matter legume percent for year, season, and treatment
Treatments
D/Gf D/Rf G/Pf Ff
(Xx) (X2) (X3) % BW kg ha Reps
1978 1978 1979 1979
W/St D/St W/St D/St
%
7
14
3.3
100
1
33
24
19
18
21
14
3.3
100
1
18
22
18
15
7
42
3.3
100
1
46
23
10
30
21
42
3.3
100
1
25
19
9
14
7
14
6.6
100
1
21
19
19
13
21
14
6.6
100
1
46
27
25
31
7
42
6.6
100
1
43
19
9
6
21
42
6.6
100
1
30
20
11
9
7
14
3.3
300
1
36
32
19
30
21
14
3.3
300
1
27
26
21
25
7
42
3.3
300
1
40
22
14
16
21
42
3.3
300
1
43
20
12
9
7
14
6.6
300
1
36
27
19
28
21
14
6.6
300
1
40
30
32
30
7
42
6.6
300
1
43
30
9
11
21
42
6.6
300
1
42
20
6
7
1
0
1.6
0
1
38
36
23
18
28
0
1.6
0
1
32
38
28
26
1
56
1.6
0
2
38
17
5
3
28
56
1.6
0
2
32
23
9
8
1
0
8.3
0
1
36
28
18
16
28
0
8.3
0
1
40
32
15
11
1
56
8.3
0
2
29
10
3
2
28
56
8.3
0
2
42
23
21
9


27
and that heavy grazing pressure and short rest periods almost elimi
nated the legumes from the pasture.


Table 29. Analysis of variance, regression coefficients and probabilities for legume
yield (g DM/m ) for the dry season of 1979.
RESPONSE MEAN
39.7357
ROOT MSE
16. 3974
R-SQUARE
0.74611206
COEF OF VARIATION
0.41271244
REGRESSION
nr
TYPE I SS
R- SQUARE
F-RATIO
PR OB
LINEAR
4
16647.0700
0. 4 365
15 47
0. 0001
QUADRATIC
4
10400 5934
0 2727
9 68
0 0001
CROSSPRODUCT
A
1396 9917
0. 0366
0. 87
0. 5293
TOTAL REGRESS
14
20452 6030
0. 7461
7. 56
0 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PR 8)3
LACK OF FIT
26
8463 0577
325. 5022
2. 671
0: 0532
PURE ERROR
10
1218. 0161
121.8016
TOTAL ERROR
36
9681. 8730
268.9409
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
62. 4 435
4.1476
15. 06
0. 0001
XI
1
1. 6614
1. 5653
1 06
0.2956
X2
J
-9. 7153
1. 5675
- 6 20
0. OOOl
X3
]
3. 1317
1. 5653
2. 00
0 0530
X4
1
0 00095420
1 5653
0. 00
0. 9995
XI *X1
1
-0.18768706
2. 5385
- 0. 07
0. 94 15
XJUX2
J
0 50432611
0. 85736754
0. 59
0. 5601
X2*X2
1
-9. 0169
2. 5385
-3. 55
0. 001 1
X1#X3
1
-0.46363750
0.81997094
-0. 57
0. 5753
X2#X3
1
-1. 5980
0.85736754
-1. 86
0. 0705
X3#X3
1
0.62429211
2. 5385
0. 25
0. 8071
X1 #X4
1
-0.25604583
0.81997094
-0. 31
0. 7566
X2*X4
1
-0.83306517
0.85736754
-0. 97
0. 3377
X3*X4
J
0 09435417
0.81997094
0. 12
0. 9090
X4#X4
1
0 22481294
2. 5385
0. 09
0. 9299
FACTDR
DF
SS
MEAN SQUARE
F-RATIO
PROD
XI
5
653. 0030
130. 7766
0. 49
0. 7842
X2
s
15403. 7776
3000. 7555
11. 46
0 0001
X3
>
1663.5066
332. 7173
1 24
0. 3122
X4
5
309. 3916
61. 8703
0. 23
0. 9469


Table 21. Analysi^ of variance, regression coefficients and probabilities for grass yield
(g DM/m ) for the wet season of 1978.
RESPONSE MEAN
33. 3261
ROUT MSE
62. 4033
R-SQUARE
0.32632070
COEF OF VARIATION
0.40751797
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
307 50. 0270
0. 1474
1. 97
0. 1201
QUADRATIC
4
11026. 1364
0. 0567
0. 76
0. 3599
CROSSPRODUCT
6
25490. 2350
0 1222
1. 09
0 3878
TOTAL REGRESS
14
60083. 2000
0. 3263
1. 23
0. 2074
RESIDUAL
pi
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
42242. 3704
1624. 7073
0. 165
1. 0000
PURE ERROR
10
70300 2164
9030. 8216
TOTAL ERROR
36
140531
3904. 1833
PARAMETER
Df
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
133. 4364
15. 0030
8. 44
0. 0001
XI
1
-0.81342760
5. 964 1
-0. 14
0. 8920
X2
1
12.0690
3. 9724
2. 15
0. 0379
X3
1
3. 7329
5. 964 1
0. 63
0. 5333
X4
i
-4.1201
5 964 1
-0. 69
0. 4941
X1*X1
1
-8. 2324
9. 6710
-0. 83
0. 4003
X1#X2
1
-0.60669823
3. 2667
-0. 21
0. 0347
X2*X2
1
9. 0310
9. 6710
1. 02
0. 3152
X1#X3
J
3. 5747
3 1242
1 13
0. 2373
X2*X3
J
1. 4299
3. 2667
0. 44
0. 6642
X3*X3
J
-i. 524 0
9. 67 J 0
-0. 16
0. 8737
XI *X4
1
5. 7003
3. 1242
1. 03
0. 0725
X2*X4
1
-3. 4696
3. 2667
-1. 06
0. 2952
X3*X4
1
2. 0247
3. 1242
0. 65
0. 521 1
X4*X4
1
6. 6426
9. 6718
0. 69
0. 4966
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
3
2.1702. 1795
4340 4359
1. 11
0 3714
X2
:>
20082. 6022
5616. 5364
1. 44
0. 2342
X3
3
J 0070. 5535
2015. 711 1
0. 52
0. 7622
X4
3
23520. 2979
5105. 6396
1 31
0. 2025
u>
ON


TABLE OF CONTENTS
PAGE
ACKNOWLEDGMENTS iii
LIST OF TABLES vii
LIST OF FIGURES xi
ABSTRACT xiii
CHAPTER I INTRODUCTION 1
CHAPTER II LITERATURE REVIEW 3
The Tropical Forages 3
Persistence of Tropical Legume-Grass Associa
tions 6
Response to Nutrients 10
Animal Productivity 14
Pasture Evaluation 16
Grazing Systems 19
Estimates of Dry Matter Production and Yield.. 20
Measuring Botanical Composition 22
Effect of the Grazing Animal on Botanical
Composition 22
Response Surface Methodology 25
CHAPTER III MATERIALS AND METHODS 28
Legume-Grass Mixture 33
Experimental Variables 33
Experimental Design 35
Field Plan of the Experiment 35
Land Preparation and Pasture Establishment.... 40
Construction of Physical Facilities 43
Collection of Data in the Three Experimental
Years 43
Pasture Measurements 44
CHAPTER IV RESULTS AND DISCUSSION 48
Effect of Lengths of Rest Period and Levels of
Grazing Pressure on Aerial Biomass (DM) 48
Effect of Lengths of Rest Period and Levels
Grazing Pressure on the Available Forage (DM). 57
Effect of Lengths of Rest Period and Levels of
Grazing Pressure on Grass Yield (DM) 65
v


107
During the wet season of 1980, which was the final season of
the experiment, the percentage grass ranged from 7 to 100%. The
linear and quadratic effects of days rest and grazing pressure
(X^) were very evident (P < 0.01) and there was also a strong
interaction between x X^. Again days grazing (X^) and level of
fertilizer (X^) had no effect. The linear components of the model
accounted for 64% of the total variation, while the quadratic and
interaction effects represented 7 and 19%, respectively (Appendix
Table 40).
Visual estimation of percentage grass was recorded on the
same dates that grass yields were taken. Some rather drastic
changes occurred over time and the effects of treatments from the
beginning of the experiment through the last grazing season are
summarized below.
During the first wet season (May-June) of 1978, no effects of
treatments were recorded, whereas in the last wet season of 1980,
both linear and quadratic effects (P < 0.01) of days rest (X^) and
grazing pressure (X^) as well as a significant interaction between
X£ x X^ were very evident. While it is not certain that two years
are sufficient to produce stable associations, it is clear that both
grasses (elephantgrass and Guineagrass) are very responsive to days
rest and to grazing pressure. The relationship of these two experi
mental variables is obviously curvilinear (see Figs. 15, 16, and 17) Low
rest periods resulted in swards with 100% grass, whereas these tall-
growing grasses almost eliminated under continuous, intensive grazing.
Each of these species has a high growth capacity and the ability to


42
In August the complete area was cross-disced in order to
eliminate some weeds and obtain a good seedbed. The area was
disced again in late September to eliminate the new growth and some
weeds. In the same month, 20 small soil samples were taken from
the whole area and mixed together for soil analysis; the samples
were taken at 25 cm depth (see Table 1).
The next step was to prepare the legume seeds. Glycine and
centro seeds were scarified using sulfuric acid for a time of 6 and
8 minutes, respectively. After the seeds were washed and shade
dried, they were inoculated using an appropriate Rhizobium strain
obtained from CIAT (Centro Internacional de Agricultura Tropical).
In early October of 1977, a mixture of glycine and centro was sown
using a small grain seeder with a four row capacity at the rates of
-1
3.0 and 6.0 kg ha respectively. The mixture was sown in rows
spaced 1.4 m apart. A day after the sowing, a mixture of preemer
gence herbicides was applied (Linuron 0.75 kg ha ^ + 2 L ha ^ of
Laso) to control both grasses and broadleaf weeds coming from seed.
Two weeks after the legumes were sown, the grasses were planted
vegetatively, using plant divisions for guineagrass and small pieces
of stems with two or three buds for elephantgrass. Both grasses
were planted alternatively between the rows of the legumes, so that
the distance between grasses and legumes was 0.70 m. It was neces
sary to make one hand weeding using machetes. Four irrigations of
4 mm each were necessary at 12 day intervals. The first irrigation
was immediately after the legumes were sown.


Table 3.continued
Treatments
Reps
Size of
Exp. Unit
S,
Totalf
Area
Required
per Treat.
No.
Pasture
No.
Days
Grazing
(xx)
Days
Rest
(x2)
Grazing
Pressure
(X ) % BW
Fertilizer
Level ,
(X^) kg ha
25
10
1
0
1.6
400
1
2000
2000
26
29
28
0
1.6
400
1
2000
2000
27
12,47
1
56
1.6
400
2
500
500
28
7,40
28
56
1.6
400
2
750
750
29
14
1
0
8.3
400
1
4000
4000
30
49
28
0
8.3
400
1
4000
4000
31
28,46
1
56
8.3
400
2
500
1000
32
3,21
28
56
8.3
400
2
2000
2000
33
5
1
28
5.0
200
1
500
500
34
51
28
28
5.0
200
1
2000
2000
35
42
14
0
5.0
200
1
3000
3000
36
17
14
56
5.0
200
1
750
750
37
11
14
28
1.6
200
1
750
750
38
2
14
28
8.3
200
1
1500
1500
39
1
14
28
5.0
0
1
750
750
40
9
14
28
5.0
400
1
750
750
41
19,39,41
14
28
5.0
200
3
750
2250
fTotal area = 73000 m2.


97
the yields of weeds were low during the first season, there presence
in response was noteworthy. During the dry season of 1978, there
was a quadratic response to P fertilizer and it appeared that some
species of weeds were the only species which responded positively
to P. There was also some evidence that the yield of weeds was
responding during the first dry season to the length of rest period
(X2), since there was a decline in the weed population as the length
of rest period increased. This may have been due to the competition
from the small-growing tropical grasses (Fig. 12).
During the wet season of 1979, the effect of days rest (X^) and
grazing pressure (X^) resulted in a decline of the yield of weeds
(P < 0.01). There was also a strong interaction between these two
variables (P < 0.01). This interaction suggests that the yield of
weeds is reduced by increasing the length of the rest period in asso
ciation with low grazing pressures. When shorter rest periods or
continuous grazing are combined with high grazing pressure, there
was an increase in the yield of weeds. It was evident that short
rest periods and high grazing pressures had a detrimental effect
upon the pasture. Weeds were encouraged due to the opening of the
sward and by the selective grazing of the more palatable species.
Most of the weedy species were not consumed nor were they much affected
by the trampling by the grazing animals. The mean dry matter yield
for the weeds was 94 kg ha \
During the dry season of 1979, increases in the days grazing (X^),
days rest (X2), and grazing pressure (X^) resulted in a decline in
the yield of weeds (P < 0.01). The experimental variables, X2 and X^,


9
concluded that the dominance of climbing legumes such as glycine and
Siratro is generally enhanced by light grazing pressure and also by
long intervals between grazing periods.
Grazing experiments conducted in the wet tropics of Ecuador by
Berrezueta (1975), Chavez (1974), INIAP (1979) and Zapata (1981)
showed that guineagrass-centro pastures and guineagrass-glycine pas
tures are very persistent and productive mixtures even if heavy grazing
pressures are applied. They also observed that rest periods of over
28 days during the wet season favored the companion grass and reduced
drastically the amount of legume. Finally they concluded that the
guineagrass is dominant, especially at the beginning of the wet season,
probably due to the large amount of nitrogen stored in the organic
matter during the six to seven months dry season. The high growth rate
of this grass decreases slowly and reaches the lowest rate of growth
at the end of the dry season while the legumes appeared to be more
productive during the dry season. The legumes are the main source
of feed for the grazing animals during the dry season. A study of
eight grass-legume associations was also made from 1971 through 1973
using grazing animals. Paragrass (Brachiaria mutica)-glycine and
guineagrass-centro mixtures were shown to be the most compatible
and the best accepted by the grazing animals, while tropical kudzu
(Pueraria phaseoloides) and Calopogonium mucunoides were unpalatable
species, a characteristic that could determine the dominance of this
species over the companion grasses (INIAP, 1974).


163
, and W. W. Bryan. 1973. Effects of soils, fertilizers,
and stocking rates on pastures and beef production. Aust. J.
Exp. Agrie. Anim. Husb. 13:530-533.
Falade, J. A. 1975. The effect of phosphorus on growth and mineral
composition of five tropical grasses. East Afr. Agrie. For. J.
40:342-350.
Farfan, C. 1974. Efecto de practicas culturales en la produccin
de semillas de plantas forrajeres tropicales. Tesis Ingeniero
Agronomo. Universidad Technica de Manabi. Portoviejo, Manabi,
Ecuador. 33 p.
Febles, G., and C. Padilla. 1972. Efecto del pastoreo en asocia
ciones de gramineas y leguminosas tropicales. Revista Cubana
de Ciencias Agricolas. 6:450-408.
Fox, R. H. 1979. Soil pH, aluminum saturation, and corn grain yield.
Soil Sci. 127:330-334.
Fox, R. L., N. K. Nishimoro, S. R. Thompson, and J. R. de la Pena.
1974. Comparative external phosphorus requirements of plants
growing in tropical soils. X Int. Cong. Soil Sci. 4:232-239.
Franca, G. E., and M. M. Carvalho. 1970. Ensayo exploratorio da
fertilizacao da cinco leguminosas tropicales en un solo do
cerrado. Pesquisa Agropecuaria Brasileira. 5:147-153.
Funes, F., and C. Perez. 1976. Estudios agronmicos en soya
perenne. I. Comparacin de variedades de Glycine wightii y
Teramnus labialis en condiciones de corte y pastoreo. Revista
Cubana de Ciencias Agricolas. 10:205-209.
Gardner, A. L. 1972. Estudio sobre los mtodos agronmicos para
la evaluacin de pasturas. Centro de Investigaciones y
Enseanza para la zona templada. Instituto Interamericano
de Ciencias Agricolas (IICA), Montevideo, Uruguay. 48 p.
Gartner, J. A., J. E. Ferguson, R. W. Walker, and E. A. Goward. 1974.
Evaluating perennial grass-legume on the Atherton tableland of
North Queensland. Queensl. J. Agrie. Anim. Sci. 1:1-17.
Gomez, D. T. 1978. Establishment methods and comparative persis
tence of five tropical legumes in grass sods. Ph.D. Dissertation.
University of Florida, Gainesville, Florida. 143 p.
Grof, B. 1970. Yield attributes of some species and ecotypes of centro-
sema in north Queensland. Queensl. J. Agrie. Anim. Sci. 4:85-89.
, and W. A. T. Harding. 1970. Dry matter yields and animal
production on guineagrass (Panicum maximum) on the humid tropical
coast of North Queensland. Trop. Grassld. 1:77-84.
Hall, R. L. 1970. Pasture development in the spear grass region at
Westwood in Fitzroy Basin. Trop. Grassld. 4:77-84.


26
control variables, such as rates of P. They also suggested that
the range of values determines the experimental region, and the func
tional relationship is called the response surface.
Factorial arrangements of treatments provide good information,
but they require greater numbers of experimental units and more
physical resources; with the same amount of resources using response
surface methodology, it is possible to obtain results from a much
greater number of variables and levels within each variable. These
designs, such as the rotatable central composite, non-rotatable central
composite and San Cristobal, have been used in grazing trials and have
proved their usefulness in obtaining valuable data (Maraschin, 1975;
Mott, 1977; Serrao, 1976; Villasmil et al., 1975).
Maraschin (1975) and Serrao (1976), both using a central composite
design, studied the effect of three variables: grazing days (1, 3.5,
7, 10.5, and 14), resting periods (0, 14, 28, 42, and 56 days), and
dry matter residue left after grazing (500, 1000, 1500, 2000, and
2500 kg DM ha ^) upon the botanical composition of Cynodon dactylon-
Desmodium intortum-Macroptilium atropurpureum-Lotononis bainesii-
Trifolium repens mixture. The great advantage of this design is that
it used only 24 treatments instead of 125 that the complete factorial
would have required for a single replication, to obtain the coefficients
needed to evaluate the grazing management systems. Serrao (1976) re
ported that the most important factors in determining dry matter yield
from pasture and also legume percentage maintained in the mixture were
grazing pressure and rest periods. He found that the legume content of
the mixture was increased with an increase in length of rest period


118
Grazing pressure appeared as a significant factor (P < 0.01)
only during the dry season of 1978 where legume percentage was more
responsive to higher levels of grazing pressure. While the grazing
pressure decreased, the legume percentage also showed a declining
trend. During the remaining four seasons of the experiment, no
significant effects of this variable were observed.
The effects of rest period (X^) and grazing pressure (X^) on
legume percentage is presented in Figs. 18 and 20 for the wet season
of 1978 and the wet season of 1980, respectively. The contours of
the response surface of percentage legume appears in Fig. 19 for the
wet season of 1980.
Percentage legume appears to be always higher than the actual
legume yield. This may be explained by the plants growth habit,
distribution on the soil surface and upon the associated grases,
and finally by their morphological characteristics (broadleaves)
which leads the observer to overestimate the actual percentage of
tropical legumes. This is evident if we compare the response
surfaces for legume yield and percentage legume in Figs. 11 and 20.


FIGURE PAGE
16 Contours of grass percentage as affected by length
of rest periods and levels of grazing pressure in
the wet season of 1980 109
17 Effect of rest period and levels of grazing pressure
upon grass percentage for the wet season of 1980... 110
18 Effect of rest period and grazing pressure upon
legume percentage for the wet season of 1978 119
19 Contours of legume percentage as affected by length
of rest period and levels of grazing pressure in
the wet season of 1980 120
20 Effect of rest period and grazing pressure upon
legume percentage for the wet season of 1980 121
xii


56
is 22 C. The third important factor is solar radiation which is
also higher during the rainy season, reaching values of 500-600 hours
of sunlight while from July to November this value averages 350-400
hours of sunlight. These three factors acting together are the main
determinants of total aerial biomass production.
Another very important factor which must be taken into considera
tion is nitrogen. This element, at the beginning of the wet season,
is rapibly mobilized from the soil organic matter, which has accumu
lated in the soil during the dry season. Sanchez (1977) has indicated
that N in the tropics is rapidly utilized by actively growing plants
with the occurrence of the first rains at the beginning of the wet
season but that the remaining N is slowly released during the rest of
the season. These three environmental factors plus the availability
of N have a great impact upon the aerial biomass production.
The most important experimental variables influencing the aerial
biomass production were days of rest (X^) and grazing pressure (X^).
Each variable behaved independently and there was a direct relationship
between the number of days rest which permitted the plants a better
opportunity to accumulate reserves for more rapid regrowth and devel
opment after each grazing period.
A similar trend was also found for grazing pressure which was
directly related to the amount of forage removed by the grazing animals
which in turn was related to the high removal and damage to the growing
points and axillary buds. Grazing pressure was related to the rate of
recovery and subsequently the final production. Hodgson and Ollerenshaw
(1969) mentioned that if grazing pressure is increased while resting


Table 20
. Analysi^ of variance, regression coefficients and probabilities for available forage
(g DM/m ) for the wet season of 1980.
RESPONSE MEAN
392. 0004
ROOT MSE
140. 3500
R-SQUARE
0 64206730
COEF OF VARIATION
0 36970430
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
1201016
0. 5652
14. 21
0. 0001
QUADRATIC
4
68540. 3460
0. 0323
0. 81
0. 5264
CROBSPRODUCT
6
94763. 7606
0. 0446
0. 75
0. 6153
TOTAL REGRESS
14
1364330
0. 6421
4. 61
0 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROP
LACK OF FIT
26
504704
19411. 6913
0. 759
0. 7276
PURE ERROR
10
200062
250B6. 2299
TOTAL ERROR
36
760566
21126. 8409
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PR OB
INTERCEPT
1
314. 61 OB
36. 7613
8. 56
0. 0001
XI
1
-7. 061 1
13. 8738
-0. 54
0. 5091
X2
1
76. 0399
13. 8932
5. 47
0. 0001
X3
1
64. 9473
13. 8738
4. 68
0. 0001
X4
1
-13. 9012
13. 8738
-1. 00
0. 3230
XI *X1
1
11. 9732
22. 4909
0. 53
0. 5979
X1*X2
1
-4. 2700
7. 5990
-0. 56
0. 5777
X2X2
1
0. 0232
22. 4909
0. 39
0. 6973
X1*X3
1
2. 9101
7. 2675
0. 40
0. 6907
X2X3
1
-12. 2149
7. 5990
-1. 61
0. 1167
X3*X3
1
-3. 9600
22. 4989
-0. 10
0. 8610
X1 #X4
1
4. 6092
7. 2675
0. 64
0. 5255
X2*X4
1
2. 0100
7. 5990
0. 33
0. 7430
X3#X4
1
6. 9129
7. 2675
0. 95
0. 3478
X4*X4
1
6. 2902
22. 4989
0. 20
0. 7811
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
36351. 6503
7270. 3301
0. 34
0. 8826
X2
5
70320B
140658
6. 66
0. 0002
X3
5
487486
97497. 2963
4. 61
0. 0023
X4
5
50697. 4944
10139. 4989
0. 40
0. 7008


Days rest
56
Grazing pressure (kg DM on offer/100 kg BW)
Fig. 16. Contours of grass percentage as affected by levels of rest
periods and levels of grazing pressure in the wet season
of 1980.
109


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.
t /? A
L. R. McDowell
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.
hlA
W. R. Ocumpaugh
Associate ProfesWor 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.
April 1983 K-C/K cj/i
Dean, College of Agricul/ture
f/W
cuyti
Dean for Graduate Studies and
Research


7
and even exclude the C-3 legumes from the mixture. Tow (1967) showed
that green panic (Panicum maximum var trichoglume) was much more
productive at all light intensities and higher root temperatures
than glycine when both species were tested under controlled environ
mental conditions. Roberts (1974) also studied some of the above
factors and included some others which are associated with the
stability of legume-grass mixtures. These were palatability of the
grass and legume, maximum height of the grass, legume ability to grow
under the shade projected by the companion vegetation, and the capacity
to withstand trampling. He also suggested that continuous grazing
helps the legume to compete more effectively with the grass due to
more frequent defoliation than under a rotational grazing system.
Kretschmer (1974) reported that in general grasses have a better
range of adaptation and also a more vigorous growth habit that allows
them to compete always at an advantage over most tropical forage
legumes. Lack of legume persistence is attributed to the use of
unadapted species and cultivars, improper or no fertilization, incor
rect rhizobium and overgrazing (t'Mannetje, 1978).
Growth habit and leaf morphology of species that compose a
mixture are important characteristics which have direct effects upon
compatibility and persistence, due especially to light interception
ability of each individual species. Santhirasegaram (1976)
in the humid tropic of Peru reported that in a well-managed guinea-
grass-centro pasture, the persistence of the legume was due to its
viney growth habit enabling it to climb the stems and leaves of this
tall and aggressive grass. Thus, the legume can intercept sufficient


6
adaptability, seed production and disease and insect resistance.
These cultivars are 'Malawi' for lower altitudes, 'Cooper' for medium
altitudes up to 1200 m and 'Tinaroo' the highest forage yielder
grows well from 50 to 1800 m of altitude, producing large amounts
of seed from 800-1500 m (INIAP, 1979).
Persistence of Tropical Legume-Grass Associations
Serrao (1976) suggested that the first requirement for successful
use of high-yielding pasture legumes was their adaptation to local
climatic conditions. Secondly, nutrient requirements must be met to
insure high yield and maintenance. And finally, they must persist
under heavy grazing to secure a long lasting beneficial contribution
to the companion grass, to the soil, and to the grazing animal.
Gomez (1978) has suggested some other factors which can affect the
persistence of tropical legumes when they are growing in association
with grasses. Among the most important are (1) environmental factors
such as light, temperature and moisture; (2) growth habits of each
species growing in the mixture; (3) nodulation ability and capacity
for nitrogen fixation; (4) edaphic factors such as pH, nutrient
availability, form of supply; (5) frequency and intensity of defolia
tion by grazing animals; (6) ability to survive during long drought
periods; (7) seed production capacity; and (8) pest and disease toler
ance. Ludlow and Wilson (1970) reported that tropical grasses achieve
up to three times the photosynthetic rate when compared with tropical
legumes. This characteristic obviously gives ecological advantages
to C-4 grasses, affording them the opportunity to grow faster, dominate


Table 10.continued.
Treatments
D/Gf
(xx)
D/Rf
(x2)
G/Pf
(X3) % BW
Ft -1
kg ha
1978
1978
1979
1979
1980
No.
Reps
w/sf
D/Si
w/st
D/S|
w/si
. 25
1
0
1.6
400
1
33
41
Vo
34
9
9
26
28
0
1.6
400
1
38
45
24
43
13
27
1
56
1.6
400
2
35
17
5
3
2
28
28
56
1.6
400
2
40
18
6
9
1
29
1
0
8.3
400
1
22
26
21
19
18
30
28
0
8.3
400
1
40
38
26
21
26
31
1
56
8.3
400
2
38
14
7
3
0
32
28
56
8.3
400
2
40
17
9
6
0
33
1
28
5.0
200
1
53
28
17
18
26
34
28
28
5.0
200
1
33
29
28
30
25
35
14
0
5.0
200
1
43
36
24
37
18
36
14
56
5.0
200
1
36
21
14
0
0
37
14
28
1.6
200
1
33
29
30
27
30
38
14
28
8.3
200
1
36
26
23
23
32
39
14
28
5.0
0
1
43
38
21
33
22
40
14
28
5.0
400
1
21
31
20
19
17
41
14
28
5.0
200
3
37
25
24
21
23
+D/G
tw/s
= days grazing, D/R = days rest, G/P =
= wet season, D/S = dry season.
grazing pressure,
and F =
fertilizer
rate.


Days rest
56
42
28
14
0
\
\ \
v >
\ \
\ 900
1200 \
1500 \ \
600
1.6 3.3 5.0 6.6 8.3
Grazing pressure (kg DM on offer/100 kg BW)
Fig. 13. Contours of yield weed (DM) as affected by length of rest period
and levels of grazing pressure in the wet season of 1980.
100


Table 7.continued.
Treatments
1978
1978
1979
1979
1980
D/Gf
(xx)
D/Rf
(x2)
G/Pf
(X3) % BW
F+ -i
kg ha
No.
Reps
w/st
D/St
w/st
D/St
w/st
25
1
0
1.6
400
1
810
450
kg ha
450
180
180
26
28
0
1.6
400
1
700
380
240
440
220
27
1
56
1.6
400
2
1000
400
140
40
30
28
28
56
1.6
400
2
1180
380
190
150
30
29
1
0
8.3
400
1
530
570
650
640
630
30
28
0
8.3
400
1
1040
890
910
640
980
31
1
56
8.3
400
2
900
570
420
100
0
32
28
56
8.3
400
2
980
680
560
110
0
33
1
28
5.0
200
1
1330
730
480
490
680
34
28
28
5.0
200
1
560
780
970
780
770
35
14
0
5.0
200
1
1130
560
410
570
320
36
14
56
5.0
200
1
1200
830
650
0
0
37
14
28
1.6
200
1
620
780
740
640
580
38
14
28
8.3
200
1
940
720
800
690
870
39
14
28
5.0
0
1
1140
850
510
780
510
40
14
28
5.0
400
1
440
650
610
520
420
41
14
28
5.0
200
3
1020
540
580
740
530
tD/G
jw/s
= days grazed, D/R
= wet season, D/S :
= days rest, G/P =
= dry season.
grazing
pressure, F =
fertilizer.


The author is indebted to graduate students, Mr. Luis R.
Rodriguez and his wife, Teresinha Rodriguez for their friendship
and hospitality. Thanks are also due to graduate student Juan
Herbas and his wife for their friendship.
The author wishes to thank Mrs. Pat French for typing the final
copy of this manuscript.
Special acknowledgment is due to the authors wife, Maggie,
and their two daughters, Alexandra and Carolina, for their invaluable
help, love and devotion and finally to my parents and dear sisters,
Cecilia and Lucy, for their love and financial help.
iv


ACKNOWLEDGMENTS
The author expresses his sincere thanks and appreciation to
Dr. Gerald 0. Mott, chairman of the supervisory committee, for his
understanding, deep concern and wise guidance throughout the field
work and graduate program. Special thanks are given to Dr. W. R.
Ocumpaugh, Dr. J. E. Moore, Dr. 0. C. RueIke, and Dr. L. R. McDowell,
who willingly served as members of the committee.
Special appreciation is due to Dr. Raul de la Torre, General
Coordinator for Animal Production Research at INIAP-Ecuador, who
helped design and advise during the field experiment. The author
also expresses his gratitude to Ing. Carlos Cortaza, former Director
of E. E. T. Pichilingue and to Ecuador's Instituto Nacional de Inves
tigaciones Agropecuarias for the physical and financial support of
this study, especially to the staff of the Programa de Pasto y
Ganadera in the E. E. T. Pichilingue. Outstanding among those
whose friendship and physical labors are responsible for this study
are Ing. Carlos Cadene, Agr. Jorge Molina, and Seores Cerapio Diaz,
Emilio Ortega, Jorge Benitez, Alejo Briones and the late Maximo Soria.
Special appreciation is due to Dr. A. I. Khuri, Dr. Michael
Conlon and to the graduate student Jose M. Gallo for their outstanding
contributions and help in the statistical analyses of the results of
this research.
Special thanks are offered to both Dr. and Mrs. G. 0. Mott for
their generosity to the author and his family throughout their stay
in Gainesville.
iii


96
Yield of Weeds (DM) for the Wet Season of 1980
The yield of weeds for the 1980 wet season is presented in
Table 8. The yield of weeds varied from 0 to 2600 kg DM ha
The yield of weeds was affected by days rest grazing pressure
(X^), and fertilizer level (X^) (P < 0.01). There was also a
quadratic component for grazing pressure (X^) (P < 0.01). The
interactions between X^ x X^, X^ x X^, and X^ x X^ were also
significant at the 1% level of probability. The linear, quadratic,
and interaction components of the model accounted for 45, 9, and 30%
of the total variation, respectively (Appendix Table 35). The yield
of weeds tends to decrease as the length of the grazing period and
rest period increases, as was also the case with increasing grazing
pressure. A positive trend was also observed as the fertilizer
level was increased.
Observations and Summary of Yield of Weeds
During the five seasons during which this experiment was
conducted from May, 1978 to May, 1980, some of the most drastic
changes which occurred were observed in the yield of weeds.
During the first wet season (May-June, 1978), the main effects
and interactions which occurred can probably be explained on the
basis of the seed reserves of certain species of weeds at the
experimental site. The fact that the application of fertilizer
appeared to have an effect while no other experimental variable
manifested itself is significant. It is quite well known that certain
weed species respond readily to fertilizer treatments and although


Table 1. Soil analysis of experimental site (1978).
Depth of sample: 0-25 cm
pH: 6.30
Percent organic matter: 4.96
Element ppm
Phosphorus
62.5
Potassium
598.8
Sulfur
18.4
Zinc
13.2
Manganese
258.5
Copper
20.9
Boron
12.6


11
tropical legumes tolerant to low available soil P either absorb P
at a faster rate or are able to tranlocate it to the tops more
rapidly than do species not tolerant to low P availability (Salinas
and Sanchez, 1976; Andrew, 1978). Considerable yield response to
P fertilizers was reported by Jones and Freitas (1970) with four
tropical legumes (Stylosanthes guianensis, centro, glycine, and
Siratro). Similar results were obtained by Franca and Carvalho
(1970) in greenhouse studies, using five tropical legumes (glycine
var 'Common,' glycine var 'Tinaroo,' Siratro, centro, and Pueraria
phaseoloides var 'Javanica' Benth.). In both cases the P deficiency
was reflected in decreased nodule weight and N fixation capability.
Snyder and Kretschmer (1974) obtained small linear increases in dry
matter yields of Siratro, 'Cook' stylo, centro, and Desmodium hetero
carp on (L.) DC when lime was applied in 500 kg ha ^ increments up
to 3000 kg ha ^ without P fertilization. When the same levels of
lime were used together with 45 kg ha ^ of P the response in yield
was linear up to 2000 kg ha ^ of lime and curvilinear thereafter.
Estimation of P requirements for plant growth should be based on the
amount of P needed to give at least 95% of maximum growth (Ozanne and
Shaw, 1976). Neme and Lovadini (1967) working with glycine found
that a combination of 120 kg of P^us 6 metric tons of lime ha ^
gave a large increase in yield. Werner (1979) reported substantially
increased yields of centro to P and K fertilization. Palacios (1976)
in Ecuador found that centro responded positively to P fertilization
during the establishment period, but yield response was not related
to the amount of P applied to the soil during the sowing time. Falade


o
V
<
o
v
O


Table 27. Analysis of variance, regression coefficients and probabilities for legume
yield (g DM/m ) for the dry season of 1978.
RESPONSE MEAN
64 0002
ROOT MSE
17. D7B7
R-SQUARE
0.45776632
COEF OF VARIATION
0. 30347053
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROB
LINEAR
4
7703. 3037
0. 3125
3 17
0. 0021
QUADRATIC
4
403. 7327
0. 0170
0. 32
0. 8637
CROSSPRODUCT
6
3212. 7068
0. 1262
1. 40
0. 2426
TOTAL REGRESS
14
11650. 2236
0. 4570
2. 17
0. 0300
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
6746. 5077
257 4011
0 368
0. 7003
PURE ERROR
JO
7033.4233
705. 3423
TOTAL ERROR
36
13777.7312
383. 3314
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
6B. 6697
4. 7518
13. 87
0. 0001
XI
1
2. 0200
1. 0688
1. 51
0. 1400
X2
1
-0.31717234
1. 0714
-0. 17
0. 8664
X3
1
7. 2177
1.0608
3. 06
0. 0005
X4
1
-2. 3317
1. 8688
-1. 35
0. 1837
XI *X1
1
-0.02206417
3. 0306
-0. 01
0. 7742
X1#X2
1
O.65606710
1.0236
0. 64
0. 5236
X2*X2
1
-1. 5526
3. 0306
-0. 31
0 6116
X1*X3
J
2. 3432
0.77874257
2. 37
0. 0220
X2*X3
i
- 0. 37078768
1. 0236
-0. 56
0. 5803
X3*X3
1
-0 14011772
3. 0306
-0. 05
0. 7634
XI *X4
J
-1. 2075
0.77874257
-1. 24
0. 2246
X2*X4
1
-0. 52022254
1. 0236
-0. 51
0. 6144
X3*X4
1
-0 37347176
0.77874257
-0. 38
0. 7051
X4*X4
J
-0.17011772
3. 0306
-0. 06
0. 7503
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
4144.6772
828.7374
2. 16
O.' 0801
X2
5
487. 5805
77.7177
0. 26
0. 9343
X3
5
8107. 3546
1621. 4707
4. 23
0. 0040
X4
5
1688. 4720
337. 6746
0. 88
0. 5037
142


Table 41. Analysis of variance, regression coefficients and probabilities for visual
estimation legume for the wet season of 1978.
RESPONSE MEAN
36. 2464
ROOT MSE
9 2274
R-SQUARE
0. J 27477113
COEF OF VARIATION
() 25463003
REGRESSION
Il
TYPE I MS
R SQUARE
F-RAT10
PR on
LINEAR
4
43. 4022
0. 0124
0. 13
0. 9714
QUADRATIC
4
157. 4004
0. 04 40
0. 46
0 7631
CROSBPRODUCT
6
247.1394
0. 0703
0. 40
0. 0161
TOTAL REGRESS
14
440 0301
0 1275
0. 30
0 9737
RESIDUAL
DK
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
1969. 3390
75. 7430
0. 690
0. 7852
PURE ERROR
10
1097. 2222
109 7222
TOTAL ERROR
36
3066. 5612
05. 1023
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
37.2015
2. 3343
15. 97
0. 0001
XI
1
O.06029446
0.00095255
0. 08
0. 9306
X2
1
0.10466557
0.00210642
0. 21
0. 0354
X3
1
0.30275024
0.00095255
0. 43
0. 6665
X 4
1
0.20930290
0.00095255
0. 33
0. 7444
X1*X1
1
1. 3743
1. 4206
0. 90
0. 3356
X1#X2
1
0.03757209
0. 40251759
0. 00
0. 9384
X2*X2
1
0. 14420096
1. 4206
0. 10
0. 9201
X1*X3
1
0. 44903333
0.46147117
0. 97
0. 3362
X2*X3
1
- 0. 06557337
0.40251759
-0. 14
0. 0927
X3*X3
1
- 0. 60905230
1.4206
-0. 40
0. 6325
XI *X4
1
-0. 17233333
0.46147117
-0. 37
0. 7110
X2*X4
1
0.35219074
0.40251759
0. 73
0 4702
X3*X4
J
-0.51603333
0.46147117
- 1. 12
0. 2701
X4#X4
1
-1. 3141
1 4206
-0. 92
0. 3630
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
175. 4002
35. 0016
0. 41
0. 0374
X2
5
52. 1961
10 4392
0. 12
0. 9065
X3
5
223 6090
44.7300
0. 53
0. 7556
X4
r*
263. 5059
52 7012
0. 62
0. 6063


112
It is also of interest to note that the experimental variables
included in the model accounted for over 90% of the total variabi
lity in the percentage of grass in the mixture.
Visual Estimation of Percentage Legume
The visual estimate of percentage of legumes in the mixture
is found in Table 10 for each treatment combination and for the five
seasons during which this experiment was conducted.
For the first experimental season (May-June of 1978), no signi
ficant effects due to the treatments were found. The analysis of
variance is presented in Appendix Table 41. As expected, the second
order model accounted for only about 12% of the total variability in
percentage legume.
During the dry season of 1978, the percentage legume varied
from 10 to 45%. Linear effects of days grazing (X^) (P < 0.05) and
days rest (X^) and grazing pressure (X^) (P < 0.01) began to appear
in the percentage legume. The effect of X^ was nil and no inter
actions among the experimental variables were found. The linear
components of the model accounted for 62% of the total variation,
while the quadratic and interaction effects represented 1 and 5%,
respectively (Appendix Table 42).
The percentage legume tended to increase with increasing length
of grazing period while a negative relationship was observed as the
length of the rest period increased. The percentage legume also
decreased as the grazing pressure decreased. The analysis of variance
is presented in Appendix Table 42.


162
, J. P. Sharpe, and K. P. Haydock. 1971. Some factors
affecting the growth of Lotononis (Lotononis bainesii). Aust.
J. Exp. Agrie. Anim. Hush. 11:29-34.
Chaverra, H. 1979. Manejo de potreros. Pastos y ganado para la
costa Atlntica. Instituto Colombiano Agropecuario (ICA),
Colombia. 138 p.
Chavez, J. E. 1974. Evaluacin de la associacion guinea-centrosema
sometida a pastoreo en pocas seca y hmeda en Pichilingue.
Tesis Ingeniero Agronomo. Unviversidad Central del Ecuador,
Quito, Pichincha, Ecuador. 47 p.
Correa, M. P. 1926. Dicdionario de plantas uteis do Brasil a
das exticas cultivadas. Rio de Janeiro, Imprensa Nacional
Brasil. 5:552-554.
Conway, A. 1965. Grazing management in relation to beef production. 8th
Proc. Int. Grassld. Cong., Sao Paulo, Brasil. P. 1601-1607.
Cowan, R. T., I. J. R. Bayford, and T. G. Stobbs. 1975. Effects of
stocking rates and energy supplementation on milk production
from tropical-grass-legume pasture. Aust. J. Exp. Agrie. Anim.
Husb. 15:740-744.
, and P. O-Grady. 1976. Effects of presentation yield of
a tropical grass-legume pasture on grazing time and milk yield
of Freisian cows. Trop. Grassld. 10:213-212.
, R. J. Moss, and I. J. R. Bayford. 1974. Milk
and fat yields of Jersey and Freisian cows grazing tropical
grass-legume pastures. Trop. Grassld. 8:177-179.
Davis, H. 1967. Influence of soil and management on the botanical
composition of 20 years-old relcaimed bill pastures in mid Walls.
J. Range Manage. 20:241-247.
Draolu, E. A., and G. W. Nabusin-Napalu. 1980. Response of pure and
mixed swards of stylo and panicum to frequency and height of
cutting. East. Afr. Agrie. For. J. 42:559-563.
Echandi, 0. 1956. Utilizacin de los potreros a travez del pastoreo
y del corte para alimentacin en establo. Mesa Redonda Regional
Sobre Forrajes en Centro America. Turrialba, Costa Rica.
1:52-54.
Epstein, E. 1972. Mineral nutrition of plants: Principles and
perspectives. John Wiley and Sons, New York. 432 p.
Evans, T. R. 1970. Some factors affecting beef production from
subtropical pastures in the coastal lowland of Southeast Queens
land. XI Int. Grassld. Cong. Proc., Surfer's Paradise, Australia.
P. 803-807.


91
Effect of Lengths of Rest Period and Levels of Grazing
Pressure on the Yield of Weeds (DM)
The Incidence of weed populations would be exptected to be
related to the seed reserves of different weed species in an experi
mental site. Whether weeds appear under certain environmental cir
cumstances will to a great extent be determined by the presence and
abundance of seed of the different weed species. The results of
this experiment appear to be no exception to this rule as the effect
of the various grazing management systems was not nearly as consis
tent as for the two species of grasses and legumes. This will
become evident as the results and the analyses are examined.
Yield of Weeds (DM) for the Wet Season of 1978
The yield of weeds for the wet season of 1978 for each treatment
combination is presented in Table 8. The yield of weeds varied
from 0 kg DM ha ^ to 210 kg DM ha ^ during the first one and one-half
months of this trial.
During the first few weeks of the trial there appeared to be an
effect of fertilizer upon the weed population (P < 0.05), but there
was no linear effect of days grazing, days rest, and grazing pressure
during this period upon the yield of weeds. The quadratic effects
of days rest (X£) and fertilizer level (X^) was also apparent during
this period (P < 0.05). An interaction between and X^ also
appeared (P < 0.05) and X_ and X. (P < 0.01). These differences in
j 4
interactions were probably due to seed reserves in the soil over
which we exerted very little control at the beginning of the experiment.


Table 31. Analysis of variance, regression coefficients and probabilities for yield
of weeds (g DM/m ) for the wet season of 1978.
RESPONSE MEAN
6. J 219
ROOT MSE
d.7756
R-SGUARE
0 54421312
COEF OF VARIATION
0 70335035
REGRESSION
Dl
f YPE I PS
R SQUARE
FRAT 10
PROD
LINEAR
4
302. 7052
0 1667
3 29
0. 0213
QUADRATIC
4
259. 0903
0 1431
2. 03
0. 0309
CROSBPRODUCT
6
425. 0630
0 2344
3. 09
0. 0153
TOTAL REGRESS
Id
908. 5305
0 5442
3. 07
0. 0034
RESIDUAL
nr
SS
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
655. 8601
25. 2254
1. 466
0. 2689
PURE ERROR
10
172. 0561
17. 2056
TOTAL ERROR
36
027. 9162
22. 9977
PARAMETER
nr
ESTIMATE
STD DEV
T-RAT10
PROB
INTERCEPT
i
7. 0163
1. 2129
5. 78
0. 0001
XI
]
0.69596000
0. 45774109
1. 52
0. 1371
X2
i
0.25523599
0.45830220
0. 56
0. 5011
X3
i
0.61406273
0.45774109
1. 34
0. 1076
X4
i
0.96055564
0. 4 5774109
2. 12
0. 0413
XI *X1
i
1. 1360
0.74231076
1. 53
0. 1347
X1#X2
i
0.04793793
0.25071512
0 19
0. 0494
X2#X2
i
-1.4473
0.74231076
-1. 95
0. 0590
X1*X3
i
0 40445033
0.23977945
1. 69
0. 1003
X2*X3
i
0.61450030
0.25071512
2. 45
0. 0192
X3X3
i
1. 3060
0. 74231076
1. 07
0. 0700
X1*X4
j
0.20712500
0.23977945
0. 87
0. 3889
X2*X4
i
0.14090743
0. 25071.512
0 56
0 5774
X3*X4
i
0.70112500
0.23977945
2. 92
0. 0059
X4#X4
i
-1.4090
0.74231076
- 2. 01
0. 0524
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
200. 1056
40. 021 1
1. 74
0. 1504
X2
5
239. 3300
47.0676
2. 03
0. 0904
X3
5
507. 0646
117 4129
5. 11
0. 0012
X4
r>
444.6547
00.9309
3. 87
0. 0066
146


69
Grass Yield (DM) for the Wet Season of 1979
The grass yield for the wet season of 1979 for all treatment
combinations is presented in Table 6. During this season the grass
yield varied from 560 kg DM ha ^ to 6530 kg DM ha ^ for treatment
combinations 17 [1 day grazing, 0 days rest (continuous grazing),
1.6 kg DM on offer/100 kg BW, and 0 kg ha ^ of superphosphate] and
23 (1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and
0 kg ha ^ of superphosphate), respectively. Grass yield was again
influenced by days rest and grazing pressure (X^) (P < 0.01),
while the experimental variables days grazing and fertilizer level
were not significant. There were no interactions between experimen
tal variables during this season.
The linear components of the model accounted for 72% of the total
variation, while the quadratic and interaction effects represented
only 2 and 3%, respectively (Appendix Table 23). The grass yield
was positively related to increasing lengths of rest period and nega
tively related to increasing amounts of forage on offer to the grazing
animals. For each 14 days increase in the rest period, there was an
increase of 800 kg ha ^ of grass yield; on the other hand, for each
unit (1.6 kg DM) decrease in grazing pressure, there was an increase
of 430 kg of grass yield. The analysis of variance is presented in
Appendix Table 23.
Grass Yield (DM) for the Dry Season of 1979
The grass yield for the dry season of 1979 for each treatment
combination is presented in Table 6. Grass yield varied from 210


65
growing points close to the soil surface as compared with guineagrass
and elephantgrass, each of which has apical meristems and lateral
buds located much higher from the crown. The two grasses included
in the current study are more sensitive to levels of defoliation
and short rest periods because these grasses do not have rhizomes
or stolons to store reserves. The legumes included in the current
study are climbing and twining tropical legumes which no doubt are
more sensitive to close defoliation than many other species.
Effect of Lengths of Rest Period and Levels of Grazing
Pressure on Grass Yield (DM)
The effects of lengths of rest period and levels of grazing
pressure on grass yield are presented in Table 6. The grass yield
is that available before each grazing cycle and each 56 days for
the continuous grazing treatments.
Grass Yield (DM) for the Wet Season of 1978
During this first experimental season no significant effects
were shown by any of the four variables (X^, X^, X^, X^) (Table 6
and Appendix Table 21).
Grass Yield (DM) for the Dry Season of 1978
The grass yield for the dry season of 1978 for each treatment
combination is given in Table 6. The grass yield varied from 400
kg DM ha ^ tO 4550 kg DM ha \ which was found on treatments 26
[28 days grazing, 0 days rest (continuous grazing), 1.6 kg DM on


81
only 6 and 7% of the total variation, respectively (Appendix Table 28).
During the 1979 wet season, the yield of legumes began to show a
pattern associated with the length of the rest period. For each 14
days increase in rest period, there was a decrease of 56 kg ha ^ of
legume yield. Also, during this season, grazing pressure began to
show an effect upon legume yield. For each unit (1.6 kg DM) decrease
in grazing pressure, there was an increase of 65 kg ha ^ of legume
yield. The analysis of variance is presented in Appendix Table 28.
Legume Yield (DM) for the Dry Season of 1979
The legume yield for the dry season of 1979 for each treatment
combination is presented in Table 7. Legume yield varied from 0 kg
DM ha ^ to 910 kg DM ha ^ for treatments 36 (14 days grazing, 56 days
rest, 5.0 kg DM on offer/100 kg BW, and 200 kg ha ^ of superphosphate)
and 6 (21 days grazing, 14 days rest, 6.6 kg DM on offer/100 kg BW,
and 100 kg ha ^ of superphosphate), respectively.
Legume yield was affected by length of rest period (X^) (P < 0.01)
and by grazing pressure (X^) (P < 0.05). Days grazing and fertilizer
levels had no significant effect and there were no interactions among
the experimental variables. For length of rest period (X^) there
was both a linear and quadratic effect upon legume yield (P < 0.01).
The linear and quadratic components of this experimental variable
accounted for 43 and 27% of the total variation, while the inter
actions among the experimental variables accounted for only 3%,
respectively (Appendix Table 29).


Table 16. Analysis of variance, regression coefficients and probabilities for available
forage (g DM/m ) for the wet season of 1978.
RESPONSE MEAN
246.0056
ROOT MSE
72. 4 536
R-SQUARE
0. 37774512
COEF OF VARIATION
0 27442454
REGRESSION
DF
TYPE I SS
LINEAR
-1
62605. 3377
QUADRATIC
4
15514. 7602
CROSBPRODUCT
6
37601. 2002
TOTAL REGRESS
14
115001
RESIDUAL
DF
SS
LACK OF FIT
26
71515. 1042
PURE ERROR
10
117468
TOTAL ERROR
36
180783
PARAMETER
DF
ESTIMATE
INTERCEPT
1
225. 2208
XI
1
-0.92600026
X2
1
17. 5255
X3
1
7. 5467
X4
1
-3. 2638
XI *X1
1
-7. 7256
XI *X2
1
-0. 07258068
X2*X2
1
15. 8160
X1*X3
1
5. 0666
X2#X3
1
1. 4203
X3X3
1
-5. 1840
X1*X4
1
6. 9022
X2*X4
1
-4.4027
X3*X4
1
0.03708333
X4#X4
1
3. 3577
FACTOR
DF
SS
XI
5
32373. 4814
X2
5
61408. 3773
X3
5
20403. 4411
X4
5
30843. 0311
R-SQUARE
F-RAT10
PR OB
0 2057
2. 77
0. 0316
0. 0507
0. 74
0. 5717
0. 1234
1. 17
0. 3317
0. 3777
1. 58
0. 1343
MEAN SQUARE
F-RATIO
PROB
2750. 5807
0. 234
0. 7706
11746. 8008
5249. 5307
STD DEV
T-RATIO
PROB
18. 3246
12. 27
0. 0001
6. 7157
-0. 13
0. 8742
6. 7254
2. 82
0. 0078
6. 7157
1. 07
0. 2824
6. 9157
-0. 47
0. 6398
11. 2151
-0. 67
0. 4753
3. 7877
-0. 02
0. 9848
11. 2151
1. 41
0. 1671
3. 6227
1. 40
0. 1705
3. 7877
0. 37
0. 7077
11. 2151
-0. 46
0. 6467
3. 6227
1. 93
0. 0617
3. 7877
-1. 16
0. 2528
3. 6227
0. 01
0. 9715
11. 2151
0. 30
0. 7664
MEAN SQUARE
F-RATIO
PROB
6474. 6763
1. 23
0. 3138
12281. 6755
2. 34
0. 0614
4080. 6882
0. 78
0. 5725
6168. 6062
1. 18
0. 3404
131


Table 45. Analysis of variance, regression coefficients and probabilities for visual
estimation legume for the wet season of 1980.
RESPONSE MEAN
11. 1137
ROUT MSE
6. 3540
R-SGUARE
0.75230103
COEF OF VARIATION
0 57173120
REGRESSION
1>I
TYPE I SS
R-SGUARE
F-RATIO
PROP
LI NEAR
4
2707. 3303
0. 4612
16 76
0. 0001
QUADRATIC
4
1578. 0067
0. 2609
9. 77
0. 0001
CROSSPRODUCT
6
130. 0670
0. 0223
0. 54
0. 7740
TOTAL REGRESS
14
4416.2930
0. 7524
7. 01
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PR OB
LACK OF FIT
26
1304.4804
S3. 2492
7. 720
0. 0007
PURE ERROR
10
60. 9763
6. 0776
TOTAL ERROR
36
14 53. 4 567
40. 3730
PARAMETER
or
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
19 3163
1.6070
12. 02
0. 0001
XI
1
0. 4 1 117677
0.60647572
0. 68
0. 5021
X2
1
-4.1517
0.60734510
-6. 84
0. 0001
X3
1
0.23409952
0.60649572
0. 39
0. 7008
X4
1
0.55973235
0.60649572
0. 92
0. 3622
X1*X1
1
0.32311172
0.90354356
0. 33
0. 7444
X1*X2
]
-0.22612757
0.33219139
-0. 68
0. 5004
X2X2
1
-3. 0602
0 98354356
-3. 92
0. 0004
X1*X3
1
0. 01033929
0.31770109
0. 06
0. 9543
X2*X3
J
-O. 07016130
0.33219139
-0. 21
0. 8339
X3*X3
1
1.7370
0.98354356
1. 77
0. 0854
X1 *X4
1
0.09626307
0.31770189
0. 30
0. 7636
X2*X4
1
-0.34437662
0 33219139
-1. 04
0. 3060
X3wX4
1
0 39726309
0.31770107
1. 25
0. 2192
X4*X4
]
-1. 1352
0 90354356
-1. 15
0. 2560
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
37.0387
7. 4077
0. 18
0. 9670
X2
5
2645. 3108
529. 0622
13. 10
0. 0001
X3
5
196.0710
39.2142
0. 97
0. 4484
X4
5
100. 9017
36. 1963
0. 90
0. 4939


Table 25 Analysis of variance, regression coefficients and probabilities for grass
yield (g DM/m ) for the wet season of 1980.
RESPONSE MEAN
363.4776
ROUT MSE
1 47. 0421
R-SGUARE
0 671)06000
COEF OF VARIATION
0 40404237
REGRESSION
DP
TYPE I SS
R-SGUARE
F-RATIO
PROD
LINEAR
4
1424030
0. 5733
16. 47
0. 0001
QUADRATIC
4
121067
0 0507
1.41
0. 2506
CROSSPRODUCT
A
76325. B507
0 0310
0. 57
0. 7373
TOTAL REGRESS
J 4
1623031
0. 6757
5 36
0 0001
RESIDUAL
l>F
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
2 A
524064
20175 5280
0. 775
0. 6768
PURE ERROR
10
253006
25300. 5700
TOTAL ERROR
36
770367
21621. 3738
PARAMETER
m
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
264. 0106
37. 1871
7. 12
0. 0001
XI
1
-8. 5772
14.0352
-0. 61
0. 5437
X2
1
87.0574
14. 0547
6. 17
0. 0001
X3
1
61. 2220
14.0352
4. 36
0. 0001
X4
J
16. 3700
14.0352
-1. 17
0. 251 1
XI *X1
1
0. 6004
22. 7607
0. 38
0. 7052
X1*X2
1
-3. 0056
7. 6874
-0. 50
0. 6170
X 2 X 2
1
17.6076
22. 7607
0. 06
0. 3746
XI #X3
1
2. 0014
7. 3521
0. 38
0. 7054
X2*X3
1
-10 5737
7. 6874
-1. 38
0. 1775
X3*X3
1
-7.3154
22.7607
-0. 32
0. 7478
X1 #X4
1
4. 1417
7. 3521
0. 56
0. 5767
X2*X4
1
3. 7002
7. 6074
0. 52
0. 6071
X3*X4
J
5. 7500
7. 3521
0. 81
0. 4237
X4*X4
1
7. 3512
22.7607
0. 4 1
0. 6836
FACTOR
Di
SS
MEAN SQUARE
F-RATIO
PROB
XI
li
32207. 0672
6441.7738
0. 30
0. 7108
X2
5
706124
181225
8. 38
0. 0001
X3
5
431172
86234.3237
3. 77
0. 0056
X4
5
54835. 6707
10767.1381
0. 51
0. 7688
140


N>
hO
ro
ro
ro
M
H*
H*
h-
M
h-*
H
H
h-4
2
4S
LO
ro
H*
o
VO
00
ON
Ln
4^
LO
ro
H*
o
VO
00
O'
Ln
4^
LO
ro
M
o
o
ro
ro
ro
ro
ro
ro
ro
ro
ro
ro
ro
ro
X
00
H
00
M
CO
H1
00
H*
I1
'O
H*
I-*
\l
h-4
H-*
'O
h-*
M
h-*
o
b
H
Ln
Ln
Ln
Ln
4N
4>
h-*
I*
4S
4N
I-*
4^
4^
h-*
h-*
>
4^
I*
H*
X
o
ON
ON
o
O
On
ON
o
o
ro
ro
4^
ro
ro
4N
4^
ro
ro
4^
4N
ro
ro
4^
P-
ro
pa
&3
h
rt
3
fD
P
rt
CO
00
oo
00
00
H*
M
H
H*
ON
ON
ON
ON
LO
LJ
LO
LO
ON
ON
ON
ON
LJ
LO
LO
LO
X
LO
o
LJ
u>
LO
D
ON
ON
ON
ON
ON
ON
ON
ON
LO
LO
LO
LO
ON
ON
ON
ON
LO
LJ
LO
LJ
S'?
t-
W
PT
L0
LO
LO
LO
LO
LO
LO
LO
M
M
H*
h-4
M
M
M
M
(TO
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
Tl
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
XT
t-
03
H
N3
ro
H*
ro
ro
h-1
h-1
H*
H*
M
h-*
H
M
H*
M
f*
h-*
H*
M
M
pa
(D
*0
CO
s;
(-*
h-
H
(-*
VO
On
ON
ro
vO
00
1-*
LO
VO
4>
LO
ON
>
M
Ln
C/3
'O
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
++
oo
o
M
M
M
vO
u>
4>
4N
Ln
-P>
VO
ro
ON
M
Ln
00
LJ
cn
'-J
o
o
o
o
o
o
o
O
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
++
00
7
TO
s
1-*
LO
LO
ro
ro
I*
h-*
00
VO
4S
h-*
o
00
Ln
00
ro
o
ro
f-*
4>
LO
VO
ON
XT
03
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
P>
W*
VO
o
M
Ln
ro
(-*
ro
H*
"O
'*>
VO
H*
h-1
ro
LO
H
VO
ON
4N
f*
LO
o
o
Ln
ON
4>
VO
03
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
++
VO
M
s:
1*
H*
00
ro
4>
4^
4>*
ro
VO
VO
ro
-C'
VO
ON
H-*
VO
LO
00
ro
P-
h-i
h->
ro
ro
h-*
ro
03
oo
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
O
o
o
o
o
o
++
o
Z6
Table 8. Dry matter means for weeds for year, season, and treatment combination


Table 15. Analysis of variance, regression coefficients and probabilities for
aerial biomass (g DM/m ) for the wet season of 1980.
RESPONSE MEAN
417.1036
ROOT MSE
147. 3207
R-SQUARE
0. 52770332
COEF OF VARIATION
0.35313162
REGRESSION
DF
TYPE I SS
R-SGUARE
F-RAT10
PROB
LINEAR
4
713378
0. 4312
8. 22
0 0001
QUADRATIC
4
133039
0. 0804
1. 53
0 2135
CROSSPRODUCT
6
26824. 4504
0. 0162
0. 21
0. 9727
TOTAL REGRESS
14
073262
0. 5278
2. 87
0. 0054
RESIDUAL
DF
SS
MEAN SQUARE
F-RAT10
PROB
LACK OF FIT
26
517574
19983. 6280
0. 763
0. 7235
PURE ERROR
10
261748
26174. 7019
TOTAL ERROR
36
781322
21703. 3927
PARAMETER
DF
ESTIMATE
STD DEV
T-RAT10
PROB
INTERCEPT
1
321. 7268
37. 2595
8. 63
0. 0001
XI
1
-10. 4159
14. 0618
-0. 74
0. 4637
X2
1
55. 4276
14. 0815
3. 94
0. 0004
X3
1
45. 2128
14. 0618
3. 22
0. 0028
X4
1
-6. 7517
14. 0610
-0. 48
0. 6340
XI *X1
1
13. 0832
22. 0030
0. 61
0. 5465
X1X2
1
-3. 2825
7. 7020
-0. 43
0. 6725
X2*X2
1
7. 3623
22. 0038
0. 32
0. 7487
X1#X3
1
3. 7180
7. 3660
0. 50
0. 6168
X2X3
1
-2. 2792
7. 7020
-0. 30
0. 7690
X3*X3
1
5. 9623
22. 8038
0. 26
0. 7952
X1*X4
1
4. 0870
7. 3660
0. 66
0. 5113
X2*X4
1
-1. 1345
7. 7020
-0. 15
0. 8837
X3*X4
1
3. 6838
7. 3660
0. 50
0. 6200
X4*X4
1
5. 9707
22. 8038
0. 26
0. 7949
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
44832.3549
8966. 4710
0. 41
0. 8365
X2
5
347008
69401. 6054
3. 20
0. 0172
X3
5
246633
49326. 5173
2. 27
0. 0679
X4
5
23434. 6732
4686. 9346
0. 22
0. 9534
130


35
For this experiment grazing pressure was expressed as the
amount of dry matter on offer per 100 kg body weight (BW) and
by the residual dry matter in kg ha left after each grazing.
Each experimental variable was studied at five levels. There
fore, the treatments comprised a factorial type of experiment of
4
four factors, each at five levels (5 factorial) (Table 2).
Experimental Design
Due to the large number of experimental units required to
4
conduct a 5 factorial (626 treatment combinations without replica
tions) a response surface design, namely, a modified central compo
site non-rotatable design was used. The number of design points
(treatment combinations) was determined from the following formula:
(see Table 2)
4 4
No. of design points = 2 (1) 4- 2 (2) + (2 x 4) + 1 = 41
Certain treatments were replicated twice (central point was replicated
thrice) and these are indicated in Table 2. The total number of
experimental units was 51.
Field Plan of the Experiment
In order to estimate the size of experimental units for each
treatment, the following formula was used: (see Table 3)
NdR
b DG
2
where S = size of experimental unit in m ,
N = kg body weight/pasture/day (assumed 300 kg BW),
d = number of days pasture is grazed during cycle,


Table 23. Analysis^ of variance, regression coefficients and probabilities for grass yield
(g DM/m ) for the wet season of 1979.
RESPONSE MEAN
346. 4752
ROOT MSE
96. 5940
R-SQUARE
O 77301777
COEF OF VARIATION
0. 271379271
REGRESSION
DF
TYPE I OS
R-SQUARE
F-RATIO
PROD
LINEAR
4
1067382
0. 7107
28. 60
0. 0001
QUADRATIC
4
47265. 5552
0 0318
1. 27
0 301 1
CROSSPRODUCT
6
34536. 8369
0. 0233
0. 62
0. 7153
TOTAL REGRESS
J 4
1 149105
0. 7730
0. 80
0. 0001
RESIDUAL
DP'
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
170007
6569. 5116
0. 398
0. 9712
PURE ERROR
10
165092
16509. 2395
TOTAL ERROR
36
335900
9330. 5471
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
277.3786
24.4302
11. 35
0. 0001
XI
1
-9.8372
9. 2200
-1. 07
0. 2931
X2
1
80.3952
9. 2329
8. 71
0. 0001
X3
1
43. 8425
9. 2200
4. 76
0. 0001
X4
1
-5. 7950
9. 2200
-0 63
0 5336
XI *X1
1
4. 644 5
14.9519
0. 31
0. 7579
XI *X2
1
-7. 5150
5. 0500
-1. 49
0. 1454
X2*X2
1
8. 3716
14.9519
0. 56
0. 5790
X1*X3
1
-2.4081
4. 8297
-0. 50
0. 6211
X2*X3
1
-2. 4641
5. 0500
-0. 49
0. 6286
X3*X3
1
-6.0409
14.9519
-0. 40
0. 6886
X1*X4
1
4.1761
4.8297
0. 86
0.3929
X2*X4
1
1. 3914
5. 0500
0. 28
0. 7845
X3*X4
1
2. 0305
4.8297
0. 42
0. 6767
X4*X4
J
11. 6716
14.9519
0. 78
0. 4401
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
53860. 9942
10772. 1988
1. 15
0. 3502
X2
5
737651
147530
15. 01
0. 0001
X3
5
224753
44950. 5041
4. 82
0 0018
X4
5
10083. 8137
3616. 7627
0. 39
0 8540
138


CHAPTER II
LITERATURE REVIEW
In tropical regions animal production on pastures is dependent
upon the quality and quantity of forage available throughout the year.
In the humid tropics the seasonal growth of pastures is greatly
influenced by the wet and dry periods. At the beginning of the
wet season very rapid growth occurs and frequently a large amount of
forage accumulates. As the season progresses the rate of growth
decreases rapidly and approaches zero during the dry season. The
concentration of nitrogen and minerals is relatively high at the
beginning of the wet season and also falls rapidly as the season
progresses. In addition there is a progressive increase in fiber and
decrease in digestibility from the beginning of the wet season
through the remainder of the year (Paladines and de Alba, 1963).
Some tropical pasture species seem to be well adapted to extreme
environmental conditions; however, their potential to supply feed for
cattle production may be limited not only by seasonal changes but
also by soil fertility (Tergas, 1968), mechanisms of adaptation
(Gartner et al., 1974), and grazing management (Mott, 1960; Evans,
1970; Stobbs, 1969).
The Tropical Forages
Panicum maximum Jacq. has been described by Humphreys (1980) and
Bogdan (1977) as a densely tufted perennial grass with relative drought
and poor soil tolerance. It is of high nutritive value when young and
combines well with other tropical pasture species (Hudgens, 1973;
3


15
by 40% when it was grown in mixtures with guineagrass, and the amount
of N fixed by the legume was estimated to be 100 kg N ha ^ yr \ In
northern Australia, Norman (1970) showed that gain per animal grazing
legume-grass mixtures was linearly related to the proportion of legume
in the pasture. Also, animal liveweight gain during the dry season
was related to the number of days during which the animals had been
grazing the legumes. He also reported that animals grazing on grass
pastures gained 60 kg head \ while those allowed to graze grass-legume
pastures gained 280 kg head \ During the period of 112 days of the
dry season, the first group on grass pastures lost almost 40 kg animal \
while the second gained 60 kg animal \ due to the companion legume,
Stylosanthes humilis. In Ecuador, Chavez (1974) reported that a guinea-
grass-centro mixture produced 536.5 kg liveweight gain ha ^ year ^ with
the beneficial effect of the legume being more apparent during the dry
season, when crude protein content of the guineagrass alone was below
7% and crude protein content for the grass-legume pastures over 10%.
Similar results were reported by INIAP (1979) on guineagrass-glycine
var 'Tinaroo' mixture, which produced 458 kg of liveweight ha ^ year \
In both cases the results were obtained under rotational grazing using
the put-and-take technique developed by Mott (1960).
According to Minson and Milford (1967), tropical legumes maintain
adequate nutritive values for a longer period of time than most tropical
grasses, when each were under the same management system. The critical
level of crude protein required in a pasture before intake is reduced
by protein deficiency is estimated to lie between 6.0 and 8.5%. Even
highly N-fertilized tropical grasses at late growth stages may have


Ill
use more efficiently the environmental factors such as light,
temperature, moisture, nutrients, and space, producing a highly
competitive situation, and finally eliminating the less agressive
or unadapted companion species.
The degree of defoliation as represented in this experiment
by the intensity of grazing pressure also had a marked effect upon
the percentage grass. Again the relationship is quadratic as shown
by Figs. 15, 17, and 18.
Harris (1978) mentioned that close, continuous defoliation leads
to a more species-rich association dominated by species with prostrate,
rhizomatous, stoloniferous or basal rosette habit, while tall species
tend to disappear. The dominance or suppression of tall-growing
species could largely be controlled by the degree of defoliation
permitting or not permitting light penetration to levels where
prostrate species dispose their leaf canopies.
The percentage grass at the beginning of the experiment was
about 60%, while during the last wet season (1980), the mean
percentage grass was 79% which included a spread of from 7 to 100%
grass, depending upon the treatment combination. Low grazing pres
sure in combination with long rest periods produced swards with
almost 100% grass, whereas combinations of high grazing pressure
and short rest periods or continuous grazing resulted in almost a
complete elimination of the grass component. These results were
obtained notwithstanding that it is well known that these two species
are very well adapted to the environmental conditions of the region.


17
optimum grazing pressure. McMeekan (1956) considered the stocking
rate as the most powerful factor, influencing the efficiency of pasture
conversion to animal products on a per unit area basis. Petersen et al.
(1965) developed quantitative expressions which related grazing pressure
and carrying capacity with output per animal. Their quantitative theory
suggests that animal gain is constant as the stocking rate is increased
up to a "critical point," where the grazeable forage is equal to the
amount of forage consumed by grazing animals. If the stocking rate is
increased beyond the critical point, then the gain per animal decreases,
and the animal gain per unit area also decreases. Conway (1965), using
three stocking rates, 1.0, 1.75, and 2.5 animals per acre, found that
by increasing the stocking rate from 1.0 to 1.75 animals per acre,
liveweight gain per animal decreased. However, with 2.5 animals per
acre, the liveweight per animal was drastically reduced resulting also
in a reduction of liveweight per acre. Jones (1979) in Australia
studied the effect of five stocking rates, namely 0.8, 1.3, 1.8, 2.3,
and 2.8 an ha ^ in combination with three resting periods of 17, 39,
and 50 days using a randomized complete block design with two replica
tions. The area allocated to each treatment varied from 0.02 to 0.24
ha, using just one animal per experimental unit. Increasing the stocking
rate had the greatest influence upon the persistence of the legume which
in this case was Siratro. Also, the resting period became important
to the legume productivity under heavy grazing pressure conditions.
Results suggest that under heavy grazing pressure a longer rest period
would allow the legume to increase or maintain its reserves of nutrients
for subsequent growth. Echandi (1956) suggested that carrying capacity


Hours
Fig. 4. Solar radiation at Estacin Experimental Pichilingue during the period
of 1978-1980.
LO
N5


Table 7. Dry matter means for legume by year, season, and treatment combination
Treatments
1978
1978
1979
1979
1980
D/G
(\)
D/R
(x2)
G/P
(X ) % BW
F
No.
kg ha ^
Reps
W/S
D/S
W/S
D/S
W/S
1
7
14
3.3
100
1
640
620
kg ha
560
510
480
2
21
14
3.3
100
1
410
590
510
330
420
3
7
42
3.3
100
1
1240
590
250
710
150
4
21
42
3.3
100
1
580
410
280
350
130
5
7
14
6.6
100
1
520
670
900
640
560
6
21
14
6.6
100
1
990
740
660
910
730
7
7
42
6.6
100
1
1150
680
220
110
20
8
21
42
6.6
100
1
670
870
320
310
280
9
7
14
3.3
300
1
1210
590
500
700
430
10
21
14
3.3
300
1
780
680
900
770
940
11
7
42
3.3
300
1
1010
650
340
320
80
12
21
42
3.3
300
1
600
470
500
210
30
13
7
14
6.6
300
1
890
640
400
770
520
14
21
14
6.6
300
1
1380
840
810
650
600
15
7
42
6.6
300
1
810
990
350
470
470
16
21
42
6.6
300
1
1240
530
190
150
0
17
1
0
1.6
0
1
650
440
270
270
300
18
28
0
1.6
0
1
610
430
460
370
260
19
1
56
1.6
0
2
860
540
190
40
0
20
28
56
1.6
0
2
850
540
490
190
0
21
1
0
8.3
0
1
870
850
620
630
370
22
28
0
8.3
0
1
710
930
490
370
110
23
1
56
8.3
0
2
1150
420
80
40
20
24
28
56
8.3
0
2
1320
1160
950
300
60


21
height, weight, and other factors differ from one species to another
(Kennedy, 1972). For many years, hand cutting and weighing of above
ground vegetative parts has been the most popular and useful method
for estimating forage yield (Heady, 1970) and for estimating grazing
pressure (Mott, 1960). Other methods, such as the simple disk meter,
described by Bransby (1975), work on the principle of measuring the
height of a disk supported by the resistance and compression of the
vegetation. Santillan (1976) proved that the simple disk meter was
very satisfactory for use in tropical species.
Mott (1974) suggested that estimation of total yield or yield of
components is based upon the following relationship: yield per unit
area = f (density, height). The total yield of an area of vegetation
is related to the density and height of individual components. Ground
cover and sward height have been used on different types of grassland
to estimate dry matter yield.
Where pasture vegetation is utilized by grazing animals, the
amount of feed present at any one time may be only one of the factors
associated with the intake by the grazing animals. Of interest to
the pasture scientist is an estimate of that portion of the pasture
which is consumed by the grazing animal since they are very selective
of plant species and plant parts, which makes for a more complex
situation (t'Mannetje, 1978).
A double sampling technique is probably one of the best ways to
ensure a more precise yield determination. Eye-estimation in combina
tion with a few harvested samples which act as a control on the obser
ver's accuracy is one of the simplest forms of estimating total forage
present or annual production potential of a pasture (t'Mannetje, 1978).


47
(4) Legume yield (DM),
(5) Yield of weeds (DM),
(6) Grass percentage, and
(7) Legume percentage.
The above parameters were statistically analyzed for the partial
wet season (May-June) of 1978, the dry season of 1978, wet and dry
seasons of 1979, and wet season of 1980. The data were processed
using the programs RSREG for response surface design, and 63D for
plotting the three-dimensional graphs of the Statistical Analysis
System of the Northeast Regional Data Center of the University of
Florida.


Table 19. Analysis of variance, regression coefficients and probabilities for available
forage (g DM/m ) for the dry season of 1979.
RESPONSE MEAN
RODT MSE
R-SGUARE
COEF OF VARIATION
290. 0490
90 0370
O 60679401
0 32000537
REGRESSION
LINEAR
QUADRATIC
CROSSPRODUCT
TOTAL REGRESS
RESIDUAL
LACK OF FIT
PURE ERROR
TOTAL ERROR
PARAMETER
INTERCEPT
XI
X4
X1*X1
X1*X2
X2*X2
X1#X3
X2*X3
X3*X3
X1#X4
X2*X4
X3*X4
X4*X4
FACTOR
XI
X2
X3
X4
DP
TYPF I SS
R-SGUARE
F-RATIO
PROD
4
476562
0. 5416
12. 40
0. 0001
4
3527. 1072
0. 0040
0. 09
0. 9845
6
538B7. 0220
0. 0612
0. 93
0. 4B25
14
533977
0. 6068
3. 97
0. 0004
DF
SS
MEAN SQUARE
F-RATIO
PROB
26
201300
7742. 3085
0. 535
0. 9027
10
144719
14471.9289
36
346019
9611. 6475
DF
ESTIMATE
STD DEV
T-RATIO
PROB
1
278. 0900
24. 7955
11. 22
0. 0001
1
0. 56256621
9. 3577
0. 06
0. 9524
1
32. 8397
9. 3710
3. 50
0. 0012
1
57. 1874
9. 3579
6. 11
0. 0001
1
-5. BOO5
9. 3579
-0. 62
0. 53B7
1
3. 3742
15. 1755
0. 22
0. 8253
1
4. 8603
5. 1255
0. 95
0. 3493
1
-2. 3591
15. 1755
-0. 16
0. 8773
1
3. 7646
4. 9020
0. 77
0 4475
1
-7. 2502
5. 1255
-1. 41
0. 1658
1
-1. 3409
15. 1755
-0. 07
0. 9301
1
5. 5716
4. 9020
1. 14
0. 2632
1
2. 5185
5. 1255
0. 49
0. 6261
1
3. 7439
4. 9020
0. 76
0. 4500
1
4. 7659
15. 1755
0. 31
0. 7553
DF
SS
MEAN SQUARE
F-RATIO
PROB
5
28398. 1763
5679. 6393
0. 59
0. 7069
5
140251
29650. 2621
3. 08
0. 0203
5
371753
74350. 6865
7. 74
0. 0001
5
23685. 1872
4737. 0374
0. 49
0. 7794
134


46
dry matter yield and botanical composition. In September, five
sampling units out of 30 disk readings were randomly selected and
clipped for the above determinations. These samples were later
hand separated into their components, placed inside cloth bags
properly identified and dried for 20 hours at 72C. The sum of
the dry weights of the components yielded the total dry weight of
the sample. Sickles were used for harvesting the forage samples.
The clipped samples were used to adjust through regression analysis
the 15 or 30 disk-meter readings of dry matter yield. The forage
meter readings and visual estimates of dry weight were used as
independent variables in regression equations to generate the
regression coefficients needed for calibration of the disk meter
and for adjustment of the visual sample estimation. The visual
estimate of percent yield was made for the component grasses
(guineagrass and elephantgrass), legumes (centro and glycine),
and weeds. Although the above grasses and others such as Paspalum
fasciculatum, P. paniculatum and Eleusine indica were present in
scattered, small patches in some pastures, they were included in
the weed components. Likewise, some native Desmodiums were accounted
for in the legume component.
The response of the pasture mixture to the experimental variables
was measured in terms of the following parameters:
(1) Aerial biomass (DM) kg ha ^ = grass (DM) + legume (DM) +
weeds (DM),
(2) Available forage (DM) kg ha ^ = grass (DM) + legume (DM),
(3) Grass yield (DM),


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of shcolarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
77"
G. 0. Mott, 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.
7 Qk
n u-.
J. E. Moore
/y 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.
GL
0. C. Ruelke
Professor of Agronomy


Table 12. Analysis of variance, regression coefficients and probabilities
biomass (g DM/m ) for the dry season of 1978.
RESPONSE MEAN
304. 7018
ROOT MSE
71. 7 J 00
R-SQUARE
0.76482802
CGEF OF VARIATION
0.23504627
REGRESSION
DF
TYPE 1 SS
R-SQUARE
F-RATIO
LINEAR
4
582507
0. 7359
28. 16
QUADRATIC
4
8183 1417
0. 0103
0. 40
CROSSPRODUCT
6
14735. 0435
0. 0186
0. 47
TOTAL REGRESS
14
605425
0. 7648
8. 36
RESIDUAL
DF
SB
MEAN SQUARE
F-RATIO
LACK OF FIT
26
82791. 6925
3184. 2959
0. 308
PURE ERROR
to
103366
10336 5774
TOTAL ERROR
36
186157
5171 0413
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
INTERCEPT
1
274. 4606
18. 1871
15. 09
XI
1
-2. 5748
6. 8638
-0. 38
X2
1
43. 1704
6. 8735
6. 28
X3
1
50. 7477
6. 8638
7. 39
X4
1
-12. 7720
6. 8638
-1. 89
X1*X1
1
3. 6619
11. 1310
0. 33
X1*X2
1
-0. 07950613
3. 7595
-0. 02
X2*X2
1
7. 3209
11. 1310
0. 66
X1#X3
1
4.4033
3. 5955
1. 22
X2#X3
1
3. 5380
3. 7595
0. 74
X3#X3
1
-1. 0006
11. 1310
-0. 09
X1*X4
1
0. 56158148
3. 5955
0. 16
X2*X4
1
1. 2143
3. 7575
0. 32
X3*X4
1
-2. 0808
3. 5955
0. 58
X4*X4
1
-2.8006
11. 1310
-0. 25
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
XI
5
9278. 2281
1855. 6456
0. 36
X2
5
212945
42589. 0883
8. 24
X3
5
346557
69311. 4215
13. 40
X4
5
21010. 6802
4202. 1360
0 81
for aerial
PR OB
0. 0001
0. 8104
0. 8224
0. 0001
PROB
0. 9924
PROB
0. 0001
0. 7076
0. 0001
0. 0001
0. 0664
0. 7441
0. 9032
0. 5149
0. 2207
0. 3529
0. 9289
0. 8768
0. 7486
0. 5664
O. 8028
PROB
0. 8731
0. 0001
0. 0001
0. 5485
127


60
rest, 5.0 kg DM on offer/100 kg BW, and 200 kg ha ^ of superphosphate)
and 19 (1 day grazing, 56 days rest, 1.6 kg DM on offer/100 kg BW,
and 0 kg ha ^ of superphosphate), respectively.
Considering the individual linear effects, days rest (X^) was the
only experimental variable affecting available forage (P < 0.01). The
other three experimental variables did not affect the available forage.
Individual quadratic effects or interactions were not significant.
The lack of interaction indicates days rest is acting independently
over available forage, which means that the amount of available forage
increases as the length of the rest period increases.
For each 14 days of increase in days rest (X^), there was 190 kg
ha ^ increase in the available forage.
Available Forage for the Dry Season of 1978 (DM)
Available forage varied from 790 kg DM ha ^ to 4880 kg DM ha \
corresponding to treatment combinations 26 [28 days grazing, 0 days
rest (continuous grazing), 1.6 kg DM on offer/100 kg BW, and 400 kg ha
of superphosphate] and 32 (28 days grazing, 56 days rest, 8.3 kg DM on
offer/100 kg BW, and 400 kg ha ^ of superphosphate), respectively.
During the first dry season, the lowest values of available forage
were found in those treatment combinations of short rest periods or
continuous grazing with the highest grazing pressure, while the highest
values were the longest rest periods and the lowest grazing pressures.
The linear components of the model accounted for 74% of the total
variation, while the quadratic effects and interactions represented
only 1 and 1% of the total variation, respectively (Appendix Table 17).


on
H*
TO
r| T3
3 i-(
H-* O
(1) Hi
PACIFIC OCEAN
)RY SAVANNAH
o
n>
i-i =r
3 3
H-
3 rf
m 3"
3 f-i
I| CD
I-1 (T>
3 3
3 3
C3 rr
C
rr r-i
3 3
3 t1
T3
3 r-j
"( 3
3 TO
rr I*-
C O
r( 3
3 CO
M
O
c
3
CL
O
i-i
cn
rr
o
Sl
t-L
3
TO
3"
3
3
3
<
3
TO
3
<
3
N
O
3
3
Cfl
H*
3
DRY TROPICAL FOREST
(600-1800mm) (24-60 in.)
WET TROPICAL FOREST
l500-3000mm (60-120 in.)
.WET SUBTROPICAL FOREST
A 1200-2500 mm (48-100 in.)
DRY-WET COLD GRASSLAND
VEGETATION
VERY COLD TUNDRA
CHIMBORAZO
, DRY COLD GRASSLANDS
i I (NATIVE)
600mm ( 24 in.)
RY-WET HIGH GRASSLANDS
1 '(CULTIVATED)
1 DRY VALLEYS (NATIVE S
1 CULTIVATED GRASSES)
DRY COLD GRASSLANDS
(NATIVE)
TUNGURAGUA
WET COLD GRASSLANDS
, (NATIVE)
VERY WET SUBTROPICAL FOREST
(2500-3500mm)( 100-140 in.)
. : 1
VERY WET TROPICAL FOREST
(3000-6000mm (120-240in.)
oj ^ ui
o o o
O o o ^
O o o o
00 OJ CO o A o5
bbobon|
6Z


te£
uve
5etcetV
taSe
6-6
Gta^v ottet
5.0
**
,ecetV
taSe
T)aVs
3d aTV
d £
taS
ioS 9
teSs
u*e
et ^8-
is-
Effect of resL ,
for the wet season o
%
i


12
(1975), comparing six levels of P (0, 15, 30, 60, 120, and 180 mg/
2 kg of soil/pot), found that P concentration in guineagrass and
elephantgrass was increased with the addition of P. The same response
was reported by Vicente-Chandler (1975) from Puerto Rico, where 80%
of the maximum dry matter production of pearl millet [Pennisetum ameri-
canum (L.) K. Schum.] was obtained when soil pH was raised to 5.5 and
115 ppm of P were added. At a pH of 4.8, twice as much P was needed
to produce the same forage yield. Fox (1979) reported that the standard
P requirement was a relative soil requirement, not an absolute plant
requirement. Fox et al. (1974) from Hawaii reported large increases
in P uptake by several pasture species, once soils which had high
P-fixing capacities were limed to pH 5.0 and 6.0. Phosphorus require
ments of soils can range from zero to more than 2220 kg P ha \
After N and P, S is considered by many as the next most important
element needed for tropical legume growth. Tergas (1977) noted the
significance of S on the growth and nodulation of several different
tropical forage legumes. Siratro and centro dry matter and nodule
weight increased as S was increased. Sanchez (1977) mentioned that
S deficiencies are widespread throughout the tropics and that some
pasture legumes are more susceptible to S deficiency than most grasses.
Medina (1969) reported S deficiencies in some crops growing in
the littoral region of Ecuador. Also, a strong response from guinea-
grass, paragrass, glycine, and centro was observed when S was applied
alone or supplied by the ordinary superphosphate or by ammonium
sulphate (INIAP, 1980).


40
R = kg dry matter offered/kg body weight/day,
D = number of days in cycle, and
2 2
G = growth rate in kg/m /day (assumed .04 kg/m /day).
As an example, treatment 14 days of grazing, 28 days of rest, and
5.0 kg DM on offer/100 kg BW/day is calculated as follows:
300 x 14 x .05 -,orri 2
S = To A~n7w = 1250 m
42 x 0.004
This figure was rounded to 1500 m to avoid the inconvenience of odd
2
pasture sizes. All pastures were 500 m or some multiple (Table 3).
In making this calculation it was assumed that one animal weighing
approximately 300 kg was used to graze the pasture during the allotted
grazing period and designated grazing pressure. It was also assumed
2
that the growth rate was 4 g/ra /day. In estimating the size (S) of
2
the experimental unit the smallest was set at 500 m in order to
avoid difficulty in handling the steers in units of smaller size.
A total of 7.3 hectares were required for the 51 experimental units
in this study (Table 3). These pastures were randomly distributed
as shown in Fig. 5. The sizes of the experimental pastures varied
2 2
from 500 m to 4000 m ; the larger areas were for the continuously
grazed treatments (Fig. 5).
Land Preparation and Pasture Establishment
Land preparation of the experimental area began in May 1977,
after the existing vegetation (common guineagrass) was partially
eliminated by an application of a grass killer herbicide (Glyfosate) .
At the beginning of July the 6-year-old guineagrass pasture was plowed
under.


71
The other two experimental variables, and X^, were not significant
and no interactions were found between the experimental variables.
The linear components of the model accounted for 59% of the total
variation, while the quadratic and interaction effects represented
only 5 and 3% of the total variation, respectively (Appendix Table
25). An increase of one unit (14 days) of rest increased the grass
dry matter yield by 870 kg ha \ while a decrease in the grazing
pressure increased the grass dry matter yield by 610 kg ha The
analysis of variance is presented in Appendix Table 25.
Summary of Grass Yield (DM)
During the first rainy season (1978), with less than 2 months
duration from May 12 to June 30, the experimental variables had no
effect upon grass yield.
Beginning with the dry season of 1978 through the last wet
season of 1980, the grass yield was positively related to increasing
lengths of rest period. The two grasses, guineagrass and elephant-
grass, used in this experiment are tall growing species with a high
capacity for dry matter production and each of them are favored by
long rest periods. As the length of the rest period was increased
the dominance of each of these grasses was evident, leaving very
little space for other species, such as legumes and weeds. The grass
yield also increased with time as these two species of grasses became
securely established. The mean grass yield for the wet season of
1978 was 1530 kg DM ha \ while for the last wet season of 1980,
the grass yield was 3630 kg DM ha It is evident that the amount


Table 22.
Analysis^ of variance, regression coefficients and probabilities for grass yield
(g DM/m ) for the dry season of 1978.
RESPONSE MEAN
236. 7073
ROOT MSE
60. 4161
R-SQUARE
0. 77736774
COEF OF VARIATION
0.27603200
REGRESSION
IM-
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
512740
0. 74 10
27. 75
0. 0001
QUADRATIC
4
14000 7664
0. 0203
0. 82
0. 5174
CROSSPRODUCT
6
11070. 4024
0 0160
0 43
0. 8527
TOTAL REGRESS
.1 4
537711
0 7774
8 78
0. 0001
RESIDUAL
nr
SS
MEAN SQUARE
F-RATIO
PROP
LACK OF FIT
26
66227. 4006
2547. 2047
0. 270
0. 7746
PURE ERROR
to
07024. 1534
8702. 4153
TOTAL ERROR
36
154054
4277. 2656
PARAMETER
1)1
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
200. 2170
16.5447
12. 10
0. 0001
XI
1
-5. 7000
6. 2440
-0. 71
0. 3673
X2
J
44.4427
6. 2527
7. 11
0. 0001
X3
I
43. 1747
6. 2440
6. 72
0. 0001
X4
1
-10.6236
6. 2440
-1. 70
0. 0775
XI *X1
J
2. 5604
10.1258
0. 25
0. 8012
X1*X2
1
0. 61524 317
3 4200
- 0. 18
0. 8502
X2*X2
1
10. 0406
10.1258
0. 77
0. 3200
X1*X3
J
1. 0140
3. 2700
0. 55
0. 5026
X2*X3
1
4. 0337
3. 4200
1. 18
0. 2457
X3*X3
1
-1. 7452
10.1258
~0. 17
0. 8641
XI *X4
1
1.0000
3. 2708
0. 55
0. 5853
X2*X4
J
1. 7770
3. 4200
0. 52
0.6060
X3*X4
1
-1. 7500
3. 2700
-0. 54
0. 5757
X4X4
1
-1 1627
10.1258
-0. 11
0. 7072
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PR OB
XI
5
7300. 0002
14 77. 7760
0. 35
0. 8818
X2
5
227031
4 5766. 2735
10. 74
0. 0001
X3
5
255375
51074. 7677
1 1. 74
0. 0001
X4
1
14 767. 7671
2773. 7730
0. 70
0. 6273
137


Fig. 8. Effect of rest period and grazing pressure upon grass yield (DM) for the wet
season of 1980.
Ln


1300
Fig.
Legume yield
kg ha~l
1300
1100
1100
900
Grazing pressure
3.3 (kg DM on offer/100 kg BW)
Days rest
1 Effect of rest period and grazing pressure upon legume yield (DM) for the
wet season of 1978.
00


95
grazing period and fertilizer level had no effect upon the yield
of weeds. The linear and interactions components of the model
accounted for 35 and 17% of the total variation, respectively
(Appendix Table 33).
The yield of weeds tended to decrease as the length of the rest
period increased from continuous grazing to 56 days of rest, and a
similar trend was observed as the grazing pressure increased from
8.3 to 1.6 kg DM on offer/100 kg BW. Each unit of rest (14 days)
decreased the yield of weeds by 50 kg ha ^ and for each unit
increase in grazing pressure the yield of weeds was decreased by
43 kg ha \ The analysis of variance is presented in Appendix
Table 33.
Yield of Weeds (DM) for the Dry Season of 1979
The yield of weeds during this season varied from 0 to 1730 kg
DM ha ^ (see Table 8).
The number of days grazing (X^), days rest and grazing
pressure (X^) affected the yield of weeds (P < 0.01). There were
also significant interactions between X^ x X^, X^ x X^, and X^ x X^.
The linear and the interaction components of the model accounted
for 35 and 30% of the total variation, respectively, while the
quadratic component represented only 6% (Appendix Table 34). The
yield of weeds decreased as the length of grazing period, length of
rest period, and the grazing pressure was increased.


LITERATURE CITED
Acosta-Soliz, M. 1967. Divisions fitogeograficas y formaciones
geobotnicas del Ecuador. Instituto Ecuatoriano de Ciencias
Naturales, Quito, Ecuador. 271 p.
Alarcon, E., and J. Lotero. 1970. Establecimiento, fertilizacin
y manejo de las principales gramineas y leguminosas forrajeras
en dos pisos trmicos de Colombia. Instituto Colombiano
Agropecuario (ICA), Colombia. 31 p. (mimeo).
Andrew, C. S. 1978. The effect of sulfur on the growth, sulfur
and nitrogen concentrations, and critical sulfur concentrations
of some tropical and temperatre pasture legumes. Aust. J. Agrie
Res. 28:807-809.
, and M. F. Robins. 1969. The effect of phosphorus on the
growth and chemical composition of some tropical pasture legumes
1. Growth and phosphorus critical percentages. Aust. J. Agrie.
Res. 20:665-674.
Berrezueta, L. G. 1975. Evaluacin de gramineas solas y asociadas
con leguminosas, sometidas a pastoreo, en Sando Domingo de
los Colorados. Tesis Ingeniero Agronomo. Universidad
Tcnica de Manabi, Portoviejo, Manabi, Ecuador. 42 p.
Betancourt, J. R. 1969. Respuesta del pasto guinea (Panicum maximum
Jacq.) a la fertilizacin nitrogenada. Tesis Ingeniero Agronomo
Universidad de Guayaquil, Guayaquil, Guayas, Ecuador. 40 p.
Blue, W. G., and L. EL Tergas. 1969. Dry season deterioration of
forage quality in the wet-dry tropic. Soil Crop Sci. Soc. Fla.
Proc. 29:224-227.
Bogdan, A. V. 1977. Tropical pasture and fodder plants. In Tropical
Agricultural Series. Longman Croup Limited, London. 475 p.
Bransby, D. I. 1975. A simple instrument for standing pasture
yield in situ. M. S. Thesis. University of Missouri, Columbia,
Missouri. 33 p.
Bryan, W. W., and T. R. Evans. 1973. Effects of soil fertilizers
and stocking rate on pastures and beef production on the Wallum
of Southeastern Queensland. I. Botanical composition and
chemical effects on plants and soils. Aust. J. Exp. Agrie.
Anim. Husb. 13:516-529.
161


117
growth. Rest periods of 28 days appeared to be most appropriate
for grass-legume yield and balance because the higher values for
legume percentage at the end of the wet season of 1980 were
recorded on pastures with 28 days rest followed by those which
had a rest period of 14 days or continuous grazing. Pastures
with 42 or 56 days rest showed zero or very low legume percentages.
These results are somewhat contrary to those found by Maraschin
(1975) and Serrao (1976) who reported that Desmodium intortum in
creased with the length of rest period when grown in association
with Cynodon dactylon. The results obtained by these authors can
probably be explained by the fact that the competitive grass is a
low-growing species which did not have nearly the destructive effect
upon the associated legume. Zapata (1981) and INIAP (1980) reported
that guineagrass-glycine pastures with more than six years main
tained an acceptable balance of both species with 25% or more of
legume when these pastures were subjected to a rotational grazing
system of 28 days grazing and 28 days rest.
In this experiment, the percentage legume was also responsive
to length of grazing (P < 0.05) for the dry season of 1978 and the
wet and dry season of 1979, but this effect disappeared during the
last season of 1980. Both of the legumes in this study, centro and
glycine, tended to increase as the length of grazing was increased
especially under the continuous grazing where due to frequent
defoliation, the grass-legume percentage was more stable but declined
during the entire course of the experiment.


116
upon percentage legume. No interactions among the experimental
variables were observed. The linear and quadratic components of
the model accounted for 46 and 27% of the total variation, respec
tively, while the interaction effects represented only 2% (Appendix
Table 45).
The percentage of legume tended to decrease up to a certain
length of rest period and then increased again. The analysis of
variance is presented in Appendix Table 45.
A summary of the effects of the experimental variables upon
percentage legume is discussed below.
The overall percentage of legume in the mixtures decreased
with time but the effect of the experimental variables appeared to
reach some stability by the end of the fifth grazing period. The
experimental variable having the greatest effect upon the percentage
legume was the length of rest period (X^) if we examine the trends
from the wet season of 1978 through the wet season of 1980.
During the wet season of 1980, it is clearly evident that the
length of rest period is the factor most influential in determining
the percentage legume. The highest percentage legume appears to
occur during rest periods of 14 to 28 days and under moderate condi
tions of grazing pressure. Both extremely short or long rest periods
were highly detrimental to the legume population with the greatest
destruction occurring for rest periods of 42 days or more. This
situation can probably best be explained by the aggressiveness of
both of the companion grasses especially during the wet season when
the environmental conditions were more favorable for maximum grass


in widespread use in the lowlands of Ecuador. They are 'common' for
grazing, 'Hybrid 534' and 'Mexican' for cutting.
5
Centrosema pubescens Benth, centro, a true tropical legume has
been described by Humphreys (1980) as a creeping, twining perennial
legume native of South America. Grof (1970) reported that the genus
Centrosema contains about 70 species growing naturally in the tropical
areas of Central and South America. Moore (1962) stated that this
tropical legume grows well in humid areas and must be considered as
a basic component of pastures under these conditions. Studies conducted
at Pichilingue, Ecuador, by Hudgens (1973), Chavez (1974), Rolando (1974)
and INIAP (1980) indicated that this legume performs well in associa
tion with guineagrass and it is also highly persistent under grazing
and produces large amounts of seed (Farfan, 1974). In the last few
years 132 native ecotypes of Centrosema have been selected and tested
in Ecuador. A few of these have been distributed to farmers and are
showing some advantages over the Australian commercial cultivars such
as higher DM yield, better tolerance to insects and disease, and
better adaptation to Ecuador conditions (INIAP, 1979).
Neonotonia wightii (R. Grah. ex Wight and Arn.) Lackey, glycine,
also known as perennial soybeans, has been described by Humphreys (1980)
as a perennial plant, slender, twining, and with long stems having
some capacity for rooting at the nodes.
Glycine was first introduced into Ecuador in 1966, but it was not
used in pastures until 1973 when it was shown to be one of the best
legumes for humid and subhumid areas. Three cultivars have been
distributed to farmers and these were selected for persistence,


Days rest
Fig. 7. Contours of grass yield (DM) as affected by length of rest
period and levels of grazing pressure in the wet season of
1980.
4>-


33
Table 1 presents the soil analysis on the experimental site
used in this research.
Legume-Grass Mixture
The legume-grass mixture selected for the experiment consisted
of Malawi glycine [Neonotonia wightii (R. Grah x Wightii and Arn.)
Lackey], commercial centro (Centrosema pubescens Benth.), guineagrass
(Panicum maximum Jacq.), and Hybrid 534 elephantgrass (Pennisetum
purpureum Schumach.).
Guineagrass and elephantgrass were considered because both are
very common and useful grass species in the littoral region of Ecua
dor, and also because of their high-yielding capacity, drought tolerance,
and ability to grow in mixture with some tropical legumes. Centro is
a native legume and may be the most widespread in the lowlands ex
tending from dry tropical forest to wet tropical forest (600-2500 mm
of rainfall). Glycine had been selected from previous experiments
as a very persistent and productive legume.
Experimental Variables
The experimental variables were
(1) Days grazing (X^1, 7, 14, 21, 28;
(2) Days rest (X^)0, 14, 28, 42, 56;
(3) Grazing pressure (X^)dry matter on offer per 100 kg body
weight (BW); and
(4)
Phosphorus (Po0_) levels (X.).
2 5 4


BIOGRAPHICAL SKETCH
Raul A. Santillan was born December 6, 1943, in Riobamba,
Chimborazo province, Ecuador, to Rigoberto and Itala Santillan.
From May 1962 to January 1968 he attended the Universidad de
Guayaquil, Ecuador, and received the Ingeniero Agronomo degree.
For six months he worked in the Programa Nacional del Banano and
in August of 1968 he joined the Instituto Nacional de Investiga
ciones Agropecuarias (INIAP), working in tropical pastures at
the Extacion Experimental Pichilingue. In 1972 he was awarded
a grant to receive a year's training in cattle production at the
Centro Internacional de Agricultural Tropical (CIAT) in Cali,
Colombia. In 1974 he received the Master of Science degree in
agriculture. In 1981 he continued his studies toward the degree
of Doctor of Philosophy in agronomy.
Raul A. Santillan is married to the former Maggie Moreno and
they have two daughters, Alexandra and Carolina. The author is
a member of the Asociacin Ecuatoriana de Produccin Animal and
Asociacin Latinoamericana de Produccin Animal.
170


element is necessary for development of enzymes related to N fixation,
nitrate reduction and legume nodulation (Andrew, 1978; Epstein, 1972).
Animal Productivity
In Pichilingue, Ecuador, Zapata (1981) observed that liveweight
gain of steers grazing on common guineagrass, improved Guineagrass
Brachiaria humidicola, and improved guineagrass-glycine pastures were,
in the order given, 0.645, 0.678, 0.741, and 0.857 kg animal ^ day
and an annual liveweight gain of 338.9, 421.9, 458.3, and 540.4 kg ha
respectively. The same author also noted a difference in liveweight
gain due to the breed of animal, being 0.598, 0.691, and 0.903 kg an ^
day ^ for red criollo, braham, and braham x holstein crosses, respec
tively. Similar results were found at 700 m altitude by Cowan et al.
(1974), but in this case milking cows were grazed on guineagrass-
glycine and kikuygrass (Pennisetum clandestinum) pastures without any
supplementation. Milk production averaged 9.06 and 12.54 kg cow ^ day
for jersey and holstein cows, respectively. In Australia, Grof and
Harding (1970) found that a guineagrass-centro pasture yielded 36%
more liveweight than Guineagrass alone, over a two year period. Hall
(1970) reported that animal production on unimproved native grass was
less than 9 kg ha ^ when compared to guineagrass-Siratro mixture which
yielded 112 kg liveweight ha \
In Costa Rica, Kretschmer (1971) reported that centro in mixtures
with guineagrass increased forage yield by 20% during the dry season
and by 30% during the wet season, when compared with grass alone.
Tergas (1976) reported that centro increased total forage production


83
The quadratic effect of rest period (Xgives a better rep
resentation of the legume yield which maximizes within the range of
14 and 28 days. As the length of the rest period is either decreased
or increased, the legume yield is decreased.
As the grazing pressure decreases, the amount of legume yield
increased by 37 kg ha ^. The analysis of variance is presented in
Appendix Table 30.
Observations and Summary of Legume Yield (DM)
Some very significant changes occurred in the legume population
of these experimental pastures, some of which were recorded and
subjected to analysis, whereas others were observations made over
the five seasons of this trial. Drastic changes occurred as
a result of the treatments imposed from the end of the wet season
of 1978 to the end of the wet season of 1980. Some of these
observations are recorded here.
When the experiment was initiated on May 12 of 1978, the
average legume yield for the period until June 30 was about 928 kg
ha During this initial period, only rest period showed any effect
(P < 0.01) upon legume yield. It is difficult to find any satisfac
tory explanation for the effect of any experimental variable over
this short period of time, but differences in the growth rate of
grasses and legumes or the short-term effects of insects like red
spider on centro could be a partial explanation (Fig. 9).
During the first dry season (July to December, 1978), the legume
yield declined from 640 kg ha which indicated a linear response to


Table 40. Analysis of variance, regression coefficients and probabilities for visual
estimation grass for the wet season of 1980.
RESPONSE MEAN
79.6997
ROUT MSE
0. 0092
R-SQUARE
0. 70924900
COEF OF VARIATION
0 11052931
REGRESSION
R!
TYPE I SR
R-SQUARE
F-RATIO
PROD
LINEAR
4
19927.7405
0. 64 74
64 20
0 0001
QUADRATIC
4
2173 1120
0. 0712
7 07
0. 0003
CROSSPRODUCT

5868.9576
0. 1907
12. 60
0. 0001
TOTAL REGRESS
14
27990 0101
0. 9072
25. 76
0. 0001
RESIDUAL
I)F
SS
MEAN SQUARE
F-RATJO
PR OB
LACK OF FIT
26
2674. 2102
102. 8542
8. 612
0. 0006
PURE ERROR
10
1 19. 4346
11.9435
TOTAL ERROR
36
2793. 6447
77. 6012
PARAMETER
nr-
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
l
78. 3931
2. 2200
35. 17
0. OOOl
XI
i
O 67629662
0.84083794
0. 80
0. 4265
X2
i
12.0220
0.84201562
14. 28
0. 0001
X3
l
7. 6069
0.84083794
9. 14
0. 0001
X4
i
-1. 5528
0.84083794
-1. 85
0. 0730
X1*X1
i
- O. 62127855
1. 3636
-0. 46
0. 6514
X1#X2
i
-0.15065786
0.46054591
-0. 33
0. 7455
X 2 & X 2
l
4. 5662
1. 3636
3. 35
0. 0019
X1#X3
i
- 0. 37193049
0.44045785
-0. 84
0. 4040
X2#X3
i
-3. 91 10
0.46054591
-8. 49
0. 0001
X3*X3
l
-6. 4338
1. 3636
-4 72
0. 0001
X1*X4
i
- 0 09173611
0.44045785
-0. 21
0. 8362
X2*X4
i
0.74262532
0.46054591
1. 61
0. 1156
X3#X4
i
0.06593056
0 44045785
0. 15
0. 8818
X4*X4
i
1. 4412
1 3636
1 06
0. 2976
FACTOR
F)F
SS
MEAN SQUARE
F-RATIO
PROB
X1
5
125. 7336
25. 1467
0. 32
0. 0952
X2
5
22755. 8100
4551.1636
58. 65
0. 0001
X3
5
1 1147. 1645
2229. 4329
28. 73
0. 0001
X4
5
4 54. 1268
90. 8254
1. 17
0. 3426


Fig.
5. Field plan of the experimental pastures.


Table 32. Analysi^ of variance, regression coefficients and probabilities of weeds
(g DM/m ) for the dry season of 1978.
RESPONSE MEAN
3. 0263
ROOT MSE
3. 7770
R-SQUARE
0. 42106717
COEF OF VARIATION
1 0376
REGRESSION
Dl
TYPE I OS
R SQUARE
F-RATIO
PR OB
LINEAR
4
163. 2320
0. 1657
2. 58
0. 0537
QUADRATIC
4
217.6507
0. 2212
3. 44
0. 0177
CROSSPRODUCT
6
33 4224
0. 0340
0. 35
0. 7040
TOTAL REGRESS
14
414.3050
0 421 1
1. 87
0. 0653
RESIDUAL
l)f-
OS
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
547. 2373
21.0477
7. 400
0. 0004
PURE ERROR
10
22. 3715
2. 2371
TOTAL ERROR
36
567. 6300
15. 0231
PARAMETER
nr
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
5. 5717
1. 0060
5. 54
0. 0001
XI
1
0. 20576412
0.37760477
0. 75
0. 4563
X2
1
-0.75512573
0.38021678
-2. 51
0. 0166
X3
1
0.33527535
0.37768477
0. 08
0. 3031
X4
1
0.16343657
0.37768477
0. 43
0. 6674
X1*X1
1
1. 1156
0. 61572855
1. 01
0. 0784
XJUX2
1
-0.12033204
0.20776204
-0 58
0. 5664
X2*X2
1
-1.1671
0. 61572855
-1. 70
0. 0661
X1*X3
1
0.24603807
0.17887116
1. 24
0. 2241
X2*X3
1
0.07470037
0.20776204
0. 36
0.7208
X3*X3
1
0.88466860
0.61572055
1. 44
0. 1574
X1 X4
1
-0 02768887
0.17887116
-0. 15
0. 8822
X2*X4
1
-0. 04 521077
0.20776204
-0. 22
0. 8271
X3#X4
1
0.04352222
0.17887116
0. 22
0. 8280
X4*X4
1
-1.4476
0.61572855
-2. 35
0. 0243
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
07. 7077
17. 5415
1. 11
0. 3730
X2
5
167. 8434
33. 7607
2. 15
0. 0820
X3
5
76. 5036
15. 3167
0. 77
0. 4503
X4
5
71. 6502
IB.3300
1. 16
0. 3404


25
the legume percentage was 8.7, 10.5, and 15.6 for the 0.5, 1.0, and
1.5 animals ha \ respectively, when it was measured with the point
quadrat method.
Jensen and Schumacher (1970) observed that the percentage of
botanical species is not only affected by grazing animals, but also
by environmental conditions such as season, precipitation, rain distri
bution, temperature, and solar radiation. Taking into consideration
most of the above factors, Tothill (1978) noted that if the primary
objective of the investigation is to obtain an estimate of botanical
composition of pastures in assessing animal production, weight of
species is the most suitable value to measure. If rainfall interception
or photosynthesis are under study, then cover may be the more appropri
ate parameter. He also mentioned that the importance of this distinc
tion is that number, weight and cover measurements are comparable in
time and space, but are independent of the mode of sampling since they
are measured directly and expressed in relation to a unit area. Shaw
and Bryan (1976) mentioned that the proportion of species on a weight
basis is generally the most useful where the main interest is in pasture
production and where samples are cut for yield determination, and botan
ical composition can be determined by hand-separating the sample into
component species. They also added that this is the most precise and
satisfactory method for yield and botanical composition determinations.
Response Surface Methodology
Littell and Mott (1975) indicated that the purpose of response
surface methodology is to estimate the functional relationship-between
a response variable such as yield and an experimental variable or


Fig. 15. Effect of rest period and grazing pressure upon grass percentage
for the wet season of 1978.
108


depending on the specified dry matter on offer and days of grazing.
In order to estimate the grazing pressure, the following formula
was used:
45
G/P
(A r + gd) S
RNd
where G/P = grazing pressure in terms of 100 kg BW,
A = available forage before grazing,
r = residue for a specified grazing pressure,
g = growth rate-preceding rest period,
d = number of days for grazing period,
2
S = size of the experimental unit in m ,
R = kg dry matter offered/kg body weight/day, and
N = kg body weight/pasture/day.
A double sampling technique was employed. As soon as possible,
before and after each grazing (during the first 3 1/2 months only),
2
15 areas measuring 1.0 m were randomly selected. From September 30,
area measurements of the same size were taken, corresponding to the
2
circular frame of a forage disk meter (one m ) similar to the one
described by Bransby (1975) and used in double-sampling measurements.
In each sampling unit the percent dry matter yield of each component
of the mixture was visually estimated, followed by an estimate of the
dry matter yield in situ. The disk meter was lowered on the forage
and after a settling time of approximately 5 to 10 seconds, the disk
height (in centimeters) was read off a graduated scale mounted on the
center shaft of the disk meter and recorded.
During the first three and one-half months, from the 15 sampling
units, three were randomly selected and clipped at ground level for


42. Analysis of variance, regression coefficients and probabilities for visual
estimation legume
for the
dry season of
1978.
RESPONSE MEAN
24.7270
ROOT MSE
0. 6336
R-SQUARE
0 67207201)
COEF OF VARIATION
0 22703300
REGRESSION
nr
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
2304.0417
0. 6177
18 15
0. 0001
QUADRATIC
4
1)0 1010
0. 0135
0. 40
0 0108
CROSSPRODUCT
6
217.7667
0. 0571
1. 15
0. 3521
TOTAL REGRESS
J 4
2073. 7606
0. 6726
5. 77
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
048 0003
32 6176
1 108
0 4 558
PURE ERROR
10
274 0023
27. 4 502
TOTAL ERROR
36
1142. 0606
31 7378
PARAMETER
DI
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
26. 1120
1. 4248
18. 33
0.0001
XI
1
1. 0483
0.53773255
1. 75
0. 0571
X2
1
-4. 1701
0.53848570
-7. 74
0. 0001
X3
1
-1. 4325
0. 53773255
-2. 66
0. 0115
X 4
J
0.23160654
0.53773255
0. 43
0. 6672
X1*X1
I
0. 53540682
0.87203152
-0. 61
0. 5431
X1*X2
1
0.01017712
0.27452825
-0. 03
0. 7726
X2*X2
1 -
0.71576238
0.87203152
-0. 82
0. 4170
X1#X3
I
0. 37427583
0.28168154
1. 33
0. 1723
X2*X3
J
0.47552347
0.27452825
1 61
0. 1151
X3*X3
1
O. 50452373
0.87203152
0. 58
0. 5665
X1*X4
1
0.22137306
0.28168154
-0. 77
0. 4370
X2*X4
1
0.40760255
0. 27452825
-1 38
0. 1747
X3*X4
1
0.03767028
0. 28168154
-0. 13
0. 8743
X4*X4
1
0.71387873
0. 87203152
0 82
0. 4 184
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
218. 1588
43.6318
1. 37
0. 2567
X2
5
2083. 4265
416.6853
13. 13
0. 0001
X3
5
316 7246
63. 3447
2. 00
0. 1028
X4
5
102. 2545
20. 4507
0. 64
0. 6674


APPENDIX


Dedicated to my wife, Maggie
my daughters, Alexandra and
Carolina, my father and to
the memory of my mother


30
Fig. 20. Effect of rest period and grazing pressure upon legume percentage
for the wet season of 1980.
121


Table 43. Analysis of variance, regression coefficients and probabilities for visual
estimation legume
for the
wet season of
1979.
RESPONSE MEAN
J;>. 7705
ROOT MSE
3. I 4 57
R-SGUARE
0.73420710
COEF OF VARIATION
0 32163041
REGRESSION
in
TYPE I SS
R SQUARE
F-RATIO
PR OB
LINEAR
4
2300. 7410
0. 6157
22. 53
0. 0001
QUADRATIC
4
163. 6376
0. 0427
1. 36
0 2049
CROSBPRODUCT
A
371.0273
0. 0750
2. 34
0 0520
TOTAL REGRESS
14
2926. 2207
0. 7543
7. 07
0. 0001
RESIDUAL
1)1
SO
MEAN SQUARE
F-RATIO
PR OB
LACK OF FIT
26
006. 2337
31. 0091
2. HO
0. 1003
PURE ERROR
10
146.9001
14. 6700
TOTAL ERROR
36
733. 2230
26. 4704
PARAMETER
nr
ESTIMATE
STD DEV
T-RAT10
PR OB
INTERCEPT
i
19. 3260
1. 3014
15. 00
0. 0001
XI
i
O. 71147042
0 471 161.31
1. 06
0. 0717
X2
i
-4. 2033
0 47184723
-8. 33
0. 0001
X3
l
0.20111020
0.47116131
-0. 41
0. 6046
X 4
i
0.44320033
0. 49116131
0. 91
0. 3707
XI *X1
i
-0 07376003
0. 77630775
-0. 07
0. 7267
XI *X2
i
0.26777/73
0.26702013
1. 00
0 3258
X2*X2
i
-1. 0327
0.77630773
-132
0. 174 3
X1#X3
i
0.47373417
0.23720603
1 85
0. 0727
X2*X3
i
0 62321641
0.26702013
2. 32
0. 0263
X3X3
1
0.77002240
0. 77630775
0. 77
0. 3274
XI X4
l
-0 34779383
0.25720603
-1. 33
0. 1047
X2*X4
l
-0.40614503
0.26702013
-1. 51
0. 1378
X3*X4
i
0.10216230
0. 25720603
~0. 40
0. 6737
X4#X4
i
-0.60272752
0.77650773
-0. 76
0. 4540
FACTOR
l)F
SS
MEAN SQUARE
F-RATIO
PROD
XI
3
270. 04 51
57. 7670
2. 26
0.0675
X2
3
2232. 3011
446.4602
16. 06
0. 0001
X3
3
264. 7032
52 7766
2. 00
0. 1017
X4
3
134. 3330
26 0700
1 01
0.4233


Table 30. Analysis of variance, regression coefficients and probabilities for legume
yields (g DM/m ) for the wet season of 1980.
RESPONSE MEAN
27. 4 J 07
ROOT MSE
17. 4201
R-SQUARE
0. 74733003
COEF OF VARIATION
0. 59237260
REGRESSION
14
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
4
17044. 54 till
0 4547
16. 33
0 OOOl
QUADRATIC
4
10312. 2713
0. 24 10
8. 65
0 0001
CROSSPRODUCT
6
2327. 7577
0. 0534
1. 28
0. 2716
TOTAL REGRESS
14
32686. 7767
0. 7473
7. 67
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
10272. 0712
375. 0004
5. 764
0. 0027
PURE ERROR
JO
662. 4674
66.2467
TOTAL ERROR
36
10734. 3507
303. 7377
PARAMETER
ni
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
47. 8002
4.4070
1 1. 30
0. 0001
XI
J
1.0300
1. 6635
0. 62
0. 5366
X2
1
-11.0174
1. 6638
-6. 61
0. OOOl
X3
]
3. 7233
1. 6635
2. 24
0. 0314
X4
1
2. 4607
1. 6635
1. 48
0. 1463
X1*X1
J
3. 2727
2. 6777
1. 22
0. 2302
X1#X2
1
-0.41436730
0.71114597
-0. 45
0. 6320
X2*X2
1
-10.7064
2. 6777
-4. 00
0. 0003
X1#X3
1
0.11367063
0.87140367
0. 13
0. 0767
X2#X3
1
-1 6412
0.91114577
- 1. 80
0. 0800
X3*X3
1
3. 3467
2. 6777
1. 24
0 2228
XI X4
1
0 51726170
0.87140367
0. 57
0. 5565
X2*X4
1
-1.4774
0 71114577
-1. 62
0. 1136
X3*X4
1
0 76270476
0.87140367
1. 11
0. 2765
X4*X4
1
-3 0531
2. 6777
-1. 13
0. 2632
FACTOR
nr
SB
MEAN SQUARE
F-RATIO
PROB
XI
3
707. 6312
141. 5262
0. 47
0. 7770
X2
5
20528. 3677
4105. 6735
13. 52
0. OOOl
X3
3
2003.7424
360. 7405
1. 85
0. 1205
X4
0
2004. 5205
400.7037
1. 32
0. 2777
-N
Ln


Table 2. Modified central composite non-rotatable design with four experimental (X) variables,
at five levels each, and 41 design points.
Treatments
Reps
Days in
cycle
No.
X1
Coded
X2 X3
X4
Days
Grazing
(x1)
Days
Rest
(x2)
Grazing
Pressure
(X3) % BWR
Fertilizer
Level 1
(X^) kg ha
1
-1
-1
-1
-1
7
14
3.3
100
1
21
2
1
-1
-1
-1
21
14
3.3
100
1
35
3
-1
1
-1
-1
7
42
3.3
100
1
49
4
1
1
-1
-1
21
42
3.3
100
1
63
5
-1
-1
1
-1
7
14
6.6
100
1
21
6
1
-1
1
-1
21
14
6.6
100
1
35
7
-1
1
1
-1
7
42
6.6
100
1
49
8
1
1
1
-1
21
42
6.6
100
1
63
9
-1
-1
-1
1
7
14
3.3
300
1
21
10
1
-1
-1
1
21
14
3.3
300
1
35
11
-1
1
-1
1
7
42
3.3
300
1
49
12
1
1
-1
1
21
42
3.3
300
1
63
13
-1
-1
1
1
7
14
6.6
300
1
21
14
1
-1
1
1
21
14
6.6
300
1
35
15
-1
1
1
1
7
42
6.6
300
1
49
16
1
1
1
1
21
42
6.6
300
1
63
17
-2
-2
-2
-2
1
0
1.6
0
1
Cont
18
2
-2
-2
-2
28
0
1.6
0
1
Cont
19
-2
2
-2
-2
1
56
1.6
0
2
56
20
2
2
-2
-2
28
56
1.6
0
2
84
21
-2
-2
2
-2
1
0
8.3
0
1
Cont
22
2
-2
2
-2
28
0
8.3
0
1
Cont
23
-2
2
2
-2
1
56
8.3
0
2
56
24
2
2
2
-2
28
56
8.3
0
2
84
u>
ON


CHAPTER V
SUMMARY AND CONCLUSIONS
A grazing trial was conducted at the Estacin Experimental
Tropical Pichilingue, belonging to Instituto Nacional de Investi
gaciones Agropecuarias (INIAP) and located 7 km from Canton
Quevedo, at Io 06' S Lat. and 79 21' W Long. The soil is classified
as Torripsamments. The experiment covered an area of 7.3 ha, which
was subdivided into 51 individual pastures, each large enough to be
grazed by at least one animal during the designated grazing period.
2
Pasture sizes ranged from 500 to 4000 m ; the larger area provided
for the continuous grazing treatments.
Land preparation for the experimental area began in May of
1977, after existing vegetation was eliminated and plowed under.
In early October of the same year, a mixture of glycine and centro
was sown at the rates of 3.0 and 6.0 kg ha ^ in rows spaced 1.4 m
apart. Two weeks later guineagrass and elephantgrass were planted
vegetatively between the rows of the legumes.
Four experimental variables were studied at each of five levels,
namely, days grazing: 1, 7, 14, 21, and 28 days; days rest: 0, 14,
28, 42, and 56 days; grazing pressure: 1.6, 3.3, 5.0, 6.6, and 8.3
kg dry matter on offer/100 kg BW; and levels of phosphorus fertili
zation: 0, 100, 200, 300, and 400 kg ha ^ of superphosphate.
A modified non-rotatable central composite response surface
design was used which included 41 selected treatment combinations.
Certain treatments were replicated which accounted for the 51 experi
mental units.
122


Table 5. Available forage (DM) by year, season, and treatment combination.
Treatments
D/Gt
D/Rf
G/Pf
Ft -i
kg ha
1978
1978
1979
1979
1980
No.
(x1)
(x2)
(X ) % BW
Reps
W/S
D/S
W/S
D/S
W/S
1
7
14
3. 3
100
1
1640
2270
kg ha
2230
2080
1550
2
21
14
3.3
100
1
1950
2720
3120
2380
3150
3
7
42
3.3
100
1
2590
2760
3850
2650
3140
4
21
42
3.3
100
1
1830
2510
3360
1800
3920
5
7
14
6.6
100
1
2330
3600
4660
4210
4070
6
21
14
6.6
100
1
2130
2990
3070
2870
2750
7
7
42
6.6
100
1
2620
4020
4240
3050
3970
8
21
42
6.6
100
1
1910
4640
5810
4720
5720
9
7
14
3.3
300
1
3150
1860
3100
2430
2540
10
21
14
3.3
300
1
2730
2810
4230
3030
3540
11
7
42
3.3
300
1
2300
2820
3610
2710
2830
12
21
42
3.3
300
1
1520
2230
4440
2470
5940
13
7
14
6.6
300
1
2290
2620
2840
3000
2520
14
21
14
6.6
300
1
2960
2870
2600
2260
2320
15
7
42
6.6
300
1
1790
4370
6130
5980
7230
16
21
42
6.6
300
1
3030
3000
3680
2720
3680
17
1
0
1.6
0
1
1650
1200
830
690
650
18
28
0
1.6
0
1
1990
1070
1380
760
560
19
1
56
1.6
0
2
3720
3600
5750
36 70
7510
20
28
56
1.6
0
2
2390
2260
4040
2760
3820
21
1
0
8.3
0
1
2280
3500
3930
4640
6120
22
28
0
8.3
0
1
1800
3530
3810
3940
4840
23
1
56
8.3
0
2
3380
4790
6620
3820
5450
24
28
56
8.3
0
2
3400
5700
5900
4440
6620


Temperature (C)
Fig. 3. Temperature recorded at Estacin Experimental Pichilingue during the
period 1978-1980.
u>


Table 2.continued
Treatments
Reps
Days in
cycle
No.
X1
Coded
X2 X3
X4
Days
Grazing
(x1)
Days
Rest
(x2)
Grazing
Pressure
(X3) % BWR
Fertilizer
Level ..
(XA) kg ha 1
25
-2
-2
-2
2
1
0
1.6
400
1
Cont
26
2
-2
-2
2
28
0
1.6
400
1
Cont
27
-2
2
-2
2
1
56
1.6
400
2
56
28
2
2
-2
2
28
56
1.6
400
2
84
29
-2
-2
2
2
1
0
8.3
400
1
Cont
30
2
-2
2
2
28
0
8.3
400
1
Cont
31
-2
2
2
2
1
56
8.3
400
2
56
32
2
2
2
2
28
56
8.3
400
2
84
33
-2
0
0
0
1
28
5.0
200
1
28
34
2
0
0
0
28
28
5.0
200
1
56
35
0
-2
0
0
14
0
5.0
200
1
Cont
36
0
2
0
0
14
56
5.0
200
1
70
37
0
0
-2
0
14
28
1.6
200
1
56
38
0
0
2
0
14
28
8.3
200
1
56
39
0
0
0
-2
14
28
5.0
0
1
56
40
0
0
0
2
14
28
5.0
400
1
56
41
0
0
0
0
14
28
5.0
200
3
56
LO


Table 4. Aerial biomass production (DM) by year, season, and treatment.
Treatments
D/Gt Wit Wit fTT 1978 1978 1979 1979 1980
No. (X1) (X2) (X3) % BW kg ha 1 Reps W/S D/S W/S D/S W/st
kg ha
1
7
14
3.3
100
1
1790
2410
3090
2870
2470
2
21
14
3.3
100
1
1960
2730
3210
2530
3360
3
7
42
3.3
100
1
2630
2760
3850
2650
3140
4
21
42
3.3
100
1
1890
2530
3490
1860
3940
5
7
14
6.6
100
1
2360
3690
4830
4460
4500
6
21
14
6.6
100
1
2170
3050
3120
2880
2760
7
7
42
6.6
100
1
2710
4020
4240
3050
3980
8
21
42
6.6
100
1
1940
4640
5830
4720
5770
9
7
14
3.3
100
1
3150
1910
3320
2530
2970
10
21
14
3.3
300
1
2850
2920
4440
3230
4020
11
7
42
3.3
300
1
2300
2880
3610
2710
3830
12
21
42
3.3
300
1
1590
2230
4430
2500
5740
13
7
14
6.6
300
1
2290
2620
2850
3010
2550
14
21
14
6.6
300
1
3150
2940
2690
2310
2410
15
7
42
6.6
300
1
1870
4400
6130
5980
7240
16
21
42
6.6
300
1
3050
3100
3740
2780
3680
17
1
0
1.6
0
1
1750
1240
1210
1480
1920
18
28
0
1.6
0
1
1990
1120
1760
1270
1460
19
1
56
1.6
0
2
3750
3620
5760
3670
7510
20
28
56
1.6
0
2
2430
2260
4060
2790
3820
21
1
0
8.3
0
1
2280
3540
3940
4090
6270
22
28
0
8.3
0
1
1800
3580
3910
3960
4870
23
1
56
8.3
0
2
3380
4800
6630
3850
5450
24
28
56
8.3
0
2
3460
5750
5940
4450
6620


PAGE
Effect of Lengths of Rest Period and Levels of
Grazing Pressure Upon Legume Yield (DM) 76
Effect of Lengths of Rest Period and Levels of
Grazing Pressure on the Yield of Weeds (DM).... 91
Visual Estimation of Forage Components 103
CHAPTER V SUMMARY AND CONCLUSIONS 122
APPENDIX 126
LITERATURE CITED 161
BIOGRAPHICAL SKETCH 170
vi


165
Kretschmer, A. E. 1971. New legumes for the Latin American
tropics. Agrie. Res. Center, Fort Pierce, Florida Rpt. P. 11-31.
. 1974. Distribution, introduction and evaluation of tropi
cal pasture species. Agrie. Res. Center Rpt. FL-1974-8, Fort
Pierce, Florida. P. 4-10.
Littell, R. C., and G. 0. Mott. 1975. Computer assisted design
and analysis of response surface experiments in agronomy. Soil
Crop Sci. Soc. Fla. Proc. 34:94-97.
Ludlow, M. M., and G. L. Wilson. 1970. Studies on the productivity
of tropical pasture plants. II. Growth analysis, photosynthesis,
and respiration of 20 species of grasses and legumes in a controlled
environment. Aust. J. Agrie. Res. 21:183-194.
Manhaes, S. S., and J. Dobereiner. 1968. Effeito do fosforo, temp
eratura, a unidade do solo na nodulacao e no desmvolvimiento de
dua variedade de soya perene (Glycine wightii). Pesquisa Agro
pecuaria Brasileira. 3:215-221.
Maraschin, G. E. 1975. Response of a complex tropical pasture mixture
to different grazing management systems. Ph.D. Dissertation.
University of Florida, Gainesville, Florida. 148 p.
Matches, A. G. 1970. Pasture research methods. Proc. Nat. Conf.
Forage Quality, Evaluation and Utilization, University of
Nebraska, Lincoln, Nebraska.
Mcllroy, R. J. 1972. Introduction al cultivo de los pastos tropi
cales. (Trad, de al 2a. Ed. Inglesa), UTEHA, Mexico.
168 p.
Mclvor, J. G., R. J. Jones, C. J. Gardiner, and W. H. Winter. 1981.
Development of legume based pastures for beef production in dry
tropical areas of Northern Australia. XIV Int. Grassld. Cong.
Proc., University of Kentucky, Lexington, Kentucky. 491 p.
McMeeckan, C. P. 1956. Grazing management and animal production.
VII Int. Grassld. Cong. Proc., New Zealand. 145:156.
Medina, K. D. 1969. Estudio de la disponibilidad de hierro, man
ganeso y zinc de la zona de Quevedo, Ecuador. Tesis Ingeniero
Agronomy, Universidad de Guayaquil, Guayaquil, Guayos, Ecuador.
95 p.
Milford, R., and K. P. Haydock. 1965. The nutritive value of protein
in subtropical pasture species grown in southeast Queensland.
Aust. J. Exp. Agrie. Anim. Husb. 5:13-17.
Minson, D. J., and R. Milford. 1967. The voluntary intake and
digestibility of diets containing different proportions of le
gumes and mature pangolagrass (Digitaria decumbens). Aust. J.
Exp. Agrie. Anim. Husb. 7:546-551.


Fig. 12. Effect of rest period and grazing pressure upon yield of weed (DM) for the
wet season of 1978.


CHAPTER IV
RESULTS AND DISCUSSION
Only two of the four experimental variables included in this
experiment, namely, lengths of rest period and levels of grazing
pressure are discussed in this section. The other two variables,
days grazing and fertilizer level had negligible effects upon the
response of the pasture sward. Each of the response variables will
be discussed in a separate section beginning in the wet season of
1978 and ending in the wet season of 1980.
Effect of Lengths of Rest Periods and Levels of Grazing
Pressure on Aerial Biomass (DM)
The effect of the lengths of rest period and levels of grazing
pressure on the aerial biomass is presented for each of the five
seasons in Table 4. The total biomass is the average amount of dry
matter present before each grazing period for the rotational grazing
treatment combinations and for each grazing period of 56 days for
continuous grazing.
Biomass Production (DM) for the Wet Season of 1978
During the first wet season only lengths of rest periods indicated
an effect (P < 0.01) on the aerial biomass produced (Appendix Table 11).
The linear components of the model accounted for 20% of the total var
iation, while the quadratic effects and interactions represented only
4 and 13% of the total variation, respectively (Appendix Table 11).
48


15
16
17
18
19
20
21
22
23
24
25
26
PAGE
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m2) for the
wet season of 1980 130
Analysis of variance, regression coefficients and
probabilities for available forage (g DM/nT) for the
wet season of 1978 131
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m^) for the
dry season of 19 78 132
Analysis of variance, regression coefficients and
probabilities for available forage (g DM/m^) for the
wet season of 1979 133
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m ) for the
dry season of 1979 134
Analysis of variance, regression coefficients and
probabilities for available forage (g DM/m2) for the
wet season of 1980 135
Analysis of variance, regression coefficients and
probabilities for grass yield (g DM/m^) for the wet
season of 1978 136
Analysis of variance, regression coefficients and
probabilities for grass yield (g DM/m2) for the dry
season of 1978 137
Analysis of variance, regression coefficients and
probabilities for grass yield (g DM/m^) for the wet
season of 1979 138
Analysis of variance, regression coefficients and
probabilities for grass yield (g DM/m2) for the dry
season of 1979 139
Analysis of variance, regression coefficients and
probabilities for grass yield (g DM/m^) for the wet
season of 1980 140
Analysis of variance, regression coefficients and
probabilities for legume yield (g DM/m2) for the
wet season of 1978 141
viii


Table 5.continued.
Treatments
1978
1978
1979
1979
1980
D/Gf
(xx)
D/Rf
(x2)
G/Pf
(X ) % BW
n -i
kg ha
No.
Reps
w/st
D/S
w/st
D/St
w/st
25
1
0
1.6
400
1
1990
1130
kg ha
950
400
410
26
28
0
1.6
400
1
1810
790
990
840
340
27
1
56
1.6
400
2
2320
2450
4260
2080
3360
28
28
56
1.6
400
2
2700
2630
4090
2560
4030
29
1
0
8.3
400
1
1690
2610
3140
3620
5530
30
28
0
8.3
400
1
2350
2420
3530
3110
4040
31
1
56
8.3
400
2
2410
4840
6870
3660
7510
32
28
56
8.3
400
2
2540
4880
6330
5830
6160
33
1
28
5.0
200
1
2460
2660
3070
2760
2890
34
28
28
5.0
200
1
1420
2280
3160
2650
3220
35
14
0
5.0
200
1
2640
1700
1790
1650
1820
36
14
56
5.0
200
1
3130
3710
4410
1300
4030
37
14
28
1.6
200
1
1760
1850
1970
1860
1360
38
14
28
8.3
200
1
2460
2730
3550
3170
3470
39
14
28
5.0
0
1
2860
2550
3130
2720
3150
40
14
28
5.0
400
1
1920
2070
3390
2800
2500
41
14
28
5.0
200
3
2290
2340
2950
2380
2800
tD/G = days grazing, D/R = days rest, G/P = grazing pressure, and F = fertilizer.
W/S = wet season, D/S = dry season.


44
animals were 16 months old and their average weight was 268 kg.
Every 28 days all animals were removed from the pastures and from
the reserve pasture. They were taken to the main corral for tick
treatments and every 56 days they were weighed in order to have a
basis for estimating stocking rates for the next grazing cycle.
At the end of the first experimental year, the pool of animals
was removed from all the pastures and showed average weights of 428 kg.
A new pool of younger animals replaced the first group for the second
year, having an average weight of 302 kg. The animals for the second
experimental year were from the same herds as those of the first year
and were 18 months old.
Pasture Measurements
Dry Matter Determination Before and After Grazing
Dry matter ha ^ estimates before (on offer) and after (residual)
each grazing period were made in order to apply the required grazing
pressure and to determine the net dry matter production.
Stocking rate was determined for grazing pressure-grazing period
combination on the basis of the total dry matter available before
each grazing plus the estimated growth rate during the grazing period.
Growth rate during the grazing period was assumed to be the same as
that of the previous rest period. The accuracy of this technique for
determining stocking rate was checked, using the residual dry matter
left on the field after each grazing period. Grazing pressure was
also based on visual observations of the amount of dry matter during
each grazing period, in order to add or remove animals from pastures


63
Available Forage for the Wet Season of 1980 (DM)
The average amount of available forage (DM) for each treatment
combination during the wet season of 1980 is presented in Table 5.
The available forage varied from 340 kg DM ha ^ to 7510 kg DM
ha \ corresponding to treatments: 26 [28 days grazing, 0 days rest
(continuous grazing), 1.6 kg DM on offer/100 kg BW, and 400 kg ha ^
of superphosphate], and 31 (1 day grazing, 56 days rest, 8.3 kg DM
on offer/100 kg BW, and 400 kg ha ^ of superphosphate), respectively.
Only the two experimental variables days rest (X^) and grazing
pressure (X^) were found to be affecting the available forage (P <0.01).
The other two variables and the interactions were not significant.
The linear components of the model accounted for 57% of the
total variation, while the quadratic effects and interactions repre
sented only 3 and 4% of the total variation, respectively (Appendix
Table 20). The other experimental variables, X^ and X^, had no
significant effect upon the available forage.
As the rest period was increased by 14 days there was an
increase of 760 kg ha ^ of available forage. On the other hand, each
unit (1.6 kg DM) decrease in grazing pressure resulted in an increase
of 650 kg ha ^ of available forage. The analysis of variance for the
wet season of 1980 is presented in Appendix Table 20.
Summary of Available Forage (DM)
Available forage may be considered a more important response
variable than aerial biomass since available forage takes into con
sideration only the components of the vegetation considered edible


19
all treatments" (1960, p. 601-602). Serrao (1976) suggested that
variable or fixed stocking rates can be used in continuous or rota
tional grazing systems. He concludes that the put-and-take system is
more appropriate when plant and animal relationships are to be measured.
Grazing Systems
Continuous Grazing
Heady (1970) noted there is much confusion in the definition of
grazing systems which are used for describing the day to day provision
of livestock feed from a wide variety of sources such as conserved
forage and by direct use of pasture by grazing animals. He also
defined continuous grazing as a grazing system in which the animals
have unrestricted access to any part of the pasture through a grazing
period, which can be a season, a year or more. Spedding (1965) sug
gested a pasture under a continuous grazing system could be called
correctly grazed when the amount of removed forage by the grazing
animals was equal to the amount of forage daily yield. Continuous
grazing is the most commonly used system in the tropics, especially
on vast areas, far away from the main consumer centers and in many
cases areas without any suitable highway system (Chaverra, 1979).
Rotational Grazing
A rotational grazing system is defined by Heady (1970) and
Heath (1978) as a system in which the animals are allowed to graze
the pasture for variable periods of time, normally with a heavy stocking
rate, during which the pasture is grazed and ungrazed several times
during a grazing season or year.


166
Moore, A. W. 1962. Symbiotic nitrogen fixation in a grazed tropical
grass-legume pasture. Nature. 185:638-640.
Mott, G. 0. 1960. Grazing pressure and the management of pasture
production. VIII Int. Grassld. Cong. Proc., University of Reading,
England. P. 601-611.
. 1973. Evaluating forage production. j[n M. E. Heath, D. E.
Metcalfe, and R. F. Barnes (eds.) Forages. Iowa State Univer
sity Press, Ames, Iowa. P. 126-147.
. 1974. Nutrient recycling in pastures. In 0. A. Mays (ed.)
Forage fertilization. The American Society of Agronomy, Madison,
Wisconsin. P. 47-54.
. 1977. Grazing management of tropical legume-grass associa
tions. 11th Ann. Conf. on Livestock and Poultry in Latin America.
University of Florida, Gainesville, Florida. A-35.
. 1981. Potential productivity of temperate and tropical
grassland systems. XIV Int. Grassld. Cong. Proc., University
of Kentucky, Lexington, Kentucky. P. 35-41.
, and H. L. Lucas. 1952. The design, conduct and interpre
tation of grazing trials on cultivated and improved pastures.
VI Int. Grassld. Cong. Proc., Pennsylvania State University,
State College, Pennsylvania. P. 1380-1385.
, and J. E. Moore. 1970. Forage evaluation in perspective.
Proc. Nat. Cong. Forage Quality, Evaluation and Utilization.
University of Nebraska, Lincoln, Nebraska.
Neme, N. A., and L. A. C. Lovadini. 1967. Efeito de adubos fosfatados
e calcario na producao de forragem de soja perene (Glycine
javanica L.) em "terra de Cerrado." Bragantia. 26:365-371.
Norman, M. J. T. 1970. Relationships between liveweight gain of
grazing beef steers and availability of Townsville lucerne. XI
Int. Grassld. Cong. Proc., Surfer's Paradise, Australia. P. 829-832.
Ozanne, P. G., and T. C. Shaw. 1976. Phosphate sorption by soils as
a measure of the phosphate requirement for pasture growth. Aust.
J. Agrie. Res. 8:601-612.
Palacios, C. 1976. Efectos de la nodulacion, pepetizacion y nutri
cin mineral en la fifacion de nitrgeno por Centrosema pubescens
Benth. a nevel de invernadero. Tesis Ingeniero Agronomo. Uni
versidad de Guayaquil, Guayaquil, Guayas, Ecuador. 31 p.
Paladines, 0. L., and J. de Alba. 1963. Aceptacin de forrajes
tropicales por el ganado. Turrialba. 13:194-196.
Paredes, 0. 1974. Evalucaion de diferentes sistemas de pastoreo con
bovinos en pasto guinea (Panicum maximum Jacq.). Tesis Ingeniero
Agronomo. Universidad de Guayaquil, Guayaquil, Guayas, Ecuador.
37 p.


20
The terms grazing period and rest period are commonly used in
the rotational grazing system (Heady, 1970). According to Heady (1970)
the grazing period is the portion of the grazing time during which
grazing takes place and rest period is the time during which the pasture
is not grazed. Comparing both continuous and rotational grazing
systems on animal production, Stobbs (1969), in Africa, found that
animal production was slightly higher for rotational grazing when it
was carried out in three paddocks, but it was lower than continuous
grazing when it was done in six paddocks. From Australia, Grof and
Harding (1970) reported that a mixed Guineagrass-centro pasture with
a carrying capacity of 3.5 an ha ^ produced a liveweight gain of 934
kg ha for 2 years and 1075 kg ha ^ for 2 years for continuous and
rotational grazing, respectively. From Pichilingue, Ecuador, Paredes
(1974) reported that during the rainy season there were no differences
in stocking rate between continuous and rotational grazing, but during
the dry season continuous grazing under variable stocking rate was
superior to rotational grazing with fixed or variable stocking rate.
He also found that continuous grazing with variable stocking rates
gave the highest average daily gain of 0.457 kg an ^ day \
Estimates of Dry Matter Production and Yield
Estimation of production and yield are major problems in grazing
experiments, because of the heterogeneity of the pasture sward and the
amount of time required to obtain an adequate sample. One must decide
the number of samples, the area to be sampled, and choose an adequate
technique for sampling a highly variable population where cover, density,


94
During the first season a significant linear, quadratic, and inter
action effects were found which accounted for 16, 14, and 23% of
the total variation, respectively (Appendix Table 31).
Yield of Weeds (DM) for the Dry Season of 1978
The yield of weeds for the dry season of 1978 is presented in
Table 8 for all the treatment combinations. The yield of weeds
varied from 0 kg DM ha ^ to 190 kg DM ha ^. There is a linear
effect of days rest on the yield of weeds (P < 0.01), but the
linear effects of X^, X^, and X^ were not significant. A quadratic
effect of X2 was also noted (P < 0.06) and no interactions between
the experimental were found. The linear and quadratic components
of the model accounted for 16 and 22% of the total variation, respec
tively, while the interaction effects represented only 3% of the
total variation (Appendix Table 32). The yield of weeds tended to
decrease as the length of the rest period was increased from 0 to
56 days, but this response was very small.
Yield of Weeds (DM) for the Wet Season of 1979
The yield of weeds for the wet season of 1979 for each treatment
combination is presented in Table 8. The yield of weeds varied
from 0 to 860 kg DM ha \ It was during this period that the effects
of rest period and grazing pressure upon the yield of weeds began to
emerge. There was a linear effect of rest period (X^) and grazing
pressure (X^) upon the yield of weeds (P < 0.01). There was also a
significant interaction of X^ and X^ (P < 0.01). The length of


99
appeared to be the principal determinants of the yield of weeds.
There were also interactions between x X^, X^ x X^, and X^ x X^,
which indicated that the yield of weeds was reduced by increasing
the rest period with short grazing period and also by increasing the
rest period with low grazing pressure. When long grazing periods
are combined with short rest periods, there is an increase in the
yield of weeds and this also occurred with short rest periods and
high grazing pressure. These last two interactions depressed the
amount of grass and legume dry matter which favored the weeds.
There was an increase in the average yield of weeds to 125 kg DM ha ^
which suggested that the weed problem was increasing with time.
During the wet season of 1980, the linear effects of rest period
(X2), grazing pressure (X^), and P fertilizer (X^) were very evident
(P < 0.01). There was also a quadratic effect of X^ (P < 0.01).
The interactions between X2 x X^, X2 x X^, and X^ x X^ were also
significant (P < 0.01). From the results of these analyses, it
appeared that days rest and grazing pressure were the principal
determinants of the yield of weeds (see Figs.13 and 14). The inter
actions indicate that the yield of weeds declined as the rest period
increased in combination with low grazing pressures, while the yield
of weeds increased when short rest periods were conbined with high
grazing pressures. Weeds apparently respond positively to increasing
levels of P. It was also observed that when short rest periods were
combined with high fertilizer levels and high grazing pressures, that
the yield of weeds increased considerably reaching values of 2600 and
2470 kg DM ha \ Near zero levels of weeds were found with treatments


53
from 1210 to 6630 kg DM ha \ corresponding to the treatments 17
[1 day grazing, 0 rest period (continuous grazing), 1.6 kg DM on
offer/loo kg BW, and 0 kg ha ^ of superphosphate] and 31, respectively
(1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 400
kg ha of superphosphate) .
During the wet season of 1979 both length of rest period and
level of grazing pressure had an effect upon biomass production (P< 0.01),
while no quadratic effects and interactions were noted. The linear
components of the model accounted for 67% of the total variation, and
only 2 and 1% of the total variation was represented by the quadratic
and interactions. As is evident, the biomass production is increased
as the length of the rest period is increased. The higher grazing
pressures also decrease the biomass production and the greatest biomass
production was reached when a long rest period and low grazing pressure
were imposed. The analysis of variance for the wet season of 1979 is
presented in Appendix Table 13.
Biomass Production (DM) for the Dry Season of 1979
Biomass production during the dry season of 1979 for each treat
ment combination is presented in Table 4. Biomass production varied
from 1090 to 5980 kg DM ha \ corresponding to treatments 26 [28 days
grazing, 0 days rest (continuous grazing), 1.6 kg DM on offer/100 kg
BW, and 400 kg ha ^ of superphosphate] and 32 (28 days grazing, 56 days
rest, 8.3 kg DM on offer/100 kg BW, and 400 kg ha ^ of superphosphate).
Again, the lowest values of biomass on offer ha ^ corresponded to
treatment combinations of short rest periods or continuous grazing


Table 39. Analysis of variance, regression coefficients and probabilities for visual
estimation grass for the dry season of 1979.
RESPONSE MEAN
77.7079
ROOT MSE
G. 331 5
R-SQUARE
0.87537303
COEF OF VARIATION
0 J0721296
REGRESSION
DF
TYPE I ns
R-SQUARE
F-RATIO
PROD
LINEAR
4
13662. 9551
0. 6014
47. 21
0 0001
QUADRATIC
4
312. 1462
0. 0156
1 12
0. 3604
CROSSPRODUCT
6
3577. 1064
0 1704
8. 59
0.0001
TOTAL REGRESS
14
17552. 2077
0. 0754
18. 06
0. 0001
RESIDUAL
nr-
SB
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
2237. 8107
06.0699
3. 297
0.0263
PURE ERROR
10
261. 0844
26. 1084
TOTAL ERROR
36
2498. 9031
69. 4140
PARAMETER
DP
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
73. 8873
2. 1072
35. 06
0. 0001
XI
1
O.21486152
0.79524585
0. 27
0. 7806
X2
1
9. 6975
0.79635968
12. 18
0. 0001
X3
J
6 3346
0.79524585
7. 97
0. 0001
X4
J
-0.19943027
0.79524585
-0. 25
0. 8034
XI *X1
1
-0 30512404
1.2896
-0. 24
0. 8143
X1#X2
- 0. 60567784
0. 43557409
-1. 39
0.1729
X2*X2
1
2. 2365
1. 2896
1. 73
0. 0914
X1#X3
1
-0.26963333
0. 41657525
-0. 65
0. 5216
X2*X3
1
-3. 0109
0.43557409
-6. 93
0. 0001
X3#X3
1
-2. 0239
1. 2096
-1. 57
0. 1253
X1*X4
1
0.24034167
0.41657525
0. 58
0. 5676
X2*X4
1
0.39065051
0.43557409
0. 90
0. 3758
X3#X4
1
-0. 03567500
0.41657525
~0. 09
0. 9322
X4*X4
1
0 42404183
1.2896
0. 33
0. 7442
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROD
XI
5
191.6928
38 3386
0. 55
0. 7355
X2
5
14132.3564
2826. 4713
40. 72
0. 0001
X3
5
6213.5553
1242 7111
17. 90
0. 0001
X4
5
86.7630
17 3926
0. 25
0 9368


126
Table 11.
biomas^ifcM/^T "8resslon efficients and probabilities
lomass (g DM/m ) for the wet season of 1978.
for aerial
RESPONSE MEAN
202 2070
ROUT MSE
72.0626
R-SQUARE
0 30071262
COEF OF VARIATION
0 20007736
REGRESSION
DF
TYPE I SS
R-SQUARE
F-RATIO
PROD
LINEAR
T
6T066. 707T
0. 2076
3. 02
0. 0303
QUADRATIC
T
12307. 0554
0 0401
0. 50
0 6766
CROSSPRODUCT
6
T 1030. 5274
0 1330
1. 27
0 2073
TOTAL REGRESS
IT
117494
0. 3007
1. 50
0. 1327
RESIDUAL
DF
SB
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
70362. T03B
2706. 2474
0. 224
0. 7770
PURE ERROR
JO
120760
12075 7774
TOTAL ERROR
36
191122
5300 7572
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
232 2371
10 4200
12. 60
0. 0001
XI
J
0.2300T730
6. 7548
-0. 03
0. 7730
X2
J
17. 7000
6. 7645
2. 04
0 0074
X3
1
0. 1610
6. 7540
1. 17
0. 2403
XT
J
-2. 2703
6. 7540
-0. 33
0. 7433
X 1 K- X I
1
-6. 0076
11 2704
- 0. 50
0. 5627
XI *X2
1
-0. 02465075
3. 0073
-0. 01
0. 7747
X2*X2
1
IT. 3607
11 2704
1. 27
0. 2100
X1*X3
J
5. T710
3. 6431
1. 50
0. 1417
X2*X3
1
2. 03T7
3. 0073
0. 53
0. 5765
X3*X3
1
-3. 7777
11. 2704
-0. 34
0. 7303
XI X4
1
7. J 7 J T
3. 6431
1. 77
0 0561
X 2 X 4
1
-T. 2617
3. 0073
-1. 12
0. 2706
X3&X4
1
0 7T020033
3. 6431
0. 20
0. 8401
X4#X4
1
1. 0607
11. 2704
0. 17
0. 0693
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
S
34470. 0023
6075 7765
1. 30
0 2061
X2
5
60702. 7108
12100. 5832
2. 27
0. 0657
X3
5
2T567. T311
4913. 0062
0. 73
0 4750
XT
5
30200. T 1 60
6040. 0834
1. 14
0. 3584


43
In early January of 1978, the P fertilizer was applied taking
into consideration the respective levels of simple superphosphate
kg ha ^ (whose composition was 18.2% ^2^5 anc* 14% elemental S) ; at
the same time a mixture of micronutrients was applied to the whole
area. The composition of this mixture was 3 kg Zn chelate, 4 kg
Fe chelate, 3 kg CuSO^, 3 kg Borax, and 0.8 kg Molybdenum nitrate
Construction of Physical Facilities
In February, the total area was surveyed for the purpose of
locating the fence lines, and four-strand wire fences were built.
During the second half of March, the entire area was moderately
grazed and the remaining vegetation was mowed at 15 cm height.
Wooden mineral boxes were built which were used to supply the fol
lowing formula: 50% of sodium chloride + 50% of mineralized salt
containing 7% P as dicalcium phosphate, 0.48% Zinc sulfate, 0.12%
manganese carbonate, 0.14% copper sulfate, 0.32% ferrous sulfate,
and 0.006% cobalt chloride. Water tanks were provided in each experi
mental pasture. An area of approximately 10 ha adjacent to the experi
ment was available as reserve pasture for 50 animals used for adjusting
the stocking rates in the experiment.
Collection of Data in the Three Experimental Years
On May 12, 1978, experimental grazing was initiated. Fifty-six
Criollo and Holstein/Brahman steers were used to graze the experi
mental pastures. At the start of the experimental period, the


76
intensity of plant removal was the major determinant of grass yield.
This was true for both guineagrass and elephanatgrass and the effect
of severe defoliation was to reduce the vigor and the regrowth
capacity of these two species. At this point, some bare soil areas
were observed and weed invasion occurred. This was very evident under
short rest periods when combined with high grazing pressure.
The animals had a tendency to selectively graze the grass species
during the wet seasons and to selectively graze the legumes during the
dry seasons. Similar results have been reported by Humphreys (1978) who
observed a selective consumption of Panicum maximum in preference to
Stylosanthes guianensis while in the dry season, S_^_ guianensis was
well-consumed.
It is well known that tall-growing grasses are very susceptible
to a higher degree of defoliation, especially if these species are
subjected to low cutting heights or high grazing pressures. Most of
the leaves, growing points, and axillary buds are located much higher
than on short-growing species. The two tall-growing species in this
study have long internodes with the axillary buds located far apart
on the stem so they are very vulnerable to intense defoliation.
Effects of Lengths of Rest Period and Levels of Grazing
Pressure Upon Legume Yield (DM)
Since one of the main objectives of this research was to
determine the optimum combinations of the components of grazing
management which would favor the legume component of the pasture
and permit a higher order of persistence, the legume yield and the
changes which occurred during the five seasons are of primary interest.


Fig. 6. Effect of rest period and grazing pressure upon grass yield (DM) for the wet
season of 1978.
vj
LO


Table 24. Analysi^ of variance, regression coefficients and probabilities for grass yield
(g DM/m ) for the dry season of 1979.
RESPONSE MEAN
2137. 1133
ROOT MSE
70. 4 4 60
R-SQUARE
0 640767D4
COEF OF VARIATION
0. 37773430
REGRESSION
DP
TYPE I BS
R-SQUARE
F-RATIO
PROD
LINEAR
4
552655
0. 5670
14. 26
0. 0001
QUADRATIC
4
20B01. 1470
0. 0215
0. 54
0. 7003
CROSSPRODUCT
6
40003. 0361
0 0502
0. 04
0. 5480
TOTAL REGRESS
14
622340
0. 64 00
4. 57
0. 0001
RESIDUAL
DP
SS
MEAN SQUARE
F-RATIO
PROB
LACK OF FIT
26
201047
7763. 4236
0. 528
0. 7072
PURE ERROR
10
147047
14704.7300
TOTAL ERROR
36
340070
7671.6223
PARAMETER
1)1
ESTIMATE
SID DEV
T-RATIO
PROD
INTERCEPT
1
215. 6553
24.0704
8. 66
0. 0001
XI
J
-1. 0700
7. 3767
-0 12
0. 7076
X2
1
42. 5550
7. 4077
4 52
0. 0001
X3
1
54. 0557
7. 3767
5. 75
0. 0001
X4
1
-5. 8074
7. 3767
-0. 62
0. 5403
XI *X1
1
3. 5617
15. 2305
0. 23
0. 0165
X1 *X2
1
4. 3560
5. 1460
0. 85
0. 4027
X2*X2
1
6. 6570
15. 2305
0. 44
0. 6648
X1*X3
1
4.2202
4. 7223
0. 86
0. 3760
X2*X3
1
-5. 6522
5. 1460
-1 10
0. 2774
X3*X3
1
-1. 7652
15. 2305
-0. 13
0. 8781
X1*X4
1
5 0277
4.7223
1. 18
0. 2442
X2*X4
J
3. 35 J 6
5 1460
0. 65
0. 5171
X3*X4
1
3. 6475
4. 7223
0. 74
0 4632
X4*X4
J
4 541 1
15. 2305
0 30
0. 7674
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
X 1
5
203B7. 3765
5677.4753
0. 57
0. 7107
X2
5
224254
44850. 7307
4. 63
0. 0023
X3
5
336753
67370. 5377
6. 75
0. 0001
X4
5
25023. 7376
5164.7077
0 53
0. 7477


77
Legume Yield (DM) for the Wet Season of 1978
Legume yields for the wet season of 1978 for each treatment
combination are presented in Table 7. Legume yield varied from
410 kg DM ha ^ to 1380 kg DM ha ^ for treatments 2 (21 days grazing,
14 days rest, 3.3 kg DM on offer/100 kg BW, and 100 kg ha ^ of super
phosphate) and 14 (21 days grazing, 14 days rest, 6.6 kg DM on offer/
100 kg BW, and 300 kg ha ^ superphosphate), respectively.
Length of rest period was the only factor which appeared to
have much effect upon legume yield (P < 0.01). The other three
experimental variables, X^, X^, and X^, and all of the interactions
were not significant. The linear components of the model accounted
for 16% of the total variation, while the quadratic and interaction
components represented only 4 and 8% of the total variation, respec
tively (Appendix Table 26). For each unit (14 days) increase in
the rest period, there was an increase of only 60 kg of legume yield.
The analysis of variance is presented in Appendix Table 26.
This lack of response to the experimental variables was expected
since the wet season of 1978 was very short, so the experimental
variables did not have time to create differences in response.
Legume Yield (DM) for the Dry Season of 1978
Legume yield for each treatment combination during the dry season
of 1978 is presented in Table 7. The legume yields varied from 380
kg DM ha ^ to 1160 kg DM ha ^ for treatments 28 (28 days grazing, 56
days rest, 1.6 kg DM on offer/100 kg BW, and 400 kg ha ^ of super
phosphate) and 24 (28 days grazing, 56 days rest, 8.3 kg DM on offer/


4
Chavez, 1974; Rolando, 1974). It is indigenous to tropical Africa
where it is dominant over large areas, particularly under humid and
subhumid conditions. In the coastal area of Ecuador it is known as
guinea, cauca, saboya, chilena. Betancourt (1969) indicated that
guineagrass was the most widespread pasture grass in the lowlands of
Ecuador. Acosta-Soliz (1967) reported that this grass grows from sea
level up to 1400 m.
Guineagrass is very popular because of its adaptation to the
dry season, resistance to fire, high production of forage, capability
of establishment by seed or by division of plant crowns, and finally
due to its ability to persist under heavy use and abuse (INIAP, 1980).
Recently the Instituto Nacional de Investigaciones Agropecuarias
(INIAP) of Ecuador has tested 124 introductions of guineagrass in
small grazing trials, mainly for adaptation and persistence. New
cultivars will be released to the farmers as soon as sufficient seed
is available for large scale use (INIAP, 1979).
Pennisetum purpureum Schumach, elephantgrass, is a tall-growing
species, up to 5 m in height. It has been described by Bogdan (1977),
Correa (1926), and Mcllroy (1972).
Acosta-Soliz (1967) and INIAP (1980) stated that elephantgrass
grows from sea level up to 2200 m in the warm valleys of the Sierra
region. In Ecuador it is mainly used as a pasture grass with some
farmers reporting pastures up to 40 years old still under grazing
conditions. Under cutting frequencies of 45 days it produces up to
80 tons DM ha ^ yr ^ when receiving 400 kg of N and irrigation during
the dry season (INIAP, 1972). There are three cultivars which are


LIST OF FIGURES
FIGURE PAGE
1 Profile of three natural regions of Ecuador, showing
the main vegetative zones in relation to rainfall
and temperature 29
2 Rainfall recorded at Estacin Experimental Pichi-
lingue during the period 1978-1980 30
3 Temperature recorded at Estacin Experimental Pichi-
lingue during the period 1978-1980 31
4 Solar radiation recorded at Estacin Experimental
Pichilingue during the period 1978-1980 32
5 Field plan of the experimental pastures 41
6 Effect of rest period and grazing pressure upon
grass yield (DM) for the wet season of 1978 73
7 Contours of grass yield (DM) as affected by length
of rest period and levels of grazing pressure in
the wet season of 1980 74
8 Effect of rest period and grazing pressure upon grass
yield (DM) for the wet season of 1980 75
9 Effect of rest period and grazing pressure upon
legume yield (DM) for the wet season of 1978 84
10 Contours of legume yield (DM) as affected by length
of rest period and levels of grazing pressure in the
wet season of 1980 88
11 Effect of rest period and grazing pressure upon
legume yield (DM) for the wet season of 1980 89
12 Effect of rest period and grazing pressure upon
yield of weed (DM) for the wet season of 1978 98
13 Contours of yield of weed (DM) as affected by length
of rest period and levels of grazing pressure in the
wet season of 1980 100
14 Effect of rest period and grazing pressure upon yield
of weed (DM) for the wet season of 1980 101
15 Effect of rest period and grazing pressure upon grass
percentage for the wet season of 1978 108
xi


Table 37. Analysis of variance, regression coefficients and probabilities for visual
estimation grass for the dry season of 1978.
RESPONSE MEAN
73.4706
ROOT MSE
6. 0078
R-SQUARE
0.72630300
COEF OF VARIATION
0.0024730?
REGRESSION
DF
TYPE I SS
R -SQUARE
F-RATIO
PROB
LINEAR
4
3107.5442
0. 6432
21. 16
0.0001
QUADRATIC
4
73. 2107
0 0193
0. 63
0 6411
CROSBPRODUCT
6
308. 6044
0. 0639
1. 40
0. 2410
TOTAL REGRESS
14
350?. 4392
0 7264
6. 83
0. 0001
RESIDUAL
r>r:
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
1016. 3421
39. 0901
1. 279
0. 3544
PURE ERROR
10
305. 6046
30 5605
TOTAL ERROR
36
1321. 7467
36.7207
PARAMETER
Dl
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
71.3133
1. 5326
46. 53
0. 0001
XI
1
-1.130?
0. 57040717
-1. 96
0. 0DB4
X2
1
4. 0573
0. 57921729
8. 3?
0. 0001
X3
1
1. 5104
0. 57840717
2. 63
0. 0126
X4
1
0.31379940
0.57040717
-0. 54
0. 5908
XI X1
1
0.32543601
0.93797285
0. 35
0. 7306
X1*X2
J
0.01770201
0.31680665
0. 06
0. 9508
X2*X2
1
1. 1727
0.93797285
1. 25
0. 2193
X1*X3
1
-0.49009213
0.30298820
-1. 62
0. 1145
X2*X3
1
-0. 59456164
0.31680665
-1. 88
0. 0687
X3*X3
1
-1. 0730
0 937972B5
-1. 14
0. 2602
X1*X4
1
0.22422639
0.30290820
0. 74
0. 4641
X2*X4
1
0 41474540
0.31630665
1 31
0 1988
X3&X4
1
0.01466250
0.30298820
0. 05
0. 7617
X4#X4
1
-0.15372906
0. 93799285
-0. 16
0. 8707
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
271.3077
54.2775
1. 40
0. 2211
X2
5
2859.9234
571.9847
15. 58
0. 0001
X3
5
446.7100
8?.3420
2. 43
0. 0534
X4
5
05. 14 64
17.0293
0. 46
0 0005


Table 34. Analysis of variance, regression coefficients and probabilities for yield
of weeds (g DM/m ) for the dry season of 1979.
RESPONSE MEAN
12.5467
ROUT MSE
10. 0158
R-SQUARE
0.72051147
COEF OF VARIATION
1. 4359
REGRESSION
OF
TYPE I SS
RSQUARE
F-RATIO
PROD
LINEAR
4
15464. 2724
0. 3593
11. 91
0. 0001
QUADRATIC
4
2601 7238
0 0623
2 07
0 1050
CROSSPRODUCT
6
13208. 1525
0. 3069
6. 78
0. 0001
TOTAL REGRESS
14
31354.1400
0. 7205
6. 90
0. 0001
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
11516. 3712
442. 9374
26. 345
0. 0001
PURE ERROR
10
168. 1203
16.0128
TOTAL ERROR
36
11684 4995
324. 5694
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
6. 4747
4. 5565
1. 42
0. 1639
XI
1
-4.6936
1. 7196
-2. 73
0. 0090
X2
1
-9. 31 74
1 7220
-5. 41
0 0001
X3
J
-9.5965
1. 7196
-5. 58
0 0001
X4
1
0.07423712
1 7196
0. 51
0. 6143
X1*X1
J
1. 7270
2. 7807
0. 62
0 5396
X1 *X2
1
2. 6713
0.94187308
2. 84
0. 0074
X2*X2
1
-1.4397
2. 7887
-0. 52
0. 6080
X1 #X3
1
1. 0730
0.90079053
2. 08
0. 0448
X2*X3
1
4. 7916
0.94187308
5. 09
0. 0001
X3*X3
1
5. 0403
2. 7887
1. 81
0. 0786
XI X4
1
-1.1428
0.90079053
-1. 27
0. 2127
X2*X4
1
-0.55544461
0.94187308
-0. 59
0 5591
X3*X4
1
-0.63014167
0.90079053
-0. 70
0. 4087
X4*X4
J
-1. 5126
2. 7887
-0. 54
0. 5909
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROD
XI
5
5943. 0961
1180.6192
3. 66
0 0008
X2
5
20701. 2000
4156.2418
12. 01
0 0001
X3
Ei
16774. 9107
3394 9021
10. 46
0 0001
X4
5
929. 6552
105 9310
0. 57
0 7203
149


24
the grass and legume during the three experimental years. Similar
results were reported from Africa by Stobbs (1969).
Cowan et al. (1975), in the Atherton Tableland, Queensland,
Australia, found a highly significant correlation coefficient between
milk yield ha ^ and the amount of the legume glycine present in the
pasture. They studied four stocking rates, 1.3, 1.6, 1.9, and 2.5
cows ha *. On the other hand, while stocking rate was increased
the milk production ha ^ also was increased but the percentage of dry
matter of legume present in the pasture decreased when the stocking
rate increased. Cowan and O'Grady (1976) reported the same trend
under similar conditions in another study carried out later at the
same location.
Some investigations carried out in Cuba have shown that under
heavy stocking rates the legumes fail to persist in the mixture.
Febles and Padilla (1972), using a stocking rate of 4 cows ha \ found
that a mixture of guineagrass and the legumes glycine, Siratro, Stylo-
santhes guianensis, Desmodium intortum, and Desmodium uncinatum, did
not persist more than 36 weeks. Funes and Perez (1976), using six ani
mals ha \ also found that the three commercial cultivars of glycine
(Tinaroo, Cooper, and Clarence) failed to persist under those conditions
and that at 36 weeks the proportion of weedy species was drastically
increased. Their final conclusion was that it is better to use light
stocking rates and that the legumes are useful in those areas in which
the grazing management is extensive. On the other hand, there are some
results that show a positive effect of stocking rate upon the propor
tion of legume in the pasture. Vilela (1979), in Brazil, found that


Table 6. Grass yields (DM) by year, season, and treatment combination
Treatments
1978
1978
1979
1979
1980
D/Gf
(x.^
D/Rf
(x2)
G/Pf
(X ) % BW
Ft -i
kg ha
No.
Reps
w/st
D/S
w/st
D/Sj
w/st
1
7
14
3.3
100
1
990
1650
kg ha
1670
1560
1060
2
21
14
3.3
100
1
1540
2130
2600
2050
2720
3
7
42
3. 3
100
1
1340
2160
3590
1930
2980
4
21
42
3.3
100
1
1240
2100
3080
1440
3780
5
7
14
6.6
100
1
1810
2930
3760
3570
3500
6
21
14
6.6
100
1
1130
2940
2410
1950
2020
7
7
42
6.6
100
1
1460
3340
4010
2940
3950
8
21
42
6.6
100
1
1240
3770
5490
4400
5440
9
7
14
3.3
300
1
1940
1270
2590
1720
2110
10
21
14
3.3
300
1
1950
2130
3330
2250
2590
11
7
42
3.3
300
1
1290
2170
3270
2390
3750
12
21
42
3.3
300
1
920
1750
3910
2260
5900
13
7
14
6.6
300
1
1400
1970
2440
2220
2000
14
21
14
6.6
300
1
1630
2020
1790
1600
1720
15
7
42
6.6
300
1
980
3380
5750
5510
6760
16
21
42
6.6
300
1
1790
2460
3490
2560
3680
17
1
0
1.6
0
1
1000
760
560
420
350
18
28
0
1.6
0
1
1380
640
910
390
300
19
1
56
1.6
0
2
1860
3060
5560
3620
7150
20
28
56
1.6
0
2
1470
1720
3920
2570
3820
21
1
0
8.3
0
1
1400
2640
3310
4010
5750
22
28
0
8.3
0
1
1090
2600
3320
3570
4730
23
1
56
8.3
0
2
2080
3370
6530
3790
5440
24
28
56
8.3
0
2
2070
4550
4940
4140
6370


124
dry matter on offer was increased). No interactions between these
two factors were found, suggesting that longer rest periods and
lower grazing pressures independently were required to maintain
high dry matter production.
Grass (DM) yield and grass percentage appeared to be highly
sensitive to short rest periods and high grazing pressures. The
negative influence of increased grazing pressure upon the grass
component was partially offset by increasing the rest periods.
Legume (DM) yields and legume percentages were highly sensitive
to lengths of rest period and less sensitive to high grazing pressures.
Short rest periods favored legume content, but its greatest contribu
tion appeared to be near the middle range of rest periods. This
response apparently is independent of length of grazing period.
Legume (DM) yield and legume percentage showed a slight decline
during the five experimental seasons, while grass yield (DM) and
grass percentage increased during the same period of time.
The weed component was reduced by long rest periods and by low
grazing pressures. Interactions occurred between these two variables.
Both high grazing pressures and short rest periods were responsible
for increasing the yield of weeds (DM). The results of this study
suggest that a grazing management system which combines a moderate
rest period and a moderate level of grazing pressure would be an
optimum management strategy to maintain a low amount of weeds, a
large amount of legumes and high yields of available forage. The
other two variables, days grazing and levels of phosphorus fertiliza
tion had negligible effects upon the response of the pasture sward.


57
periods are decreased, the frequency and severity of defoliation is
increased affecting directly the subsequent regrowth and the total
dry matter yields. Harris (1978) reported that although the function
of reserves, availability of growing points, and uptake characteristics
are influenced by the level of stubble biomass, the relationship
between stubble biomass and growth rate relates to the amount of
photosynthetic tissues.
Effect of Lengths of Rest Period and Levels of Grazing
Pressure Upon Available Forage (DM)
Available forage in this context represents the sum of the grass
and legume component but is exclusive of the weeds.
The effect of the lengths of the rest period and grazing pressure
upon available forage is presented in Table 5 for the five seasons of
this experiment. The available forage is the average of the estimated
amount of dry matter present before each grazing for the rotational
grazing treatment combinations and for 56 days for the continuous
treatment.
Available Forage for the Wet Season of 1978
During the first wet season, only the length of the rest period
had an effect on available forage (P < 0.01). The linear components
of the model accounted for only 20% of the total variation, while the
quadratic effects and interactions represented 5 and 12%, respectively
(Appendix Table 16).
The means of total available forage varied from 1420 kg DM ha ^
to 3720 kg DM ha (Table 5) for treatments 34 (28 days grazing, 28 days


167
Petersen, R. G., H. L. Lucas, and G. 0. Mott. 1965. Relationship
between rate of stocking and per animal and per acre performance
on pasture. Agron. J. 57:27-30.
Reuter, D. J. 1975. The recognition and correlation of trace element
deficiencies. In D. J. D. Nichlas and A. R. Egan (eds.) Trace
elements in soil-plant-animal systems. Academic Press, New York.
Roberts, C. R. 1974. Some problems of establishment and management
of legume-based tropical pastures. Trop. Grassld. 8:61-67.
Rolando, C. X. 1974. Comportamiento al pastoreo de tres gramineas
forrajeras, con fertilizacin nitrogenada y en asociacin con
centrosema, bajo cinco periodos de descanso. Tesis Ingeniero Agrn
omo. Universidad de Guayaquil, Guayaquil, Guayas, Ecuador. 46 p.
Russell, J. S. 1978. Comparative soil tolerance of some tropical
and temperate legumes and tropical grasses. Aust. J. Exp. Agrie.
Anim. Husb. 16:103-106.
Salinas, J. G., and P. A. Sanchez. 1976. Soil-plant relationships
affecting varietal difference in tolerances to low available
soil phosphorus. Ciencia e Cultura. 28:156-168.
Sanchez, P. A. 1977. Properties and management of soils in the
tropics. John Wiley and Sons, New York.
Santhirasegaram, K. 1976. Manejo de praderas de leguminosas y
gramineas en un ecosistema de selva illuviosa tropical en Peru.
In E. Bornemisza and A. Alvarado (eds.) Manejo de suelos en la
America tropical. North Carolina State University, Raleigh,
North Carolina. P. 455-466.
Santillan, R. A. 1971. Capacidad de carga en el pasto puntero
[Hyparrhenia rufa (Nees) Stapf.] sometido a pastoreo continuo.
Tesis Ingeniero Agronomo. Universidad de Guayaquil, Guayaquil,
Guayas, Ecuador. 31 p.
. 1976. Estimating forage yield with a disk meter. M.S.
Thesis. University of Florida, Gainesville, Florida. 43 p.
Serrao, E. A. 1976. The use of response surface design in the agronomic
evaluation of a grass-legume mixture under grazing. Ph.D. Disserta
tion. University of Florida, Gainesville, Florida. 232 p.
Servicio Nacional de Hidrologia y Meterologia del Ecuador. 1980.
Boletines climatolgicos. Ano 19, No. 209-220, Quito, Ecuador.
Shaw, N. H., and W. W. Bryan. 1976. Tropical pasture research:
Principles and methods. Commonwealth Agrie. Bureax. Alden
Press, Oxford, England.


103
Visual Estimation of Forage Components
Visual estimations of forage components were made as part of
a double sampling procedure which was checked against the botanical
separations. This section presentes the results for the visual
estimations of the percentage grass and the percentage legume. These
estimations were made and adjusted statistically with the use of a
linear regression model for each of the 41 treatments and for each of
the five seasons.
Visual Estimation of Percentage Grass
The visual estimations of the percentage grass on a dry matter
basis are given in Table 9 for each treatment combination and for
each season. In the wet season of 1978, the percentage grass varied
from 50 to 80% which represented the variation that occurred during
the first two and one-half months of the experiment. There were no
significant effects of treatment during this first season. The ana
lysis of variance is presented in Appendix Table 36.
In the dry season of 1978, the visual estimates are also presented
in Table 9, and the percentage grass varied from 52 to 89%. The
linear effects of rest period and grazing pressure (P < 0.01) became
evident during the first dry season, but the effects of days grazing
and fertilizer level were not significant. Also there were no inter
actions between the experimental variables. The linear components
of the model accounted for 64% of the total variation, while the
quadratic and interaction effects represented 1 and 6% of the total
variation, respectively (Appendix Table 37). During the 1979 season,


16
values below these points (Ventura et al., 1975). Tropical forage
legumes, on the other hand, retain higher crude protein levels during
the dry season, even in advanced maturity stages (Milford and Haydock,
1965). Selectivity by grazing animals is a possible explanation for
higher performance of grazing cattle during the dry-season periods in
tropical regions, especially when available forage meets or exceeds
animal requirements (t'Mannetje, 1974).
Pasture Evaluation
Mott and Moore (1970) developed a five-phase scheme for forage
evaluation. Such a scheme involves quantity and quality determinations.
Phase: I. Introduction and breeder's lines,
II.Small plot clipping trials,
III.Mob grazing experiments, forage response to grazing animals,
IV.Animal response, effect of forage on animal output,
V.Forage-livestock feeding systems.
Forage quality, in vitro organic matter digestibility (IVOMD) is
taken into consideration in the first three phases, while phases IV
and V involve animal response in nutrient digestibility, performance
per animal and production per hectare.
Definitions for stocking rate, grazing pressure, and carrying
capacity were given by Mott (1960). Stocking rate is defined as the
number of animals per unit area of land, the term bearing no relation
ship to the amount of forage. Grazing pressure is the amount of forage
dry matter on offer per animal per day and carrying capacity, also
called grazing capacity, is defined as the stocking rate at the


Fig. 14.
Effects of rest period and grazing pressure upon yield of weed (DM) for the
wet season of 1980.
o


Table 18. Analysis of variance
forage (g DM/m ) for
RESPONSE MEAN
ROOT MSE
R-SGUARE
COEF OF VARIATION
REGRESSION
DF
LINEAR
4
QUADRATIC
4
CROSSPRODUCT
6
TOTAL REGRESS
14
RESIDUAL
DF
LACK OF FIT
26
PURE ERROR
10
TOTAL ERROR
36
PARAMETER
DF
INTERCEPT
1
XI
J
X2
1
X3
1
X4
1
X1*X1
1
X1#X2
1
X2*X2
1
X1#X3
1
X2*X3
1
X3X3
1
xi*x4
1
X2*X4
1
X3*X4
1
X4*X4
1
FACTOR
DF
XI
5
X2
5
X3
5
X4
5
regression coefficients and probabilities for available
the wet season of 1979.
393. 84 55
104.1711
0.73064654
0.26449741
TYPE I SS
R-SQUARE
F-RATIO
PROB
1006949
0. 694 3
23. 20
0. 0001
30055. 6043
0 0207
0. 69
0. 6021
22692. 5744
0. 0156
0. 35
0. 9061
1059697
0. 7306
6. 98
0. 0001
SS
MEAN SQUARE
F-RATI0
PROB
181696
6988. 2905
0. 334
0. 9880
208963
20896. 2721
390650
10851 6187
ESTIMATE
STD DEV
T-RATIO
PROB
334. 7751
26. 3464
12. 71
0. 0001
-5. 6979
9. 9432
0. 57
0. 5702
74. 7019
9. 9571
7. 51
0. 0001
50. 3946
9. 9432
5. 07
0. 0001
-4.4736
9. 9432
-0. 45
0. 6555
5. 1526
16. 1247
0. 32
0. 7512
-6. 3854
5. 4461
-1. 17
0. 2487
4. 8651
16. 1247
0. 30
0. 7646
0. 33773750
5. 2086
-0. 06
0. 9487
-2. 1599
5 4461
-0. 40
0. 6940
-3. 6512
16. 1247
-0. 23
0. 8221
3. 1430
5. 2086
0. 60
0. 5500
0.76786291
5. 4461
0. 14
0. 8887
2. 1536
5. 2086
0. 41
0. 6817
8. 7676
16. 1247
0. 54
0. 5900
SS
MEAN SQUARE
F-RATIO
PROB
29967. 6100
5993. 5220
0. 55
0. 7355
638164
126433
11. 65
0 0001
295145
59028. 9042
5. 44
0. 0008
11212. 5205
2242. 5041
0. 21
0. 9575
133


106
the percentage grass tended to increase as the length of rest period
increased and to decrease as the grazing pressure increased. The
analysis of variance is presented in Appendix Table 37.
For the wet season of 1979, the percentage grass varied from
34 to 96%. Again the length of the rest period (X2) and grazing
pressure (X^) affected the percentage grass (P < 0.01) and there were
also significant interactions for X^ x X^ and X2 x X^ (P < 0.01).
The linear components of the model accounted for 61% of the total
variation, while the quadratic and interaction effects represented
2 and 20% of the total variation, respectively (Appendix Table 38).
Again, the percentage grass tended to increase as the length of the
rest period increased, while increases in grazing pressure tended to
reduce the percentage grass. The analysis of variance is presented
in Appendix Table 38.
During the dry season of 1979, the spread in the percentage
grass has now increased from 15 to 99%. The days rest (X2) and
grazing pressure (X^) were even more evident during the fourth season
and there were also significant interactions between X2 x X^ (P < 0.01).
The effect upon percentage grass of days grazing and fertilizer
level were nil.
The linear components of the model in the dry season of 1979
accounted for 68% of the total variation, while the quadratic and
interaction effects represented 1 and 17%, respectively (Appendix
Table 39). Again the percentage grass tended to increase as the
length of rest period increased, and to decrease as the grazing
pressure was increased.


1
2
3
4
5
6
7
8
9
10
11
12
13
14
LIST OF TABLES
PAGE
Soil analysis of experimental site (1978) 34
Modified central composite non-rotatable design with
four experimental (X) variables, at five levels each,
and 41 design points 36
Modified central composite non-rotatable design with
four experimental (X) variables, at five levels each,
41 design points, 51 experimental units with their
respective area 38
Aerial biomass production (DM) by year, season, and
treatment 49
Available forage (DM) by year, season, and treatment
combination 58
Grass yields (DM) by year, season, and treatment
combination.' 66
Legume yields (DM) by year, season, and treatment
combination 78
Yields of weed (DM) by year, season, and treatment
combination 92
Visual estimation of grass percentage by year, season,
and treatment combination 104
Visual estimation of legume percentage by year,
season, and treatment combination 113
Analysis of variance, regression coefficients and
probabilities for aeiral biomass (g DM/m ) for the
wet season of 1978 126
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m2) for the
dry season of 1978 127
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m2) for the
wet season of 1979 128
Analysis of variance, regression coefficients and
probabilities for aerial biomass (g DM/m ) for the
dry season of 1979 129
vii


Rainfall (mm)
Fig. 2. Rainfall recorded at Estacin Experimental Pichilingue during
the period of 1978-1980.
u>
o


70
kg DM ha-1 to 572 kg DM ha 1 for the following treatment combinations:
25 [1 day grazing, 0 days rest (continuous grazing), 1.6 kg DM on
offer/100 kg BW, and 400 kg ha ^ of superphosphate] and 32 (28 days
grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 400 kg ha
of superphosphate), respectively. Again the grass yield was increased
by increasing the days of rest and decreasing the grazing pressure
(P < 0.01). The variables X and were not significant and no
interactions were found between the experimental variables. The
linear components of the model accounted for 57% of the total varia
tion, while the quadratic and interaction effects represented only
2 and 5% of the total variation, respectively (Appendix Table 24).
For each 14 days increase in rest period, there was a corres
ponding increase of 420 kg ha ^ of grass yield, while each unit
(1.6 kg DM) decrease in grazing pressure increased the grass yield
by 540 kg ha \ The analysis of variance is presented in Appendix
Table 24.
Grass Yield (DM) for the Wet Season of 1980
The grass yield (DM) for the wet season of 1980 for each treat
ment combination is presented in Table 6. Grass yield varied from
120 kg DM ha ^ to 7150 kg DM ha for treatments 26 [28 days grazing,
0 days rest (continuous grazing), 1.6 kg DM on offer/100 kg BW, and
400 kg ha ^ of superphosphate] and 19 (1 day grazing, 56 days rest,
1.6 kg DM on offer/100 kg BW, and 0 kg ha ^ of superphosphate),
respectively. Grass dry matter yield was again affected by increasing
the number of days rest and decreasing the grazing pressure (P < 0.01).


Fig. 11. Effect of rest period and grazing pressure upon legume yield (DM) for the
wet season of 1980.
00
VO


Table 28.
Analysis of variance, regression coefficients and probabilities for legume
yield (g DM/m ) for the wet season of 1979.
RESPONSE MEAN
47.3703
ROUT MSE
23. 1 144
R-SQUARE
0. 40320032
COEF OF VARIATION
0 48795867
REGRESSION
Df
TYPE I SS
R-SQUARE
F-RATIO
PROR
LINEAR
4
12762. 1104
0. 34 03
6. 07
0. 0008
QUADRATIC
4
2231. 4 555
0. 0600
1. 04
0. 3901
CROSBPRODUCT
6
2790.5191
0. 0750
0. 87
0. 5259
TOTAL REGRESS
14
17984. 0851
0. 4832
2. 40
0. 0172
RESIDUAL
nr
SS
MEAN SQUARE
F-RATJO
PROD
LACK OF FIT
26
13604. 5041
523. 2502
0. 929
0. 5856
PURE ERROR
10
5629. 4893
562. 9489
TOTAL ERROR
36
19233. 9934
534. 2776
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIG
PROD
INTERCEPT
1
57. 3965
5. 8460
9. 82
0. 0001
XI
1
4. 1393
2. 2063
1. 88
0. 0683
X2
1
-5. 6133
2. 2094
-2. 54
0. 0155
X3
1
6. 5521
2. 2063
2. 97
0. 0053
X4
1
1. 3222
2. 2063
0. 60
0. 5527
XI *X1
1
O. 50007593
3. 5779
0. 14
0. 8879
XI #X2
J
1. 1295
1.2084
0. 93
0. 3562
X2*X2
1
-3.5065
3. 5779
-0 98
0. 3336
X1*X3
1
2. 0704
1. 1557
1 79
0. 0816
X2*X3
1
0.30411686
1. 2084
0. 25
0. 8027
X3#X3
2. 3897
3. 5779
0. 67
0. 50B4
X1 *X4
1
-1. 0332
1. 1557
-0. 89
0. 3773
X2*X4
1
-0 62354713
1. 2004
-0. 52
0. 6090
X3*X4
1
0.12312500
1. 1557
0. 11
0. 9157
X4*X4
1
-2. 9040
3. 5779
-0. 81
0. 4223
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROD
XI
5
5317. 1780
1063. 4356
1. 99
0. 1036
X2
5
4695. 2836
939. 0567
1. 76
0. 1466
X3
5
7402. 6295
1480. 5259
2. 77
0. 0323
X4
0
1044.7816
200. 9563
0. 39
0. 8516
143


8
solar radiation. They also suggested that the ideal type of twining
tropical legume should have a strong stoloniferous or rhizomatous
growth habit in order to withstand frequent defoliation and heavy
damage by grazing animals. Whiteman (1969) pointed out that frequent
defoliation results in low yields of twining legumes such as glycine
and Siratro, whereas slight defoliation or absence of defoliation
results in a higher contribution of the legumes to the total pasture
production. In Africa, Draolu and Nabusin-Napulu (1980) studied the
effect of cutting intervals of 3, 6, 12, and 24 weeks and cutting
heights of 3.8, 7.5, 15.0, and 30.0 cm in mixed swards of guinea-
grass and Stylosanthes guianensis. They concluded that DM production
was greatly reduced under the lowest and most frequent cutting treat
ment. The amount of legume in the mixture also decreased as the
cutting intervals were increased and the legume was particularly
sensitive to close defoliation. In Australia, Mclvor et al. (1981)
found the same linear tendency with Desmodium intortum and Setaria
sphacelata mixtures under different cutting heights of 8 and 20 cm
and cutting intervals of 3, 6, and 9 weeks. The growth response of
desmodium was markedly depressed at the lowest cutting height of 8 cm
and the shortest cutting interval of 3 weeks. They concluded that
cutting at the height of 20 cm at intervals of 6 to 9 weeks was
necessary for the persistence of the desmodium in the mixture.
Bryan et al. (1971) indicated that the short growth habit of the
tropical legumes Stylosanthes humilis and Lotononis bainesii, which
may be shaded by taller companion grasses, benefits from heavy grazing
pressure which allows light penetration into the canopy. They also


Table 14. Analysis of variance, regression coefficients and probabilities for aerial
biomass (g DM/m ) for the dry season of 1979.
RESPONSE MEAN
311. 3709
ROOT MSE
97.7679
R-SQUARE
0. 53220424
COEF OF VARIATION
0.31396663
REGRESSION
DF
TYPE J SS
R-SQUARE
F-RATIO
PROB
LINEAR
4
333107
0. 4528
0. 71
0. 0001
QUADRATIC
4
9693. 7000
0. 0132
0. 25
0. 9056
CROSBPRODUCT
6
48811. 0601
0. 0663
0. 05
0. 5396
TOTAL REGRESS
14
37.1613
0. 5323
2. 93
0. 0047
RESIDUAL
DF
SS
MEAN SQUARE
F-RA110
PROB
LACK OF FIT
26
197777
7683. 7481
0. 532
0. 9043
PURE ERROR
10
144331
14433. 0972
TOTAL ERROR
36
344100
9558. 5673
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROB
INTERCEPT
1
2B4. 5735
24. 7269
11. 51
0. 0001
XI
1
-4. 131 1
9. 3320
-0. 44
0. 6606
X2
1
23. 5222
9. 3451
2. 52
0. 0164
X3
1
47. 5909
9. 3320
5. 10
0. 0001
X4
1
-4. 9342
9. 3320
-0. 53
0. 6002
X1*XI
1
5. 1012
15. 1335
0. 34
0. 7380
X1 *X2
1
7. 5316
5. 1113
1. 47
0 1493
X2#X2
1
-3. 7988
15. 1335
-0. 25
0. 8032
X1 #X3
1
5. 6375
4. 8884
1. 15
0. 2564
X2*X3
1
-2. 4 505
5. 1113
-0. 48
0. 6334
X3#X3
1
3. 7075
15. 1335
0. 24
0. 8079
X1#X4
1
4. 4289
4. 8BB4
0. 91
0. 3710
X2*X4
1
1. 9631
5. 1113
0. 38
0. 7032
X3*X4
1
3. 1137
4. 8884
0. 64
0. 5282
X4*X4
1
3. 2533
15. 1335
0. 21
0. 8310
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
42399. 9627
8479. 9925
0. 89
0. 4998
X2
5
85209. 7709
17041. 99B2
1. 78
0. 1412
X3
5
276393
55278 6810
5. 78
0. 0005
X4
5
15389. 0544
3077 8109
0. 32
0. 8964
129


Table 36. Analysis of variance, regression coefficients and probabilities for visual
estimation grass for the wet season of 1978.
RESPONSE MEAN
60. 1 JOS
ROOT MSE
10. 1617
R-SQUARE
0 10023747
COEF OF VARIATION
0.16704773
REGRESSION
1)1
TYPE I SS
R-SQUARE
F-RATIO
PRO0
LINEAR
4
1 06. 1627
0. 0344
0. 3B
0 8220
QUADRATIC
4
473. 4 330
0. 1044
1. 15
0. 3506
CROSBPRODUCT
6
187. 7741
0. 04 15
0. 30
0. 7310
TOTAL REGRESS
14
017. 5870
0. 1003
0. 57
0. 0736
RESIDUAL
DF
SS
MEAN SQUARE
F-RATIO
PROD
LACK OF FIT
26
2472. 0233
75. 0470
0. 782
0. 7076
PURE ERROR
10
1225. 3102
122. 5310
TOTAL ERROR
36
3717.3335
103. 2573
PARAMETER
DF
ESTIMATE
STD DEV
T-RATIO
PROD
INTERCEPT
1
57. 0547
2. 5700
22. 51
0. 0001
XI
1
-0. 47430047
0.76773462
~0. 47
0. 6277
X2
1
-0.07773062
0.97127312
-0. 08
0. 7367
X3
1
-0.37320743
0.96773462
-0 41
0. 6876
X4
1
-0.77750772
0.76773462
-0. 82
0. 4164
X1*X1
1
~2 0474
1. 5727
-1. 30
0. 2007
X1*X2
1
0.07748421
0.53125507
0. 15
0. 8847
X2*X2
1
0.06726673
1. 5727
0. 55
0. 5048
X1#X3
1
-0.53700333
0.50808283
-1. 06
0. 2757
X2*X3
1
-0.14866703
0. 53125507
-0. 28
0. 7012
X3#X3
1
-0 30273327
1. 5727
-0. 24
0. 8071
X1*X4
0.13241667
0. 50800203
0. 26
0. 7757
X2*X4
1
0.25414045
0 53125507
-0. 48
0 6353
X3*X4
1
0.27758333
0 50000203
0. 55
0 5082
X4#X4
J
2. 5337
1. 5727
1. 61
0. 1157
FACTOR
DF
SS
MEAN SQUARE
F-RATIO
PROB
XI
5
323. 2526
64.6505
0. 63
0. 6807
X2
5
65. 7067
13.1413
0. 13
0. 7053
X3
5
100. 0564
37.6113
0 36
0. 8676
X4
0
4 33. 7177
86. 7836
0. 84
0. 5300


164
Hardy, F. 1960. Report on a visit to the Riverine belt of Ecuador.
Inter-American Institute of Agricultural Science. Turrialba,
Costa Rica. Rpt. No. 37. 112 p.
Harris, W. 1978. Defoliation as a determinant of the growth,
persistence and composition of pasture. Iri J. R. Wilson (ed.)
Plant relations in pastures. CSIRO, Brisbane, Australia. P. 67-84.
Heady, H. F. 1970. Grazing systems: Therms and definitions. J.
Range Manage. 23:81-86.
Heath, M. E. 1978. Grassland agriculture. In M. E. Heath, D. S.
Metcalfe, and R. F. Barnes (eds.) Forages. Iowa State Univer
sity Press, Ames, Iowa. P. 13-20.
Hodgson, J., and J. H. Ollerenshaw. 1969. The frequency and severity
of defoliation of individual tillers in serstocket swards. J.
Br. Grassld. Soc. 24:81-86.
Holdridge, L. R. 1967. Life zone ecology. Tropical Science Center,
San Jose, Costa Rica. 86 p.
Hudgens, R. E. 1973. The compatibility, persistence and nutritive
value of grass-legume associations in the wet-dry tropics of
coastal Ecuador. M. S. Thesis. University of Florida, Gaines
ville, Florida. 120 p.
Humphreys, L. R. 1978. Tropical pasture and fodder crops. In
W. J. A. Payne (ed.) Intermediate tropical agriculture series.
Longman Group Limited, London. P. 7-78.
. 1980. A guide to better pasture for the tropics and
sub-tropics. Wright Stephenson and Co., Australia. 96 p.
INIAP (Instituto Nacional de Investigaciones Agropecuarias). 1972-
1980. Informes tcnicos anuales del programa de pastos y gana
dera de la Estacin Experimental Tropical Pichilingue. Quito,
Ecuador.
Jensen, P. N., and C. M. Schumacher. 1970. Changes in prairie plant
composition. J. Range Manage. 22:57-60.
Jones, R. M. 1979. Effect of stocking rate and grazing frequency
on a Siratro (Macroptilium atropurpureum)/Setaria anceps cv
Nandy pasture. Aust. J. Exp. Agrie. Anim. Husb. 19:318-320.
Jones, M. B., and L. M. M. de Freitas. 1970. Respostas de quatro
leguminosas tropicais a fostoro, potassio e calcario num Latos-
solo Vermelho-Amarelo de campo cerrado. Pesquisa Agropecuaria
Brasileira. Serie Agr. 5:91-99.
Kennedy, R. K. 1972. The sickledrat: a circular quadrat modifica
tion useful in grassland studies. J. Range Manage. 25:312-313.


168
Snyder, G. H., and A. E. Kretschmer. 1974. Tropical legume response
to lime and superphosphate in oldsmar fine sand. Soil Crop Sci.
Soc. Fla. Proc. 34:63-66.
Spedding, C. R. W. 1965. The physiological basis of grazing manage
ment. J. Br. Grassld. Soc. 24:81-86.
Stobbs, T. H. 1969. Beef production from pastures: Leys in Uganda.
J. Br. Grassld. Soc. 24:87-91.
Tergas, L. E. 1968. The effect of nitrogen fertilization on the
movements of nutrients from a tropical grass under soil moisture
stress in a hot savanna. Ph.D. Dissertation. University of
Florida, Gainesville, Florida. 118 p.
1976. Factores que afectan la persistencia de las
leguminosas en associaciones con gramineas tropicales. In
Conferencia Anual Sobre Ganaderia y Avicultura en America
Latina. University of Florida, Gainesville, Florida.
. 1977. Importancia del azufre en la nutricin mineral
de leguminosas forraheras tropicales. Turrialba. 27:63-69.
t'Mannetje, L. 1974. Relations between pastures attributes and live
weight gains on a subtropical pasture. XII Int. Grassld. Cong.
Proc., Moscow, Russia. P. 299-304.
. 1976. An introduction to grassland vegetation and its
measurement. In L. t'Mannetje (ed.) Measurement of grassland
vegetation and animal production. Commonwealth Agrie. Bureaux.
Famhan Royal, Bucks, England. P. 1-7.
. 1978. The role of a legume based pasture for beef produc
tion in the tropics. IV World Conf. Anim. Prod., Buenos Aires,
Argentina.
Tothill, J. C. 1978. Comparative aspects of the ecology of pastures.
In J. R. Wilson (ed.) Plant relations in pastures. Commonwealth
Scientific and Industrial Research Organization, Brisbane, Austra
lia.. P. 385-404.
, and M. L. Petersen. 1962. Botanical analysis and sampling:
Tame pastures. In Pasture and range research techniques. Ithaca,
New York. P. 109-134.
Tow, P. G. 1967.
and legume.
Controlled climate comparisons of a tropical grass
Neth. J. Agrie. Sci. 15:141-154.
Ventura, M., J. E. Moore, 0. C. Ruelke, and D. E. Franke. 1975.
Effect of maturity and protein supplementation of voluntary
intake and nutrient digestibility of pangla digitgrass hays.
J. Anim. Sci. 40:769-774.


87
During the month of June and part of July of 1979, a severe
attack of red spider was observed, especially on centro, while
glycine appeared to be more tolerant and was damaged less than centro
which led to the dominance of glycine during this whole season.
Linear and quadratic effects of rest period (X2) were observed (P < 0.01)
while the linear effect of grazing pressure (X^) was noted (P < 0.05).
Field observations suggested that rest period was a greater determinant
of legume survival, productivity, and persistence than was grazing
pressure.
For the wet season of 1980 (January-June), there was also a
decrease in the average legume yield at 294 kg ha In the last
wet season, some pastures showed a complete dominance of grass,
completely excluding both legumes and weeds. These changes were
observed on treatments with the longest rest period which was con
firmed by the analysis which showed rest period (X^) with both linear
and quadratic effects (P < 0.01). The effect of grazing pressure
(X^) was linear (P < 0.05) which suggested that rest period was the
major determinant for the survival and productivity of the legumes
(Figs. 10 and 11).
The treatments with continuous grazing or short rest periods
in combination with the highest grazing pressures drastically affected
the yields of legumes, grasses, and weeds. This was especially true
for the planted grasses and legumes, which were in some cases almost
completely eliminated by the close and frequent defoliation which
removed the major portion of the young active leaf material and apical
meristems leading to a reduction in rate of recovery and ability to


Table 6.continued.
Treatments
1978
1978
1979
1979
1980
D/Gf
(x1)
D/Rf
(x2)
G/Pf
(X3) % BW
F+ -i
kg ha
No.
Reps
w/st
D/S$
w/st
D/Si
w/st
25
1
0
1.6
400
1
1180
670
kg ha
490
210
220
26
28
0
1.6
400
1
1110
400
740
390
120
27
1
56
1.6
400
2
1320
2040
4120
2040
3330
28
28
56
1.6
400
2
1520
2250
3890
2410
4010
29
1
0
8.3
400
1
1160
2040
2480
2970
2890
30
28
0
8.3
400
1
1300
1510
2620
2460
3060
31
1
56
8.3
400
2
1440
4260
6530
3580
7480
32
28
56
8.3
400
2
1590
4180
5780
5720
6160
33
1
28
5.0
200
1
1130
1900
2580
2270
2210
34
28
28
5.0
200
1
860
1490
2250
1860
2450
35
14
0
5.0
200
1
1500
1140
1380
1080
1500
36
14
56
5.0
200
1
1930
2870
3760
3300
4040
37
14
28
1.6
200
1
1140
1060
1230
1220
780
38
14
28
8.3
200
1
1390
2010
2750
2470
2600
39
14
28
5.0
0
1
1710
1700
2610
1930
2630
40
14
28
5.0
400
1
1470
1420
2780
2280
2080
41
14
28
5.0
200
3
1270
1780
2370
1640
2290
fD/G = days grazing, D/R = days rest,G/P= grazing pressure, F = fertilizer
W/S = wet season, D/S = dry season.


123
The collection of data started in May 1978 and ended in June
1980. Aerial biomass (DM), available forage (DM), grass yield (DM),
legume yield (DM), yield of weeds (DM), grass percentage and legume
percentage were estimated for each grazing cycle by a double-sampling
procedure. During the first 31/2 months, 15 random observations of
2
one m each were taken with a forage disk meter. From these sampling
units three were randomly selected and clipped at ground level for
actual yield determinations (DM) and percent composition. From
September 1978 the number of samples taken was increased to 30 units
of the same size and five sampling units out of the 30 were randomly
selected and clipped for the above determinations. These samples
were later hand separated into their components and dried for 20
hours. The sum of the dry weights of the components yielded the
total dry weight of the sample. These values were used for aerial
biomass (DM), available forage (DM), and for the individual component
yields (DM). Estimations for growth during grazing were made to
achieve the total available forage (DM).
Visual estimates of botanical composition were also taken in
order to determine the amount of the individual components in terms
of percentage. The visual estimate of percent yield was made for
the components grasses, legumes, and weeds.
From the results of this experiment based on the information
obtained from the responses which were measured, the following con
clusions appear to be justified.
Aerial biomass (DM) and available forage (DM) were increased
as the rest periods increased and grazing pressure decreased (forage


80
100 kg BW, and 0 kg ha ^ of superphosphate), respectively. The legume
yield was increased as the result of reducing the grazing pressure
(P < 0.01). There was an interaction between days grazing (X^) and
grazing pressure (X^) but this interaction did not occur again during
the remaining three seasons. There appeared to be no linear effect
for days grazing (X^) days rest (X^), and fertilizer level (X^).
The linear components of the model accounted for only 31% of the
total variation, while the quadratic effects represented only 1% and
the interactions 12% (Appendix Table 27).
During the dry season of 1978, there was a tendency for the
legume yield to increase as the grazing pressures decreased.
Legume Yield (DM) for the Wet Season of 1979
The legume yield for the wet season of 1979 for each treatment
combination is presented in Table 7. The legume yield during the
second wet season varied from 80 kg DM ha 1 to 970 kg DM ha ^ for
treatments 23 (1 day grazing, 56 days rest, 8.3 kg DM on offer/100
kg BW, and o kg ha ^ of superphosphate) and 34 (28 days grazing,
28 days rest, 5.0 kg DM on offer/100 kg BW, and 200 kg ha of
superphosphate), respectively.
Only the length of rest period (X£) and grazing pressure (X^)
were found to have an effect upon legume yield (P < 0.01). There
were no effects of X^ and X^ nor any of the interactions between
the experimental variables.
The linear components of the model accounted for 34% of the
total variation, while the quadratic and interaction effects represented


55
(1 day grazing, 56 days rest, 8.3 kg DM on offer/100 kg BW, and 400
kg ha of superphosphate). During this particular season, the amount
of biomass production was again most affected by the length of rest
period and by the levels of grazing pressure. The highest values of
biomass production corresponded to the treatment combinations of long
rest periods and low grazing pressures. For each 14 days increase in
rest period, there was an increase of 550 kg DM ha ^ of biomass produc
tion, while for each unit (1.6 kg DM) decrease in grazing pressure
there was a 450 unit increase in biomass production. The analysis of
variance for the wet season of 1980 is presented in Appendix Table 15.
Summary of Biomass Production (DM)
In comparison of aerial biomass production (DM) among seasons
and years, large differences, especially between wet and dry seasons,
are observed. The principal environmental factors involved in these
differences between seasons are precipitation, temperature, and solar
radiation (see Fig. 2 of Materials and Methods). Every year the rainy
season begins in the second half of December, initially with light
showers and then increasing in amount and intensity reaching the
highest peak of precipitation in February or March. After this time
the amount and duration of the showers decreases until it reaches
almost zero during the second half of June. About 90% of the total
precipitation falls from December to June with the remaining months
almost completely dry. The second important factor is temperature
which is always higher during the rainy season reaching an average
of 26C during the wet season, while the average during the dry season


2
pasture. They increase or maintain soil fertility due to their
ability to fix N, improve the quality of the diet grazed by the
animals, provide better seasonal distribution of the forage through
out the year, especially where alternate wet-dry seasons occur.
Therefore, a system which combines adequate grazing management
practices with high-yielding grasses and legumes growing in mixtures
which provide feed during the whole year is needed to increase
carrying capacity and the final output of the land.
The objectives of this research were
1) To determine the response of a tropical legume-grass
mixture to various treatment combinations of grazing
management factors length of grazing period, length of
rest period, grazing pressure, and levels of phosphorus
fertilization.
2) To determine the proper management strategy for adapted
forage species to attain the optimum legume contribution
as a component in the mixture, and
3) To measure the results of legume-grass mixture in terms
of aerial biomass, available forage, and botanical compo
sition of the forage on offer.


RESPONSE OF A TROPICAL LEGUME-GRASS ASSOCIATION
TO SYSTEMS OF GRAZING MANAGEMENT AND LEVELS OF
PHOSPHORUS FERTILIZATION
BY
RAUL ALONSO SANTILLAN
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
1983