Citation
Germplasm Characterization of Arachis pintoi Krap. and Greg. (Leguminosae)

Material Information

Title:
Germplasm Characterization of Arachis pintoi Krap. and Greg. (Leguminosae)
Creator:
CARVALHO, MARCELO AYRES
Copyright Date:
2008

Subjects

Subjects / Keywords:
Callus ( jstor )
Crop reports ( jstor )
Flowers ( jstor )
Forage ( jstor )
Forage evaluation ( jstor )
Genetic diversity ( jstor )
Peanuts ( jstor )
Plant morphology ( jstor )
Seed pods ( jstor )
Species ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Marcelo Ayres Carvalho. Permission granted to University of Florida to digitize and display 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.
Embargo Date:
12/18/2004
Resource Identifier:
57722311 ( OCLC )

Downloads

This item has the following downloads:

carvalho_m ( .pdf )

carvalho_m_Page_132.txt

carvalho_m_Page_006.txt

carvalho_m_Page_087.txt

carvalho_m_Page_091.txt

carvalho_m_Page_111.txt

carvalho_m_Page_035.txt

carvalho_m_Page_141.txt

carvalho_m_Page_102.txt

carvalho_m_Page_123.txt

carvalho_m_Page_136.txt

carvalho_m_Page_076.txt

carvalho_m_Page_113.txt

carvalho_m_Page_144.txt

carvalho_m_Page_153.txt

carvalho_m_Page_021.txt

carvalho_m_Page_088.txt

carvalho_m_Page_109.txt

carvalho_m_Page_025.txt

carvalho_m_Page_138.txt

carvalho_m_Page_067.txt

carvalho_m_Page_024.txt

carvalho_m_Page_129.txt

carvalho_m_Page_090.txt

carvalho_m_Page_126.txt

carvalho_m_Page_075.txt

carvalho_m_Page_114.txt

carvalho_m_Page_115.txt

carvalho_m_Page_063.txt

carvalho_m_Page_078.txt

carvalho_m_Page_083.txt

carvalho_m_Page_062.txt

carvalho_m_Page_057.txt

carvalho_m_Page_038.txt

carvalho_m_Page_085.txt

carvalho_m_Page_079.txt

carvalho_m_Page_004.txt

carvalho_m_Page_116.txt

carvalho_m_Page_018.txt

carvalho_m_Page_007.txt

carvalho_m_Page_147.txt

carvalho_m_Page_046.txt

carvalho_m_Page_059.txt

carvalho_m_Page_061.txt

carvalho_m_Page_086.txt

carvalho_m_Page_047.txt

carvalho_m_Page_001.txt

carvalho_m_Page_117.txt

carvalho_m_Page_036.txt

carvalho_m_Page_142.txt

carvalho_m_Page_010.txt

carvalho_m_Page_030.txt

carvalho_m_Page_106.txt

carvalho_m_Page_029.txt

carvalho_m_Page_049.txt

carvalho_m_Page_043.txt

carvalho_m_Page_124.txt

carvalho_m_Page_023.txt

carvalho_m_Page_016.txt

carvalho_m_Page_143.txt

carvalho_m_Page_042.txt

carvalho_m_Page_066.txt

carvalho_m_Page_125.txt

carvalho_m_Page_146.txt

carvalho_m_Page_121.txt

carvalho_m_Page_072.txt

carvalho_m_Page_048.txt

carvalho_m_Page_103.txt

carvalho_m_Page_093.txt

carvalho_m_Page_094.txt

carvalho_m_Page_040.txt

carvalho_m_Page_119.txt

carvalho_m_Page_044.txt

carvalho_m_Page_068.txt

carvalho_m_Page_045.txt

carvalho_m_Page_112.txt

carvalho_m_Page_137.txt

carvalho_m_Page_082.txt

carvalho_m_Page_051.txt

carvalho_m_Page_002.txt

carvalho_m_Page_031.txt

carvalho_m_Page_071.txt

carvalho_m_Page_013.txt

carvalho_m_Page_026.txt

carvalho_m_Page_019.txt

carvalho_m_Page_149.txt

carvalho_m_Page_060.txt

carvalho_m_Page_127.txt

carvalho_m_Page_118.txt

carvalho_m_Page_032.txt

carvalho_m_Page_098.txt

carvalho_m_Page_148.txt

carvalho_m_Page_110.txt

carvalho_m_Page_135.txt

carvalho_m_Page_133.txt

carvalho_m_Page_041.txt

carvalho_m_Page_122.txt

carvalho_m_Page_104.txt

carvalho_m_Page_131.txt

carvalho_m_Page_107.txt

carvalho_m_Page_108.txt

carvalho_m_Page_056.txt

carvalho_m_Page_154.txt

carvalho_m_Page_034.txt

carvalho_m_Page_150.txt

carvalho_m_Page_139.txt

carvalho_m_Page_089.txt

carvalho_m_Page_097.txt

carvalho_m_Page_008.txt

carvalho_m_Page_027.txt

carvalho_m_Page_005.txt

carvalho_m_Page_095.txt

carvalho_m_Page_152.txt

carvalho_m_Page_092.txt

carvalho_m_Page_022.txt

carvalho_m_Page_052.txt

carvalho_m_Page_140.txt

carvalho_m_Page_037.txt

carvalho_m_Page_130.txt

carvalho_m_Page_053.txt

carvalho_m_Page_058.txt

carvalho_m_Page_073.txt

carvalho_m_Page_099.txt

carvalho_m_Page_033.txt

carvalho_m_Page_084.txt

carvalho_m_Page_077.txt

carvalho_m_Page_081.txt

carvalho_m_Page_050.txt

carvalho_m_Page_080.txt

carvalho_m_Page_151.txt

carvalho_m_Page_017.txt

carvalho_m_Page_055.txt

carvalho_m_Page_014.txt

carvalho_m_Page_064.txt

carvalho_m_Page_145.txt

carvalho_m_Page_020.txt

carvalho_m_Page_105.txt

carvalho_m_Page_054.txt

carvalho_m_Page_003.txt

carvalho_m_pdf.txt

carvalho_m_Page_101.txt

carvalho_m_Page_134.txt

carvalho_m_Page_015.txt

carvalho_m_Page_065.txt

carvalho_m_Page_039.txt

carvalho_m_Page_011.txt

carvalho_m_Page_128.txt

carvalho_m_Page_028.txt

carvalho_m_Page_069.txt

carvalho_m_Page_074.txt

carvalho_m_Page_012.txt

carvalho_m_Page_009.txt

carvalho_m_Page_096.txt

carvalho_m_Page_070.txt

carvalho_m_Page_100.txt

carvalho_m_Page_120.txt


Full Text












GERMPLASM CHARACTERIZATION OF Arachispintoi Krap. and Greg.
(LEGUMINOSAE)















By

MARCELO AYRES CARVALHO


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

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Marcelo Ayres Carvalho

































To my wife, Aline Varandas, who bravely decided to follow me in pursuing this dream.















ACKNOWLEDGMENTS

The author would like to acknowledge the people of Brazil for financially

supporting him during his Ph.D. program, through a scholarship awarded by the National

Council for Scientific and Technological Development (CNPq).

He also would like to thank the Brazilian Agricultural Research Corporation

(EMBRAPA) in the person of its Director-president, Dr. Clayton Campanhola, for the

financial support and continuous effort to improve the training and quality of its

employees.

Marcelo thanks his parents, Magaly and Milton; sisters Patricia and Juliana;

grandparents Orlando; Maria Analia, Policarpo, and Araci; and all his family for being

always present when needed; for helping build his character; and finally, for guiding him

through his life in the right direction, and making him a better human being.

He is deeply grateful to the Quesenberry family for their support and kindness

while he lived in Gainesville. He is especially thankful to Dr. Ken Quesenberry, his

advisor, for the guidance and assistance required for the conclusion of his Ph.D. program.

He would like to acknowledge the members of his committee: Dr. Sollenberger, Dr.

Gallo-Meagher, Dr. Blount, and Dr. Lyrene, for their numerous and valuable

contributions to this project and, more important, to the development of this student.

He is grateful to Dr. Roy Pittman from the USDA-NPGS/PI Station, located in

Griffin, Ga, who generously made the germplasm of A. pintoi available for this research.









He would like to thank the research assistants Loan Ngo, Judy Dampier, Jeff Seib,

and Richard Fethiere for their assistance in the laboratory, greenhouse and field activities.

Thanks are also extended to fellow graduate students Dr. Eastonce Gwata, Dr. Liana

Jank, Dr. Eduardo Carlos, Jose Dubeux, Joao Vendramini, Gabriela Luciani, and Sindy

Interrante for their friendship and support during the difficult times of this journey.

Marcelo would like to thank Dr. Pizarro, Dr. Mario Soter, Dr. Jose Valls, Dr. Zoby,

Dr. Lazarini, and many others for their encouragement and for being examples of a

positive attitude during his professional life.

Finally, he would like to thank all his friends in Brazil and here in Gainesville for

their support and encouragement.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ....... .. .............. viii

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

A B S T R A C T .............................................. ..........................................x iii

CHAPTER

1 IN TRODU CTION ....................................................... ..... ...... .. ........ .. 1

H ypotheses ................................................ 4
O bjectiv e s ..................................................................................................... .4
G o a ls ....................................................................................................... . 4

2 LITERATURE REVIEW .............................................................5

Plant Genetic Resources Characterization......................... ...............
The Genus Arachis .................................................... 12
The Forage Potential of the Genus Arachis .................................................. 14
A ra ch is p in to i ...............................................................18
Botanical characteristics..................................18
Agronomic characteristics.............. ............ ......... 19
Arachispintoi germplasm characterization.................... ...............22

3 MOLECULAR CHARACTERIZATION OF Arachispintoi GERMPLASM ..............27

In tro d u ctio n ........................................................................................................... 2 7
M materials an d M eth od s ......................................................................................... 30
Parent Plants ................................... .......................... ... ......... 31
Tissue Culture Regenerated Plants ............... ...............................32
P ro to c o l 1 ............................................................................................... 3 4
P ro to c o l 2 ............................................................................................... 3 5
R esu lt an d D iscu ssion s .......................................................................................... 3 6
Parent Plants .................................................. 36
Tissue Culture Regenerated Plants ............... ...............................44
Sum m ary and C onclu sions ................................................................................... 53









4 MORPHOLOGICAL CHARACTERIZATION OF Arachispintoi GERMPLASM.....56

Intro du action ...................... .. .. ......... .. .. ...............................................56
M materials and M methods ....................................................................... ..................57
R result and D iscu ssions .............................. ........................ .. ...... .... ............ 60
Sum m ary and C onclu sions .............................................................. .....................82

5 AGRONOMIC EVALUATION OF Arachis pintoi GERMPLASM ...........................84

In tro d u ctio n ........................................................................................................... 8 4
M material and M methods ........................................................... .. ............... 86
The G erm plasm .................................. ......... .... ...............86
Field Evaluation.......... .......................... ...... ............ 87
Adaptation and forage dry matter yield..................................................87
F orage nutritive value........................................................ ............... 89
Seed produ action ......... .......................................................... .. .... .. ... 89
Nematode Response Evaluation......... ..................... ..................... 89
Results and Discussion ..................... .................... .............. .. ............ 92
Adaptation and forage dry matter yield......... ....................................92
F orage nutritive value...................... .. ............................... ............... 101
Seed production ......... ........................................ ...... .. .... . ......... 103
Nematode response evaluation..... .......... ........................................ 106
Sum m ary and Conclusions ......................................................... .............. 113

6 C O N C L U SIO N S....... ............................................................................... .......... 114

APPENDIX

A LIST OF A. pintoi GERMPLASM EVALUATED AT UF ................................... 119

B CTAB DNA EXTRACTION PROTOCOL ..........................................................121

C MORPHOLOGICAL DESCRIPTORS CORRELATION TABLE.........................123

D CLIMATOLOGICAL DATA AT THE FORAGE RESEARCH UNIT IN
GAINESVILLE-FL, DURING THE PERIOD OF THE AGRONOMIC
EVALUATION FIELD TRIAL. ........................................ ......................... 127

LIST OF REFERENCES ......... ...................................... ........ .. ............... 129

BIO GR A PH ICA L SK ETCH .................................... ........... ......................................139















LIST OF TABLES


Table p

2-1 Proposed gene pools of the genus Arachis based on Arachis hypogaea breeding
perspective ............... ..... .. ...... ..................... .... .... ......... 15

3-1 List of Operon Technologies primers used to amplify Arachispintoi DNA
reg io n s .............................................................................. 3 3

3-2 Tissue culture protocols used to regenerate Arachis pintoi plants......................... 35

3-3 List of Operon primers, number and size of amplified bands, and number of
bands accession-specific ofArachispintoi germplasm genomic DNA...................36

3-4 Characteristics of RAPD patterns of the 35 Arachispintoi germplasm accessions.38

3-5 Gene frequency of the 100 RAPD locus ofArachispintoi germplasm accessions .40

3-6 Nei's gene diversity and by Shannon-Weaver diversity index of RAPD loci .........42

3-7 Analysis of variance table of callus rating and weight produced from
Arachispintoi leaf discs incubated on two different tissue culture media..............46

3-8 Callus rating and weight of Arachispintoi tissue culture incubated in two
different protocols ...................... .... ................ ................. .... ....... 46

3-9 Callus rating and callus weight of Arachispintoi callus induced on Protocol 1......48

3-10 Callus rating and callus weight of Arachispintoi callus induced on Protocol 2......49

3-11 Number of regenerated plants of Arachispintoi accessions cultured on two
different protocols ...................... .... ................ ................. .... ....... 52

4-1 Morphological descriptors applied to Arachispintoi accessions.............................58

4-2 Morphological characteristics of Arachispintoi germplasm accessions .................61

4-3 Correlations among morphological descriptors ofArachispintoi germplasm ........67

4-4 Simpson and Shannon-Weaver diversity indices for Arachispintoi
m orphological descriptors ............................................ ..... ........................ 71









4-5 Vector loadings and percentage of variation explained by the first five principal
components for morphological characteristics ofArachispintoi.............................73

4-6 Morphological characteristics of Arachispintoi accession groups obtained by the
clu ster an aly sis ..................................................................... 7 9

5-1 Gall index, gall size, percent galled area and damage index values ......................91

5-2 Winter survival evaluations ofArachispintoi at the forage research unit in
Gainesville-FL ............. ...... .. .................. ........ ...... ....... 93

5-3 Plot coverage and plant height before forage dry matter yield evaluations of
Arachispintoi at the forage research unit in Gainesville-FL in 2003 ....................95

5-4 Analysis of variance table of forage dry matter yield (FDMY) evaluations of
Arachispintoi at the forage research unit in Gainesville-FL in 2003 ....................96

5-5 Forage dry matter yield ofArachispintoi germplasm at the Forage Research
U nit near G ainesville, FL in 2003 ........................................ ........................ 97

5-6 Analysis of variance table of the annual forage dry matter yield (FDMY) of
Arachispintoi at the forage research unit in Gainesville-FL in 2003 ....................99

5-7 Total forage dry matter yield ofArachispintoi at the Forage Research Unit
near G ainesville, FL in 2003 ............................................................................ 100

5-8 Analysis of variance table of crude protein (CP) and in vitro organic matter
digestion (IVOMD) of eight-week regrowth ofArachispintoi at the Forage
Research Unit near Gainesville, FL in 2003 .................................. ............... 101

5-9 Crude protein (CP) and in vitro organic matter digestibility (IVOMD) of eight-
week regrowth of Arachispintoi at the Forage Research Unit near Gainesville,
F L in 2 0 0 3 ..............................................................................................................1 0 3

5-10 Analysis of variance table of seed production ofArachispintoi at the Forage
Research Unit near Gainesville, FL in 2003 and 2004 ................. .............. 104

5-11 Seed production ofArachispintoi at the Forage Research Unit near Gainesville,
F L in 2003 and 2004 ............................................ ........... ..... ........105

5-12 Analysis of variance table ofArachispintoi germplasm reaction to
M eloidogyne arenaria ......... ................. ...................................... ............... 107

5-13 Reaction ofArachispintoi germplasm to Meloidogyne arenaria race 1 ...............108

5-14 Analysis of variance table ofArachispintoi germplasm reaction to
M eloidogyne javanica ....................... ......... ...... ... ..................................... 109

5-15 Reaction of Arachis pintoi germplasm to Meloidogyne javanica........................ 110









5-16 Analysis of variance table ofArachispintoi germplasm reaction to
M eloidogyne incognita ........... ................. ................. ................. .....................

5-17 Reaction of Arachis pintoi germplasm to Meloidogyne incognita......................... 112

A-1 List of A. pintoi germ plasm evaluated at UF ........................................................ 119

C-1 M orphological descriptors correlation table .................................. ............... 124

D-l Climatological data at the forage research unit in Gainesville, FL, during the
period of the agronom ic evaluation field trial....................................................... 127
















LIST OF FIGURES


Figure pge

2-1 Area of natural occurrence ofArachispintoi. ............... ................................18

2-2 Arachispintoi plant characteristics ............................................... ............... 20

3-1 Ratings scale applied to callus pieces ofArachispintoi explants incubated in two
different protocols. ........................................... ... .... ........ ......... 33

3-2 Leaflet cutter and callus induction dishes of Protocol 1 and 2 ofArachispintoi
accessions. ............................................................................34

3-3 RAPD band profile of 24 accessions of Arachispintoi amplified by primer A4 ....39

3-4 Dendogram illustrating the genetic relationships among 35 Arachispintoi
accessions based on Nei's genetic distance obtained by 100 RAPD markers
resolved by 8 random primers and generated by the UPGMA method .................45

3-5 Callus growth of Arachispintoi PI 604812 on two different tissue culture media..47

3-6 Shoot regeneration of Arachispintoi accession PI 604812 subculture in medium
M S+lg L1 BA............................ ............. ....... ..... .........50

3-7 Representation of the different steps in the process used to tissue culture Arachis
p in to i .............................................................................................. 5 0

4-1 Flower standard colors ofArachispintoi germplasm. .........................................63

4.2 Flower sizes and hypanthium colors displayed by Arachispintoi germplasm. .......63

4.3 Stem characteristics ofArachispintoi germplasm ................................................64

4-4 Leaflet characteristics ofArachispintoi germplasm.............................................65

4-5 Seed characteristics ofArachispintoi germplasm. ................................................ 66

4-6 Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC 1 and PC2. ............................................... .............................. 74

4-7 Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC1 and PC3 ...................... ......... .............. ............... 75









4-8 Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC2 and PC3 ................................................................. .....................76

4-9 Dendogram of 35 Arachispintoi accessions based on morphological descriptors
and the first nine principal components. ...................................... ............... 78

4-10 Group 1 representative accession (PI 497541)............................................ ............ 80

4-11 Group 2 representative accession (PI 604814)............................................. 80

4-12 Group 3 representative accession (PI 604798)............................................. 81

4-13 Group 4 representative accession (PI 604817)............................................. 81

5-1 Forage dry matter yield (FDMY) of Arachis pintoi and Arachis glabrata cultivars
at the forage research unit in Gainesville-FL in 2003................... .................98















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

GERMPLASM CHARACTERIZATION OF Arachis pintoi Krap. and Greg.
(Leguminosae)

By

Marcelo Ayres Carvalho

December 2004

Chair: Kenneth H. Quesenberry
Major Department: Agronomy

Arachispintoi Krap. and Greg. is a herbaceous, perennial legume, exclusively

native to Brazil. It is considered a multiple use legume, being grown for forage; ground

cover in fruits orchards, forest, and low tillage systems; erosion control; and ornamental

purposes. Although several cultivars have been released in different countries, little is

known about the genetic diversity of the germplasm stored at genebanks. Our objective

was to characterize and evaluate the genetic diversity of the germplasm of 35 accessions

of Arachispintoi at molecular, morphological, and agronomic levels. A. pintoi accessions

were used to study the genetic diversity at the molecular level using RAPD markers.

Concurrently, two tissue culture protocols were evaluated for their organogenesis ability,

and capacity to generate somaclonal variation. From the original 18 primers tested,

amplifications were obtained with eight, which amplified 100 polymorphic bands.

Average genetic distance was estimated as 0.36, indicating that a large amount of genetic

diversity exits among the accessions. The accessions were grouped by their genetic









similarities into four distinct groups. Callus induction was achieved on two different

Murashige and Skoog basal protocols, and shoot regeneration was achieved for several

accessions on both media. Regenerated plants recovered in both protocols presented no

differences in their RAPD band profile. Morphological characterization using data from

stems, leaves, flowers, pegs, pods, and seeds demonstrated that the germplasm presented

great morphological variability. Principal component analysis was able to discriminate

the accessions in terms of three dimensions, and Cluster analysis differentiated four

distinct groups. Average dry matter yield in 2003 was 4.35 Mg ha-1, and ranged from zero

to 9.10 Mg ha-1. Average crude protein was 180 g kg-1 of DM and ranged from 130 to

220 g kg-1. The average value of in vitro organic matter digestion was 670 g kg-1 of DM

and ranged from 600 to 730 g kg-1. Some accessions produced high seed yields reaching

values above 1.00 Mg ha-1. Average seed production was 0.32

Mg ha-l in 2003, and 0.43 Mg ha-l in 2004. A. pintoi germplasm presented high levels of

resistance to the nematode species Meloidogyne arenaria, M. javanica, and M. incognita.














CHAPTER 1
INTRODUCTION

Biodiversity can be defined as the total variation found within all living organisms

and their habitats. It can be accessed at three different levels: communities (environment),

species, and genes. When accessing biodiversity at the species level, we are interested in

observing differences among individuals or populations of that particular species. This

can be referred to as the genetic diversity of the species. Thus, we can consider genetic

diversity as a form of biodiversity.

Genetic diversity is associated with the degree of differentiation among individuals

in a population at their genetic material level. The genetic material corresponds to the

DNA, genic or cytoplasmic, called the genotype. Expression of genes contained in the

DNA is the result of interaction between the environment and the genotype, which is

called the phenotype. Genetic diversity is important because it will enable evolution and

adaptation of the species to an ever changing environment. Thus it is essential for the

long-term survival of a species.

Since very early in the history of the world, humans have exploited the genetic

diversity of plants, primarily as sources of food, and then to improve their land races and

cultivars. With the advance of modern agriculture, plant breeders around the world began

to collect genetic diversity of the most important food crops and to store these materials

at national research institutes and local institutions. By the late 1960s, the Food and

Agriculture Organization (FAO) of the United Nations (UN) was concerned about the

problem of conserving plant genetic diversity, and a series of conferences were arranged.









As the result of these conferences, in 1974 the International Board for Plant Genetic

Resources (IBPGR) was established. Coordinated actions were then made to increase

collection efforts and to increase the creation of germplasm banks at national and global

levels. As a consequence of these efforts, about 1300 ex-situ genebanks were created.

Today, they together conserve an estimated 6.1 million accessions (Hawkes et al. 2000).

Although actions to increase conservation of genetic diversity were very successful,

the use and characterization of these resources were not. In 1996, FAO held the Fourth

International Technical Conference on Plant Genetic resources in Leipzig/Germany.

During this event, a global plan for the conservation and sustainable use of plant genetic

resources for food and agriculture was prepared and formally adopted by more than 150

countries. Among other issues, a call was made for expanding characterization and

evaluation, and for increasing the number of core collections to facilitate the use of the

stored germplasm (FAO, 1996).

Information obtained in studies of germplasm can be used for accession

classification, analysis of genetic diversity, and studies of the genetic divergence. Studies

of genetic diversity are important because they are a tool for genetic improvement

allowing the efficient use of the available germplasm of a species.

The genus Arachis has great importance at the world level. Arachis hypogaea L.

(peanut), the species with greatest importance in the genus, is cultivated commercially in

more than 80 countries, supplying food with high protein levels and oil of excellent

quality. According to FAO, world production of peanut is about 23 million metric tons

annually.









Because of the importance of the peanut as a food crop, great importance is given

to the whole genus, and thousands of accessions of the common peanut and its wild

relatives have been collected and stored in genebanks, especially those located at

ICRISAT, USDA-NPGS and EMBRAPA.

Although A. hypogaea is the most important species of the genus, some of its wild

relatives also have agronomic potential. Some of these species are used as ground cover

or forage crops.

According to Kerridge (1994), the major research priorities in the genetics of

species with potential as forage crops in the genus Arachis are:

* More thorough studies of germplasm in the sections Caulorhizae and
Rhizomatosae.

* Rigorous characterization of genetic variability at the molecular, physiological, and
agronomic levels.

* Development of molecular markers for use in genotype identification and studies of
breeding behavior.

* Quantification of genetic variation, inheritance of important traits, and
identification of sources of traits of agronomic interest.

* Survey of accessible sources of genetic resistance for disease and insect pest
resistance, and survey of diseases in natural populations.

Arachispintoi is a perennial species of the section Caulorhizae, and it is considered

by many as the most promising of the wild species of the genus. It is a multiple use

legume, which has been used predominantly as a forage crop, with released cultivars in

several countries.

The importance of the genus, the agronomic potential presented by Arachispintoi,

and the lack of information about its germplasm (stored at the USDA-NPGS) were the

primary justification for this research.









The hypothesis, objective, and goals of this work are presented below.

Hypotheses

Genetic diversity is limited in the Arachispintoi germplasm stored at the USDA-

NPGS germplasm bank.

Objectives

To characterize the germplasm ofArachispintoi at molecular, morphological, and

agronomic levels.

Goals

* To quantify the genetic variability of accessions of Arachispintoi at the molecular,
morphological, and agronomic levels.

* To supply basic information to the breeding programs that are using the germplasm
to develop new improved cultivars.













CHAPTER 2
LITERATURE REVIEW

Plant Genetic Resources Characterization

Economic exploration of plants is essential for maintenance of a wide base of their

genetic resources, because the same ones compose the vital patrimony of the species. The

most important way to increase a species' productivity is to know the variability in its

existing germplasm. Incorporating new organisms into the group of useful plants, by

direct use or through programs of genetic improvement, requires use of available genetic

resources. These resources are collected to be used in breeding programs and not just to

be conserved (Patemiani, 1988).

According to Smith and Linington (1997) the cost to collect a single germplasm

accession in its country of origin and incorporate that accession into a local genebank is

about US $870. In addition, the Keystone Center (1991) estimated the cost to maintain an

individual germplasm accession is US $50 per year. The value of the material itself and

the aggregate costs to acquire and maintain the tremendous amounts of genetic diversity

stored in the genebanks around the world are enough to justify germplasm

characterization and evaluation.

After the IBPGR was established in 1974, great attention was given to collecting

germplasm of the most important food crops. This was particularly important because of

the changes brought about by the green revolution to producers everywhere. Land races

were rapidly being replaced by improved cultivars and the risk of losing these genetic

resources was great. Furthermore, the population explosion and the expansion of









agricultural lands to supply the needs of expanding urban populations also increased the

risk of genetic erosion in some ecosystems. As a result of these efforts, FAO (1998)

estimated that 6.2 million accessions of 80 different crop species were stored in 1,320

genebanks and related facilities in 131 countries. These numbers show that today,

germplasm conservation is well recognized issue, and is deemed important by most of the

world community.

However, plant genetic resource conservation is much broader than just collection

and storage. Characterization and evaluation should be an integral part of the general

scheme of genetic resource conservation, but because of the great effort made to increase

the number of accessions stored in the genebanks, these activities often have been

relegated to a second level. This was not intentional, but reflects the size that germplasm

collections have reached and the cost of complete evaluation. For instance, according to

Holbrook et al. (1993), the number of common peanut genotypes stored in the USDA-

NPGS genebank is about 7,500. However, the most recent data from the Germplasm

Resources Information Network (GRIN) account for 9,232 accessions.

Germplasm characterization consists of studies of eco-geographic and demographic

adaptation (Martins, 1984), and according to Solbrig (1980) involves mostly the

parameters of the vital cycle of the organism, genetic and physiological studies, plant

pathology, and yield evaluation, among other studies. Characterization often also

involves taxonomic confirmation and should produce an easy and quick way to

differentiate the germplasm, using highly heritable and visible traits (Hawkes et al.,

2000). Breeding programs should begin only after appropriate germplasm

characterization (Cameron, 1983).









Improved knowledge of the available germplasm provides essential information for

its more intense use. Characterization and evaluation of available resources will allow the

establishment of nuclear collections (core collections) that, by definition, embrace the

maximum genetic diversity contained in cultivated species and in the related wild

species, with the minimum possible number of accessions (Frankel and Brown, 1984).

Another positive result of characterization is the detection of duplications, which is

a serious problem especially when we have in mind the size of the collections and the

resources needed to maintain them. Plucknett et al. (1987) estimated that as few as 35%

of the accessions in world collections are actually distinct.

In summary, characterization is the best way to understand the variability contained

in a germplasm collection and to increase use of the germplasm by plant breeders. It is

also important in monitoring the genetic stability of the germplasm storage processes.

Characterization of germplasm can be based on molecular, biochemical, morphological,

and agronomic features.

The use of biochemical and molecular characterization can be a more precise way

to discriminate among accessions in germplasm collections. This is especially true in

species where the application of phenotypic morphological descriptors is delayed by very

slow growth or delay in reaching the reproductive stage, when many markers tend to be

evidenced (Valls, 1988).

During the past 25 years, many improvements in molecular biology techniques

have allowed the direct application of these methods in studies of genetic

characterization. According to Karp et al. (1997), the advent of the polymerase chain

reaction (PCR) permitted the development of several molecular technologies that can be









used with great success in detecting, characterizing and evaluating genetic diversity.

Several different types of molecular markers were generated after the advent of the PCR,

and they differ among themselves in the way they resolve genetic differences; in the type

of information that they generate; in the taxonomic levels where they should be applied;

and finally in costs, labor requirement, and training.

Although an extensive number of molecular markers are available, DNA-based

markers have greatly overtaken markers based on proteins or enzymes, because the latter

can be influenced by the environment.

According to Ford-Lloyd (2001), DNA-based markers can be classified into three

different categories:

* Non-PCR-based methods: Restriction fragment length polymorphism analysis
(RFLP) and variable number of tandem repeats (VNTRs) are some examples of this
category.

* Arbitrary (or semi-arbitrary) primed techniques: This is a PCR-based category
that uses random primers during the PCR reaction. The most well known and
widely used of these methods is the "Random Amplified Polymorphic DNA"
(RAPD).

* Site-targeted PCR: In this class, primers that amplify specific regions of the DNA
are used during the PCR reaction. Examples are single nucleotide polymorphism
(SNP's), microsatellite repeats (VNTR's), and many others.

Karp et al. (1997) also indicated that DNA-based markers can be useful in defining

an accession identity; defining the degree of similarity among individuals of an accession

or a group of accessions; and defining the presence of a particular allele or nucleotide

sequence in an individual, population, or taxon. Among all DNA-based markers, RAPD

have been extremely popular with plant genetic scientists. Reiter (2001) revealed that in a

recent bibliographic search more than 2,000 publications were listed where RAPDs

markers had been used.









According to Reiter, the popularity of this marker arises from the fact that the method is

very simple and the cost required for its application is low.

RAPDs are the product of PCR reaction where a single, short, sequence-arbitrary

primer (oligonucleotide 10mer usually) is used. Amplification happens by setting a low

annealing temperature (35-37C typical) in the thermal-cycling program. Amplification

products are separated by size in an electrophoresis agarose gel. A DNA fragment pattern

of low complexity is observed after ethidium bromide staining. Fragment patterns of

individuals of the same species are almost identical (with some exceptions where one

band is present in one individual and absent in another). These fragment polymorphisms

are heritable and can be classified as a new category of genetic marker (Williams et al.,

1990).

Most RAPD bands are amplification polymorphisms due a nucleotide base change

at one of the priming sites or an insertion/deletion event within the amplified region.

DNA amplification from heterozygous individuals at RAPD marker loci are normally

identical to the homozygous parent. Thus, RAPD markers are typically dominant

markers. Dominant markers are less informative than co-dominant markers for genetic

mapping in some types of segregating populations (Williams et al., 1990).

When working with several accessions amplified using the same 10mer primer,

similar DNA fragment patterns may be observed. However, it is not possible to know if

all fragments of the same size class have the same DNA sequence (and thus are allelic).

Bands from related germplasm lines are likely to be allelic. However, as genetic distance

increases so does the probability of non-allelism. Fragment allelism is an important









concern when RAPDs are used for population genetic studies, varietal fingerprinting, and

some forms of marker-assisted selection (Reiter, 2001).

Another criticism regarding RAPD is incomplete pattern reproducibility (Westman

and Kresovich, 1997). Any sequence-arbitrary amplification method will be considerably

less robust than conventional PCR for the following reasons:

* Multiple amplicons are present competing for available enzyme and substrate

* Low-stringent thermal-cycling permits mismatch annealing between primer and
template

To overcome these problems, it is important to create optimal and consistent

amplification conditions, and to optimize the quantity and quality of reaction components

(DNA, MgC1, Taq polymerase, etc).

Characterization of genetic variability can use conventional methods of evaluation

complemented by biochemical and molecular methods (Hulli, 1987; Marcon, 1988). It is

recognized that morphological data, despite its importance, is not always easily

understood at the level of genes (Simpson and Withers, 1986). The phenotypic expression

of botanical and agronomic characteristics, which in general have polygenic inheritance,

is the result of genotype x environment interactions (phenotypic plasticity). On the other

hand, simple verification of the existence of molecular markers does not imply

knowledge of their expression in the phenotype. Thus, phenotypic characterization of

accessions should still be used in germplasm evaluation. Thus morphological and

molecular methods present different advantages, and the complementation between them

can produce benefits that would not be obtained in separate analyses (Hulli, 1987).

Morphological characterization is the simplest way to gather information about

germplasm; and before molecular marker techniques became popular and accessible, it









was the primary way for plant genetic scientists to improve their knowledge about the

resources available to them.

The term morphological characterization refers to a broad subject that involves

taxonomy, botany, genetics and other disciplines. Sometimes, the first step in

characterizing a germplasm accession involves taxonomic confirmation. After this step,

highly heritable and visible botanical features are used to prepare a list of descriptors that

will be applied to the plants. Usually, qualitative traits that show little environmental

influence are preferred, but sometimes quantitative traits are also used. Previous

knowledge about the morphology and phenology of the species is required (Hawkes et

al., 2000).

The importance of morphological features was demonstrated by Mendel's work.

Mendel used simple morphological traits and studied their inheritance in hybrids

produced from distinct homozygous parents. The outcome of his work was the

demonstration that traits are transferred from parents to offspring by factors later called

genes. The discipline of genetics (and a new world) was opened by a work based on

morphological traits. Nowadays, Mendel's work and morphological traits are still used to

understand the mechanism of inheritance and genetics of traits in many different species.

Morphological descriptors must be universal and simple to score. The universality

of the descriptor will allow plant breeders and researchers working with plant genetic

resources to exchange information about germplasm accessions generated in different

places. To achieve this goal the International Plant Genetic Resources Institute (IPGRI),

in collaboration with crop specialists, published guides to descriptor selection and also

descriptor lists for more than 100 different species, including the most important food









crops and some wild relatives of the same (Hawkes et al., 2000). The IPGRI

(www.ipgri.cgiar.org) lists species with published descriptors, and these descriptors lists

can ultimately be used to generate germplasm catalogues and to build computational

databases of different species that can be accessed by the World Wide Web, contributing

to increased use of plant genetic resources.

Another level of germplasm characterization is agronomic research. This type of

research measures useful traits related to agronomic performance. These usually include

quantitative traits related to yield, disease resistance, and environmental stress tolerance

are measured. Although it seems insignificant, some authors make a distinction between

agronomic characterization and evaluation. Hawkes et al. (2000) suggested that

evaluation involves traits important for crop improvement, which may often be affected

by their interaction with the environment. They also suggested that one big difference

between characterization and evaluation is that the first should be carried out by the

germplasm curator and the latter by plant breeders and plant pathologists working outside

the genebank.

Therefore, it is important to exercise caution when using agronomic data generated

in different places at different times. The use of standard checks will permit a more

extensive comparison of the germplasm, but will never replace local evaluations,

especially when a more detailed analysis is needed (Ford-Lloyd and Jackson, 1986).

The Genus Arachis

The genus Arachis has great importance at the world level. Arachis hypogaea L.,

the most important species in the genus, is cultivated commercially in more than 80

countries, supplying food with high protein levels and oil of excellent quality. According

to FAO, world production of peanut reaches about 23 million tons annually and assumes









great agronomic importance in developing countries. These countries are responsible for

80% of the world production with 67% being produced in the semi-arid tropics (Singh

and Singh, 1992).

The genus Arachis, with diploid (2n=2x=20) and tetraploid (2n=4x=40) species,

belongs to the Fabaceae family, Papilionoideae subfamily, Stylosanthinae subtribe,

Aeschynomeneae tribe. It possesses herbaceous, annual and perennial species, leaves

with 3 or 4 leaflets; papilionaceous corolla; tubular hypanthium, sessile ovaries and

underground fruits (Gregory et al., 1980; Krapovickas and Gregory, 1994).

The genus has natural occurrence exclusively in South America, extending from

east of the Andes Mountains, south of Amazonia, north of the Planicie Platina and

northwest of Argentina, until the northeast of Brazil. It is believed that the genus

originated in the Amambai Mountain, near the border of Mato Grosso do Sul-Brazil and

Paraguay, from where it was dispersed over an area of 4000 km in diameter (Krapovickas

and Gregory, 1994).

Krapovickas and Gregory (1994) describe 69 species, with 39 exclusively from

Brazil. Valls and Simpson (1994), however, assert that the genus has about 80 species.

The genus can be divided into nine groups or sections; one of them with two series:

* Section Arachis
* Section Caulorrhizae
* Section Erectoides
* Section Extranervosae
* Section Heteranthae
* Section Procumbentes
* Section Rhizomatosae
o Series Prorhizomatosae
o Series Rhizomatosae
* Section Trierectoides
* Section Triseminatae









The importance of A. hypogaea as a major food crop generated a need for a better

understanding of the wild species of the genus, because of the potential to use them in

breeding programs of the common peanut. Valls and Simpson (1994) reviewed the two

different proposed gene pools of the genus (Table 2-1). Some degree of cross-ability is

present among some species, and this could represent a chance to move genes from one

species to others. Examples of introgression of genes from wild species to the common

peanut are presented by Simpson and Starr (2001), Starr et al. (1990), and Mallikarjuna

and Sastri (2002). Cultivar "Spancross" (Starr et al., 1990), generated by the crossing of

A. hypogaea and A. monticola; and cultivar "COAN" (Simpson and Starr, 2001), a runner

market type derived from a backcross introgression pathway involving a complex

interspecific amphiploid hybrid (A. cardenasii x A. diogoi) using cv. "Florunner"

(Norden et al., 1969) as the recurrent parent are real examples of the potential of the wild

species in breeding programs of the common peanut.

Parallel to the interest in improvement of cultivated peanut, the acknowledgment of

the forage potential of some wild species of the genus brought a general interest in these

species. This interest generated an effort to recollect most of the available genetic

diversity. According to Valls and Pizarro (1994) more than 30 expeditions were

conducted between 1981 and 1993, which greatly impacted the number of accessions of

perennial Arachis species stored in germplasm banks.

The Forage Potential of the Genus Arachis

Many species of the genus Arachis produce forage high in quantity and quality,

comparable or superior to species of other tropical legumes used commercially as forage

crops (Valls and Simpson, 1994). Species of the genus Arachis, belonging to the sections

Rhizomatosae, Arachis, Erectoides, Procumbentes, Caulorrhizae and Triseminatae have









been evaluated for forage in Australia since 1950. Many of those species have presented

great potential as forage crops, mainly because of persistence under grazing (Cook et al.,

1994).

Table 2-1. Proposed gene pools of the genus Arachis based on Arachis hypogaea
breeding perspective
Gene Pools Wynne and Halward (1989) Singh et al. (1990)
Primary Cultivated varieties of A. Arachis hypogaea and A.
hypogaea monticola
Landraces of A. hypogaea*
Breeding lines derived from
the above
Secondary Arachis monticola Diploids in section Arachis
Other wild tetraploid forms in cross-compatible with A.
section Arachis (as yet hypogaea
uncollected)
Tertiary Diploids in section Arachis Species in all sections
cross-compatible with A.
hypogaea
Quaternary Diploid and tetraploid species
of other sections of the genus


The forage importance of the common peanut (A. hypogaea) has been long

recognized, especially because of the great nutritional value of its stem and leaves (Cook

and Crosthwaite, 1994). Although long recognized, breeding programs neglected this fact

and selections based on the forage potential were not emphasized, a point that seems to

have changed in recent years during which evaluation of the forage production of A.

hypogaea has been conducted. Gorbet et al. (1994) reported dry matter yields (DMY) of

9 Mg ha-1 in a 140 days growth period in research done in Florida (USA). Crude protein

values ranged from 140 to 200 g kg-1 and IVOMD ranged from 600 to 720 g kg-.

Also in the southeastern USA, Muir (2002) studied the forage production and

nutritive value of eight warm-season legumes in north-central Texas under two dairy

manure compost rates, and two harvest regimes during 2 years. He reported that Arachis









hypogaea produced over 2.5 Mg ha-1 yr- of high quality forage and was one of the most

productive. He concluded that although not highly productive under dryland conditions,

groundnut can contribute with both forage and seeds for livestock and wildlife.

Pizarro et al. (1996a) evaluated 34 germplasm lines of A. hypogaea in a clay oxisol

in Brazil, and reported DMY that varied from 0.4 to 2 Mg ha-l in a 180 growth days

period. In Nigeria, Larbi et al. (1999) investigated variation in forage and seed yields,

crude protein (CP), neutral detergent fiber (NDF), in situ degradation of dry matter (DM),

and nitrogen (N) of 38 late-maturing groundnut cultivars. Average yield of forage was

4550 Mg ha-l and seed production was 1.25 Mg ha-'. Crude protein, NDF, and DM

degradation of leaf and stem varied among cultivars. Forage and seed yields were

significantly correlated (r = 0.53), but seed yield was poorly correlated with forage

quality indices examined. They suggested that plant breeders could select genotypes with

higher seed and forage yield, and better forage quality.

Also in Nigeria, a 3-yr study was carried out with 11 peanut varieties to select

superior lines for use in crop livestock systems. Crude protein (CP) concentration ranged

from 148 to 216 g kg-', with seven varieties recording forage yields above 5 Mg ha1.

Mean seed yield (over 3 yr) varied significantly from 0.73 to 1.68 Mg ha-l (Omokanye et

al., 2001).

Moreover, other species of the genus have also shown great potential to be used as

a forage crop, principally because they have great persistence under grazing. Cook and

Crosthwaite (1994) reported that stands of some species of the genus have been grazed

for more than 30 yr. In Brazil where the genus originated, native populations have been

grazed and utilized by producers for many years. Cameron et al. (1989) argued that wild









species of the genus Arachis are excellent alternatives to improve tropical and subtropical

pastures, particularly due to this strong persistence under grazing. Kretschmer and

Wilson (1988) pointed to superior forage and seed production, and adaptation to wetlands

as some of the important characteristics of an Arachis genotype from section

Procumbentes (Pantanal peanut) evaluated by them and later described as A. kretschmeri

by Krapovickas and Gregory (1994).

Among the several species of the genus that can be utilized as forage crops and

ultimately will impact the quality of the pastures, we can point to Arachis glabrata

Benth., A. pintoi Krap. & Greg.; and A. repens Handro as the ones with the highest

potential (Grof, 1985; Valls, 1992).

Cultivars "Florigraze" (Prine et al., 1986) and "Arbrook" (Prine et al., 1990) of

Arachis glabrata were released by the University of Florida, and they are used in an area

of approximately 10,000 ha throughout the southeastern USA. These cultivars are

primarily used to produce hay with high nutritional quality that is consumed by the dairy

and beef cattle and by racing horses. Although A. glabrata has shown excellent forage

characteristics, the fact that it produces few seeds, and consequently that its propagation

is exclusively vegetative, causes establishment difficulties. This has prevented its

utilization over larger areas (French et al., 1994).

According to some authors, however, the species with the most forage potential is

A. pintoi, a endemic species in Brazil. Arachispintoi produces high dry herbage yield

with excellent quality, as well as high seed production (Valls and Simpson, 1994). The

geographical distribution of the species spreads over an area that is part of the states of







18


Goias, Bahia, and Minas Gerais, and extends to the Atlantic coast of Brazil, where the

original accession ofArachis pintoi was collected (Figure 2-1).

750 70W.Gr 65 60 55 50 45 40 35


75o 70oW.Gr .65o


60o 55o 50o 45o 40o 35o 30o


Figure 2-1. Area of natural occurrence of Arachis pintoi (Valls, J.F.M. and E.A. Pizarro. 1994.
Collection of wild Arachis germplasm. p. 19-27. In P.C. Kerridge and B. Hardy
(ed.). Biology and Agronomy of Forage Arachis, CIAT, Cali, Colombia).

Arachis pintoi

Botanical characteristics

A. pintoi is a herbaceous, perennial legume, with a low stoloniferous growth habit.

Growth heights range from 20 to 40 cm, with axonomorfus roots, without enlargements

(Figure 2-2). The first branch is erect with attached low branches, rooting at the nodes,









and stems are cylindrical, angular, and hollow (Krapovickas and Gregory, 1994). The

leaves are alternate, compound, with four obovate leaflets (50 mm length x 32 mm

width). Stipules have the basal portion attached to the petiole, and measure 10-15 mm

length x 3 mm length with the free portion measuring 10-12 mm length x 2.5 mm length

in the base (Krapovickas and Gregory, 1994).

A. pintoi shows indeterminate and continuous flowering. The inflorescence is

axillary, in very short spikes, with four to five flowers, covered by the joined portion of

the stipules. The flowers are sessile, protected by two bracts. The hypanthium is well

developed, and can reach 10 cm in length, with silky hair. The calyx is bilabiate, with

silky hairs. The corollas are yellow in the typical condition. Standard petals are 11 mm

long x 13 mm wide, with yellow nerve and keel petals of 6-7 mm of length. Four oblong

anthers and four spherical anthers are typically present. The species is considered

normally self-pollinated (Krapovickas and Gregory, 1994). The flowers can be yellow,

orange, cream and white (Valls, 1992).

The small fruit ofA. pintoi are located underground in an articulated legume form,

with each articulation classified as an indehiscent capsule, which usually contains a

single seed (Cook et al., 1990). The pericarp is flat and resistant, covered with fine hairs

that retain the soil. It presents two distinct segments, each one with a seed (Krapovickas

and Gregory, 1994). The number of chromosomes is 2n=2x=20 (Fernandez and

Krapovickas, 1994).

Agronomic characteristics

A. pintoi grows well in tropical areas from sea level to heights of 1800 m, with

1500 to 3500 mm of annual rainfall (Rincon et al., 1992; Valls and Simpson, 1994).























































Figure 2-2. Arachis pintoi plant characteristics (Krapovickas, A. and W.C. Gregory. 1994.
Taxonomy of the genus Arachis (Leguminosae). Bonplandia 8: 1-186)

According to Rao and Kerridge (1994), A. pintoi CIAT 17434 (PI 338447) exhibits good

adaptation to acid soils, with optimal growth occurring at pH 5.4. It is also tolerant to









high soil Al concentrations (70%), and requires soil organic matter higher than 3% for

normal growth. A. pintoi (PI 338447) grows in a variety of soil textures ranging from

sandy to clay, although optimum growth occurs in sandy soils (Argel and Pizarro, 1992).

A. pintoi (PI 338447) is able to absorb P in soils with very low concentrations of

this element, or in situations when relatively insoluble forms are applied. It has low

response to N, Cu, Mo, Fe, S, and K (Rao and Kerridge, 1994). It is adapted to poorly-

drained soils, but also grows well in well-drained soils having long periods without

precipitation (Pizarro and Rinc6n, 1994).

Under defoliation, A. pintoi (PI 338447) has a good initial regrowth and high light

interception capacity, with forage DMY, estimated between 30 and 40 d, of 0.08 Mg ha-1

day-1 (Fisher and Cruz, 1994). A. pintoi (PI 338447) grows better under shade than in full

sun conditions. Plants under full sun presented lower leaf area and less above-ground

biomass than counterparts maintained under 30, 50, and 70% shaded conditions. The

below-ground biomass was not different in the same circumstances. This characteristic

provides A. pintoi with the ability to be used as a ground cover in coffee (Coffea Arabica

L.) and fruit groves (CIAT, 1991).

Arachispintoi (PI 338447) nodulated with native Bradyrhizobium strains in

Colombia, although the nodules were not active. Inoculation with selected strains was

more efficient and demonstrated superior plant growth. Small doses of N fertilizer

(50 kg ha-1) increased the initial infection and sped the nodulation process. Rates of N

fixation of A. pintoi ranged from 0.07 to 0.20 Mg ha-1 yr- (Thomas, 1994).

Pizarro and Rincon (1994), reported that A. pintoi CIAT 17434 (PI 338447), 18748

(PI 604858), 18749 (PI 604859) and 18750 (PI 604815), growing in a subtropical









environment in Pelotas-RS, Brazil, were resistant to severe frosting. The authors stated

that, although the plants presented freezing burning symptoms that affected their

development, regrowth of herbage was achieved after the first rainfall.

Arachispintoi germplasm characterization

In 1991, Arachispintoi had a very narrow genetic base, approximately 30

accessions. Starting from that date, intense collection of materials of the species was

initiated by various projects. The germplasm base was enlarged to more than 150

accessions (Valls et al., 1994).

The large number of accessions today available implies in a need for discrimination

among them, because they can present different agronomic performance. Knowledge of

the genetic variability of A. pintoi will be obtained through an appropriate

characterization of the accessions. Characterization and evaluation of the wide range of

A. pintoi germplasm should be carried out according to the priorities and strategies for

handling the genetic resources of Arachis (Valls, 1988).

A. pintoi has been spread world wide by the international germplasm exchange

network that consists of international centers under the International Plant Genetic

Resources Institute (IPGRI) and the national germplasm centers. Despite this world wide

distribution of A. pintoi, much of the evaluation research was conducted using only the

original accession (PI 338447). As a result of these works, A. pintoi was released as a

commercial cultivar in 9 countries, including Australia, USA, Costa Rica, Honduras,

Colombia and recently in Brazil. Most of these released cultivars however represent

accession PI 338447. Recently new cultivars have been released which originate from

accessions other than PI 338447.









According to Pizarro and Carvalho (1996), CIAT distributed about 1.2 Mg of seeds

of A. pintoi accession (PI 338447) to Europe, Africa, Asia, Southeast Asia and North,

Central and South America. A total of 61 countries were supplied with seeds.

Even when more than one accession has been evaluated at a location, much of this

characterization of germplasm has been with small numbers of accessions and based

primarily on agronomic evaluation in different locations, with great emphasis on herbage

production. Most of these works were conducted in South America, Central America, and

Australia.

In South America, under the coordination of the International Center for Tropical

Agriculture (CIAT), the International Tropical Pastures Evaluation Network (RIEPT)

evaluated A. pintoi accession CIAT 17434 (PI 338447) in several places of Colombia,

inside the savanna ecosystem. In that network, this accession presented poor adaptation

when compared with other legumes. The DMY accumulated in 12 wk of regrowth, using

the methodology of RIEPT (Toledo, 1982), varied from 0 to 0.47 Mg ha-1 during the

rainy season, and from 0 to 0.25 Mg ha-1 in the dry season (Pizarro and Rinc6n, 1994).

In Brazil, Valentim (1994) evaluated the adaptation and forage production of

accessions BRA-013251 (PI 338447) and BRA-015121 (PI 604858) in Rio Branco-AC

(North). The results showed excellent adaptation of the accessions to the environmental

conditions in Acre state. The average DMY of the accessions during 16 wk of growth in

the rainy period of 1992 was 4.6 Mg ha-1 and 5.3 Mg ha-1 during the period of minimum

precipitation. In 1993, the average production was of 6.1 Mg ha-1 in the rainy season and

4.2 Mg ha-1 in the period of minimum rainfall. In mixed pastures with Paspalum spp. in

humid areas of low fertility in Campo Grande-MS, A. pintoi accession BRA-015598 (PI









604815) yielded 0.6 to 1.3 Mg ha-1 in the dry season and 0.2 to 0.5 Mg ha-1 during the

rainy season (Fernandes et al., 1992).

Pizarro et al. (1996b) evaluating forage legumes mixed with Brachiaria decumbens

Stapf. in Uberlindia-MG, obtained, with the accessions BRA-013251 (PI 338447),

015121 (PI 604858), 015598 (PI 604815) and 031143, dry matter production of 2.2, 0.6,

2.3 and 4.4 Mg ha-l, respectively, in the period of maximum precipitation during

1994/1995. Data obtained in Planaltina-DF by EMBRAPA-CPAC, starting from 9

accessions in a lowland area, showed great variability of forage production of A. pintoi in

those conditions. The total DMY varied from 5 to 13 Mg ha-1 in the first year of

evaluation and from 3 to 11 Mg ha-1 in the second year (Pizarro et al., 1992; Pizarro et al.,

1993).

In Veracruz, Mexico, A. pintoi CIAT 17434 (PI 338447), showed average DMY of

2.0, 1.2 and 0.8 Mg ha-l in three cuttings of 12-wk regrowth during the rainy season, in

two years of evaluation (Valles et al., 1992). In Guapiles, Costa Rica, the accessions

CIAT 17434 (PI 338447), 18744 (PI 476132), 18747 (PI 497574), 18748 (PI 604858)

and18751 ofA. pintoi presented DMY of 4.1, 4.9, 4.0, 3.8 and 3.7 Mg ha-l, respectively,

in 2-yr of evaluation (Argel, 1994). In the USA, Kretschmer and co-workers have

evaluated a number of seed-propagated accessions of wild species of Arachis for use on

the poorly drained soils of South Florida and identified an A. pintoi accession (IRFL

4222) that was persistent under grazing (French et al., 1994).

In Australia, only three accessions were initially available for evaluation, CPI

58113 (PI 338447), CPI 28273 and CPI 93472, with the last two being considered mostly

the same. CPI 58113 was evaluated in several locations and presented great adaptation









and persistence under grazing. Subsequently it was released as cultivar "Amarillo" in

1987, and since then it has gained increasing acceptance with producers as a forage crop

(Cook et al., 1994). These authors reported that in 1991, 9 Mg of seed were produced and

commercialized.

In seven locations in Central and West Africa, A. pintoi CIAT 17434 (PI 338447)

showed production that varied from 0.6 to 3.2 Mg ha-1 in the rainy season and 0.1 to 2.0

Mg ha-1 during the dry season, confirming its wide adaptation (Stur & Ndikumana, 1994).

The same authors reported yields ranging from 0.5 to 4.5 Mg ha-1 yr- for Amarillo, in

evaluations conducted at three locations in the dry areas of Fiji.

Although the majority of the characterization work conducted with the germplasm

ofA. pintoi has been agronomically based, some examples of germplasm characterization

at other levels can be found in the literature. Moncato (1997), working with

approximately 45 accessions ofA. pintoi, applied a series of morphological descriptors to

describe the variability of this germplasm. The accessions showed great variability in

morphological traits. Oliveira et al. (1999) demonstrated the morphological variability

and inheritance of flower color. The yellow flower is dominant over the orange flowers.

Mass et al. (1993) used 60 morphological descriptors to characterize and

demonstrate the variability of eight accessions. The accessions were organized into two

groups representing the plant types, one homogeneous, and the other divided into four

distinct subgroups. These results pointed to a lack of continuous patterns of variation in

morphology and highlighted the need for further germplasm collection.

Paganella and Valls (2002) applied a list of 12 descriptors to evaluate seven

cultivars and 13 accessions of A. pintoi. The objective of this work was to review the









origin of the cultivars and check inconsistencies in the literature about the germplasm

accessions that gave rise to the commercial cultivars. The results confirm that five of

seven cultivars derived from the original or first accession of the species collected (PI

338447).

Perennial Arachis germplasm has also been evaluated at the molecular DNA level.

Genetic variation within and among accessions of the genus Arachis representing

sections Extranervosae, Caulorrhizae, Heteranthae, and Triseminatae was evaluated using

RFLP and RAPD markers. RAPD markers revealed a higher level of genetic diversity

than RFLP markers, both within and among the species (Galgaro et al., 1998).

Gimenes et al. (2000) working with sixty-four accessions of section Caulorrhizae

utilized RAPD analysis to characterize the genetic variation and the phylogenetic

relationships. A total of 104 fragments of DNA of different sizes were generated, and 97

were polymorphic in the accessions tested. Despite the large number of polymorphic

fragments detected, the mean number of unique genotypes detected by each RAPD

primer was low. However, when data from all primers were considered together, all

accessions were uniquely fingerprinted.

Genetic characterizations of accessions ofA. pintoi using isoenzymes, RFLP and

RAPD molecular markers were also done by Valente et al. (1998), Bertozo and Valls

(1996), Bertozo & Valls (1997a), Bertozo & Valls (1997b), and Carvalho et al. (1998)

with a small set of germplasm accessions, different from those used in this research.














CHAPTER 3
MOLECULAR CHARACTERIZATION OF Arachispintoi GERMPLASM

Introduction

Until the late 1960s most genetics studies were associated with genes that

controlled morphological features, which were in general easily identified. These features

appropriately named morphological markers and were very important to the

understanding of gene action and expression, and for the construction of the first genetic

linkage maps. However, the low number of morphological markers linked to important

agronomic traits made their use inefficient for the purpose of genetic improvement.

Further, morphological markers were generally available for few species, which were

used as model systems for genetic studies. For the great majority of crop species and their

wild relatives few or no morphological markers were available (Ferreira and Grattapaglia,

1998).

By definition, a molecular marker is every single molecular phenotype expressed

by a particular gene. The nucleotide sequence and function of the marker could be known

or unknown. A molecular marker can only be considered a genetic marker after its

mendelian segregation is observed in segregating populations (Ferreira and Grattapaglia,

1998).

Currently, several molecular biology techniques are available for the study of

genetic variability at the DNA level. These techniques provide for the possibility of

identification of innumerable molecular markers that theoretically could cover the whole

genome of a species. These markers can be used for many different applications









including fingerprinting, genetic mapping, phylogenetic relationships, genetic diversity

studies, and plant breeding.

Among all DNA-based markers, Random Amplified Polymorphic DNA (RAPD)

has been the most popular with plant geneticists. Reiter (2001) revealed that in a recent

bibliographic search there were more than 2,000 publications were in which RAPD

markers had been used. According to the same author, the popularity of this marker arises

from the fact that the method is very simple and the cost required for its application is

low. RAPD markers have also been the marker of choice in several genetic studies with

the common peanut (Arachis hypogaea L.) and its wild relatives.

Hilu and Stalker (1995) used RAPD markers to access genetic relationships

between the common peanut and wild species of the section Arachis of the genus. These

authors reported 132 polymorphic bands that were useful for separating species and

accessions, and for evaluation of the genetic diversity presented by the germplasm.

Chang et al. (1999) accessed the genetic diversity of A. hypogaea cultivars released

in Taiwan using RAPD markers. They were successful in estimating genetic distances

among varieties and grouping them in accordance with their genetic similarities. The

mean genetic distance among cultivars was 0.411, and they were classified into six

groups.

RAPD markers have also been used to characterize wild species of Arachis. Nobile

et al. (2004) studied the genetic variation within and among species of the section

Rhizomatosaea of Arachis using RAPD markers. These authors reported that by using

110 polymorphic RAPD bands they were able to describe the genetic diversity and draw

a dendogram displaying the similarities among accessions and species of this section.









Another very important technique in genetic resources research is tissue culture.

The use of in vitro techniques is particularly critical to species with vegetative

propagation and recalcitrant seeds. Other potential applications of tissue culture

techniques are: micropropagation, long-term storage, germplasm exchange, embryo

culture of interspecific hybrids, induction of mutations, production of synthetic seeds, and

genetic transformation. Another advantage of plants originated in vitro is that they are

generally free of pathogens and diseases.

Although in vitro techniques are useful for genetic resources conservation, attention

must be given to the genetic stability of the systems used in this process. Tissue culture-

induced genetic variation or somaclonal variation is defined as the variation that arises

during the period of dedifferentiated cell proliferation that takes place between the

explant culture and recovering of regenerated plants. Such genetic variation has been

observed among regenerants of several species and they usually present mutations that

include: chromosome breakage, changes in ploidy number, single base changes, changes

in copy numbers of repeated sequences, increased transposon mutagenegis, sister

chromatid exchange and alteration in DNA methylation (Hawkes et al., 2000).

Because in vitro techniques have a wide spectrum of applications to genetic

resources conservation and plant breeding, the development of protocols optimized to a

particular species should have high priority. Several factors may affect the efficiency of

the process and should be considered when research is conducted in this field. Genotypes,

source and age of explant, hormone concentration in the medium, and day length are

some factors that affect the success of regeneration of plants from callus tissue culture

(Flick et al., 1983).









Tissue culture regeneration protocols for A. pintoi were proposed by Rey et al.

(2000) based on a single genotype, and Ngo and Quesenberry (2000) based on a small

number of accessions. In terms of genetic resources it is very critical to evaluate the

efficiency of protocols in terms of plant regeneration among different genotypes and also

the preservation of the genetic characteristics of the germplasm regenerated or stored

with these protocols. Additionally, in terms of plant breeding, it is important to assess the

genetic diversity of the germplasm with respect to their plant regeneration abilities. It is

also important to estimate the capacity of the protocols to generate somaclonal variations,

which ultimately could generate mutations and gene diversity.

The objectives of this research were:

* To characterize the Arachispintoi germplasm accessions stored at the USDA-
NPGS germplasm banks using RAPD markers.

* To evaluate the organogenic regeneration ability of these A. pintoi accessions with
two tissue culture protocols.

* To study the variation in RAPD band profile of plants regenerated compared to the
parent plants.

Materials and Methods

Accessions of Arachispintoi stored in the Southern Regional PI Station of the

National Plant Germplasm System (NPGS) located at Griffin-GA were transferred to the

University of Florida in 2001 and 2002 (Appendix A) A subset composed of 35

accessions was used to study the genetic diversity of this collection at the molecular level

using RAPD markers.

In addition to the RAPD characterization of the parent plants, 25 out of 35

accessions were selected randomly and evaluated for their organogenesis ability using

two tissue culture protocols. Extent of generated somaclonal variation was accessed by









comparing the RAPD band profile of the regenerated tissue culture plants to the parent

plants.

Parent Plants

DNA was extracted from leaves of single plants stored at the Agronomy

Department greenhouse using a modified CTAB protocol first proposed by Rogers and

Bendich (1985). In this protocol, 0.1 g of ground tissue was mixed with 400 pl of 2x

CTAB extraction buffer in a 2.5 ml eppendorf tube and incubated in a 65C water bath

for 60 min. After that 400 pl of a chloroform:isoamyl alcohol (24:1) was added and the

mixture was centrifuged for 5 min. The supernatant was then transferred to a new tube

where 1/10 volume of 10% CTAB was combined, and 400 pl of a chloroform:isoamyl

alcohol (24:1) was added again. The solution was centrifuged once more, and the

supernatant was transferred to a new tube where an equal volume of CTAB precipitation

buffer was added. The mixture was then centrifuged, the supernatant was removed and

the DNA pellet was rehydrated in 100 [il of high salt TE buffer resting in a water bath.

Ten minutes latter the DNA was reprecipitated with 0.6 volumes of isopropanol and

centrifuged for 15 min at 10,000 rpm. Finally the pellet was washed with 80% ethanol,

dried, and resuspended in 50 pl of DDW (Appendix B).

DNA concentrations were analyzed by measuring absorbance at 260 nm, and DNA

quality was determined by spectrophotometer readings of the ratio of 260/280 nm.

Working solutions were prepared by diluting the DNA stock solutions with DDW and

standardized to concentrations of 25 ng of DNA per pl.

Eighteen primers of ten nucleotides length from the Operon Technologies kit

(Table 3-1) were used to amplify regions of genomic DNA under thermocycling

conditions proposed by Gimenes et al. (2000). Thermocycling conditions were: 40 Cycles









of 92C/1 min, 35C/1 min, and 72C/2 min. Twenty-five [l of PCR reaction mixtures

were prepared by adding 17.3 il ddH20, 2.5 il PCR Buffer (500 mM KC1, 100 mM

Tris-HCl pH 8.3, 15 mM mg MgCl2, and 0.1% gelatin), 1.0 pl MgC1, 1.0 pl of 2.5mM

dNTP solution, 1.0 pl Primer, 0.2 pl Taq polymerase (5 units/pl), and 2.0 pl DNA

solution.

Amplification products were separated by electrophoresis on 1.5% agarose gels,

and the 1kb DNA ladder (Promega Coorporation) was used as a standard. Banding

patterns were visualized by staining the gels in ethidium bromide solutions and viewing

under UV radiation. Images were captured with Quantity One quantification software

version 4.1.1 from BIO-RAD Laboratories and then bands were scored.

For each combination of accession and primer, five PCR reactions were prepared.

Only bands present in at least three out of five gels were considered. RAPD bands were

scored as presence (1) or absence (0) of homolog bands for all accessions and a

phenotypic binary matrix was produced. This matrix was used to perform genetic

analysis using the software POPGENE version 1.32

(http://www.ualberta.ca/-fyeh/index.htm). Allele frequency, number of polymorphic loci,

Nei's genetic distance (Nei, 1972), Nei's genetic diversity index (Nei, 1973), and

Shannon-Weaver's genetic diversity index (Shannon and Weaver, 1949) were the

parameters calculated. Genetic distance (D= -In I) was later used as a criterion for

differentiation among accessions to prepare a cluster analysis.

Tissue Culture Regenerated Plants

Two protocols were used to access their organogenesis ability, and capacity to

generate somaclonal variation. Protocol 1 was proposed by Rey et al. (2000) and Protocol

2 proposed by Ngo and Quesenberry (2000).









Table 3-1. List of Operon Technologies* primers used to amplify Arachispintoi DNA
regions
Primer number Nucleotide sequence
A4 5'-AATCGGGCTG-3'
A10 5'-GTGATCGCAG-3'
A12 5'-TCGGCGATAG-3'
A15 5'-TTCCGAACCC-3'
B4 5'-GGACTGGAGT-3'
B5 5'-TGCGCCCTTC-3'
B10 5'-CTGCTGGGAC-3'
B16 5'-TTTGCCCGGA-3'
C2 5'-GTGAGGCGTC-3'
C4 5'-CCGCATCTAC-3'
D4 5'-AATCGGGCTG-3'
D13 5'-GGGGTGACGA-3'
E4 5'-GTGACATGCC-3'
E5 5'-TCAGGGAGGT-3'
E8 5'-AATCGGGCTG-3'
E15 5'-ACGCACAACC-3'
G5 5'-CTGAGACGGA-3'
G15 5'-ACTGGGACTC-3'
Operon Technologies, Alameda, CA, USA


To compare these two protocols, callus rating and weight and number of

regenerated plants were used. The experiment was conducted using a completely

randomized design with 50 treatments and 3 replications. Quantity of callus produced

was rated based on a 1 to 5 scale, where 1 = no callus growth, and 5 = largest amount of

callus (Figure 3-1). In addition, callus weight was measured with a digital scale placed in

the laminar flow hood to prevent contamination.


Figure 3-1. Rating scale applied to callus pieces of Arachispintoi explants incubated in
two different protocols.









Leaflet pieces from adult plants were the tissue source for callus initiation. Leaflets

were surface sterilized by immersion in 70% ethanol for 1 to 2 min, followed by

immersion in a solution of commercial bleach (0.9% sodium hypochlorite, final

concentration) plus one drop of Tween for 1 to 2 min. Leaflets were then washed three

times with autoclaved distilled water. Circles of approximately 19.4 mm2 area of the

median portion of the laminae were cut and placed with the abaxial side down on the

media in 60 mm x 15 mm petri dishes (Figure 3-2).

















Figure 3-2. Leaflet cutter and callus induction dishes of Protocol 1 and 2 of Arachis
pintoi accessions.

Protocol 1

The callus induction medium consisted of major and minor salts, as well as

vitamins according to Murashige and Skoog (1962), with 3% sucrose, 0.7% agar, and 10

mg L-1 NAA (1-naphthaleneacetic acid) + 1 mg L-1 BA (6-benzylaminopurine). The pH

of the medium was adjusted to 5.7 with 0.1 N KOH or HC1 prior the addition of agar. To

induce shoots, small pieces of callus (30 mg fresh wt.) were transferred to fresh medium

composed of MS basal salts and vitamins+lg L-1 BA. For root induction, the regenerated









shoots were transferred to a medium composed of MS basal salts and vitamins plus

0.01mg L-1 NAA (Table 3-2).

Plantlets resulting from rooted shoots were rinsed gently under running tap water to

remove adhering cultured media and immediately planted in pots containing potting mix.

These plants were acclimatized in a humidity box and then placed in the greenhouse.

Protocol 2

The MS basal salts and vitamins medium (Murashige and Skoog, 1962) was the

base for all media used in this protocol. For callus induction the MQC medium, which is

a modification of the MS medium proposed by Wofford et al. (1992) for tropical

legumes, was the medium of choice. In this medium a 2:1 auxin/cytokinin ratio was

employed, with a final concentration of 2 mg L-1 of IAA (indole-3-acetic acid) and 1 mg

L-1 of kinetin. The pH was adjusted to 5.8 prior to autoclaving.

For shoot development, medium IBA which was composed of MS basal salts and

vitamins plus 0.1 mg L-1 of IBA (3-Indol butyric acid) was employed. For root induction,

the regenerated shoots were transferred to medium MS basal salts and vitamins with

growth regulators.

Plantlets resulting from rooted shoots were rinsed gently under running tap water to

remove adhering cultured media and immediately planted in pots containing potting mix.

These plants were acclimatized and then placed in the greenhouse.


Table 3-2. Tissue culture protocols used to regenerate Arachispintoi plants
Protocol Callus induction Shoot development Root induction
Media Media Media
1 MST MS+lg L- BA MS+O.OlmgL- NAA
2 MQCff MS+0.lmg L-' IBA MSNHfTf"
f MS + 3% sucrose, 0.7% agar, and 10 mg L-1 NAA + 1 mg L-1 BA
ft MS + 3% sucrose, 0.7% agar, and 2 mg L-1 of IAA and 1 mg L-1 of kinetin
ftN MS basal salts and vitamins and no hormones










DNA was extracted from regenerated plants in accordance with the protocol

described earlier. PCR reactions and band separation and scoring were also identical to

that used for the parent plants. RAPD band profile of the regenerated tissue culture plants

was compared to the parent plants.

Result and Discussions

Parent Plants

Of the 18 primers evaluated, eight were able to generate reproducible and reliable

bands. Primers A4, B4, B5, C2, D4, D13, E4, and G5 amplified 100 different bands. The

number and size of amplified bands for each primer is shown in Table 3-3.

Table 3-3. List of Operon primers, number and size of amplified bands, and number of
bands accession-specific ofArachispintoi germplasm genomic DNA
Primer Number of Size of amplified bands (bp) Number of bands
number amplified bands accession-specific

A4 15 250,550,670,750,870,1000, 1100, 1350, 0
1500,1700,1900,2100,2500,3000,3100

B4 12 700,850,900,1000,1200,1400,1600,1750, 1
2000,2300,2700,3000

B5 11 500,750, 1000, 1100, 1250, 1400, 1700, 3
2000,2500, 3000,3500

C2 15 465,520,600,700,750,850,1000,1300, 2
1500,1750,2000,2500,2750,3000,3500

D4 08 580, 650, 750, 910, 1200, 1350, 1500, 2000 1

D13 16 370,500,650,750,850,1000,1200,1400, 2
1500,1700,2000,2100,2300,2500,2800,
3310

E4 10 600,900,1000,1100,1250,1500,1800, 1
2100, 2500,2800

G5 13 350,450,500,650,750,850, 1000, 1100, 1
1250, 1350, 1750,2000,2500,2750

The average number of amplified bands per primer was 12.5, with primer D4

amplifying eight fragments and primer D13 amplifying 16 fragments. The size of these

100 fragments ranged from 250 bp to 3500 bp. From the 100 bands amplified, 98









presented polymorphism with the only two exceptions being Primer C2-1000 bp and

primer E4-1250 bp. The average presence of bands per accession was 32, ranging from

20 to 44 bands.

The observed number of amplified bands was significantly variable for each primer

and accession analyzed. Primer A4 amplified 15 different bands with average bands per

accession of 5.4 and a range of 2 to 10 bands. Primer B4 amplified 12 fragments with

average bands per accession of 4.3 and a range of 1 to 10. Primer B5 amplified 11 bands

with an average band per accession of 3.5 and range of 1 to 7. Primer C2 amplified 15

bands and average bands per accession of 5.2 and a range of 1 to 10. Primer D4 amplified

only 8 bands and the average bands per accession was 1.7 with a range of 1 to 4. Primer

D13 displayed 16 bands with average bands per accession of 5.2 and a range of 2 to 11.

Primer E4 showed 10 different fragments with an average of 4 bands per accession and a

range of 2 to 8. Finally, primer G5 presented 13 bands with average bands per accession

of 3 and a range of 1 to 7 bands.

Ten bands were unique to an individual germplasm accession. Accession PI

604856 was discriminated by band C2-465bp, accession PI 604858 by band E4-600bp,

accession PI 604810 by band C2-520bp, accession PI 604799 band B5-1250bp, PI

604809 by band B5-1100bp, accession PI 604807 by band G5-1750bp, accession CIAT

22256 by D4-1200bp, accession CIAT 22159 by band D13-3310bp, CIAT 22152 by band

B4-700bp, accession CIAT 22265 by band D13-2800bp, and finally accession CIAT

22260 by band B5-500bp. Table 3-4 summarize the results obtained with RAPD markers

among the 35 germplasm accessions ofA. pintoi analyzed in this research. Figure 3-3

displays stained eletrophoresis gels of primer A4 and 24 accessions.









Table 3-4. Characteristics of RAPD patterns of the 35 Arachispintoi germplasm
accessions
Total number of screened primers 18
Number of polymorphic primers 08
Total number of bands amplified 100
Size of the amplified bands 250 bp 3500 bp
Minimum and maximum number of bands per primer 08 (D4) 16 (D13)
Average bands per primer 12.5
Total number of polymorphic bands 98
Total number of monomorphic bands 02
Average number of bands per accession 32
Minimum and maximum number of bands per accession 20 44
Number of accession-specific bands 10

According to Nei (1973), when a large number of loci are examined to evaluate the

genetic variability of a population, the amount of variation is measured by the proportion

of polymorphic loci and average heterozygosity per locus. Also according to Nei a locus

is called polymorphic when the frequency of the most common allele (xi) is equal to or

less than 0.95, in cases where the sample size is smaller than 50.

In Table 3-5, information about the gene frequency of each allele at every RAPD

locus is presented. Great variability in gene frequency was observed for each different

RAPD locus. Using the criterion described above to characterize polymorphic loci we

can observe that 11 loci presented the frequency of the most common allele higher than

0.95, and then were classified as monomorphic. Therefore, we can conclude that 89 out

of 100 loci, or 89% of the RAPD loci, was the proportion of polymorphic loci.

Although the proportion of polymorphic loci is a good estimation of genetic variability, a

more precise and appropriate measure of gene diversity is obtained by the use of the gene

diversity statistics (Nei, 1987).





















































Figure 3-3. RAPD band profile of 24 accessions ofArachispintoi amplified by primer
A4 (Operon Technologies).
A. lane 1-PI604856, 2-PI604857, 3-PI604858, 4-PI604798, 5-PI604803, 6-PI604805,
7-PI604812, 8-PI604810, 9-PI604811,10-PI604799, 11-PI604800, 12-PI604809.
B. lane 1-PI604817, 2-PI604815, 3-PI604814, 4-PI497541, 5-PI604813, 6-PI604801,
7-PI604804, 8-PI604808, 9-PI604807, 10-PI476132, 11-PI497574, 12-PI604859











Table 3-5. Gene frequency of the 100 RAPD locus of A. pintoi germplasm accessions
Allele Loci/Gene Frequency
A4-250 A4-550 A4-670 A4-750 A4-870 A4-1000 A4-1100
0 0.4722 0.8611 0.6667 0.7222 0.5278 0.8056 0.5833
1 0.5278 0.1389 0.3333 0.2778 0.4722 0.1944 0.4167

A4-1350 A4-1500 A4-1700 A4-1900 A4-2100 A4-2500 A4-3000
0 0.3333 0.3889 0.6111 0.6667 0.8611 0.6667 0.7222
1 0.6667 0.6111 0.3889 0.3333 0.1389 0.3333 0.2778

A4-3100 B4-700 B4-850 B4-900 B4-1000 B4-1200 B4-1400
0 0.6667 0.0278 0.6944 0.8889 0.7500 0.5000 0.6944
1 0.3333 0.9722 0.3056 0.1111 0.2500 0.5000 0.3056

B4-1600 B4-1750 B4-2000 B4-2300 B4-2700 B4-3000 B5-500
0 0.8333 0.8889 0.3889 0.7500 0.5833 0.7500 0.9722
1 0.1667 0.1111 0.6111 0.2500 0.4167 0.2500 0.0278

B5-750 B5-1000 B5-1100 B5-1250 B5-1400 B5-1700 B5-2000
0 0.9444 0.5556 0.9722 0.0278 0.9444 0.8056 0.6667
1 0.0556 0.4444 0.0278 0.9722 0.0556 0.1944 0.3333

B5-2500 B5-3000 B5-3500 C2-465 C2-520 C2-600 C2-700
0 0.2500 0.4444 0.9444 0.9722 0.9722 0.8333 0.8056
1 0.7500 0.5556 0.0556 0.0278 0.0278 0.1667 0.1944

C2-750 C2-850 C2-1000 C2-1300 C2-1500 C2-1750 C2-2000
0 0.9444 0.4167 0.0000 0.4444 0.2778 0.6389 0.7500
1 0.0556 0.5833 1.0000 0.5556 0.7222 0.3611 0.2500

C2-2500 C2-2750 C2-3000 C2-3500 D4-580 D4-650 D4-750
0 0.6944 0.7222 0.5000 0.8611 0.8889 0.9444 0.8333
1 0.3056 0.2778 0.5000 0.1389 0.1111 0.0556 0.1667

D4-910 D4-1200 D4-1350 D4-1500 D4-2000 D13-370 D13-500
0 0.0833 0.9722 0.9444 0.9444 0.6944 0.7222 0.0833
1 0.9167 0.0278 0.0556 0.0556 0.3056 0.2778 0.9167

D13-650 D13-750 D13-850 D13-1000 D13-1200 D13-1400 D13-1500
0 0.2500 0.3611 0.7500 0.6667 0.2778 0.8056 0.8333
1 0.7500 0.6389 0.2500 0.3333 0.7222 0.1944 0.1667

D13-1700 D13-2000 D13-2100 D13-2300 D13-2500 D13-2800 D13-3310
0 0.5556 0.8056 0.9167 0.9167 0.9444 0.9722 0.9722
1 0.4444 0.1944 0.0833 0.0833 0.0556 0.0278 0.0278

E4-600 E4-900 E4-1000 E4-1100 E4-1250 E4-1500 E4-1800
0 0.9722 0.9444 0.8889 0.0833 0.0000 0.2222 0.7500
1 0.0278 0.0556 0.1111 0.9167 1.0000 0.7778 0.2500

E4-2100 E4-2500 E4-2800 G5-350 G5-450 G5-500 G5-650
0 0.5278 0.6944 0.9444 0.9444 0.7778 0.8889 0.6944
1 0.4722 0.3056 0.0556 0.0556 0.2222 0.1111 0.3056

G5-750 G5-850 G5-1000 G5-1100 G5-1250 G5-1350 G5-1750
0 0.7778 0.8611 0.6944 0.6667 0.2778 0.7500 0.9722
1 0.2222 0.1389 0.3056 0.3333 0.7222 0.2500 0.0278

G5-2500 G5-2750
0 0.9167 0.8056
1 0.0833 0.1944









Genetic diversity was estimated by Nei's gene diversity and by the Shannon-Weaver

diversity index (Shannon and Weaver, 1949). Nei's gene diversity is defined as

h = 1-_ xi 2, where xi is the frequency of ith allele.

Additionally, Shannon-Wiever's diversity index is defined as: H = -1 (pi log

pi)/log pi, where I = 1 to n, and p is the proportion of the total genotypes made up to the

ith genotype. In both indices values close to one indicate high genetic diversity.

Values of h and H among the 100 RAPD loci were extremely variable; with some

loci presenting numbers close to one and others small numbers close to zero (Table 3-6).

In general, H (Shannon-Wiever index) values were higher than the ones presented by h

(Nei index). To estimate the genetic diversity of the whole set of germplasm, the average

of both indices was calculated and was named total genetic diversity. Total h was

estimated as 0.29 + 0.16, and total H was estimated as 0.45 0.20. Both values can be

considered high, which demonstrates the great genetic variability contained in this set of

germplasm.

Another measure of genetic diversity is obtained by the genetic distance (D)

statistic. According to Nei (1972) genetic distance is calculated using the following

formula: D = -loge I, where I = Jxy/(JxJy)1/2, Jxy is the number of bands in common

among accessions x and y, and Jx and Jy is the number of bands of accessions x and y,

respectively. I values 1 when two accessions or populations have identical gene

frequencies over all loci examined, and zero when they share no alleles. Here genetic

distances were calculated between every pair of accessions and then the average genetic

distance was estimated as 0.36. This value also indicates that a great genetic diversity

exits among the germplasm evaluated in this research.











Table 3-6. Nei's gene diversity and by Shannon-Weaver diversity index of RAPD loci
Index Locus/Genetic Diversity
A4-250 A4-550 A4-670 A4-750 A4-870 A4-1000 A4-1100
h 0.50 0.24 0.44 0.40 0.50 0.31 0.49
H 0.69 0.40 0.64 0.59 0.69 0.49 0.68

A4-1350 A4-1500 A4-1700 A4-1900 A4-2100 A4-2500 A4-3000
h 0.44 0.48 0.48 0.44 0.24 0.44 0.40
H 0.64 0.67 0.67 0.64 0.40 0.64 0.59

A4-3100 B4-700 B4-850 B4-900 B4-1000 B4-1200 B4-1400
h 0.44 0.05 0.42 0.20 0.38 0.50 0.42
H 0.64 0.13 0.62 0.35 0.56 0.69 0.62

B4-1600 B4-1750 B4-2000 B4-2300 B4-2700 B4-3000 B5-500
h 0.28 0.20 0.48 0.38 0.49 0.38 0.05
H 0.45 0.35 0.67 0.56 0.68 0.56 0.13

B5-750 B5-1000 B5-1100 B5-1250 B5-1400 B5-1700 B5-2000
h 0.10 0.49 0.05 0.05 0.10 0.31 0.44
H 0.21 0.69 0.13 0.13 0.21 0.49 0.64

B5-2500 B5-3000 B5-3500 C2-465 C2-520 C2-600 C2-700
h 0.38 0.49 0.10 0.05 0.05 0.28 0.31
H 0.56 0.69 0.21 0.13 0.13 0.45 0.49

C2-750 C2-850 C2-1000 C2-1300 C2-1500 C2-1750 C2-2000
h 0.10 0.49 0.00 0.49 0.40 0.46 0.38
H 0.21 0.68 0.00 0.69 0.59 0.65 0.56

C2-2500 C2-2750 C2-3000 C2-3500 D4-580 D4-650 D4-750
h 0.42 0.40 0.50 0.24 0.20 0.10 0.28
H 0.62 0.59 0.69 0.40 0.35 0.21 0.45

D4-910 D4-1200 D4-1350 D4-1500 D4-2000 D13-370 D13-500
h 0.15 0.05 0.10 0.10 0.42 0.40 0.15
H 0.29 0.13 0.21 0.21 0.62 0.59 0.29

D13-650 D13-750 D13-850 D13-1000 D13-1200 D13-1400 D13-1500
h 0.38 0.46 0.38 0.44 0.40 0.31 0.28
H 0.56 0.65 0.56 0.64 0.59 0.49 0.45

D13-1700 D13-2000 D13-2100 D13-2300 D13-2500 D13-2800 D13-3310
h 0.49 0.31 0.15 0.15 0.10 0.05 0.05
H 0.69 0.49 0.29 0.29 0.21 0.13 0.13

E4-600 E4-900 E4-1000 E4-1100 E4-1250 E4-1500 E4-1800
h 0.05 0.10 0.20 0.15 0.00 0.35 0.38
H 0.13 0.21 0.35 0.29 0.00 0.53 0.56

E4-2100 E4-2500 E4-2800 G5-350 G5-450 G5-500 G5-650
h 0.50 0.42 0.10 0.10 0.35 0.20 0.42
H 0.69 0.62 0.21 0.21 0.53 0.35 0.62

G5-750 G5-850 G5-1000 G5-1100 G5-1250 G5-1350 G5-1750
h 0.35 0.24 0.42 0.44 0.40 0.38 0.05
H 0.53 0.40 0.62 0.64 0.59 0.56 0.13

G5-2500 G5-2750
h 0.15 0.31
H 0.29 0.49









Results obtained in this research are equivalent to those reported in the literature

for A. pintoi with a different set of germplasm accessions and primers. Gimenes et al.

(2000) used RAPD markers to evaluate the genetic variation of the A. pintoi Brazilian

germplasm collection and obtained a total of 104 different bands resolved by ten primers.

Average number of bands per primer was 10.4, ranging from 7 to 15. The proportion of

polymorphic loci was 90%, and accessions were grouped in 3 different groups based on

their genetic distances.

Bertozo et al. (1997), also working with a small germplasm collection ofA. pintoi,

reported 220 amplified bands resolved by 22 primers. Average number of bands per

primer was 7.5, ranging from three to 14 bands. The proportion of polymorphic loci

observed was 48.5%. The greatest genetic variability was observed within accessions

(0.53), while genetic variability among accessions was reported as 0.39.

Nobile et al. (2004) working with germplasm of A. glabrata (58), A. burkartii (12),

A. nitida (10), A. pseudovillosa (2), and A. lignosa (1) stated that 110 polymorphic RAPD

bands were resolved for 10 different primers. They also presented average genetic

distances of 0.30, 0.38, and 0.38, respectively to A. glabrata, A. nitida, and A. burkartii.

Hilu and Stalker (1995) worked with 26 accessions of wild species and

domesticated peanut, and reported that 10 primers resolved 132 RAPD bands. The most

variation was observed among accessions of A. cardenasii and A. glandulifera, and the

least was observed in A. hypogaea and A. monticola.

Findings in this research were also compared to results obtained with RAPD

markers to other species. Renu (2003) stated that 147 bands were determined by 15

primers when 47 germplasm lines ofPisum sativum L. were used to access the genetic









variability of the species. The proportion of polymorphic loci obtained was 87%. The

author concluded that RAPD markers are quick, easy to use and refractory to many

environmental influences, which makes the technique a very important complement to

traditional methods of germplasm characterization.

Lowe et al. (2003) studied 56 germplasm accessions of Pennisetum purpureum

using 67 RAPD bands. They concluded that the genetic diversity across all accessions

was high based on the Shannon-Weaver diversity index, which they estimated as 0.31.

The next step in this research was the grouping of the accessions based on their

genetic distances. In Figure 3-4 the dendogram constructed for the 35 A. pintoi accessions

is presented. Four groups were formed according to this dendogram. Group 1 was formed

by accessions PI 604798, 604801, 604804, 604805, 604808, 604809, 604814, 604815,

604817, 604856, and 604857. Group 2 was formed by accessions PI 497541, 604800,

604812, 604858, 604859, and CIAT 18745, 20826, 22150, 22152, 22256, 22260, and

22265. Group 3 was formed by accessions PI 476132, 497574, 604803, 604807, 604810,

604811, 604813, and CIAT 22159, 22234, 22271. Finally, accession PI 604799 was

grouped by itself in group 4.

Tissue Culture Regenerated Plants

Callus induction was achieved by both protocols from most of the A. pintoi

accessions evaluated. Callus quantity ratings ranged from 1 to 5, with the most of the

accessions having a mean rating of> 3 on at least one protocol.

There were highly significant (P < 0.001) effects of protocols, accessions, and the

interaction of protocols by accessions (Table 3-7). Protocol 1 was superior to Protocol 2

for both variables related with callus growth.









PI 604856
PI 604809
PI 604857
PI 604814
PI 604798
PI 604817 =
PI 604815
PI 604805
PI 604808
PI 604804
PI 604801
PI 604858
PI 604812
PI 604859
CIAT 22150
CIAT 22256
CIAT 22152
CIAT 22265
PI 604800
PI 497541
r CIAT 20826
CIAT 18745
CIAT 22260
PI 604803
PI 604807
CIAT 22234
PI 604810
PI 604813
PI 604811 =
PI 476132
PI 497574
CIAT 22159
CIAT 22271
PI 604799 3
=
4

Figure 3-4. Dendogram illustrating the genetic relationships among 35 A. pintoi
accessions based on Nei's genetic distance (Nei, 1972) obtained by 100
RAPD markers resolved by 8 random primers and generated by the UPGMA
method.









Mean values of callus quantity rating and callus weight for both protocols are presented

in Table 3-8. Differences in callus growth among protocols can be seen in Figure 3-5,

where dishes of accessions PI 604812 are presented.

Table 3-7. Analysis of variance table of callus rating and weight produced from Arachis
pintoi leaf discs incubated on two different tissue culture media
Source df Callus rating Callus weight
MS Pr>F MS Pr>F
Protocol 1 194.384 0.0001 4.585 0.0001
Accession 24 1.762 0.0001 0.059 0.0001
Protocol*Accession 24 1.006 0.0001 0.033 0.0001
Error 240 0.287 0.007
Total 289


Table 3-8. Callus rating and weight of Arachispintoi tissue culture incubated in two
different protocols
Protocol Callus rating Callus weight (g)
1 3.10 a* 0.333 a
2 1.46 b 0.078 b
* Means followed by the same letter in the same column were not different by Duncan's test (p<0.05).

Because the interaction of protocols by accessions was significant, average values

of callus rating and weight were presented by protocol. In Table 3-9 callus quantity rating

and weight of accessions incubated in Protocol 1 are presented. As expected, great

variability for these two variables was displayed among the accessions. Average callus

quantity rating was 3.11 and average callus weight 0.333 g. Accessions PI 604800,

604815, 604858, 604798, 604812, 604805, 604799, and CIAT 22234 were the ones with

superior values for callus quantity rating and callus weight. Average callus quantity

rating and weight were significantly lower for Protocol 2 than Protocol 1 (Table 3-8).

Overall average callus quantity rating was 1.46 and average callus weight was 0.078 g.

Accessions PI 497541 and 604857 did not produce any callus growth with Protocol 2.

Accessions, PI 604799, 604804, 604809, 604814, 604815, 604858 and CIAT 22256 had








the highest values on Protocol 2, but only PI 604799, 604815, and 604858 were in the
highest callus quantity group on both protocols (Table 3-10). Callus ratings and weight
were highly correlated (r = 0.98) when both protocols were analyzed, which validates the
callus rating scale used in this research.


Figure 3-5. Callus growth ofA. pintoi PI 604812 on two different tissue culture media.
Shoot regeneration was achieved for several accessions on both media. However,
shoot development was variable for each accession and medium, with no structures
indicative of somatic embryogenesis being detected. It seemed that callus quantity was
not correlated with shoot regeneration. In Protocol 1 shoot regeneration was obtained
from accessions PI 604856, 604857, 604805, 604811, 604809, 604814, 604818, and
CIAT 22234, 20826, 22152, and 22265.


~heyPr 1 1 Pr.~mcl~a

1b 4.gr
.... .. .
R-



.. .. ....










Table 3-9. Callus rating and callus weight ofArachispintoi callus induced with
Protocol 1
Accession Protocol 1
CIAT / PI number Callus rating Callus weight (g)
20826 3.00 cdef* 0.316 def
22152 2.17 gh 0.120 h
22234 3.50 abcde 0.430 abcd
22256 2.20 fgh 0.183 fgh
22265 3.00 cdef 0.264 efgh
22271 3.00 cdef 0.320 def
497541 3.00 cdef 0.325 def
604798 3.83 abc 0.425 abcd
604799 3.50 abcde 0.521 ab
604800 4.25 a 0.427 abcd
604803 3.33 bcde 0.281 defg
604804 3.17 cde 0.390 bcde
604805 3.50 abcde 0.559 a
604809 3.17 cde 0.327 def
604810 3.00 cdef 0.236 efgh
604811 2.17 gh 0.203 fgh
604812 3.67 abcd 0.427 abcd
604813 3.00 cdef 0.341 cdef
604814 2.83 defg 0.265 efgh
604815 4.00 ab 0.485 abc
604817 3.00 cdef 0.281 defg
604818 2.67 efg 0.234 efgh
604856 3.00 cdef 0.327 def
604857 1.83 h 0.146 hg
604858 4.00 ab 0.501 ab
*Means followed by the same letter in the column were not different by Duncan's test (p<0.05).

For Protocol 2, which yielded lower callus quantity, shoot regeneration was attained from

15 accessions: PI 604856, 604805, 604799, 604804, 604818, 604809, 604810, 604800,

604813, 604857, and CIAT 22256, 22234, 20826, 22152, and 22265. Figure 3-6

illustrates shoot regeneration of accession PI 604813 subcultured in Protocol 1 shoot

media induction (MS+lg L1 BA).

Developed shoots were transferred to root induction medium and then rooted plants

were placed in pellets in a growth chamber for acclimatization and subsequently










transferred to pots in the greenhouse. Figure 3-7 presents a picture of every stage of the

process used to tissue culture and regenerate A. pintoi accessions in this research.

Table 3-10. Callus rating and callus weight ofArachispintoi callus induced with
Protocol 2
Accession Protocol 2
PI/CIAT number Callus rating Callus weight (g)
20826 1.50 bcd 0.057 cde
22152 1.17 cd 0.065 cde
22234 1.50 bcd 0.090 bcd
22256 1.80 abc 0.057 cde
22265 1.67 bcd 0.096 bcd
22271 1.67 bcd 0.080 cde
497541 1.00 d 0.000 e
604798 1.50 bcd 0.072 cde
604799 2.00 ab 0.173 a
604800 1.67 bcd 0.103 bc
604803 1.00 d 0.050 cde
604804 1.83 abc 0.057 cde
604805 1.17 cd 0.041 de
604809 2.00 ab 0.105 bc
604810 1.00 d 0.027 e
604811 1.17 cd 0.044 de
604812 1.33 bcd 0.094 bcd
604813 1.17 cd 0.032 e
604814 2.33 a 0.142 ab
604815 1.83 abc 0.084 cde
604817 1.17 cd 0.058 cde
604818 1.50 bcd 0.055 cde
604856 1.50 bcd 0.077 cde
604857 1.00 d 0.000 e
604858 1.83 abc 0.143 ab
*Means followed by the same letter in the column were not different by Duncan's test (p<0.05).

Although, shoot regeneration was achieved for several accessions as stated before,

shoot development and root induction were not achieved in a reliable or repeatable way.

Root induction was very difficult to attain, and invariably many shoots died during this

process. The addition of 1 g L-1 of active charcoal in the root medium was tested to

evaluate its effect on root formation. However, little or no effect was achieved.


















PI 604813
Protocol 1
Figure 3-6. Shoot regeneration of A. pintoi accession PI 604812 subculture in medium
MS+lg L-1 BA (Protocol 1).
Several reports in the literature account for the developmental problems that face

organogenic tissue culture protocols of Leguminosae species. Flick et al. (1983)

suggested that plant regeneration has been very difficult to achieve among legumes, and

they continue, stating that in cases where it is attained, low frequency of regeneration is

often present. The same authors stated that forage legumes have been more difficult to

induce and form roots than seed legumes.


Figure 3-7. Representation of the different steps in the process used to tissue culture A.
pintoi.









Arachis species are among the legume species that have problems with shoot

development and root formation. Chengalrayan et al. (1995) obtained 33% average shoot

rooting in MS hormone free medium, and 52% of shoot rooting when NAA in

concentrations of 0.01, 0.05, and 0.5 mg L-1 was added to the medium with A.hypogaea

cultivar J.L. 24. However this result was specific for this particular cultivar and in general

A. hypogaea is recalcitrant in tissue culture.

Akasaka et al. (2000) were successful in inducing bud formation and shoot

development from leaf segments of A. hypogaea cv. Chico. The percentage of conversion

of buds to shoots was relatively high (34.7%). However, the highest frequency of shoot

regeneration was 14.3%, suggesting that some buds failed to grow into normal shoots and

plants.

Burtnick and Mroginski (1985) studying methods to regenerate plants from leaf

explants of A. pintoi cultured on Murashige and Skoog (MS) nutrient medium containing

different combinations of NAA (0.1-2 mg L1) and BA (0.1-3 mg L-) reported that callus

induction was nearly 100%. However calluses were usually friable and produced no

shoots or roots.

The same problems described above are also reported for other wild species of the

genus Arachis. McKently et al. (1990) were able to regenerate plants of A. glabrata using

MS medium supplemented with 3 to 5 mg L-1 of BAP. They reported, however that only

10% of shoot meristems continued growth and development into whole plants.

Due to the problems reported above, only 16 regenerated plants were obatined by

both protocols. Table 3-11 presents the number of regenerated plants of A. pintoi









accessions from protocols 1 and 2. All plants were phenotypically normal and uniform in

their appearance, with no somaclonal variation visually observed.

Table 3-11. Number of regenerated plants of Arachispintoi accessions cultured on two
different protocols
Accession Protocol
PI/CIAT number 1 2
-------------------------- no. pls -----------------------
20826 0 2
22234 1 0
22265 1 0
604799 0 1
604805 0 1
604809 4 0
604810 0 1
604811 3 0
604856 0 1
604857 1 0
Total 10 6

Differences in callus induction and growth, shoot regeneration and development,

root formation, and plant regeneration were observed among accessions. Callus induction

and growth was observed in all 25 accessions in Protocol 1, but only in 23 in Protocol 2.

Shoot induction was observed in 11 accessions in Protocol 1 and 15 in Protocol 2. Both

protocols were able to produce regenerated plants of five accessions. Although

differences in callus quantity rating and weight among protocols were observed earlier in

the tissue culture process, we can conclude that based on shoot development and plant

regeneration both protocols were equivalent. Additional research to study the shoot

development and rooting problems observed in this work should be carried out to secure

an efficient tissue culture protocol to regenerate A. pintoi plants.

Leaf tissue of the 16 regenerated plants was collected, ground, and used to extract

DNA using the protocol described earlier. No differences in RAPD band profiles were

detected between the original source plants and the tissue culture regenerated plants.









Although just a few plants were regenerated and analyzed, these results indicate that both

protocols could be suitable for in vitro plant genetic conservation ofA. pintoi accessions

since no variation in RAPD band profile was observed. Further tests on the regenerated

plants must be conducted to assure that no somaclonal variation occurred during the

tissue culture process.The application of in vitro genetic conservation has several

advantages, which include small space required to store large number of accessions, low

costs compared to growing and organizing plants annually in the field, and maintaining

large living collections of fruit trees (Hawkes et al., 2000).

Summary and Conclusions

Thirty-five A. pintoi accessions stored at the Southern Regional PI Station of the

National Plant Germplasm System (NPGS) located at Griffin, GA were used to study

genetic diversity at the molecular level using RAPD markers. Concurrently, two tissue

culture protocols, proposed by Rey et al. (2000) and Ngo and Quesenberry (2000), were

evaluated for their organogenesis ability, and capacity to generate somaclonal variation.

From the original 18 primers tested, amplifications were obtained in just eight of

them. Primers A4, B4, B5, C2, D4, D13, E4, and G5 amplified 100 different bands. The

average number of amplified bands per primer was 12.5, with primer D4 amplifying eight

fragments and primer D13 amplifying 16 fragments. The size of these 100 fragments

ranged from 250 bp to 3500 bp. From the 100 bands amplified, 98 presented

polymorphism with the only two exceptions being Primer C2-1000 bp and primer E4-

1250 bp. The average presence of bands per accession was 32, ranging from 20 to 44

bands. Ten bands were able to discriminate individual germplasm accessions. The

proportion of polymorphic RAPD loci was 89%. Genetic diversity of the whole set of

germplasm was estimated by Nei's gene diversity (h) and by Shannon-Weaver diversity









index (H). The average h was0.29 + 0.16, and average H was 0.45 0.20. Average

genetic distance was estimated as 0.36, and indicated that a great genetic diversity exits

among the germplasm evaluated in this research. Genetic distances were used to prepare

a dendogram for the 35 A. pintoi accessions, which separated them in four distinct

groups.

Callus induction was achieved on two different M.S. basal protocols after 28 d of

incubation. Analysis of variance demonstrated that Protocol 1 was superior to Protocol 2

for both variables related with callus growth. Callus rating and weight values of Protocol

1 confirmed the great variability for these two variables among the accessions. Average

callus rating was 3.11 and average callus weight 0.333 g. Accessions PI 604800, 604815,

604858, 604798, 604812, 604805, 604799, and CIAT 22234 showed the highest values.

In Protocol 2, average callus ratings and weight were significantly lower than the values

obtained with Protocol 1. Average callus rating was 1.49 and average callus weight was

0.072 g. Accessions PI 497541 and 604857 did not produce callus growth. Accessions PI

604799, 604804, 604809, 604814, 604815, 604858, and CIAT 22256 were the ones with

highest values. Shoot regeneration was achieved for several accessions on both media

with no structures indicative of somatic embryogenesis being detected. Callus growth

was not correlated with shoot regeneration. In Protocol 1 shoot regeneration was obtained

from 15 accessions, whereas in Protocol 2, shoot regeneration was attained from 18

accessions, but only PI 604856, 604818, 604809, and CIAT 20826 and 22234

regenerated shoots on both protocols. Root induction was very difficult to obtain, and

invariably many shoots died during this process. At the end, just 16 regenerated plants

were recovered between the two protocols.






55


Although differences in callus ratings and weight among protocols were observed

earlier in the tissue culture process, we conclude that based on shoot development and

plant regeneration both protocols were similar. RAPD band profiles of regenerated tissue

culture plants were similar to their parent plants. However, we should point out that the

number of regenerated plants was too small to conclusively affirm that genetic stability

will be maintained by these two protocols. Further investigations should be conduced to

definitely confirm our findings.














CHAPTER 4
MORPHOLOGICAL CHARACTERIZATION OF Arachispintoi GERMPLASM

Introduction

Morphological characterization is used to assess and understand the genetic

variability of germplasm collections. When working with a wild species of a genus of

interest, this usually is the chance to first gather basic knowledge about it.

Polymorphic, highly heritable morphological traits were originally used in early

scientific investigations of genetic diversity; such as the ones performed by Mendel and

DeVries.

In general, morphological studies did not involve sophisticated equipment or

laborious procedures, and these monogenic or oligogenic morphological traits were

simple, rapid, and inexpensive to score (Hawkes et al., 2000).

The information generated from this type of morphological characterization

research can be used to identify individual accessions based on a set of particular

phenotype traits. Such data can also be useful to estimate genetic diversity of a

germplasm collection, which will possibly impact the decision to enlarge the gene pool

by further collection trips. Additionally, this activity can generate information about the

genetic divergence among the accessions, which can be used to group accessions. Such

grouping of the germplasm accessions can be used to assemble core collections,

especially important in large germplasm banks.

Several statistical techniques may be used to assess genetic divergence. However,

when a large number of accessions are present, multivariate analysis is most appropriate.









The advantage of this technique is the ability to discriminate among accessions

considering multiple variables at the same time. Among the multivariate techniques the

most used in genetic studies are Principal Component and Cluster Analysis (Hawkes et

al., 2000).

Principal components analysis (PCA) has been widely used in genetic resource

related research. The technique can be used with several objectives:

* Quantification of genetic divergence among germplasm accessions
* Selection of divergent parental genotypes to hybridize
* Variable reduction in sets of data with great number of parameters
* Variable exclusion based on its contribution to the total variance
* Calculation of a similarity index for the purpose of grouping accessions

The goal of this research was to morphologically characterize the A. pintoi

germplasm accessions stored at the USDA-NPGS germplasm bank and to cluster the

accessions based on similarity indices.

Materials and Methods

Accessions of A. pintoi stored in the Southern Regional Plant Introduction Station

of the National Plant Germplasm System (NPGS) located at Griffin, Georgia were

transferred to the University of Florida in 2001 and 2002. A list of these accessions with

information related to the respective PI numbers and sites of collection is presented in

Appendix A.

Morphological characterization of the above accessions was accomplished by

evaluating each individual accession for a list of morphological descriptors selected from

the IBPGR/ICRISAT list of morphological descriptors for Arachis (1990 andl992). Data

from leaves, stems, flowers, pegs, pods, and seeds were collected from plants of each











accession. The list of morphological descriptors evaluated is presented in more detail in

Table 4.1.


Table 4-1. Morphological descriptors applied to Arachis pintoi accessions


Morphological descriptor
Flower/inflorescence
Flower standard width
Flower standard length
Flower standard color
Flower standard crescent
Flower wing width
Flower wing length
Flower keel length
Flower hypanthium length
Flower hypanthium width
Flower hypanthium color
Flower hypanthium hairiness
Flower pollen size
Flower pollen shape
Stem internode length
Stem internode diameter
Stem color
Stem hairiness
Stem bristles
Leaflet shape
Leaflet hairiness sup. surface
Leaflet hairiness margin
Leaflet hairiness inf. surface
Leaflet bristles sup. surface
Leaflet bristles margin
Leaflet bristles inf. surface
Leaflet length
Leaflet width
Leaflet length/Leaf width
Leaf Petiole length
Peg length
Peg width
Peg color
Peg hairiness
Pod weight
Pod length
Pod width
Pod beak
Pod reticulation
Seed width
Seed length
Seed weight

Seed color
Pod weight/Seed weight


Descriptor
code
FPI
FSW
FSL
FSC
FScr
FWW
FWL
FKL
FHL
FHW
FHC
FHH
FPSi
FPS
SIL
SID
SC
SH
SGH
LS
LHU
LHM
LHL
LGHU
LGHM
LGHL
LL
LW
LLLW
LPL
PegL
PegW
PegC
PegH
PodWe
PodL
PodW
PodB
PodR
SW
SL
SWe
Scolor

PodweSwe


# of structures
measured
10
10
10
10
10
10
10
10
10
10
10
10
10
10
50
50
50
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10

10


number
mm
mm
Color scale
Present/absent
mm
mm
mm
mm
mm
Present/absent
Present/absent
mm
IBPGR scale
mm
mm
Present/absent
Pres/abs/abun
Present/absent
Shape scale
Present/absent
Present/absent
Present/absent
Present/absent
Present/absent
Present/absent
mm
mm
mm
mm
mm
mm
Present/absent
Present/absent
grams
mm
mm
IBPGR scale
IBPGR scale
mm
mm
grams
ISCC-NBS color
chart
%


Equipment
used
Visual evaluation
Caliper
Caliper
Visual evaluation
Dissect scope
Caliper
Caliper
Caliper
Caliper
Caliper
Dissect scope
Dissect scope
Caliper
Microscope
Caliper
Caliper
Dissect scope
Dissect scope
Dissect scope
Visual evaluation
Dissect scope
Dissect scope
Dissect scope
Dissect scope
Dissect scope
Dissect scope
Caliper
Caliper
Caliper
Caliper
Caliper
Caliper
Visual evaluation
Visual evaluation
Scale
Caliper
Caliper
Dissect scope
Dissect scope
Caliper
Caliper
Scale

Dissect scope









Two plots of four plants (2 m x 2 m) were established at the Forage Research Unit

of the Agronomy Department of the University of Florida in Gainesville-FL. Six random

plants of each accession were selected from these plots and used as plant-part sources.

For each genotype, 10 stems per plant were collected, the terminal part of each

stem (1st 3 intemodes) was discarded and then five internodes were evaluated on each

stem. For the other part categories, 10 pieces were collected and evaluated. Continuous

variables were measured with a 15 cm electronic digital caliper (Chicago Brand Industrial

Inc., Fremont, CA), and categorical variables were scored under the dissecting

microscope using the standards proposed by IBPGR/ICRISAT (1990 and 1992).

The data were tabulated in a Microsoft Excel spreadsheet and the mean, standard

deviation, and range were calculated for quantitative descriptors, and the mode was

determined for qualitative ones.

Genetic variability among the accessions with respect to the morphological

descriptors was examined by calculating Simpson's (1949) and Shannon-Weaver's

(1949) diversity indices. These indices give a measure of phenotypic diversity and range

from zero to one, where one represents great genetic diversity and zero the opposite or no

genetic diversity. The indices correspond to the probability that two individuals randomly

selected from a group of populations will have the same morphological feature. The

formula to calculate both indices is presented below.

Shannon-Wiever Diversity index: H = -SUM (pi log pi)/log pi

Simpson Diversity index: D = 1-SUM (pi2)

where i= 1 to n, and p is the proportion of the total morphotypes made up of the ith

morphotype.









The data matrix was then analyzed with SAS software (SAS institute, 1989).

Phenotypic correlations among morphological descriptors were computed with the PROC

CORR procedure. After that, qualitative characteristics were transformed with PROC

PRINQUAL, and a principal component analysis was performed with the procedure

PROC PRINCOMP. Finally, a cluster analysis using the "Complete linkage method" was

prepared. Means of quantitative traits of each group were compared by using the

Newman-Keuls procedure (SAS institute, 1989).

Result and Discussions

Great morphological variability was observed among the accessions for all

descriptors, exceptions being, pollen size and shape, and bristles on the superior and

inferior leaf surface, which showed no polymorphism. In Table 4-2, the mean/mode,

standard deviation, variance and range of each morphological descriptor are listed.

Flowers arise from inflorescences located at reproductive buds under the lea axils.

According to Conagin (1959), each inflorescence produces one to nine flowers, which

will bloom in a sequence, usually with a 1-2 d interval. Numbers observed in this study

revealed accessions with a mean of two flowers per inflorescence and a maximum of four

flowers per inflorescence. Although important, these differences do not seem to have

great impact in terms of seed production.

Arachispintoi flowers are typical of papilionacea legumes possessing five petals.

They displayed one standard petal, two wing petals, and a keel that is actually formed by

the fusion of two petals. The other flower structures are the calyx and the hypanthium.










Table 4-2. Morphological characteristics ofArachis pintoi germplasm accessions
Morphological Mean/Mode Standard Variance Range
Descriptor Deviation
FPI 3.14 0.33 0.11 2.00 3.80
FSW 15.31 1.63 2.67 11.99 18.85
FSL 11.58 1.10 1.22 9.84- 14.45
FSC Yellow -
FScr Present Present/Absent
FWW 6.27 0.69 0.47 4.47 7.95
FWL 8.07 0.69 0.48 6.92 9.80
FKL 5.09 0.23 0.05 4.45 5.55
FHL 70.66 12.58 158.28 52.39 -104.79
FHW 0.97 0.10 0.01 0.77- 1.12
FHC Absent Present/Absent
FHH Present Present/Absent
FPSi 4.00 0.00 0.00
FPS Round -
SIL 29.47 7.43 55.27 18.23 52.23
SID 2.95 0.84 0.71 2.05 7.15
SC Absent Present/Absent
SH Present Present/Absent
SGH Absent Present/Absent
LS Obovate -
LHU Absent Present/Absent
LHM Absent Present/Absent
LHL Present Present/Absent
LGHU Absent Present/Absent
LGHM Present Present/Absent
LGHL Absent Present/Absent
LL 24.89 4.07 16.60 15.79 32.03
LW 15.14 2.79 7.80 9.45 20.78
LLLW 1.66 0.21 0.04 1.29 2.06
LPL 22.26 6.40 40.91 11.67 39.33
PegL 11.90 3.36 11.28 6.30 21.40
PegW 0.94 0.14 0.02 0.57- 1.23
PegC Present Present/Absent
PegH Present Present/Absent
PodWe 0.13 0.07 0.01 0.07 0.25
PodL 9.33 4.52 20.40 7.55- 14.25
PodW 4.70 2.22 4.91 4.53 6.91
PodB Moderate -
PodR Slight -
SW 3.89 1.86 3.45 3.55 -5.46
SL 7.13 3.50 12.22 5.71- 10.51
SWe 0.10 0.06 0.001 0.04- 0.18
Scolor Orange -
yellowish
PodweSwe 75.47 6.71 44.99 59.71 87.07









Most of the accessions (91%) have yellow flowers, with the 9% exception showing

a lighter shade of yellow classified as lemon yellow (Figure 4.1). This was the case of

accessions PI 604814, PI 604818, and CIAT 20826. Standard color is considered a very

good genetic marker, and so it could be used in genetic studies with A. pintoi.

Our results in terms of standard petals color are different from the ones reported by

Maass et al. (1993) who performed morphological characterization of eight accessions of

A. pintoi. They reported that lemon yellow was the standard petal color displayed by all

eight germplasm lines. Qualitative traits such standard petals color are subject to

individual interpretation and that is probably why their results are very different them the

ones reported in this research.

Also different are the results of Upadhyaya (2003) who worked with the ICRISAT

core collection (1704 accessions) of Arachis hypogaea. He stated that 97% of the

germplasm in that collection had orange standard petals. White, lemon yellow, yellow,

and dark orange were not observed in any of his accessions.

Differences in standard petal length and width, wing length and width, keel length,

and hypanthium length and width were also observed. Those features are directly related

to overall flower size. Figure 4.2 illustrates the large differences displayed by the

germplasm in relation to this group of characteristics. Also, differences in hypanthium

color were found and can be observed in Figure 4.2. The mode for this characteristic was

absence of color, but some accessions presented a distinct purple color, which is probably

associated with the presence of anthocyanin. One of the reported functions of

anthocyanin is to attract insects: since it absorbs UV radiation that is very attractive to

them (Mann, 1987).










Arachispintoi stems were hollow with reproductive and vegetative nodes

occurring along the stem length. Vegetative nodes have the ability to root which allows

the species to be vegetatively propagated, a desirable agronomic characteristic.


















Figure 4-1. Flower standard colors of Arachispintoi germplasm.


Figure 4-2. Flower sizes and hypanthium colors displayed by Arachispintoi germplasm.

Large differences were observed in relation to stem internode length and diameter,

and stem color. The average internode length was 29 mm, ranging from 18 to 52 mm.

Average internode diameter was 3 mm, varying from 2 to 7 mm. As in the case of

hypanthium color, it seems that stem internode color is determined by the presence of


1
,i
I t


' MI









anthocyanin. However, here more variants in color were present than in the former

feature, although scoring was just made in terms of presence and absence. Absence was

the mode for this characteristic with 57% of the accessions showing no purple coloration,

which is translated as a light green color. Among those that displayed stem coloration,

yellowish, pinkish, light and deep purple were observed (Figure 4-3).

A. pintoi leaves are compound with four leaflets. Usually the basal leaflets were

smaller than the distal, and exhibited an elliptic shape. As observed in flower and stem

features, large differences were also present in leaves.




















Figure 4-3. Stem characteristics of Arachis pintoi germplasm.

When examining leaflet characteristics, the features related to overall leaf size (leaflet

length and width and petiole length) and shape were the ones that showed the most

variability. Average leaflet length was 25 mm with a range of 16 to 32 mm. Leaflet width

varied from 9 to 21 mm, with an average value of 15 mm. Due to leaflet length and width

variation, an array of leaflet sizes arise (Figure 4-4). In terms of leaflet shape ten different

types were present with obovate shape being the mode, this agreed with the findings of









Maass et al. (1993). However, Upadhyaya (2003) reported different results among the

1704 accessions of the ICRISAT A. hypogaea core collection, which had elliptic leaf

shape without exceptions.

Similar to others species of the genus, A. pintoi also possesses geocarpic fruits.

Flowers self-pollinate and then lose their petals as the fertilized ovary begins to enlarge.

The budding ovary or "peg" grows under the ground, away from the plant, forming a

small stem which contains the ovary and the embryo in its tip. The embryo turns

horizontal to the soil surface and begins to mature inside a pod. In the cultivated species

(A. hypogaea), two to four or more seeds may be formed in each pod. However, in A.

pintoi, single seed pods were exclusively observed.



















Figure 4-4. Leaflet characteristics ofArachispintoi germplasm.

Pegs, pods, and seeds exhibited an array of variability among the accessions. One

of the differences is the fact that accessions PI 604804, 604813, 604817, and CIAT

22152, 22159, 22234 did not produce any of the above plant structures under the

conditions of these experiments. Additionally, differences were also present in the









accessions that produced these structures. Average peg length was 12 mm, ranging from

6 to 21 mm. With respect to the other peg features, average peg width was 0.90 mm, and

presence of peg color and hairiness were the mode.

Pod size and weight showed great divergence among the germplasm. These

characteristics were most impacted by pod length and width. Large pod size and weight

were reflected in large seed size and weight. Figure 4-5 illustrates the differences in seed

size among the genotypes. Average seed weight was 0.12 g, ranging from 0.04 to 0.18 g.

Also, the average ratio of seed weight/pod weight was 0.75, ranging from 0.59 to 0.87.














Figure 4-5. Seed characteristics ofArachispintoi germplasm.

Phenotypic correlations were calculated among morphological descriptors, and

Pearson's correlation coefficient and significance test were also calculated (Appendix C).

Pollen size and shape (FPSi and FPS) and bristles on the superior and inferior leaf surface

(LGHU and LGHL) were not included because they did not show any variability. In

addition, leaf length/leaf width ratio (LLLW) and pod weight/seed weight ratio

(PodweSwe) were left out of the correlation matrix because they were derived from two

other variables. Accessions PI 604804, 604813, 604817, and CIAT 22152, 22159, 22234

were excluded because they did not produce any pegs, pods, or seeds. In summary, 35









morphological descriptors and 28 germplasm accessions were used to calculate

correlation coefficients.

Therefore, with 26 degrees of freedom (df = N-2 or 28-2), any correlation

coefficient with an absolute value greater than 0.361 had a P value of < 0.05. From 595

correlations computed, 96 were higher than 0.361. These 96 significant correlations could

be divided into eight different groups (Table 4-3).

Table 4-3. Correlations among morphological descriptors ofArachis pintoi germplasm
Group Group Description
1 Correlations between flower features dimensions
2 Correlations between leaflet features dimensions
3 Correlations between flower features dimensions and leaflet dimensions
4 Correlations between stem diameter and leaflet dimensions
5 Correlations between flower features dimensions and pod and seed dimensions
6 Correlations between stem diameter and pod and seed dimensions
7 Correlations between leaflet features dimensions and pod and seed dimensions
8 Correlations between pod features dimensions and seed dimensions

Correlations in Groups 1, 2, and 8 are not very meaningful and could be explained

by the fact that part proportions should be maintained within each plant organ or

structure. So, flowers with large standards length will also have large standard width,

wing length and width, and keel length. For the same reason we should expect that pods

with large measures will also have large seed if an adequate development had occurred,

because the seeds are enclosed by the pod.

Correlations in Groups 3 and 4 would have agronomic importance if leaflet size is

translated into higher vegetative mass production. Likewise leaf/stem ratio could be

important assuming leaves have higher nutritive values than stem. These are very

important features in forage species. These correlations suggest that flower size and stem

diameter could be included as selection indices in A. pintoi forage selection and breeding









programs. The same idea could be applied to Groups 5, 6, and 7, where stem, flower, and

leaflet size appear to impact pod and seed size.

Although 96 correlation coefficients were considered significant (r = 0.361).

Skinner et al. (1999) suggested that only correlation coefficients with absolute values

higher than 0.71 should be considered biologically meaningful. They explain that only in

these situations is more than 50% of the variance of one trait is predicted by the other.

Reexamining the correlation coefficient table (Appendix C) under this new

criterion only 29 correlation coefficients out of 595 would be considered biologically

meaningful. In this situation only Groups 1, 2, 7, and 8 remain in place. As stated before,

Groups 1, 2, and 8, do not have great biological importance, since they could be

explained by the fact that part proportions should be maintained within each plant organ

or structure. Therefore, only correlations relative to Group 7 may have meaningful

biological importance. Examining this information more carefully, we can observe that

meaningful correlations were found between leaf length (LL) and Pod weight (Podwe);

leaf length (LL) and pod width (PodW); leaf length (LL) and seed weight (Swe); leaf

length (LL) and seed width (SW); and finally leaf length (LL) and seed length (SL).

Thus, leaf length could be used as a selection criterion in programs where increased seed

size/weight is one of the objectives.

One of the biggest problems with forage legumes species is slow establishment,

which sometimes is associated with small seed size. Having a large seed could represent

more stored reserves and higher seedling vigor, which would reduce the establishment

time. These suggestions based on the above correlations should be verified in future

research to evaluate the use of these characteristics as selection indices.









To support this hypothesis we could examine the findings of Skinner et al. (1999)

who worked with the Australian annual Medicago collection (20997 accessions)

measuring 27 traits. They reported that the traits seeds per gram and winter and spring

herbage yield were correlated (0.42 and 0.29). They also stated that seeds per gram and

seedling vigor were correlated (0.44). Upadhyaya (2003) also found correlations between

100 seed weight and yield (0.32) in the ICRISAT Arachis hypogaea core collection.

Although we can affirm that phenotypic genetic diversity was observed among the

accessions by examining the information contained in Appendix B and Table 4.2 there is

a need to quantify this diversity. To achieve this goal we should make use of a genetic

diversity index. Two of the most used diversity indices are Shannon-Weaver and

Simpson indices. These indices are often used in ecological studies where species

richness and composition of a particular community or ecosystem are evaluated.

Recently, these indices have been applied to quantify genetic diversity of germplasm

collections when phenotypic frequencies are collected. The greater the index value, the

greater the genetic diversity. In Table 4-4 the values of the indices for each

morphological descriptor are presented. The total genetic diversity was also calculated,

and it is an indication of how different the accessions are in relation to the morphological

features utilized in this research.

Diversity values were variable among traits, but in general all morphological

features expressed high genetic diversity. According to Simpson's index, leaf shape

(0.83), seed color (0.82), flower standard width (0.81), and seed length (0.75) were the

descriptors with greatest diversity. In opposition, the lowest diversity values were related









to flower standard color (0.18), flower standard crescent (0.24), and flower hypanthium

hairiness (0.24). The total Simpson's index to all morphological descriptors was 0.58.

Shannon-Weaver's diversity values were in general higher than Simpson's.

Descriptors with higher values for the Shannon-Weaver index were leaf hairiness inferior

surface (1.00), flower hypanthium color (1.00), leaf hairiness margins, and pod

reticulation (0.97). Total genetic diversity was estimated as 0.71.

Shannon-Weaver diversity index values observed in this work are much higher than

the ones observed by Upadhyaya et al. (2002), who applied 38 agromorphological

descriptors to the whole A. hypogaea ICRISAT collection (13342 accessions). These

authors found a total genetic diversity value of 0.50. Leaflet length (0.62) and shelling

percentage (0.62) were the traits showing most variation.

These values are also higher than those reported by Upadhyaya (2003) who

evaluated a core collection prepared using the results of the previous work. He obtained a

total diversity index of 0.44 in 32 agromorphological traits. He concluded that the core

collection had significant variation for the morphological and agronomic traits evaluated.

A principal components analysis (PCA) was performed with the goal of

discriminating among accessions and grouping even further. Principal components

analysis (PCA) can be used in sets of data with large number of variables. The goal of

PCA is to provide a reduced dimension model that would indicate measured differences

among groups. It also can contribute to a better understanding of the set of variables by

describing how much of the total variance is explained by each one. With this objective

the PCA was performed with the matrix of morphological data generated by applying the

list of descriptors presented in Table 4-1.









Table 4-4. Simpson and Shannon-Weaver diversity indices
morphological descriptors
Morphological Descriptor Simpson
Flower/inflorescence 0.60
Flower standard width 0.81
Flower standard length 0.68
Flower standard color 0.18


Flower standard crescent
Flower wing width
Flower wing length
Flower keel length
Flower hypanthium length
Flower hypanthium width
Flower hypanthium color
Flower hypanthium hairiness
Stem internode length
Stem internode diameter
Stem color
Stem hairiness
Stem bristles
Leaf shape
Leaflet hairiness inf. surface
Leaflet hairiness margins
Leaf bristles margins
Leaf length
Leaf width
Leaf length/Leaf width
Leaf petiole length
Peg length
Peg width
Peg color
Peg hairiness
Pod weight
Pod length
Pod width
Pod beak
Pod reticulation
Seed width
Seed length
Seed weight
Seed color
Pod weight/Seed weight
Total


0.24
0.56
0.60
0.54
0.73
0.73
0.50
0.24
0.50
0.65
0.47
0.63
0.36
0.83
0.50
0.48
0.36
0.64
0.58
0.67
0.74
0.59
0.46
0.45
0.45
0.69
0.47
0.74
0.73
0.64
0.69
0.75
0.64
0.82
0.57
0.58


for Arachis. pintoi


Shannon-Weaver
0.34
0.89
0.80
0.47
0.58
0.33
0.33
0.74
0.89
0.46
1.00
0.58
0.58
0.45
0.96
0.96
0.80
0.83
1.00
0.98
0.80
0.46
0.34
1.00
0.84
0.50
0.71
0.92
0.92
0.45
0.74
0.89
0.87
0.97
0.48
0.88
0.37
0.89
0.87
0.71









Variables pollen size and shape (FPSi and FPS) and leaf bristles superior and

inferior surface (LGHU and LGHL) were not included because they did not show any

variability. In addition, leaf length/leaf width ratio (LLLW) and pod weight/seed weight

ratio (PodweSwe) were also left out because they are derived from two other variables.

After the first analysis, the leaflet hairiness superior surface variable (LHU) was also

excluded from the analysis because it contributed little to the exploration of total

variance.

The first five principal components (PCs) were responsible for 67.7% of the total

variation (Table 4-5). Values similar to these were reported by Stalker (1990),

Upadhyaya et al. (2002), Upadhyaya (2003), and Upadhyaya et al. (2002), who worked

with wild species ofArachis, groundnut, and chickpea germplasm collections,

respectively, to explain their results. The first PC explained 30.0% of the variation, the

second accounted for 15.2%, the third for 10.1%, the fourth for 6.6%, and the fifth

explained 5.8% of the total variation.

Examining the variable loadings of the first five PCs (Table 4-5) we can clearly

observe that the characteristics of pegs, pods, and seeds are the ones with highest

contribution to PC1. Therefore, PC1 could be termed the "sexual reproduction axis".

Performing the same exam to PC2 we can conclude that the features related with flower

and leaf dimensions were the ones with highest loadings. Therefore, PC2 could be called

the "vegetative axis". Finally, examining the loading of PC3 we note that features related

to the shape, color, and hairiness of morphological structures were the ones with the most

contribution. Those are all qualitative features and because of that we could call PC3 the

"qualitative axis".










Therefore, we could state that the principal components analysis was able to

discriminate and separate the accessions in terms of these three dimensions, represented

by "sexual reproduction", "vegetative", and "qualitative" axes. This is clearly observed

when accessions were projected in two-dimensional graphs formed by PC1 and PC2, PC1

and PC3, and PC2 and PC3 (Figures 4-6, 4-7, and 4-8).

Table 4-5. Vector loadings and percentage of variation explained by the first five
principal components for morphological characteristics ofArachispintoi
Principal components
Characteristics 1 2 3 4 5
Variance explained (%) 30.01 15.15 10.09 6.63 5.76
Cumulative variance explained (%) 30.01 45.16 55.25 61.88 67.65
Flower/inflorescence 0.077 0.041 0.215 0.036 -0.019
Flower standard width -0.060 0.338 -0.085 -0.092 0.027
Flower standard length -0.068 0.369 -0.028 0.070 -0.014
Flower standard color -0.060 -0.038 0.072 0.333 0.317
Flower standard crescent 0.145 -0.117 0.056 -0.132 0.248
Flower wing width -0.065 0.331 0.006 -0.099 0.122
Flower wing length -0.080 0.350 -0.017 0.025 0.085
Flower keel length -0.084 0.213 0.146 0.200 0.141
Flower hypanthium length -0.097 0.299 0.089 0.136 -0.142
Flower hypanthium width 0.017 0.218 0.273 0.235 -0.128
Flower hypanthium color 0.023 0.013 -0.164 -0.165 -0.382
Flower hypanthium hairiness -0.009 0.087 0.133 0.124 -0.422
Stem internode length -0.030 0.079 -0.304 -0.145 0.051
Stem internode diameter -0.043 -0.036 0.157 -0.194 0.338
Stem color -0.076 -0.042 0.090 0.203 -0.274
Stem hairiness 0.015 0.051 0.438 0.104 -0.115
Stem bristles -0.033 0.198 0.013 0.085 0.112
Leaflet shape -0.105 0.074 -0.275 0.231 0.115
Leaflet hairiness margin -0.056 -0.063 0.302 -0.290 0.023
Leaflet hairiness inf. surface 0.006 -0.086 0.366 -0.180 0.139
Leaflet bristles margin 0.021 0.071 -0.186 0.285 0.290
Leaflet length 0.057 0.285 0.039 -0.310 0.011
Leaflet width 0.155 0.233 0.071 -0.245 -0.101
Leaf Petiole length 0.062 0.251 -0.138 -0.248 0.028
Peg length 0.231 0.070 -0.092 0.038 -0.165
Peg width 0.256 -0.001 0.015 0.185 -0.075
Peg color 0.138 -0.090 -0.195 -0.134 -0.140
Peg hairiness 0.109 -0.002 -0.240 0.113 -0.075
Pod weight 0.273 0.099 0.016 -0.029 0.095
Pod length 0.293 0.016 0.023 0.046 0.032
Pod width 0.291 0.033 0.047 0.079 0.035
Pod beak 0.274 0.017 0.003 -0.034 0.022
Pod reticulation 0.288 -0.016 0.024 0.111 -0.005
Seed width 0.292 0.044 0.044 0.066 0.041
Seed length 0.293 0.042 0.008 0.001 0.042
Seed weight 0.268 0.101 -0.005 -0.031 0.122
Seed color 0.287 -0.028 0.003 0.099 0.012







74










:4~ '4. 2 2271
am~* vr; L ;c


497It'1 604 '

-2,. -



22234

W34804 4 "1 2'7 22-2

-7,6--





--5-0 -2.5 O-3 25 5.0 7
Rir2
Figure 4-6. Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC and PC2.

Figure 4-6 corresponds to the plane formed by PC1 and PC2 dimensions. PCI

values are on the Y axis and PC2 on the X axis. Moving from the bottom of the Y axis

where coefficients were negative to the top where they were positive represents moving

from lower values of peg, pods, seed dimensions, and weight. Doing the same to the X

axis represented by PC2, and moving from the left (- coefficients) to the right (+

coefficients) means that we are moving from lower values of flower and leaf dimensions

to higher values. Accessions were obviously discriminated, and two groups were formed.

The group in the bottom was composed of six accessions and the other 29 in the top of

the graph. These six accessions in the bottom are the ones with no pegs, seeds, and pods.


















60498M MW4R87 604814
2.5-- ,60 ....0 1 604812 : 4604798
604801: "" i' ,
22265 604799 604807.97574 604805
22256 ,.476132 60403
20826 **604811 604811 604858
*22260 604857 604815
22271 604856 604859
0.0-- 22150 *604810
60480Q 497541
497541









: 22152 604804 604817

*22159
-7.5
8604813



-10.0--

-5.0 -2.5 0.0 25 50
R-ir-

Figure 4-7. Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC1 and PC3.


In the same way, figure 4-7 corresponds to the plane formed by PC1 and PC3


dimensions, where PC1 represents the Y axis and PC3 represents the X axis. Here, the Y


axis indicates the same tendency as in the previous figure. PC3 is the represented by the


qualitative characteristics (shape, color, and hairiness), and a more comprehensive


reading of the tendency here is difficult to achieve. However, accessions were also


discriminated with two groups formed. The group in the bottom was again composed of


the six accessions with no pegs, seeds, and pods.













6--
22159 ,22150



4- *22256
604807 604798 *

,604818
2 604818 604859
22234 604858
(. *22260 476132
*20826
0 22152 604808 604805
S 6080 648 0809 0803
22265 604799 18745 604814 *
497574 604813
604800 604811
604817
-2 *22271 604815
497541 604856
604857 604804
604812
S604810
-4


-4 -2 0 2 4
Prin3
Figure 4-8. Projection of the 35 Arachispintoi accessions in a two-dimensional graph
defined by PC2 and PC3.

Finally, figure 4-8 represents the plane formed by PC2 and PC3 dimensions, where

PC2 represents the Y axis and PC3 represents the X axis. Moving from the bottom (-

coefficients) to the top (+ coefficients) means that we are moving from smaller to bigger

flowers and leaves. Accession discrimination was also obtained. However, group

formation was more difficult to obtain, once accessions were spread in the plane.

We could state that by using this set of descriptors we accomplished the

discrimination of the A. pintoi germplasm. All the 35 morphological features used in the

PCA presented high loading values at least once when the first five PC were analyzed,

reinforcing the importance of each one as an A. pintoi descriptor.









Since discrimination was obtained by using of the genetic diversity indices and

also by the principal component analysis, the next step was to perform a grouping or

cluster analysis. The first nine principal components were used to execute a cluster

analysis using the complete linkage clusters method. The dendogram resulting from this

analysis is presented in Figure 4-9.

From the dendogram we can differentiate four distinct groups of accessions. Group

1 was composed of accessions PI 497541, 497574, 604800, 604810, 604811, 604812,

604815, 604856, 604857, and CIAT 22271. Group 2 was formed by PI 604799, 604801,

604808, 604809, 604814, 604818, CIAT 18745, 20826, 22260, and 22265. Group 3 was

composed by PI 476132, 604798, 604803, 604805, 604807, 604858, 604859, CIAT

22150, and 22256. Finally, Group 4 was composed by PI 604804, 604813, 604817, CIAT

22152, 22159, and 22234.

Morphological characteristics of each of the four groups created by the cluster

analysis are presented in Table 4-6. Based on these features we could characterize the

four groups. Group 1 was composed by accessions with small leaves, flowers, pods, pegs,

and seeds. Therefore, we could name this group as the "small type" group. In Figure 4-

10, accession PI 497541, a member of this group is presented.

Group 2 was formed by accessions with intermediate size features and could be

called as the "intermediate type" group. Accession PI 604814 is presented as a

representative of this group in Figure 4-11.

Group 3 was formed by accessions with large leaves, flowers, pegs, pods, and

seeds. It was named as the "large type" group due to their large features. In Figure 4-12

accession PI 604798 is displayed as a representative of this type.









78



















22150
222l56









60485
4?604788
42132





604814
60481A

22260
M8745

60M48I



604801

604799

497574

:60481.5



6E04812

-60485

_22271


497541


8 W i 8
d d d d T



Figure 4-9. Dendogram of 35 Arachispintoi accessions based on morphological

descriptors and the first nine principal components.


Finally, Group 4 was composed of accessions which did not produce any pegs,


pods, and seeds. Their leaves, flowers, and stems characteristics were relatively similar to










the ones displayed by the members of Group 3, which presented large sizes for these

structures. So, based on the fact that this group did not produced seeds, this group could

be called the "vegetative type" group. Accession 604817 is presented in Figure 4-13 as an

exemplar of this group.

Table 4-6. Morphological characteristics of Arachispintoi accession groups obtained by
the cluster analysis
Characteristics Group 1 Group 2 Group 3 Group 4
Quantitative descriptorsT
Flower/inflorescence 3.14 aa 3.03 a 3.34 a 3.00 a
Flower standard width 14.17 b 15.15 ab 16.23 a 16.07 a
Flower standard length 10.80 b 11.32 ab 12.40 a 12.06 a
Flower wing width 5.72 c 6.13 bc 6.81 a 6.62 ab
Flower wing length 7.53 b 7.85 b 8.64 a 8.49 a
Flower keel length 4.98 a 5.02 a 5.21 a 5.20 a
Flower hypanthium length 65.00 b 64.00 b 79.00 a 78.00 a
Flower hypanthium width 0.94 b 0.91b 1.08 a 0.94 b
Stem intemode length 26.40 a 32.27 a 28.13 a 31.94 a
Stem internode diameter 2.72 a 2.79 a 3.14 a 3.34 a
Leaflet length 20.68 b 26.61 a 27.80 a 24.69 a
Leaflet width 13.01 b 16.73 a 17.23 a 12.91 b
Leaf Petiole length 16.13 c 27.44 a 23.89 ab 21.42 b
Peg length 9.85 b 13.66 a 12.14 ab 0.00 c
Peg width 0.97 a 0.89 a 0.96 a 0.00 b
Pod weight 0.11 b 0.18 a 0.17 a 0.00 c
Pod length 10.23 b 12.07 a 11.53 a 0.00 c
Pod width 5.26 c 5.66 b 6.12 a 0.00 d
Seed width 4.17 c 4.80 b 5.15 a 0.00 d
Seed length 7.33 b 9.56 a 8.96 a 0.00 c
Seed weight 0.08 b 0.14 a 0.13 a 0.00 c
Qualitative descriptorsTf
Flower standard color Yellow Yellow Yellow Yellow
Flower standard crescent Present Present Present Absent
Flower hypanthium color Absent Present Absent Absent
Flower hypanthium hairiness Present Present Present Present
Stem color Absent Absent Absent Present
Stem hairiness Present Absent Abundant Present
Stem bristles Absent Absent Absent Absent
Leaflet shape Obovate Obovate Narrow ellipitc Obovate
Leaflet hairiness sup. surface Absent Absent Absent Absent
Leaflet hairiness margin Absent Absent Present Present
Leaflet hairiness inf. surface Present Absent Present Absent
Leaflet bristles margin Present Present Present Present
Peg color Present Present Absent
Peg hairiness Present Present Absent
Pod beak Moderate Slight Moderate
Pod reticulation Moderate Moderate Slight
Seed color Yellow brownish Orange yellowishOrange yellowish-
T Mean
tt Mode
a Differences between means of different groups were tested by Student Newman-Keuls test. Means followed by same letter are not
different atp=0.05.





80





)









PI 497541
Figure 4-10. Group 1 representative accession (PI 497541).










"-"



SPI 604814
Figure 4-11. Group 2 representative accession (PI 604814).




















PI 604798
Figure 4-12. Group 3 representative accession (PI 604798).













No Seeds



P PI 604817
Figure 4-13. Group 4 representative accession (PI 604817).









Summary and Conclusions

Thirty-five germplasm accessions ofArachispintoi were morphologically

characterized using a list of descriptors prepared by IBPGR/ICRISAT (1990 and 1992)

Data from stems, leaves, flowers, pegs, pods, and seeds were collected and comparisons

among accessions were made, based on the mean, standard deviations, and range of the

quantitative features and the mode of the qualitative characteristics. Phenotypic

correlations were conducted among descriptors, and Pearson's correlation coefficient and

significance test were also calculated. Simpson and Shannon-Weaver's diversity index

were computed for each descriptor to access the genetic diversity among the accessions

for individual descriptors. Principal component analysis was then executed to

discriminate the accessions, and finally a cluster analysis was performed to group the

germplasm in accordance with its morphological similarities.

The germplasm presented great morphological variability with all the descriptors,

except pollen size and shape, leaf bristles superior and inferior surface, showing

polymorphism.

From 595 correlations computed, 96 were statistically significant. These 96

significant correlations could be divided in eight different groups. However, when only

the biologically meaningful correlations (r > 0.71) were evaluated, the number of

significant correlations dropped to 29, and only four groups were observed. These

meaningful correlations were found between leaf length (LL) and pod weight (Podwe);

leaf length (LL) and pod width (PodW); leaf length (LL) and seed weight (Swe); leaf

length (LL) and seed width (SW); and finally leaf length (LL) and seed length (SL).

Thus, leaf length could be used as a selection criterion in programs where increased seed

size is one of the objectives.









Diversity values were variable among traits, but in general all morphological

features expressed high genetic diversity. According to Simpson's index, leaf shape

(0.83), seed color (0.82), flower standard width (0.81), and seed length (0.75) were the

descriptors with most diversity. In contrast, the lowest values were related to flower

standard color (0.18), flower standard crescent (0.24), and flower hypanthium hairiness

(0.24). The total Simpson's index to all morphological descriptors was 0.58.

Alternatively, Shannon-Weaver's values were in general higher than Simpson's,

with higher values displayed by leaf hairiness inferior surface (1.00), flower hypanthium

color (1.00), leaf hairiness margins (0.80), and pod reticulation (0.97). Total genetic

diversity was estimated as 0.71.

The first five principal components explained 67.7% of the total variation, with

PC1 explaining 30.0% of the variation, PC2 15.2%, PC3 10.1%, PC4 6.6%, and PC5

5.8% of the total variation. The principal component analysis was able to discriminate

and separate the accessions in terms of three dimensions, represented by the "sexual

reproduction", "vegetative", and "qualitative" axes.

The cluster analysis based on the first nine principal components differentiated four

distinct groups of accessions. Group 1 was composed by accessions with small leaves,

flowers, pods, pegs, and seeds. Group 2 was formed by accessions with intermediate size

features, Group 3 was formed by accessions with large leaves, flowers, pegs, pods, and

seeds, and finally, Group 4 was composed of accessions which did not produce any pegs,

pods, and seeds.














CHAPTER 5
AGRONOMIC EVALUATION OF Arachispintoi GERMPLASM

Introduction

Agronomic evaluation is a very important step in germplasm characterization

programs. Although molecular and morphological characteristics are relevant, plant

breeders and ultimately producers have their attention focused on the potential of the

plant to grow well in their environment and to produce forage, grain, or other economic

products. Thus, agronomic evaluation will always be a key component in breeding

programs.

When evaluating a species outside its original environment, it is important to assess

its adaptation to the new ecosystem. Emphasis must be given to how soils, climate, and

rainfall conditions will impact the growth of this "new species". Along with adaptation,

several agronomic characteristics can be measured. The importance of each variable will

be defined by the use of the plant and by the environment where it will be cultivated. In

the case of a forage crop such as Arachispintoi growing in a subtropical environment like

Florida, forage yield, forage nutritive value, seed production, winter survival, and

nematode resistance are just some of the characteristics that should be evaluated.

Arachispintoi is native of and well adapted to certain tropical environments.

According to Pizarro and Rincon (1994), A. pintoi was evaluated by the International

Tropical Pastures Evaluation Network (RIEPT) in Brazil, Uruguay, Bolivia, Colombia,

Peru, and Venezuela under savanna and humid tropical conditions. They concluded that it









presented a wide range of adaptation and grows best under humid tropical conditions

with total annual rainfall ranging from 2000 to 4000 mm.

Fisher and Cruz (1994), however suggested that although A. pintoi grows well

under high rainfall conditions it can tolerate periods of drought. They reported that A.

pintoi was able to maintain a large proportion of its aerial parts at the expense of root

tissue when exposed to 8 wk of water deficit. The same authors also revealed that A.

pintoi did not tolerate long periods of flooding. They concluded that 3 wk of flooding

severally restricted plant growth with severe leaf chlorosis and reduction of leaf area.

Pizarro and Rincon (1994) reported that plants growing in a subtropical

environment in Pelotas-Brazil were exposed to severe frosts (Temperature < 0C) that

reduced growth, but did not kill the plant stands. They recovered after the return of warm

and rainy conditions.

The literature has abundant A. pintoi forage yield data collected in the tropics. In

evaluations performed in Bolivia, Brazil, Ecuador, Colombia, and Peru, accession CIAT

17434 produced between 0 and 2.7 Mg ha-1 of DM during the rainy season and 0.04 to

2.8 Mg ha-1 of DM during the dry season with a growing period of 12 wk (Pizarro and

Rinc6n, 1994). In Costa Rica, Argel and Valerio (1993) reported forage yields of 7, 12

and 7 Mg ha-1 of DM for accessions CIAT 17434, 18744, and 18748, with 20 mo of

growth in Guapiles and San Isidro. In Puerto Rico, forage dry matter yields of 2.1 Mg

ha-1 of DM were harvested 16 wk after planting from the accession CIAT 17434 (Argel,

1994).

Several publications also report forage nutritive value and seed production data of

Arachispintoi in the tropics. Crude protein values ranging from 120 to 220 g kg-1 and in









vitro digestibility (IVOMD) ranging from 560 to 700 g kg-1 were reported from Argel &

Pizarro (1992); Rincon et al. (1992); Carulla et al. (1991); and Pizarro and Carvalho

(1996).

Average seed yields of A. pintoi varied from 1 to 2 Mg ha-1 when harvested at 15 to

18 mo after planting (Ferguson et al., 1992). However, in Colombia, when planted in

soils with high fertility, Ferguson (1994) reported yields of 7.3 Mg ha-l. In Australia,

Cook and Franklin (1988) reported seed yields of 1.4 Mg ha-l, 12 mo after sowing cv.

Amarillo. Cook and Lock (1993), also working with the cv. Amarillo in Australia, stated

that seed yields of 2.8 Mg ha-1 were obtained in a commercial seed crop.

As stated before, the literature has several examples of research work done in the

past where agronomic characterization of Arachispintoi was the primary goal. However,

most of this work wase done with a single germplasm accession, that latter was released

as a commercial cultivars in several different countries. These studies also have in

common the fact that most of them where executed in tropical regions. Therefore, there is

a lack of information about other accessions of A. pintoi stored in germplasm banks, there

are little or no existing data regarding the performance of the same germplasm in

subtropical conditions.

The goal of this research was to evaluate the agronomic adaptation, forage yield,

seed yield, forage nutritive value, and nematode resistance of several A. pintoi germplasm

accessions stored at the USDA-NPGS germplasm bank.

Material and Methods

The Germplasm

Germplasm of A. pintoi stored at the USDA-NPGS germplasm bank located in

Griffin, GA was transferred to the University of Florida on three different occasions. In




Full Text

PAGE 1

GERMPLASM CHARACTERIZATION OF Arachis pintoi Krap. and Greg. ( LEGUMINOSAE) By MARCELO AYRES CARVALHO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Marcelo Ayres Carvalho

PAGE 3

To my wife, Aline Varandas, who bravely decided to follow me in pursuing this dream.

PAGE 4

iv ACKNOWLEDGMENTS The author would like to acknowledge th e people of Brazil for financially supporting him during his Ph.D. program, thr ough a scholarship awarded by the National Council for Scientific and Technol ogical Development (CNPq). He also would like to thank the Brazi lian Agricultural Research Corporation (EMBRAPA) in the person of its Director-president, Dr. Clayton Campanhola, for the financial support and continuous effort to improve the training and quality of its employees. Marcelo thanks his parents, Magaly a nd Milton; sisters Pa tricia and Juliana; grandparents Orlando; Maria An alia, Policarpo, and Araci; and all his family for being always present when needed; for helping build his character; and finally, for guiding him through his life in the right directi on, and making him a better human being. He is deeply grateful to the Quesenbe rry family for their support and kindness while he lived in Gainesville. He is esp ecially thankful to Dr. Ken Quesenberry, his advisor, for the guidance and assistance requir ed for the conclusion of his Ph.D. program. He would like to acknowledge the member s of his committee: Dr. Sollenberger, Dr. Gallo-Meagher, Dr. Blount, and Dr. Lyre ne, for their numerous and valuable contributions to this project and, more impor tant, to the development of this student. He is grateful to Dr. Roy Pittman from the USDA-NPGS/PI Station, located in Griffin, Ga, who generously made the germplasm of A. pintoi available for this research.

PAGE 5

v He would like to thank the research assi stants Loan Ngo, Judy Dampier, Jeff Seib, and Richard Fethiere for their assistance in the laboratory, greenhouse and field activities. Thanks are also extended to fellow gradua te students Dr. Eastonce Gwata, Dr. Liana Jank, Dr. Eduardo Carlos, Jose Dubeux, Joao Vendramini, Gabriela Luciani, and Sindy Interrante for their friendship and suppor t during the difficult times of this journey. Marcelo would like to thank Dr. Pizarro, Dr Mario Soter, Dr. Jose Valls, Dr. Zoby, Dr. Lazarini, and many others for their encouragement and for being examples of a positive attitude during his professional life. Finally, he would like to thank all his frie nds in Brazil and here in Gainesville for their support and encouragement.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................xi ABSTRACT.....................................................................................................................xi ii CHAPTER 1 INTRODUCTION............................................................................................................1 Hypotheses....................................................................................................................4 Objectives..................................................................................................................... 4 Goals.......................................................................................................................... ...4 2 LITERATURE REVIEW.................................................................................................5 Plant Genetic Resources Characterization....................................................................5 The Genus Arachis .....................................................................................................12 The Forage Potential of the Genus Arachis .........................................................14 Arachis pintoi ......................................................................................................18 Botanical characteristics...............................................................................18 Agronomic characteristics............................................................................19 Arachis pintoi germplasm characterization..................................................22 3 MOLECULAR CHARACTERIZATION OF Arachis pintoi GERMPLASM..............27 Introduction.................................................................................................................27 Materials and Methods...............................................................................................30 Parent Plants........................................................................................................31 Tissue Culture Regenerated Plants......................................................................32 Protocol 1.....................................................................................................34 Protocol 2.....................................................................................................35 Result and Discussions...............................................................................................36 Parent Plants........................................................................................................36 Tissue Culture Regenerated Plants......................................................................44 Summary and Conclusions.........................................................................................53

PAGE 7

vii 4 MORPHOLOGICAL CHARACTERIZATION OF Arachis pintoi GERMPLASM.....56 Introduction.................................................................................................................56 Materials and Methods...............................................................................................57 Result and Discussions...............................................................................................60 Summary and Conclusions.........................................................................................82 5 AGRONOMIC EVALUATION OF Arachis pintoi GERMPLASM.............................84 Introduction.................................................................................................................84 Material and Methods.................................................................................................86 The Germplasm...................................................................................................86 Field Evaluation...................................................................................................87 Adaptation and forage dry matter yield........................................................87 Forage nutritive value...................................................................................89 Seed production............................................................................................89 Nematode Response Evaluation..........................................................................89 Results and Discussion...............................................................................................92 Adaptation and forage dry matter yield........................................................92 Forage nutritive value.................................................................................101 Seed production..........................................................................................103 Nematode response evaluation...................................................................106 Summary and Conclusions.......................................................................................113 6 CONCLUSIONS...........................................................................................................114 APPENDIX A LIST OF A. pintoi GERMPLASM EVALUATED AT UF.....................................119 B CTAB DNA EXTRACTION PROTOCOL.............................................................121 C MORPHOLOGICAL DESCRIPT ORS CORRELATION TABLE.........................123 D CLIMATOLOGICAL DATA AT THE FORAGE RESEARCH UNIT IN GAINESVILLE-FL, DURING THE PERIOD OF THE AGRONOMIC EVALUATION FIELD TRIAL...............................................................................127 LIST OF REFERENCES.................................................................................................129 BIOGRAPHICAL SKETCH...........................................................................................139

PAGE 8

viii LIST OF TABLES Table page 2-1 Proposed gene pools of the genus Arachis based on Arachis hypogaea breeding perspective................................................................................................................15 3-1 List of Operon Technologies primers used to amplify Arachis pintoi DNA regions......................................................................................................................33 3-2 Tissue culture protocols used to regenerate Arachis pintoi plants...........................35 3-3 List of Operon primers, number and size of amplified bands, and number of bands accession-specific of Arachis pintoi germplasm genomic DNA...................36 3-4 Characteristics of RAPD patterns of the 35 Arachis pintoi germplasm accessions.38 3-5 Gene frequency of the 100 RAPD locus of Arachis pintoi germplasm accessions.40 3-6 Neis gene diversity and by Shannon-W eaver diversity index of RAPD loci.........42 3-7 Analysis of variance table of ca llus rating and weight produced from Arachis pintoi leaf discs incubated on two different tissue culture media...............46 3-8 Callus rating and weight of Arachis pintoi tissue culture incubated in two different protocols....................................................................................................46 3-9 Callus rating and callus weight of Arachis pintoi callus induced on Protocol 1......48 3-10 Callus rating and callus weight of Arachis pintoi callus induced on Protocol 2......49 3-11 Number of regenerated plants of Arachis pintoi accessions cultured on two different protocols....................................................................................................52 4-1 Morphological descriptors applied to Arachis pintoi accessions.............................58 4-2 Morphological characteristics of Arachis pintoi germplasm accessions.................61 4-3 Correlations among mor phological descriptors of Arachis pintoi germplasm........67 4-4 Simpson and Shannon-Weav er diversity indices for Arachis pintoi morphological descriptors........................................................................................71

PAGE 9

ix 4-5 Vector loadings and perc entage of variation explained by the first five principal components for morphological characteristics of Arachis pintoi .............................73 4-6 Morphological characteristics of Arachis pintoi accession groups obtained by the cluster analysis.........................................................................................................79 5-1 Gall index, gall size, percent ga lled area and damage index values........................91 5-2 Winter survival evaluations of Arachis pintoi at the forage research unit in Gainesville-FL..........................................................................................................93 5-3 Plot coverage and plant height before forage dry ma tter yield evaluations of Arachis pintoi at the forage research un it in Gainesville-FL in 2003......................95 5-4 Analysis of variance table of forage dry matter yield (FDMY) evaluations of Arachis pintoi at the forage research un it in Gainesville-FL in 2003......................96 5-5 Forage dry matter yield of Arachis pintoi germplasm at the Forage Research Unit near Gainesville, FL in 2003............................................................................97 5-6 Analysis of variance table of the annu al forage dry matter yield (FDMY) of Arachis pintoi at the forage research un it in Gainesville-FL in 2003......................99 5-7 Total forage dry matter yield of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003..................................................................................100 5-8 Analysis of variance table of crude pr otein (CP) and in vi tro organic matter digestion (IVOMD) of ei ght-week regrowth of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003..........................................................101 5-9 Crude protein (CP) and in vitro orga nic matter digestibility (IVOMD) of eightweek regrowth of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003..............................................................................................................103 5-10 Analysis of variance table of seed production of Arachis pintoi at the Forage Research Unit near Gaines ville, FL in 2003 and 2004...........................................104 5-11 Seed production of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 and 2004..............................................................................................105 5-12 Analysis of variance table of Arachis pintoi germplasm reaction to Meloidogyne arenaria ............................................................................................107 5-13 Reaction of Arachis pintoi germplasm to Meloidogyne arenaria race 1...............108 5-14 Analysis of variance table of Arachis pintoi germplasm reaction to Meloidogyne javanica ............................................................................................109 5-15 Reaction of Arachis pintoi germplasm to Meloidogyne javanica ..........................110

PAGE 10

x 5-16 Analysis of variance table of Arachis pintoi germplasm reaction to Meloidogyne incognita ...........................................................................................111 5-17 Reaction of Arachis pintoi germplasm to Meloidogyne incognita .........................112 A-1 List of A. pintoi germplasm evaluated at UF.........................................................119 C-1 Morphological descript ors correlation table..........................................................124 D-1 Climatological data at the forage rese arch unit in Gainesville, FL, during the period of the agronomic evaluation field trial........................................................127

PAGE 11

xi LIST OF FIGURES Figure page 2-1 Area of natura l occurrence of Arachis pintoi ...........................................................18 2-2 Arachis pintoi plant characteristics..........................................................................20 3-1 Ratings scale applie d to callus pieces of Arachis pintoi explants incubated in two different protocols....................................................................................................33 3-2 Leaflet cutter and callus inducti on dishes of Protocol 1 and 2 of Arachis pintoi accessions.................................................................................................................34 3-3 RAPD band profile of 24 accessions of Arachis pintoi amplified by primer A4 ....39 3-4 Dendogram illustrating the ge netic relationships among 35 Arachis pintoi accessions based on Neis genetic distance obtained by 100 RAPD markers resolved by 8 random primers and generated by the UPGMA method...................45 3-5 Callus growth of Arachis pintoi PI 604812 on two different tissue culture media..47 3-6 Shoot regeneration of Arachis pintoi accession PI 604812 subculture in medium MS+1g L-1 BA..........................................................................................................50 3-7 Representation of the different steps in the process used to tissue culture Arachis pintoi. ........................................................................................................................50 4-1 Flower standard colors of Arachis pintoi germplasm..............................................63 4.2 Flower sizes and hypanthium colors displayed by Arachis pintoi germplasm........63 4.3 Stem characteristics of Arachis pintoi germplasm...................................................64 4-4 Leaflet characteristics of Arachis pintoi germplasm................................................65 4-5 Seed characteristics of Arachis pintoi germplasm...................................................66 4-6 Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC1 and PC2..........................................................................................74 4-7 Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC1 and PC3..........................................................................................75

PAGE 12

xii 4-8 Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC2 and PC3..........................................................................................76 4-9 Dendogram of 35 Arachis pintoi accessions based on morphological descriptors and the first nine principal components...................................................................78 4-10 Group 1 representative accession (PI 497541).........................................................80 4-11 Group 2 representative accession (PI 604814).........................................................80 4-12 Group 3 representative accession (PI 604798).........................................................81 4-13 Group 4 representative accession (PI 604817).........................................................81 5-1 Forage dry matter yield (FDMY) of Arachis pintoi and Arachis glabrata cultivars at the forage research unit in Gainesville-FL in 2003..............................................98

PAGE 13

xiii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GERMPLASM CHARACTERIZATION OF Arachis pintoi Krap. and Greg. ( Leguminosae ) By Marcelo Ayres Carvalho December 2004 Chair: Kenneth H. Quesenberry Major Department: Agronomy Arachis pintoi Krap. and Greg. is a herbaceous, perennial legume, exclusively native to Brazil. It is considered a multiple use legume, being grown for forage; ground cover in fruits orchards, forest, and low till age systems; erosion control; and ornamental purposes. Although several cultivar s have been released in di fferent countries, little is known about the genetic diversity of the ge rmplasm stored at genebanks. Our objective was to characterize and evaluate the geneti c diversity of the ge rmplasm of 35 accessions of Arachis pintoi at molecular, morphological, and agronomic levels. A. pintoi accessions were used to study the genetic diversity at the molecular level using RAPD markers. Concurrently, two tissue culture protocols were evaluated for their organogenesis ability, and capacity to generate somaclonal varia tion. From the original 18 primers tested, amplifications were obtained with eight which amplified 100 polymorphic bands. Average genetic distance was estimated as 0.36, indicating that a large amount of genetic diversity exits among the accessions. The accessions were grouped by their genetic

PAGE 14

xiv similarities into four dist inct groups. Callus induction was achieved on two different Murashige and Skoog basal protocols, and s hoot regeneration was achieved for several accessions on both media. Regenerated plants recovered in both protocols presented no differences in their RAPD band profile. Mor phological characteriza tion using data from stems, leaves, flowers, pegs, pods, and seeds demonstrated that the germplasm presented great morphological variability. Principal component analysis was able to discriminate the accessions in terms of three dimensions and Cluster analysis differentiated four distinct groups. Average dry matter yield in 2003 was 4.35 Mg ha-1, and ranged from zero to 9.10 Mg ha-1. Average crude protein was 180 g kg-1 of DM and ranged from 130 to 220 g kg-1. The average value of in vitro organic matter digestion was 670 g kg-1 of DM and ranged from 600 to 730 g kg-1. Some accessions produced hi gh seed yields reaching values above 1.00 Mg ha-1. Average seed production was 0.32 Mg ha-1 in 2003, and 0.43 Mg ha-1 in 2004. A. pintoi germplasm presented high levels of resistance to the nematode species Meloidogyne arenaria M. javanica and M. incognita

PAGE 15

1 CHAPTER 1 INTRODUCTION Biodiversity can be defined as the total varia tion found within all living organisms and their habitats. It can be accessed at thr ee different levels: communities (environment), species, and genes. When accessing biodiversity at the species level, we are interested in observing differences among individuals or popul ations of that particular species. This can be referred to as the genetic diversity of the species. Thus, we can consider genetic diversity as a form of biodiversity. Genetic diversity is associated with th e degree of differentiation among individuals in a population at their genetic material leve l. The genetic material corresponds to the DNA, genic or cytoplasmic, called the genot ype. Expression of gene s contained in the DNA is the result of interaction between th e environment and the genotype, which is called the phenotype. Genetic diversity is impor tant because it will enable evolution and adaptation of the species to an ever changi ng environment. Thus it is essential for the long-term survival of a species. Since very early in the history of the world, humans have exploited the genetic diversity of plants, primarily as sources of food, and then to improve their land races and cultivars. With the advance of modern agriculture, plant breeders around the world began to collect genetic diversity of the most impor tant food crops and to store these materials at national research institut es and local institutions. By the late 1960s, the Food and Agriculture Organization (FAO) of the Unite d Nations (UN) was concerned about the problem of conserving plant genetic diversity, and a series of conferences were arranged.

PAGE 16

2 As the result of these conferences, in 1974 the International Board for Plant Genetic Resources (IBPGR) was established. Coordina ted actions were then made to increase collection efforts and to increase the creation of germplasm banks at national and global levels. As a consequence of these efforts, about 1300 ex-situ genebanks were created. Today, they together conserve an estimated 6.1 million accessions (Hawkes et al. 2000). Although actions to increase c onservation of genetic divers ity were very successful, the use and characterization of these resour ces were not. In 1996, FAO held the Fourth International Technical Conference on Plan t Genetic resources in Leipzig/Germany. During this event, a global plan for the cons ervation and sustainable use of plant genetic resources for food and agriculture was prepar ed and formally adopted by more than 150 countries. Among other issues, a call was made for expanding characterization and evaluation, and for increasing the number of core collections to facilitate the use of the stored germplasm (FAO, 1996). Information obtained in studies of germplasm can be used for accession classification, analysis of genetic diversity, a nd studies of the geneti c divergence. Studies of genetic diversity are important because they are a tool for genetic improvement allowing the efficient use of the available germplasm of a species. The genus Arachis has great importance at the world level. Arachis hypogaea L. (peanut), the species with greatest importance in the genus, is cultivated commercially in more than 80 countries, supplying food with high protein levels and oil of excellent quality. According to FAO, world production of peanut is about 23 million metric tons annually.

PAGE 17

3 Because of the importance of the peanut as a food crop, great importance is given to the whole genus, and thousands of accessions of the common peanut and its wild relatives have been collected and stored in genebanks, especial ly those located at ICRISAT, USDA-NPGS and EMBRAPA. Although A. hypogaea is the most important species of the genus, some of its wild relatives also have agronomic potential. Some of these species are used as ground cover or forage crops. According to Kerridge (1994), the major re search priorities in the genetics of species with potential as forage crops in the genus Arachis are: More thorough studies of germplasm in the sections Caulorhizae and Rhizomatosae. Rigorous characterizatio n of genetic variability at th e molecular, physiological, and agronomic levels. Development of molecular markers for use in genotype identification and studies of breeding behavior. Quantification of genetic variation, i nheritance of important traits, and identification of sources of trai ts of agronomic interest. Survey of accessible sources of genetic resistance for disease and insect pest resistance, and survey of di seases in natural populations. Arachis pintoi is a perennial species of the sectio n Caulorhizae, and it is considered by many as the most promising of the wild species of the genus. It is a multiple use legume, which has been used predominantly as a forage crop, with released cultivars in several countries. The importance of the genus, the agronomic potential presented by Arachis pintoi and the lack of information about its germ plasm (stored at the USDA-NPGS) were the primary justification for this research.

PAGE 18

4 The hypothesis, objective, and goals of this work are presented below. Hypotheses Genetic diversity is limited in the Arachis pintoi germplasm stored at the USDANPGS germplasm bank. Objectives To characterize the germplasm of Arachis pintoi at molecular, morphological, and agronomic levels. Goals To quantify the genetic variability of accessions of Arachis pintoi at the molecular, morphological, and agronomic levels. To supply basic information to the breedi ng programs that are using the germplasm to develop new improved cultivars.

PAGE 19

5 CHAPTER 2 LITERATURE REVIEW Plant Genetic Resources Characterization Economic exploration of plants is essential for maintenance of a wide base of their genetic resources, because the same ones comp ose the vital patrimony of the species. The most important way to increase a species produ ctivity is to know th e variability in its existing germplasm. Incorporating new organi sms into the group of useful plants, by direct use or through programs of genetic im provement, requires use of available genetic resources. These resources are collected to be used in breeding programs and not just to be conserved (Paterniani, 1988). According to Smith and Linington (1997) th e cost to collect a single germplasm accession in its country of origin and incorpor ate that accession into a local genebank is about US $870. In addition, the Keystone Center (1991) estimated the co st to maintain an individual germplasm accession is US $50 per ye ar. The value of the material itself and the aggregate costs to acquire and maintain the tremendous amounts of genetic diversity stored in the genebanks around the wo rld are enough to justify germplasm characterization and evaluation. After the IBPGR was established in 1974, great attention was given to collecting germplasm of the most important food crops. Th is was particularly important because of the changes brought about by the green revol ution to producers everywhere. Land races were rapidly being replaced by improved cult ivars and the risk of losing these genetic resources was great. Furthermore, the population explosion and the expansion of

PAGE 20

6 agricultural lands to supply the needs of e xpanding urban populations also increased the risk of genetic erosion in some ecosystems. As a result of these efforts, FAO (1998) estimated that 6.2 million accessions of 80 different crop species were stored in 1,320 genebanks and related facil ities in 131 countries. These numbers show that today, germplasm conservation is well recognized issue, and is deemed important by most of the world community. However, plant genetic resource conservati on is much broader than just collection and storage. Characterization a nd evaluation should be an in tegral part of the general scheme of genetic resource conservation, but beca use of the great effort made to increase the number of accessions stored in the gene banks, these activities often have been relegated to a second level. Th is was not intentional, but re flects the size that germplasm collections have reached and the cost of co mplete evaluation. For instance, according to Holbrook et al. (1993), the num ber of common peanut genotypes stored in the USDANPGS genebank is about 7,500. However, the most recent data from the Germplasm Resources Information Network (GRIN) account for 9,232 accessions. Germplasm characterization consists of st udies of eco-geographic and demographic adaptation (Martins, 1984), and according to Solbrig (1980) involves mostly the parameters of the vital cycle of the organi sm, genetic and physio logical studies, plant pathology, and yield evaluation, among other studies. Characterization often also involves taxonomic confirmation and shoul d produce an easy and quick way to differentiate the germplasm, using highly he ritable and visible tr aits (Hawkes et al., 2000). Breeding programs should begin only after appropriate germplasm characterization (Cameron, 1983).

PAGE 21

7 Improved knowledge of the available germ plasm provides essential information for its more intense use. Characterization and ev aluation of available resources will allow the establishment of nuclear collections (core co llections) that, by definition, embrace the maximum genetic diversity contained in cul tivated species and in the related wild species, with the minimum possible numb er of accessions (Frankel and Brown, 1984). Another positive result of characterization is the detection of duplications, which is a serious problem especially when we have in mind the size of the collections and the resources needed to maintain them. Plucknett et al. (1987) estimated that as few as 35% of the accessions in world collectio ns are actually distinct. In summary, characterization is the best way to understand the variability contained in a germplasm collection and to increase use of the germplasm by plant breeders. It is also important in monitoring the genetic stab ility of the germplas m storage processes. Characterization of germplasm can be based on molecular, biochemical, morphological, and agronomic features. The use of biochemical and molecular char acterization can be a more precise way to discriminate among accessions in germplasm collections. This is especially true in species where the application of phenotypic mo rphological descriptors is delayed by very slow growth or delay in reaching the repr oductive stage, when many markers tend to be evidenced (Valls, 1988). During the past 25 years, many improvem ents in molecular biology techniques have allowed the direct application of these methods in studies of genetic characterization. According to Karp et al. (1997), the advent of the polymerase chain reaction (PCR) permitted the development of several molecular technologies that can be

PAGE 22

8 used with great success in detecting, characterizing and evaluating genetic diversity. Several different types of molecular markers we re generated after the advent of the PCR, and they differ among themselves in the way th ey resolve genetic differences; in the type of information that they generate; in the ta xonomic levels where they should be applied; and finally in costs, labo r requirement, and training. Although an extensive number of molecu lar markers are available, DNA-based markers have greatly overtaken markers based on proteins or enzymes, because the latter can be influenced by the environment. According to Ford-Lloyd (2001), DNA-based markers can be classified into three different categories: Non-PCR-based methods : Restriction fragment lengt h polymorphism analysis (RFLP) and variable number of tandem rep eats (VNTRs) are some examples of this category. Arbitrary (or semi-arbitrary) primed techniques : This is a PCR-based category that uses random primers during th e PCR reaction. The most well known and widely used of these methods is th e Random Amplified Polymorphic DNA (RAPD). Site-targeted PCR : In this class, primers that amplify specific regions of the DNA are used during the PCR reaction. Exampl es are single nucleotide polymorphism (SNPs), microsatellite repeat s (VNTRs), and many others. Karp et al. (1997) also indi cated that DNA-based markers can be useful in defining an accession identity; defining the degree of similarity among individuals of an accession or a group of accessions; and defining the pres ence of a particular allele or nucleotide sequence in an individual, population, or taxon. Among all DNA-based markers, RAPD have been extremely popular with plant genetic scientists. Reiter (2001) revealed that in a recent bibliographic search more than 2,000 publications were listed where RAPDs markers had been used.

PAGE 23

9 According to Reiter, the popularity of this mark er arises from the fact that the method is very simple and the cost require d for its application is low. RAPDs are the product of PCR reaction wher e a single, short, sequence-arbitrary primer (oligonucleotide 10mer usually) is used. Amplificat ion happens by setting a low annealing temperature (35-37 C typical) in the thermal-cycling program. Amplification products are separated by size in an electr ophoresis agarose gel. A DNA fragment pattern of low complexity is observed after ethidi um bromide staining. Fragment patterns of individuals of the same species are almost identical (with some exceptions where one band is present in one individual and absent in another). These fragment polymorphisms are heritable and can be classi fied as a new category of gene tic marker (Williams et al., 1990). Most RAPD bands are amplification polym orphisms due a nucleotide base change at one of the priming sites or an inserti on/deletion event within the amplified region. DNA amplification from heterozygous indivi duals at RAPD marker loci are normally identical to the homozygous parent. Thus RAPD markers are typically dominant markers. Dominant markers are less informa tive than co-dominant markers for genetic mapping in some types of segregating populations (Williams et al., 1990). When working with several accessions amplified using the same 10mer primer, similar DNA fragment patterns may be observe d. However, it is not possible to know if all fragments of the same size class have th e same DNA sequence (a nd thus are allelic). Bands from related germplasm lines are likely to be allelic. However, as genetic distance increases so does the probability of non-alle lism. Fragment allelism is an important

PAGE 24

10 concern when RAPDs are used for population ge netic studies, varietal fingerprinting, and some forms of marker-assiste d selection (Reiter, 2001). Another criticism regarding RAPD is inco mplete pattern reproducibility (Westman and Kresovich, 1997). Any sequence-arbitrary am plification method will be considerably less robust than conventional PCR for the following reasons: Multiple amplicons are present competing for available enzyme and substrate Low-stringent thermal-cycling permits mi smatch annealing between primer and template To overcome these problems, it is importa nt to create optimal and consistent amplification conditions, and to optimize the quantity and quality of reaction components (DNA, MgCl, Taq polymerase, etc). Characterization of genetic variability can use conventi onal methods of evaluation complemented by biochemical and molecula r methods (Hulli, 1987; Marcon, 1988). It is recognized that morphological data, despit e its importance, is not always easily understood at the level of genes (Simpson and Withers, 1986). The phenotypic expression of botanical and agronomic characteristics, wh ich in general have polygenic inheritance, is the result of genotype x environment interac tions (phenotypic plasticity). On the other hand, simple verification of the existen ce of molecular markers does not imply knowledge of their expression in the phenot ype. Thus, phenotypic ch aracterization of accessions should still be used in germ plasm evaluation. Thus morphological and molecular methods present different advantag es, and the complementation between them can produce benefits that would not be obt ained in separate analyses (Hulli, 1987). Morphological characterization is the simp lest way to gather information about germplasm; and before molecular marker techniques became popular and accessible, it

PAGE 25

11 was the primary way for plant genetic scient ists to improve their knowledge about the resources available to them. The term morphological characterization refe rs to a broad subject that involves taxonomy, botany, genetics and other discip lines. Sometimes, the first step in characterizing a germplasm accession involves taxonomic confirmation. After this step, highly heritable and visible botanic al features are used to prep are a list of descriptors that will be applied to the plants. Usually, qualita tive traits that show little environmental influence are preferred, but sometimes quant itative traits are al so used. Previous knowledge about the morphology and phenology of the species is required (Hawkes et al., 2000). The importance of morphological features was demonstrated by Mendels work. Mendel used simple morphological traits and studied their inheritance in hybrids produced from distinct homo zygous parents. The outcome of his work was the demonstration that traits are transferred from parents to offspring by factors later called genes. The discipline of genetics (and a new world) was opened by a work based on morphological traits. Nowadays, Mendels work and morphological traits are still used to understand the mechanism of inheritance and gene tics of traits in many different species. Morphological descriptors must be universal and simple to score. The universality of the descriptor will allow plant breeders and researchers worki ng with plant genetic resources to exchange information about germplasm accessions generated in different places. To achieve this goal the Internationa l Plant Genetic Resources Institute (IPGRI), in collaboration with crop specialists, publishe d guides to descriptor selection and also descriptor lists for more than 100 different species, including th e most important food

PAGE 26

12 crops and some wild rela tives of the same (Hawke s et al., 2000). The IPGRI ( www.ipgri.cgiar.org ) lists species with published descriptors, and these descriptors lists can ultimately be used to generate germ plasm catalogues and to build computational databases of different species that can be accessed by the World Wide Web, contributing to increased use of pl ant genetic resources. Another level of germplasm characterizati on is agronomic research. This type of research measures useful traits related to agronomic performance. These usually include quantitative traits related to yield, disease resistance, and environm ental stress tolerance are measured. Although it seems insignificant, some authors make a distinction between agronomic characterization and evaluation. Hawkes et al. (2000) suggested that evaluation involves traits important for crop improvement, which may often be affected by their interaction with the environment. Th ey also suggested that one big difference between characterization and ev aluation is that the first s hould be carried out by the germplasm curator and the latter by plant breed ers and plant patholog ists working outside the genebank. Therefore, it is important to exercise cau tion when using agronomic data generated in different places at different times. The use of standard checks will permit a more extensive comparison of the germplasm, but will never replace local evaluations, especially when a more detailed analysis is needed (Ford-Lloyd and Jackson, 1986). The Genus Arachis The genus Arachis has great importance at the world level. Arachis hypogaea L., the most important species in the genus, is cultivated commercially in more than 80 countries, supplying food with hi gh protein levels and oil of excellent quality. According to FAO, world production of peanut reaches about 23 million tons annually and assumes

PAGE 27

13 great agronomic importance in developing count ries. These countries are responsible for 80% of the world production with 67% being produced in the semi-arid tropics (Singh and Singh, 1992). The genus Arachis with diploid (2n=2x=20) and te traploid (2n=4x=40) species, belongs to the Fabaceae family, Papiliono ideae subfamily, Stylosanthinae subtribe, Aeschynomeneae tribe. It possesses herbaceous, annual and perennial species, leaves with 3 or 4 leaflets; papilionaceous coro lla; tubular hypanthium sessile ovaries and underground fruits (Gregory et al., 19 80; Krapovickas and Gregory, 1994). The genus has natural occurrence exclusiv ely in South America, extending from east of the Andes Mountains, south of Amaz onia, north of the Plancie Platina and northwest of Argentina, until the northeast of Brazil. It is believed that the genus originated in the Amambai M ountain, near the border of Ma to Grosso do Sul-Brazil and Paraguay, from where it was dispersed over an area of 4000 km in di ameter (Krapovickas and Gregory, 1994). Krapovickas and Gregory (1994) describe 69 species, with 39 exclusively from Brazil. Valls and Simpson (1994), however, asse rt that the genus has about 80 species. The genus can be divided into nine groups or sections; one of them with two series: Section Arachis Section Caulorrhizae Section Erectoides Section Extranervosae Section Heteranthae Section Procumbentes Section Rhizomatosae o Series Prorhizomatosae o Series Rhizomatosae Section Trierectoides Section Triseminatae

PAGE 28

14 The importance of A. hypogaea as a major food crop genera ted a need for a better understanding of the wild species of the genus because of the potential to use them in breeding programs of the common peanut. Va lls and Simpson (1994) reviewed the two different proposed gene pools of the genus (Table 2-1). Some degree of cross-ability is present among some species, and this could represent a chance to move genes from one species to others. Examples of introgression of genes from wild species to the common peanut are presented by Simps on and Starr (2001), Starr et al. (1990), and Mallikarjuna and Sastri (2002). Cultivar Spancross (Sta rr et al., 1990), generated by the crossing of A. hypogaea and A. monticola ; and cultivar COAN (Simpson and Starr, 2001), a runner market type derived from a backcross introgression pathway involving a complex interspecific amphiploid hybrid ( A. cardenasii x A. diogoi ) using cv. Florunner (Norden et al., 1969) as the recurre nt parent are real examples of the potential of the wild species in breeding programs of the common peanut. Parallel to the interest in improvement of cultivated peanut, the acknowledgment of the forage potential of some wild species of the genus brought a gene ral interest in these species. This interest generated an effort to recollect most of the available genetic diversity. According to Valls and Pizarro (1994) more than 30 expeditions were conducted between 1981 and 1993, which greatly impacted the number of accessions of perennial Arachis species stored in germplasm banks. The Forage Potential of the Genus Arachis Many species of the genus Arachis produce forage high in quantity and quality, comparable or superior to species of other tropical legumes used commercially as forage crops (Valls and Simpson, 1994). Species of th e genus Arachis, belonging to the sections Rhizomatosae, Arachis, Erectoides, Procumbe ntes, Caulorrhizae and Triseminatae have

PAGE 29

15 been evaluated for forage in Australia sin ce 1950. Many of those species have presented great potential as forage crops, mainly becau se of persistence under grazing (Cook et al., 1994). Table 2-1. Proposed gene pools of the genus Arachis based on Arachis hypogaea breeding perspective Gene Pools Wynne and Halward (1989) Singh et al. (1990) Primary Cultivated varieties of A. hypogaea Landraces of A. hypogaea Breeding lines derived from the above Arachis hypogaea and A. monticola Secondary Arachis monticola Other wild tetraploid forms in section Arachis (as yet uncollected) Diploids in section Arachis cross-compatible with A. hypogaea Tertiary Diploids in section Arachis cross-compatible with A. hypogaea Species in all sections Quaternary Diploid and tetraploid species of other sections of the genus The forage importance of the common peanut ( A. hypogaea ) has been long recognized, especially because of the great nutritional value of its stem and leaves (Cook and Crosthwaite, 1994). Although long recognize d, breeding programs neglected this fact and selections based on the forage potential were not emphasized, a point that seems to have changed in recent years during whic h evaluation of the forage production of A. hypogaea has been conducted. Gorbet et al. (199 4) reported dry matter yields (DMY) of 9 Mg ha-1 in a 140 days growth period in research done in Florida (USA). Crude protein values ranged from 140 to 200 g kg-1 and IVOMD ranged from 600 to 720 g kg-1. Also in the southeastern USA, Muir (2002) studied the forage production and nutritive value of eight warm-season legumes in north-central Texas under two dairy manure compost rates, and tw o harvest regimes during 2 years. He reported that Arachis

PAGE 30

16 hypogaea produced over 2.5 Mg ha-1 yr-1 of high quality forage and was one of the most productive. He concluded that although not highly productive under dryland conditions, groundnut can contribute with both forage and seeds for livestock and wildlife. Pizarro et al. (1996a) evalua ted 34 germplasm lines of A. hypogaea in a clay oxisol in Brazil, and reported DMY that varied from 0.4 to 2 Mg ha-1 in a 180 growth days period. In Nigeria, Larbi et al. (1999) investig ated variation in fora ge and seed yields, crude protein (CP), neutral detergent fiber (NDF), in situ degradation of dry matter (DM), and nitrogen (N) of 38 late-maturing groundnut cultivars. Average yield of forage was 4550 Mg ha-1 and seed production was 1.25 Mg ha-1. Crude protein, NDF, and DM degradation of leaf and stem varied among cultivars. Forage and seed yields were significantly correlated (r = 0.53), but seed yield was poorly correlated with forage quality indices examined. They suggested that plant breeders could select genotypes with higher seed and forage yield, and better forage quality. Also in Nigeria, a 3-yr study was carried out with 11 peanut va rieties to select superior lines for use in crop livestock syst ems. Crude protein (CP) concentration ranged from 148 to 216 g kg-1, with seven varieties recordi ng forage yields above 5 Mg ha-1. Mean seed yield (over 3 yr) varied significantly from 0.73 to 1.68 Mg ha-1 (Omokanye et al., 2001). Moreover, other species of the genus have al so shown great potenti al to be used as a forage crop, principally because they ha ve great persistence under grazing. Cook and Crosthwaite (1994) reported that stands of some species of the genus have been grazed for more than 30 yr. In Brazil where the genus originated, native populations have been grazed and utilized by producers for many years. Cameron et al. (1989) argued that wild

PAGE 31

17 species of the genus Arachis are excellent alternatives to improve tropical and subtropical pastures, particularly due to this strong persistence under grazing. Kretschmer and Wilson (1988) pointed to superior forage a nd seed production, and adaptation to wetlands as some of the important characteristics of an Arachis genotype from section Procumbentes (Pantanal peanut) evaluated by them and later described as A. kretschmeri by Krapovickas and Gregory (1994). Among the several species of the genus that can be utilized as forage crops and ultimately will impact the quality of the pastures, we can point to Arachis glabrata Benth., A. pintoi Krap. & Greg.; and A. repens Handro as the ones with the highest potential (Grof, 1985; Valls, 1992). Cultivars Florigraze (Prine et al., 1986) and Arbrook (Prine et al., 1990) of Arachis glabrata were released by the University of Florida, and they are used in an area of approximately 10,000 ha throughout the southeastern USA. These cultivars are primarily used to produce hay with high nutriti onal quality that is consumed by the dairy and beef cattle and by racing horses. Although A. glabrata has shown excellent forage characteristics, the fact that it produces few seeds, and cons equently that its propagation is exclusively vegetative, causes establishment difficulties. This has prevented its utilization over larger ar eas (French et al., 1994). According to some authors, however, the sp ecies with the most forage potential is A. pintoi a endemic species in Brazil. Arachis pintoi produces high dry herbage yield with excellent quality, as well as high s eed production (Valls and Simpson, 1994). The geographical distribution of the species spreads over an area that is part of the states of

PAGE 32

18 Gois, Bahia, and Minas Gerais, and extends to the Atlantic coast of Brazil, where the original accession of Arachis pintoi was collected (Figure 2-1). Figure 2-1. Area of natura l occurrence of Arachis pintoi (Valls, J.F.M. and E.A. Pizarro. 1994. Collection of wild Arachis germplasm. p. 19-27. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia). Arachis pintoi Botanical characteristics A. pintoi is a herbaceous, perennial legume, with a low stoloniferous growth habit. Growth heights range from 20 to 40 cm, with axonomorfus roots, without enlargements (Figure 2-2). The first branch is erect with attached low branches, rooting at the nodes, 30 25 20 15 10 05 00 05 35 40 45 50 55 60 65 70W. Gr 75 05 10 15 20 25 30 05 00 75 70W. Gr 65 60 55 50 45 40 30 35 RR AP AM PA AC RO MT TO MA CE RN PB PE AL SE PI BA GO DF MG ES RJ SP MS PR SC RS

PAGE 33

19 and stems are cylindrical, angular, and hollow (Krapovickas and Gregory, 1994). The leaves are alternate, compound, with four obovate leaflets (50 mm length x 32 mm width). Stipules have the ba sal portion attached to the petiole, and measure 10-15 mm length x 3 mm length with the free porti on measuring 10-12 mm length x 2.5 mm length in the base (Krapovickas and Gregory, 1994). A. pintoi shows indeterminate and continuous flowering. The inflorescence is axillary, in very short spikes, with four to five flowers, covered by the joined portion of the stipules. The flowers are sessile, prot ected by two bracts. The hypanthium is well developed, and can reach 10 cm in length, with silky hair. The calyx is bilabiate, with silky hairs. The corollas are yellow in the typical condition. Standard petals are 11 mm long x 13 mm wide, with yellow nerve and keel petals of 6-7 mm of length. Four oblong anthers and four spherical an thers are typically present. The species is considered normally self-pollinated (Krapovickas and Gregory, 1994). The flowers can be yellow, orange, cream and white (Valls, 1992). The small fruit of A. pintoi are located underground in an articulated legume form, with each articulation classified as an i ndehiscent capsule, which usually contains a single seed (Cook et al., 1990). The pericarp is fl at and resistant, covered with fine hairs that retain the soil. It pres ents two distinct segments, each one with a seed (Krapovickas and Gregory,1994). The number of chro mosomes is 2n=2x=20 (Fernandez and Krapovickas, 1994). Agronomic characteristics A. pintoi grows well in tropical areas from s ea level to heights of 1800 m, with 1500 to 3500 mm of annual rainfall (Rincn et al., 1992; Valls and Simpson, 1994).

PAGE 34

20 Figure 2-2. Arachis pintoi plant characteristics (Krapovickas, A. and W.C. Gregory. 1994. Taxonomy of the genus Arachis (Leguminosae). Bonplandia 8: 1-186) According to Rao and Kerridge (1994), A. pintoi CIAT 17434 (PI 338447) exhibits good adaptation to acid soils, with op timal growth occurring at pH 5.4. It is also tolerant to

PAGE 35

21 high soil Al concentrations ( 70%), and requires soil organi c matter higher than 3% for normal growth. A. pintoi (PI 338447) grows in a variety of soil textures ranging from sandy to clay, although optimum growth occurs in sandy soils (Argel and Pizarro, 1992). A. pintoi (PI 338447) is able to absorb P in so ils with very low concentrations of this element, or in situations when rela tively insoluble forms are applied. It has low response to N, Cu, Mo, Fe, S, and K (Rao a nd Kerridge, 1994). It is adapted to poorlydrained soils, but also grows well in welldrained soils having long periods without precipitation (Pizarro and Rincn, 1994). Under defoliation, A. pintoi (PI 338447) has a good initia l regrowth and high light interception capacity, with forage DMY, estimated between 30 and 40 d, of 0.08 Mg ha-1 day-1 (Fisher and Cruz, 1994). A. pintoi (PI 338447) grows better under shade than in full sun conditions. Plants under full sun pres ented lower leaf area and less above-ground biomass than counterparts maintained under 30, 50, and 70% shaded conditions. The below-ground biomass was not different in th e same circumstances. This characteristic provides A. pintoi with the ability to be used as a ground cover in coffee ( Coffea Arabica L.) and fruit groves (CIAT, 1991). Arachis pintoi (PI 338447) nodulated with native Bradyrhizobium strains in Colombia, although the nodules were not active. Inoculation with se lected strains was more efficient and demonstrated superior pl ant growth. Small doses of N fertilizer (50 kg ha-1) increased the initial in fection and sped the nodula tion process. Rates of N fixation of A. pintoi ranged from 0.07 to 0.20 Mg ha-1 yr-1 (Thomas,1994). Pizarro and Rincn (1994), reported that A. pintoi CIAT 17434 (PI 338447), 18748 (PI 604858), 18749 (PI 604859) and 18750 (PI 6 04815), growing in a subtropical

PAGE 36

22 environment in Pelotas-RS, Brazil, were resi stant to severe frosti ng. The authors stated that, although the plants presented freezing burning symptoms that affected their development, regrowth of herbage wa s achieved after the first rainfall. Arachis pintoi germplasm characterization In 1991, Arachis pintoi had a very narrow genetic base, approximately 30 accessions. Starting from that date, intense co llection of materials of the species was initiated by various projects. The germplas m base was enlarged to more than 150 accessions (Valls et al., 1994). The large number of accessions today availa ble implies in a need for discrimination among them, because they can present differe nt agronomic performance. Knowledge of the genetic variability of A. pintoi will be obtained through an appropriate characterization of the accessions. Characteri zation and evaluation of the wide range of A. pintoi germplasm should be carried out accordi ng to the priorities and strategies for handling the genetic resources of Arachis (Valls, 1988). A. pintoi has been spread world wide by th e international germplasm exchange network that consists of in ternational centers under the Internati onal Plant Genetic Resources Institute (IPGRI) and the national germplasm centers. Despite this world wide distribution of A. pintoi much of the evaluation resear ch was conducted using only the original accession (PI 338447). As a result of these works, A. pintoi was released as a commercial cultivar in 9 countries, incl uding Australia, USA, Costa Rica, Honduras, Colombia and recently in Brazil. Most of these released cultiv ars however represent accession PI 338447. Recently new cultivars have been released which originate from accessions other than PI 338447.

PAGE 37

23 According to Pizarro and Carvalho (1996), CIAT distributed about 1.2 Mg of seeds of A. pintoi accession (PI 338447) to Europe, Africa, Asia, Southeast Asia and North, Central and South America. A total of 61 countries were supplied with seeds. Even when more than one accession has been evaluated at a location, much of this characterization of germplasm has been w ith small numbers of accessions and based primarily on agronomic evaluation in different locations, with great emphasis on herbage production. Most of these works were conducted in South America, Central America, and Australia. In South America, under the coordination of the International Center for Tropical Agriculture (CIAT), the International Tropi cal Pastures Evaluation Network (RIEPT) evaluated A. pintoi accession CIAT 17434 (PI 338447) in several places of Colombia, inside the savanna ecosystem. In that netw ork, this accession presented poor adaptation when compared with other legumes. The DMY accumulated in 12 wk of regrowth, using the methodology of RIEPT (Toledo, 1982), varied from 0 to 0.47 Mg ha-1 during the rainy season, and from 0 to 0.25 Mg ha-1 in the dry season (Pizarro and Rincn, 1994). In Brazil, Valentim (1994) evaluated th e adaptation and forage production of accessions BRA-013251 (PI 338447) and BRA -015121 (PI 604858) in Rio Branco-AC (North). The results showed excellent adap tation of the accessions to the environmental conditions in Acre state. The average DMY of the accessions during 16 wk of growth in the rainy period of 1992 was 4.6 Mg ha-1 and 5.3 Mg ha-1 during the period of minimum precipitation. In 1993, the aver age production was of 6.1 Mg ha-1 in the rainy season and 4.2 Mg ha-1 in the period of minimum rain fall. In mixed pastures with Paspalum spp. in humid areas of low fertility in Campo Grande-MS, A. pintoi accession BRA-015598 (PI

PAGE 38

24 604815) yielded 0.6 to 1.3 Mg ha-1 in the dry season and 0.2 to 0.5 Mg ha-1 during the rainy season (Fernande s et al., 1992). Pizarro et al. (1996b) evaluating forage legumes mixed with Brachiaria decumbens Stapf. in Uberlndia-MG, obtained, with the accessions BRA-013251 (PI 338447), 015121 (PI 604858), 015598 (PI 604815) and 031143, dry matter production of 2.2, 0.6, 2.3 and 4.4 Mg ha-1, respectively, in the period of maximum precipitation during 1994/1995. Data obtained in Planaltina-D F by EMBRAPA-CPAC, starting from 9 accessions in a lowland area, showed great variability of forage production of A. pintoi in those conditions. The total DMY varied from 5 to 13 Mg ha-1 in the first year of evaluation and from 3 to 11 Mg ha-1 in the second year (Pizarro et al., 1992; Pizarro et al., 1993). In Veracruz, Mexico, A. pintoi CIAT 17434 (PI 338447), showed average DMY of 2.0, 1.2 and 0.8 Mg ha-1 in three cuttings of 12-wk regr owth during the rainy season, in two years of evaluation (Valles et al., 1992) In Gupiles, Costa Rica, the accessions CIAT 17434 (PI 338447), 18744 (PI 476132), 18747 (PI 497574), 18748 (PI 604858) and18751 of A. pintoi presented DMY of 4.1, 4.9, 4.0, 3.8 and 3.7 Mg ha-1, respectively, in 2-yr of evaluation (Argel, 1994). In th e USA, Kretschmer and co-workers have evaluated a number of seed-propaga ted accessions of wild species of Arachis for use on the poorly drained soils of S outh Florida and identified an A. pintoi accession (IRFL 4222) that was persistent under grazing (French et al., 1994). In Australia, only three accessions were initially available for evaluation, CPI 58113 (PI 338447), CPI 28273 and CPI 93472, with the last two being considered mostly the same. CPI 58113 was evaluated in several locations and presented great adaptation

PAGE 39

25 and persistence under grazing. Subsequently it was released as cultivar Amarillo in 1987, and since then it has gained increasing acceptance with producers as a forage crop (Cook et al., 1994). These author s reported that in 1991, 9 Mg of seed were produced and commercialized. In seven locations in Ce ntral and West Africa, A. pintoi CIAT 17434 (PI 338447) showed production that varied from 0.6 to 3.2 Mg ha-1 in the rainy season and 0.1 to 2.0 Mg ha-1 during the dry season, confirming its wide adaptation (Stur & Ndikumana, 1994). The same authors reported yields ranging from 0.5 to 4.5 Mg ha-1 yr-1 for Amarillo, in evaluations conducted at three loca tions in the dry areas of Fiji. Although the majority of the characteriza tion work conducted with the germplasm of A. pintoi has been agronomically based, some ex amples of germplasm characterization at other levels can be found in the lit erature. Monato (1997), working with approximately 45 accessions of A. pintoi applied a series of mo rphological descriptors to describe the variability of this germplasm. The accessions showed great variability in morphological traits. Oliveira et al. (1999) demonstrated the morphological variability and inheritance of flower colo r. The yellow flower is domin ant over the orange flowers. Mass et al. (1993) used 60 morphological descriptor s to characterize and demonstrate the variability of eight accessi ons. The accessions were organized into two groups representing the plant types, one hom ogeneous, and the other divided into four distinct subgroups. These results pointed to a lack of continuous patt erns of variation in morphology and highlighted the need for further germplasm collection. Paganella and Valls (2002) applied a list of 12 descri ptors to evaluate seven cultivars and 13 accessions of A. pintoi The objective of this wo rk was to review the

PAGE 40

26 origin of the cultivars and ch eck inconsistencies in the l iterature about the germplasm accessions that gave rise to the commercial cu ltivars. The results confirm that five of seven cultivars derived from the original or first accession of the species collected (PI 338447). Perennial Arachis germplasm has also been evaluated at the molecular DNA level. Genetic variation within and among accessions of the genus Arachis representing sections Extranervosae, Caulorrhizae, Hetera nthae, and Triseminatae was evaluated using RFLP and RAPD markers. RAPD markers revealed a higher level of genetic diversity than RFLP markers, both within and am ong the species (Galgaro et al., 1998). Gimenes et al. (2000) worki ng with sixty-four accessions of section Caulorrhizae utilized RAPD analysis to characterize the genetic variation and the phylogenetic relationships. A total of 104 fragments of DN A of different sizes were generated, and 97 were polymorphic in the accessions tested. De spite the large number of polymorphic fragments detected, the mean number of unique genotype s detected by each RAPD primer was low. However, when data from all primers were considered together, all accessions were uniquely fingerprinted. Genetic characterizations of accessions of A. pintoi using isoenzymes, RFLP and RAPD molecular markers were also done by Valente et al. (1998), Bertozo and Valls (1996), Bertozo & Valls (1997a), Bertozo & Valls (1997b), and Carv alho et al. (1998) with a small set of germplasm accessions, diffe rent from those used in this research.

PAGE 41

27 CHAPTER 3 MOLECULAR CHARACTERIZATION OF Arachis pintoi GERMPLASM Introduction Until the late 1960s most genetics studies were associated with genes that controlled morphological features, which were in general easily identif ied. These features appropriately named morphol ogical markers and were very important to the understanding of gene action a nd expression, and for the constr uction of the first genetic linkage maps. However, the low number of morphological markers linked to important agronomic traits made their use ineffici ent for the purpose of genetic improvement. Further, morphological markers were genera lly available for few species, which were used as model systems for genetic studies. For the great majority of crop species and their wild relatives few or no morphological markers were available (Ferreira and Grattapaglia, 1998). By definition, a molecular marker is every single molecular phenotype expressed by a particular gene. The nucleotide sequence and function of the marker could be known or unknown. A molecular marker can only be considered a genetic marker after its mendelian segregation is observed in segrega ting populations (Ferreira and Grattapaglia, 1998). Currently, several molecular biology tec hniques are available for the study of genetic variability at the DNA level. These techniques provide for the possibility of identification of innumerable molecular marker s that theoretically could cover the whole genome of a species. These markers can be used for many different applications

PAGE 42

28 including fingerprinting, genetic mapping, phyl ogenetic relationships, genetic diversity studies, and plant breeding. Among all DNA-based markers, Random Amplified Polymorphic DNA (RAPD) has been the most popular with plant genetici sts. Reiter (2001) revealed that in a recent bibliographic search there were more th an 2,000 publications were in which RAPD markers had been used. According to the same author, the popularity of this marker arises from the fact that the method is very simple and the cost required for its application is low. RAPD markers have also been the marker of choice in several genetic studies with the common peanut ( Arachis hypogaea L .) and its wild relatives. Hilu and Stalker (1995) used RAPD ma rkers to access genetic relationships between the common peanut and wild species of the section Arachis of the genus. These authors reported 132 polymor phic bands that were useful for separating species and accessions, and for evaluation of the genetic diversity presented by the germplasm. Chang et al. (1999) accessed the genetic diversity of A. hypogaea cultivars released in Taiwan using RAPD markers. They were successful in estimating genetic distances among varieties and grouping them in accordan ce with their genetic similarities. The mean genetic distance among cultivars was 0.41 1, and they were classified into six groups. RAPD markers have also been used to characterize wild species of Arachis Nobile et al. (2004) studied the ge netic variation within and among species of the section Rhizomatosaea of Arachis using RAPD markers. These authors reported that by using 110 polymorphic RAPD bands they were able to describe the genetic diversity and draw a dendogram displaying the similarities among accessions and species of this section.

PAGE 43

29 Another very important technique in gene tic resources research is tissue culture. The use of in vitro techniques is particularly critical to species with vegetative propagation and recalcitrant seeds. Other potential applications of tissue culture techniques are: micropropagation, long-te rm storage, germplasm exchange, embryo culture of interspecifi c hybrids, induction of mutations, production of synthetic seeds, and genetic transformation. Another ad vantage of plants originated in vitro is that they are generally free of pathogens and diseases. Although in vitro techniques are useful for genetic resources conservation, attention must be given to the genetic st ability of the systems used in this process. Tissue cultureinduced genetic variation or somaclonal variatio n is defined as the variation that arises during the period of dedifferentiated cell pr oliferation that take s place between the explant culture and recovering of regenerated plants. Such genetic variation has been observed among regenerants of several species and they usually present mutations that include: chromosome breakage, changes in pl oidy number, single base changes, changes in copy numbers of repeated sequences, increased trans poson mutagenegis, sister chromatid exchange and alteration in DNA methylation (H awkes et al., 2000). Because in vitro techniques have a wide spectru m of applications to genetic resources conservation and plant breeding, the development of protocols optimized to a particular species should have high priority. Several factors may affect the efficiency of the process and should be considered when re search is conducted in this field. Genotypes, source and age of explant, hormone concen tration in the medium, and day length are some factors that affect the success of rege neration of plants from callus tissue culture (Flick et al., 1983).

PAGE 44

30 Tissue culture regeneration protocols for A. pintoi were proposed by Rey et al. (2000) based on a single genotype, and Ngo and Quesenberry (2000) based on a small number of accessions. In terms of genetic reso urces it is very critical to evaluate the efficiency of protocols in terms of plant regeneration among different genotypes and also the preservation of the genetic characteristic s of the germplasm regenerated or stored with these protocols. Additionally, in terms of plant breeding, it is important to assess the genetic diversity of the germplas m with respect to their plant regeneration abilities. It is also important to estimate the capacity of the protocols to generate somaclonal variations, which ultimately could generate mutations and gene diversity. The objectives of this research were: To characterize the Arachis pintoi germplasm accessions stored at the USDANPGS germplasm banks using RAPD markers. To evaluate the organogenic regeneration ability of these A. pintoi accessions with two tissue culture protocols. To study the variation in RAPD band profile of plants regenerated compared to the parent plants. Materials and Methods Accessions of Arachis pintoi stored in the Southern Regional PI Station of the National Plant Germplasm System (NPGS) locat ed at Griffin-GA were transferred to the University of Florida in 2001 and 2002 (Appendix A) A subset composed of 35 accessions was used to study the genetic diversit y of this collection at the molecular level using RAPD markers. In addition to the RAPD characterizati on of the parent plants, 25 out of 35 accessions were selected randomly and evalua ted for their organogenesis ability using two tissue culture protocols. Extent of ge nerated somaclonal variation was accessed by

PAGE 45

31 comparing the RAPD band profile of the regene rated tissue culture plants to the parent plants. Parent Plants DNA was extracted from leaves of si ngle plants stored at the Agronomy Department greenhouse using a modified CTAB protocol first proposed by Rogers and Bendich (1985). In this protocol, 0.1 g of ground tissue was mixed with 400 l of 2x CTAB extraction buffer in a 2.5 ml eppendorf t ube and incubated in a 65C water bath for 60 min. After that 400 l of a chlorofo rm:isoamyl alcohol (24:1) was added and the mixture was centrifuged for 5 min. The supernat ant was then transferred to a new tube where 1/10 volume of 10% CTAB was combin ed, and 400 l of a chloroform:isoamyl alcohol (24:1) was added again. The so lution was centrifuged once more, and the supernatant was transferred to a new tube wh ere an equal volume of CTAB precipitation buffer was added. The mixture was then centrifuged, the supernatant was removed and the DNA pellet was rehydrated in 100 l of high salt TE buffer resti ng in a water bath. Ten minutes latter the DNA was reprecipita ted with 0.6 volumes of isopropanol and centrifuged for 15 min at 10,000 rpm. Finally th e pellet was washed with 80% ethanol, dried, and resuspended in 50 l of DDW (Appendix B). DNA concentrations were analyzed by m easuring absorbance at 260 nm, and DNA quality was determined by spectrophotomet er readings of the ratio of 260/280 nm. Working solutions were prepared by dilu ting the DNA stock solutions with DDW and standardized to c oncentrations of 25 n g of DNA per l. Eighteen primers of ten nucleotides le ngth from the Operon Technologies kit (Table 3-1) were used to amplify regi ons of genomic DNA under thermocycling conditions proposed by Gimenes et al. (2000). Thermocycling conditions were: 40 Cycles

PAGE 46

32 of 92C/1 min, 35C/1 min, and 72C/2 min. Tw enty-five l of PCR reaction mixtures were prepared by adding 17.3 l ddH2O, 2.5 l PCR Buffer (500 mM KCl, 100 mM Tris-HCl pH 8.3, 15 mM mg MgCl2, and 0.1% gelatin), 1.0 l MgCl, 1.0 l of 2.5mM dNTP solution, 1.0 l Primer, 0.2 l Taq polymerase (5 units/l), and 2.0 l DNA solution. Amplification products were separated by electrophoresis on 1.5% agarose gels, and the 1kb DNA ladder (Promega Coorporati on) was used as a standard. Banding patterns were visualized by staining the gels in ethidium bromide solutions and viewing under UV radiation. Images were captured wi th Quantity One quantification software version 4.1.1 from BIO-RAD Laboratorie s and then bands were scored. For each combination of accession and primer, five PCR reactions were prepared. Only bands present in at leas t three out of five gels were considered. RAPD bands were scored as presence (1) or absence (0) of homolog bands for all accessions and a phenotypic binary matrix was produced. This matrix was used to perform genetic analysis using the software POPGENE version 1.32 ( http://www.ualberta.c a/~fyeh/index.htm ). Allele frequency, numb er of polymorphic loci, Neis genetic distance (Nei, 1972), Neis ge netic diversity inde x (Nei, 1973), and Shannon-Weavers genetic diversity inde x (Shannon and Weaver, 1949) were the parameters calculated. Genetic distance (D= -ln I) was later used as a criterion for differentiation among accessions to prepare a cluste r analysis. Tissue Culture Regenerated Plants Two protocols were used to access their organogenesis ability, and capacity to generate somaclonal variation. Protocol 1 wa s proposed by Rey et al (2000) and Protocol 2 proposed by Ngo and Quesenberry (2000).

PAGE 47

33 Table 3-1. List of Operon Technologies* primers used to amplify Arachis pintoi DNA regions Primer number Nucleotide sequence A4 5-AATCGGGCTG-3 A10 5-GTGATCGCAG-3 A12 5-TCGGCGATAG-3 A15 5-TTCCGAACCC-3 B4 5-GGACTGGAGT-3 B5 5-TGCGCCCTTC-3 B10 5-CTGCTGGGAC-3 B16 5-TTTGCCCGGA-3 C2 5-GTGAGGCGTC-3 C4 5-CCGCATCTAC-3 D4 5-AATCGGGCTG-3 D13 5-GGGGTGACGA-3 E4 5-GTGACATGCC-3 E5 5-TCAGGGAGGT-3 E8 5-AATCGGGCTG-3 E15 5-ACGCACAACC-3 G5 5-CTGAGACGGA-3 G15 5-ACTGGGACTC-3 Operon Technologies, Alameda, CA, USA To compare these two protocols, callu s rating and weight and number of regenerated plants were used. The expe riment was conducted using a completely randomized design with 50 treatments and 3 replications. Quantity of callus produced was rated based on a 1 to 5 scale, where 1 = no callus growth, and 5 = largest amount of callus (Figure 3-1). In addition, callus weight was measured with a digital scale placed in the laminar flow hood to prevent contamination. Figure 3-1. Rating scale app lied to callus pieces of Arachis pintoi explants incubated in two different protocols. 5 4 3 21

PAGE 48

34 Leaflet pieces from adult plants were the ti ssue source for callus initiation. Leaflets were surface sterilized by immersion in 70% ethanol for 1 to 2 min, followed by immersion in a solution of commercial bleach (0.9% sodium hypochlorite, final concentration) plus one drop of Tween for 1 to 2 min. Leaflets were then washed three times with autoclaved distilled water. Ci rcles of approximately 19.4 mm area of the median portion of the laminae were cut and placed with the abax ial side down on the media in 60 mm x 15 mm petr i dishes (Figure 3-2). Figure 3-2. Leaflet cutter and callus induc tion dishes of Protocol 1 and 2 of Arachis pintoi accessions. Protocol 1 The callus induction medium consisted of major and minor salts, as well as vitamins according to Murashige and Skoog ( 1962), with 3% sucrose, 0.7% agar, and 10 mg L-1 NAA (1-naphthaleneacetic acid) + 1 mg L-1 BA (6-benzylaminopurine). The pH of the medium was adjusted to 5.7 with 0.1 N KOH or HCl prior the addition of agar. To induce shoots, small pieces of callus (30 mg fresh wt.) were transferred to fresh medium composed of MS basal salts and vitamins+1g L-1 BA. For root induction, the regenerated

PAGE 49

35 shoots were transferred to a medium compos ed of MS basal salts and vitamins plus 0.01mg L-1 NAA (Table 3-2). Plantlets resulting from rooted shoots were rinsed gently under running tap water to remove adhering cultured medi a and immediately planted in pots containing potting mix. These plants were acc limatized in a humidity box and th en placed in the greenhouse. Protocol 2 The MS basal salts and vitamins medium (Murashige and Skoog, 1962) was the base for all media used in this protocol. For callus induction the MQC medium, which is a modification of the MS medium proposed by Wofford et al. (1992) for tropical legumes, was the medium of choice. In this medium a 2:1 auxin/cytokinin ratio was employed, with a final c oncentration of 2 mg L-1 of IAA (indole-3-acet ic acid) and 1 mg L-1 of kinetin. The pH was adjust ed to 5.8 prior to autoclaving. For shoot development, medium IBA which was composed of MS basal salts and vitamins plus 0.1 mg L-1 of IBA (3-Indol butyric acid) wa s employed. For root induction, the regenerated shoots were transferred to medium MS basal salts and vitamins with growth regulators. Plantlets resulting from rooted shoots were rinsed gently under running tap water to remove adhering cultured medi a and immediately planted in pots containing potting mix. These plants were acclimatized and then placed in the greenhouse. Table 3-2. Tissue culture prot ocols used to regenerate Arachis pintoi plants Protocol Callus induction Media Shoot development Media Root induction Media 1 MS MS+1g L-1 BA MS+0.01mg L-1 NAA 2 MQC MS+0.1mg L-1 IBA MSNH MS + 3% sucrose, 0.7% agar, and 10 mg L-1 NAA + 1 mg L-1 BA MS + 3% sucrose, 0.7% agar, and 2 mg L-1 of IAA and 1 mg L-1 of kinetin MS basal salts and vitamins and no hormones

PAGE 50

36 DNA was extracted from regenerated plan ts in accordance with the protocol described earlier. PCR reactions and band sepa ration and scoring were also identical to that used for the parent plants. RAPD band pr ofile of the regenerate d tissue culture plants was compared to the parent plants. Result and Discussions Parent Plants Of the 18 primers evaluated, eight were able to generate reproducible and reliable bands. Primers A4, B4, B5, C2, D4, D13, E4, and G5 amplified 100 different bands. The number and size of amplified bands for each primer is shown in Table 3-3. Table 3-3. List of Operon primers, number and size of amplified bands, and number of bands accession-specific of Arachis pintoi germplasm genomic DNA Primer number Number of amplified bands Size of amplified bands (bp) Number of bands accession-specific A4 15 250, 550, 670, 750, 870, 1000, 1100, 1350, 1500, 1700, 1900, 2100, 2500, 3000, 3100 0 B4 12 700, 850, 900, 1000, 1200, 1400, 1600, 1750, 2000, 2300, 2700, 3000 1 B5 11 500,750, 1000, 1100, 1250, 1400, 1700, 2000, 2500, 3000, 3500 3 C2 15 465, 520, 600, 700, 750, 850, 1000, 1300, 1500, 1750, 2000, 2500, 2750, 3000, 3500 2 D4 08 580, 650, 750, 910, 1200, 1350, 1500, 2000 1 D13 16 370, 500, 650, 750, 850, 1000, 1200, 1400, 1500, 1700, 2000, 2100, 2300, 2500, 2800, 3310 2 E4 10 600, 900, 1000, 1100, 1250, 1500, 1800, 2100, 2500, 2800 1 G5 13 350, 450, 500, 650, 750, 850, 1000, 1100, 1250, 1350, 1750, 2000, 2500, 2750 1 The average number of amplified bands per primer was 12.5, with primer D4 amplifying eight fragments and primer D13 amplifying 16 fragments. The size of these 100 fragments ranged from 250 bp to 3500 bp. From the 100 bands amplified, 98

PAGE 51

37 presented polymorphism with the only two exceptions being Primer C2-1000 bp and primer E4-1250 bp. The average presence of bands per accession was 32, ranging from 20 to 44 bands. The observed number of amplified bands was significantly variable for each primer and accession analyzed. Primer A4 amplified 15 different bands with average bands per accession of 5.4 and a range of 2 to 10 bands. Primer B4 amplified 12 fragments with average bands per accession of 4.3 and a range of 1 to 10. Primer B5 amplified 11 bands with an average band per accession of 3.5 and range of 1 to 7. Primer C2 amplified 15 bands and average bands per accession of 5.2 a nd a range of 1 to 10. Primer D4 amplified only 8 bands and the average bands per accessi on was 1.7 with a range of 1 to 4. Primer D13 displayed 16 bands with average bands per accession of 5.2 and a range of 2 to 11. Primer E4 showed 10 different fragments with an average of 4 bands per accession and a range of 2 to 8. Finally, primer G5 presente d 13 bands with averag e bands per accession of 3 and a range of 1 to 7 bands. Ten bands were unique to an individual germplasm accession. Accession PI 604856 was discriminated by band C2-465bp, accession PI 604858 by band E4-600bp, accession PI 604810 by band C2-520bp, accession PI 604799 band B5-1250bp, PI 604809 by band B5-1100bp, accession PI 604807 by band G5-1750bp, accession CIAT 22256 by D4-1200bp, accession CIAT 22159 by band D13-3310bp, CIAT 22152 by band B4-700bp, accession CIAT 22265 by band D13-2800bp, and finally accession CIAT 22260 by band B5-500bp. Table 3-4 summarize the re sults obtained with RAPD markers among the 35 germplasm accessions of A. pintoi analyzed in this research. Figure 3-3 displays stained eletrophoresis gels of primer A4 and 24 accessions.

PAGE 52

38 Table 3-4. Characteristics of RAPD patterns of the 35 Arachis pintoi germplasm accessions Total number of screened primers 18 Number of polymorphic primers 08 Total number of bands amplified 100 Size of the amplified bands 250 bp 3500 bp Minimum and maximum number of ba nds per primer 08 (D4) 16 (D13) Average bands per primer 12.5 Total number of polymorphic bands 98 Total number of monomorphic bands 02 Average number of bands per accession 32 Minimum and maximum number of bands per accession 20 44 Number of accession-specific bands 10 According to Nei (1973), when a large number of loci are examined to evaluate the genetic variability of a popul ation, the amount of variation is measured by the proportion of polymorphic loci and average heterozygosit y per locus. Also according to Nei a locus is called polymorphic when the freq uency of the most common allele (xi) is equal to or less than 0.95, in cases where the sample size is smaller than 50. In Table 3-5, information about the gene frequency of each allele at every RAPD locus is presented. Great variability in ge ne frequency was observed for each different RAPD locus. Using the criterion described above to characterize polymorphic loci we can observe that 11 loci presented the freque ncy of the most common allele higher than 0.95, and then were classified as monomorphic. Therefore, we can conclude that 89 out of 100 loci, or 89% of the RAPD loci, was the proportion of polymorphic loci. Although the proportion of polymorphic loci is a good estimation of genetic variability, a more precise and appropriate measure of gene diversity is obtained by the use of the gene diversity statistics (Nei, 1987).

PAGE 53

39 Figure 3-3. RAPD band prof ile of 24 accessions of Arachis pintoi amplified by primer A4 (Operon Technologies). A. lane 1-PI604856, 2-PI604857, 3-PI604858, 4-PI604798, 5-PI604803, 6-PI604805, 7-PI604812, 8-PI604810, 9-PI604811,10-PI604799, 11-PI604800, 12-PI604809. B. lane 1-PI604817, 2-PI604815, 3-PI604814, 4-PI497541, 5-PI604813, 6-PI604801, 7-PI604804, 8-PI604808, 9-PI604807, 10-PI476132, 11-PI497574, 12-PI604859 1 2 3 45678910 11 12 B A1 2 3 45678910 11 12 1kb ladder 1500b 3000b 1kb ladder 4000b 250b 500b 750b 1000b 2000b 2500b 10000 1kb ladder 1kb ladder 250b 500b 750b 1000b 1500b 2000b 2500b 3000b 4000b 10000 1kb ladder 1kb ladder

PAGE 54

40 Table 3-5. Gene frequency of the 100 RAPD locus of A. pintoi germplasm accessions Allele Loci/Gene Frequency A4-250 A4-550 A4-670 A4-750 A4-870 A4-1000 A4-1100 0 0.4722 0.8611 0.6667 0.7222 0.5278 0.8056 0.5833 1 0.5278 0.1389 0.3333 0.2778 0.4722 0.1944 0.4167 A4-1350 A4-1500 A4-1700 A4-1900 A4-2100 A4-2500 A4-3000 0 0.3333 0.3889 0.6111 0.6667 0.8611 0.6667 0.7222 1 0.6667 0.6111 0.3889 0.3333 0.1389 0.3333 0.2778 A4-3100 B4-700 B4-850 B4-900 B4-1000 B4-1200 B4-1400 0 0.6667 0.0278 0.6944 0.8889 0.7500 0.5000 0.6944 1 0.3333 0.9722 0.3056 0.1111 0.2500 0.5000 0.3056 B4-1600 B4-1750 B4-2000 B4-2300 B4-2700 B4-3000 B5-500 0 0.8333 0.8889 0.3889 0.7500 0.5833 0.7500 0.9722 1 0.1667 0.1111 0.6111 0.2500 0.4167 0.2500 0.0278 B5-750 B5-1000 B5-1100 B5-1250 B5-1400 B5-1700 B5-2000 0 0.9444 0.5556 0.9722 0.0278 0.9444 0.8056 0.6667 1 0.0556 0.4444 0.0278 0.9722 0.0556 0.1944 0.3333 B5-2500 B5-3000 B5-3500 C2-465 C2-520 C2-600 C2-700 0 0.2500 0.4444 0.9444 0.9722 0.9722 0.8333 0.8056 1 0.7500 0.5556 0.0556 0.0278 0.0278 0.1667 0.1944 C2-750 C2-850 C2-1000 C2-1300 C2-1500 C2-1750 C2-2000 0 0.9444 0.4167 0.0000 0.4444 0.2778 0.6389 0.7500 1 0.0556 0.5833 1.0000 0.5556 0.7222 0.3611 0.2500 C2-2500 C2-2750 C2-3000 C2-3500 D4-580 D4-650 D4-750 0 0.6944 0.7222 0.5000 0.8611 0.8889 0.9444 0.8333 1 0.3056 0.2778 0.5000 0.1389 0.1111 0.0556 0.1667 D4-910 D4-1200 D4-1350 D4-1500 D4-2000 D13-370 D13-500 0 0.0833 0.9722 0.9444 0.9444 0.6944 0.7222 0.0833 1 0.9167 0.0278 0.0556 0.0556 0.3056 0.2778 0.9167 D13-650 D13-750 D13-850 D13-1000 D13-1200 D13-1400 D13-1500 0 0.2500 0.3611 0.7500 0.6667 0.2778 0.8056 0.8333 1 0.7500 0.6389 0.2500 0.3333 0.7222 0.1944 0.1667 D13-1700 D13-2000 D13-2100 D13-2300 D13-2500 D13-2800 D13-3310 0 0.5556 0.8056 0.9167 0.9167 0.9444 0.9722 0.9722 1 0.4444 0.1944 0.0833 0.0833 0.0556 0.0278 0.0278 E4-600 E4-900 E4-1000 E4-1100 E4-1250 E4-1500 E4-1800 0 0.9722 0.9444 0.8889 0.0833 0.0000 0.2222 0.7500 1 0.0278 0.0556 0.1111 0.9167 1.0000 0.7778 0.2500 E4-2100 E4-2500 E4-2800 G5-350 G5-450 G5-500 G5-650 0 0.5278 0.6944 0.9444 0.9444 0.7778 0.8889 0.6944 1 0.4722 0.3056 0.0556 0.0556 0.2222 0.1111 0.3056 G5-750 G5-850 G5-1000 G5-1100 G5-1250 G5-1350 G5-1750 0 0.7778 0.8611 0.6944 0.6667 0.2778 0.7500 0.9722 1 0.2222 0.1389 0.3056 0.3333 0.7222 0.2500 0.0278 G5-2500 G5-2750 0 0.9167 0.8056 1 0.0833 0.1944

PAGE 55

41 Genetic diversity was estimated by Neis ge ne diversity and by the Shannon-Weaver diversity index (Shannon and Weaver, 1949). Neis gene diversity is defined as h = 1xi where xi is the frequency of ith allele. Additionally, Shannon-Wievers divers ity index is defined as: H = (pi log pi)/log pi, where I = 1 to n, and p is the pr oportion of the total genotypes made up to the ith genotype. In both indices values close to one indi cate high genetic diversity. Values of h and H among the 100 RAPD loci were extremely variable; with some loci presenting numbers close to one and others small numbers close to zero (Table 3-6). In general, H (Shannon-Wiever index) values were higher than th e ones presented by h (Nei index). To estimate the genetic diversity of the whole set of germplasm, the average of both indices was calculated and was na med total genetic diversity. Total h was estimated as 0.29 0.16, and total H was estimated as 0.45 0.20. Both values can be considered high, which demonstrates the great ge netic variability contai ned in this set of germplasm. Another measure of genetic diversity is obtained by the genetic distance (D) statistic. According to Nei (1972) genetic distance is calculat ed using the following formula: D = -loge I where I = Jxy/(JxJy)1/2, Jxy is the number of bands in common among accessions x and y, and Jx and Jy is the number of bands of accessions x and y, respectively. I values 1 when two accessi ons or populations have identical gene frequencies over all loci examined, and zero when they share no alleles. Here genetic distances were calculated between every pair of accessions and then the average genetic distance was estimated as 0.36. This value also indicates that a gr eat genetic diversity exits among the germplasm evaluated in this research.

PAGE 56

42 Table 3-6. Neis gene diversity and by Sha nnon-Weaver diversity index of RAPD loci Index Locus/Genetic Diversity A4-250 A4-550 A4-670 A4-750 A4-870 A4-1000 A4-1100 h 0.50 0.24 0.44 0.40 0.50 0.31 0.49 H 0.69 0.40 0.64 0.59 0.69 0.49 0.68 A4-1350 A4-1500 A4-1700 A4-1900 A4-2100 A4-2500 A4-3000 h 0.44 0.48 0.48 0.44 0.24 0.44 0.40 H 0.64 0.67 0.67 0.64 0.40 0.64 0.59 A4-3100 B4-700 B4-850 B4-900 B4-1000 B4-1200 B4-1400 h 0.44 0.05 0.42 0.20 0.38 0.50 0.42 H 0.64 0.13 0.62 0.35 0.56 0.69 0.62 B4-1600 B4-1750 B4-2000 B4-2300 B4-2700 B4-3000 B5-500 h 0.28 0.20 0.48 0.38 0.49 0.38 0.05 H 0.45 0.35 0.67 0.56 0.68 0.56 0.13 B5-750 B5-1000 B5-1100 B5-1250 B5-1400 B5-1700 B5-2000 h 0.10 0.49 0.05 0.05 0.10 0.31 0.44 H 0.21 0.69 0.13 0.13 0.21 0.49 0.64 B5-2500 B5-3000 B5-3500 C2-465 C2-520 C2-600 C2-700 h 0.38 0.49 0.10 0.05 0.05 0.28 0.31 H 0.56 0.69 0.21 0.13 0.13 0.45 0.49 C2-750 C2-850 C2-1000 C2-1300 C2-1500 C2-1750 C2-2000 h 0.10 0.49 0.00 0.49 0.40 0.46 0.38 H 0.21 0.68 0.00 0.69 0.59 0.65 0.56 C2-2500 C2-2750 C2-3000 C2-3500 D4-580 D4-650 D4-750 h 0.42 0.40 0.50 0.24 0.20 0.10 0.28 H 0.62 0.59 0.69 0.40 0.35 0.21 0.45 D4-910 D4-1200 D4-1350 D4-1500 D4-2000 D13-370 D13-500 h 0.15 0.05 0.10 0.10 0.42 0.40 0.15 H 0.29 0.13 0.21 0.21 0.62 0.59 0.29 D13-650 D13-750 D13-850 D13-1000 D13-1200 D13-1400 D13-1500 h 0.38 0.46 0.38 0.44 0.40 0.31 0.28 H 0.56 0.65 0.56 0.64 0.59 0.49 0.45 D13-1700 D13-2000 D13-2100 D13-2300 D13-2500 D13-2800 D13-3310 h 0.49 0.31 0.15 0.15 0.10 0.05 0.05 H 0.69 0.49 0.29 0.29 0.21 0.13 0.13 E4-600 E4-900 E4-1000 E4-1100 E4-1250 E4-1500 E4-1800 h 0.05 0.10 0.20 0.15 0.00 0.35 0.38 H 0.13 0.21 0.35 0.29 0.00 0.53 0.56 E4-2100 E4-2500 E4-2800 G5-350 G5-450 G5-500 G5-650 h 0.50 0.42 0.10 0.10 0.35 0.20 0.42 H 0.69 0.62 0.21 0.21 0.53 0.35 0.62 G5-750 G5-850 G5-1000 G5-1100 G5-1250 G5-1350 G5-1750 h 0.35 0.24 0.42 0.44 0.40 0.38 0.05 H 0.53 0.40 0.62 0.64 0.59 0.56 0.13 G5-2500 G5-2750 h 0.15 0.31 H 0.29 0.49

PAGE 57

43 Results obtained in this research are equiva lent to thoses reported in the literature for A. pintoi with a different set of germplasm accessions and primers. Gimenes et al. (2000) used RAPD markers to evalua te the genetic variation of the A. pintoi Brazilian germplasm collection and obtaine d a total of 104 different bands resolved by ten primers. Average number of bands per primer wa s 10.4, ranging from 7 to 15. The proportion of polymorphic loci was 90%, and accessions we re grouped in 3 diffe rent groups based on their genetic distances. Bertozo et al. (1997), also working with a small germplasm collection of A. pintoi, reported 220 amplified bands resolved by 22 primers. Average number of bands per primer was 7.5, ranging from three to 14 bands. The proportion of polymorphic loci observed was 48.5%. The greatest genetic variability was observed within accessions (0.53), while genetic variability among accessions was reported as 0.39. Nobile et al. (2004) working with germplasm of A. glabrata (58), A. burkartii (12), A. nitida (10), A. pseudovillosa (2), and A. lignosa (1) stated that 110 polymorphic RAPD bands were resolved for 10 different primer s. They also presented average genetic distances of 0.30, 0.38, and 0.38, respectively to A. glabrata A. nitida and A. burkartii Hilu and Stalker (1995) worked with 26 accessions of wild species and domesticated peanut, and reported that 10 pr imers resolved 132 RAPD bands. The most variation was observed among accessions of A. cardenasii and A. glandulifera and the least was observed in A. hypogaea and A. monticola Findings in this research were also co mpared to results obtained with RAPD markers to other species. Renu (2003) stat ed that 147 bands were determined by 15 primers when 47 germplasm lines of Pisum sativum L. were used to access the genetic

PAGE 58

44 variability of the species. The proportion of polymorphic loci obtained was 87%. The author concluded that RAPD markers are quick, easy to use and refractory to many environmental influences, which makes the t echnique a very important complement to traditional methods of germplasm characterization. Lowe et al. (2003) studied 56 germplasm accessions of Pennisetum purpureum using 67 RAPD bands. They concluded that the genetic diversity across all accessions was high based on the Shannon-Weaver divers ity index, which they estimated as 0.31. The next step in this research was th e grouping of the accessions based on their genetic distances. In Figure 3-4 the dendogram constructed for the 35 A. pintoi accessions is presented. Four groups were formed accord ing to this dendogram. Group 1 was formed by accessions PI 604798, 604801, 604804, 604805, 604808, 604809, 604814, 604815, 604817, 604856, and 604857. Group 2 was formed by accessions PI 497541, 604800, 604812, 604858, 604859, and CIAT 18745, 20826, 22150, 22152, 22256, 22260, and 22265. Group 3 was formed by accessions PI 476132, 497574, 604803, 604807, 604810, 604811, 604813, and CIAT 22159, 22234, 22271. Fina lly, accession PI 604799 was grouped by itself in group 4. Tissue Culture Regenerated Plants Callus induction was achieved by bot h protocols from most of the A. pintoi accessions evaluated. Callus quantity ratings ranged from 1 to 5, with the most of the accessions having a mean rating of 3 on at least one protocol. There were highly significant (P 0.001) effects of protoc ols, accessions, and the interaction of protocols by accessions (Table 37). Protocol 1 was supe rior to Protocol 2 for both variables related with callus growth.

PAGE 59

45 Figure 3-4. Dendogram illustrating th e genetic relationships among 35 A. pintoi accessions based on Neis genetic distance (Nei, 1972) obtained by 100 RAPD markers resolved by 8 random pr imers and generated by the UPGMA method. PI 604856 PI 604809 PI 604857 PI 604814 PI 604798 PI 604817 PI 604815 PI 604805 PI 604808 PI 604804 PI 604801 PI 604799 PI 604858 PI 604812 PI 604859CIAT 22150 CIAT 22256 CIAT 22152 CIAT 22265PI 604800 PI 497541CIAT 20826 CIAT 18745 CIAT 22260PI 604803 PI 604807 CIAT 22234 PI 604810 PI 604813 PI 604811 PI 476132 PI 497574 CIAT 22159 CIAT 22271 Grou p 1 Grou p 2 Grou p 3 Grou p 4

PAGE 60

46 Mean values of callus quanti ty rating and callus weight fo r both protocols are presented in Table 3-8. Differences in callus growth among protocols can be seen in Figure 3-5, where dishes of accessions PI 604812 are presented. Table 3-7. Analysis of variance table of callus rating and weight produced from Arachis pintoi leaf discs incubated on two different tissue culture media Callus rating Callus weight Source df MS Pr>F MS Pr>F Protocol 1 194.384 0.00014.585 0.0001 Accession 24 1.762 0.00010.059 0.0001 Protocol*Accession 24 1.00 6 0.00010.033 0.0001 Error 240 0.287 0.007 Total 289 Table 3-8. Callus rati ng and weight of Arachis pintoi tissue culture incubated in two different protocols Protocol Callus rating Callus weight (g) 1 3.10 a* 0.333 a 2 1.46 b 0.078 b Means followed by the same letter in the same column were not different by Duncans test (p<0.05). Because the interaction of protocols by accessions was significant, average values of callus rating and weight were presented by protocol. In Table 39 callus quantity rating and weight of accessions incubated in Prot ocol 1 are presented. As expected, great variability for these two variables was di splayed among the accessions. Average callus quantity rating was 3.11 and average callus weight 0.333 g. Accessions PI 604800, 604815, 604858, 604798, 604812, 604805, 604799, and CIAT 22234 were the ones with superior values for callus quantity rating and callus weight. Average callus quantity rating and weight were significantly lower for Protocol 2 than Prot ocol 1 (Table 3-8). Overall average callus quan tity rating was 1.46 and aver age callus weight was 0.078 g. Accessions PI 497541 and 604857 did not produce a ny callus growth with Protocol 2. Accessions, PI 604799, 604804, 604809, 604814, 604815, 604858 and CIAT 22256 had

PAGE 61

47 the highest values on Protocol 2, bu t only PI 604799, 604815, and 604858 were in the highest callus quantity group on both protocols (Table 3-10). Callus ratings and weight were highly correlated (r = 0.98) when both pr otocols were analyze d, which validates the callus rating scale used in this research. Figure 3-5. Callus growth of A. pintoi PI 604812 on two different tissue culture media. Shoot regeneration was achieved for severa l accessions on both media. However, shoot development was variable for each accession and medium, with no structures indicative of somatic embryogenesis being det ected. It seemed that callus quantity was not correlated with shoot regeneration. In Protocol 1 shoot re generation was obtained from accessions PI 604856, 604857, 604805, 604811, 604809, 604814, 604818, and CIAT 22234, 20826, 22152, and 22265. Protocol 1 Protocol 2PI 604812

PAGE 62

48 Table 3-9. Callus rating and callus weight of Arachis pintoi callus induced with Protocol 1 Accession Protocol 1 CIAT / PI number Callus rating Callus weight (g) 20826 3.00 cdef* 0.316 def 22152 2.17 gh 0.120 h 22234 3.50 abcde 0.430 abcd 22256 2.20 fgh 0.183 fgh 22265 3.00 cdef 0.264 efgh 22271 3.00 cdef 0.320 def 497541 3.00 cdef 0.325 def 604798 3.83 abc 0.425 abcd 604799 3.50 abcde 0.521 ab 604800 4.25 a 0.427 abcd 604803 3.33 bcde 0.281 defg 604804 3.17 cde 0.390 bcde 604805 3.50 abcde 0.559 a 604809 3.17 cde 0.327 def 604810 3.00 cdef 0.236 efgh 604811 2.17 gh 0.203 fgh 604812 3.67 abcd 0.427 abcd 604813 3.00 cdef 0.341 cdef 604814 2.83 defg 0.265 efgh 604815 4.00 ab 0.485 abc 604817 3.00 cdef 0.281 defg 604818 2.67 efg 0.234 efgh 604856 3.00 cdef 0.327 def 604857 1.83 h 0.146 hg 604858 4.00 ab 0.501 ab *Means followed by the same letter in the column were not different by Duncans test (p<0.05). For Protocol 2, which yielded lower callus quantity, shoot rege neration was attained from 15 accessions: PI 604856, 604805, 604799, 604804, 604818, 604809, 604810, 604800, 604813, 604857, and CIAT 22256, 22234, 20826, 22152, and 22265. Figure 3-6 illustrates shoot regeneration of accession PI 604813 subcultured in Protocol 1 shoot media induction (MS+1g L-1 BA). Developed shoots were transferred to root induction medium and then rooted plants were placed in pellets in a growth ch amber for acclimatization and subsequently

PAGE 63

49 transferred to pots in the greenhouse. Figure 37 presents a picture of every stage of the process used to tissue culture and regenerate A. pintoi accessions in this research. Table 3-10. Callus rating and callus weight of Arachis pintoi callus induced with Protocol 2 Accession Protocol 2 PI/CIAT number Callus rating Callus weight (g) 20826 1.50 bcd 0.057 cde 22152 1.17 cd 0.065 cde 22234 1.50 bcd 0.090 bcd 22256 1.80 abc 0.057 cde 22265 1.67 bcd 0.096 bcd 22271 1.67 bcd 0.080 cde 497541 1.00 d 0.000 e 604798 1.50 bcd 0.072 cde 604799 2.00 ab 0.173 a 604800 1.67 bcd 0.103 bc 604803 1.00 d 0.050 cde 604804 1.83 abc 0.057 cde 604805 1.17 cd 0.041 de 604809 2.00 ab 0.105 bc 604810 1.00 d 0.027 e 604811 1.17 cd 0.044 de 604812 1.33 bcd 0.094 bcd 604813 1.17 cd 0.032 e 604814 2.33 a 0.142 ab 604815 1.83 abc 0.084 cde 604817 1.17 cd 0.058 cde 604818 1.50 bcd 0.055 cde 604856 1.50 bcd 0.077 cde 604857 1.00 d 0.000 e 604858 1.83 abc 0.143 ab *Means followed by the same letter in the column were not different by Duncans test (p<0.05). Although, shoot regeneration was achieved for several accessions as stated before, shoot development and root induc tion were not achieved in a reliable or repeatable way. Root induction was very difficult to attain, and invariably many shoots died during this process. The addition of 1 g L-1 of actived charcoal in the root medium was tested to evaluate its effect on root formation. Ho wever, little or no effect was achieved.

PAGE 64

50 Figure 3-6. Shoot regeneration of A. pintoi accession PI 604812 subculture in medium MS+1g L-1 BA (Protocol 1). Several reports in the literature account for the developmental problems that face organogenic tissue cult ure protocols of Leguminosae species. Flick et al. (1983) suggested that plant regenera tion has been very difficult to achieve among legumes, and they continue, stating that in cases where it is attained, low freque ncy of regeneration is often present. The same authors stated that forage legumes have been more difficult to induce and form roots than seed legumes. Figure 3-7. Representation of the different step s in the process used to tissue culture A. pintoi PI 604813Protocol 1

PAGE 65

51 Arachis species are among the legume species that have problems with shoot development and root formation. Chengalrayan et al. (1995) obtaine d 33% average shoot rooting in MS hormone free medium, a nd 52% of shoot rooting when NAA in concentrations of 0.01, 0.05, and 0.5 mg L-1 was added to the medium with A.hypogaea cultivar J.L. 24. However this result was specific for this particular cultivar and in general A. hypogaea is recalcitrant in tissue culture. Akasaka et al. (2000) were successful in inducing bud formation and shoot development from leaf segments of A. hypogaea cv. Chico. The percentage of conversion of buds to shoots was relatively high (34.7%). However, the highest frequency of shoot regeneration was 14.3%, suggesting that some buds failed to grow into normal shoots and plants. Burtnick and Mroginski (1985) studying met hods to regenerate plants from leaf explants of A. pintoi cultured on Murashige and Skoog (MS) nutrient medium containing different combinations of NAA (0.1-2 mg L-1) and BA (0.1-3 mg L-1) reported that callus induction was nearly 100%. However calluse s were usually fria ble and produced no shoots or roots. The same problems described above are also reported for other w ild species of the genus Arachis McKently et al. (1990) were able to regenerate plants of A. glabrata using MS medium supplemented with 3 to 5 mg L-1 of BAP. They reported, however that only 10% of shoot meristems continued growth and development into whole plants. Due to the problems reported above, only 16 regenerated plants were obatined by both protocols. Table 3-11 presents the number of regene rated plants of A. pintoi

PAGE 66

52 accessions from protocols 1 and 2. All plants were phenotypically normal and uniform in their appearance, with no somacl onal variation visually observed. Table 3-11. Number of regenerated plants of Arachis pintoi accessions cultured on two different protocols Accession Protocol PI/CIAT number 1 2 ---------------------------no. pls --------------------------20826 0 2 22234 1 0 22265 1 0 604799 0 1 604805 0 1 604809 4 0 604810 0 1 604811 3 0 604856 0 1 604857 1 0 Total 10 6 Differences in callus induction and growt h, shoot regeneration and development, root formation, and plant regeneration were observed among accessions. Callus induction and growth was observed in all 25 accessions in Protocol 1, but only in 23 in Protocol 2. Shoot induction was observed in 11 accessions in Protocol 1 and 15 in Protocol 2. Both protocols were able to produce regenera ted plants of five accessions. Although differences in callus quantity rating and wei ght among protocols were observed earlier in the tissue culture process, we can conclude that based on shoot development and plant regeneration both protocols were equivale nt. Additional research to study the shoot development and rooting problems observed in th is work should be carried out to secure an efficient tissue culture protocol to regenerate A. pintoi plants. Leaf tissue of the 16 regenerated plants was collected, ground, and used to extract DNA using the protocol described earlier. No differenc es in RAPD band profiles were detected between the original source plants and the tissue culture regenerated plants.

PAGE 67

53 Although just a few plants were regenerated a nd analyzed, these resu lts indicate that both protocols could be suitable for in vitro plant genetic conservation of A. pintoi accessions since no variation in RAPD band profile wa s observed. Further test s on the regenerated plants must be conducted to assure that no somaclonal variation occurred during the tissue culture process.The application of in vitro genetic conservation has several advantages, which include small space required to store large number of accessions, low costs compared to growing and organizing pl ants annually in the field, and maintaining large living collections of fr uit trees (Hawkes et al., 2000). Summary and Conclusions Thirty-five A. pintoi accessions stored at the Southe rn Regional PI Station of the National Plant Germplasm System (NPGS) lo cated at Griffin, GA were used to study genetic diversity at the molecular level usi ng RAPD markers. Conc urrently, two tissue culture protocols, proposed by Rey et al. (2000) and Ngo and Quesenberry (2000), were evaluated for their organogenesi s ability, and capacity to generate somaclonal variation. From the original 18 primers tested, amplif ications were obtaine d in just eight of them. Primers A4, B4, B5, C2, D4, D13, E4, and G5 amplified 100 different bands. The average number of amplified bands per primer was 12.5, with primer D4 amplifying eight fragments and primer D13 amplifying 16 frag ments. The size of these 100 fragments ranged from 250 bp to 3500 bp. From the 100 bands amplified, 98 presented polymorphism with the only two exceptions being Primer C2-1000 bp and primer E41250 bp. The average presence of bands per accession was 32, ranging from 20 to 44 bands. Ten bands were able to discrimi nate individual germplasm accessions. The proportion of polymorphic RAPD loci was 89%. Genetic diversity of the whole set of germplasm was estimated by Neis gene dive rsity (h) and by Shannon-Weaver diversity

PAGE 68

54 index (H). The average h was0.29 0.16, and average H was 0.45 0.20. Average genetic distance was estimated as 0.36, and indi cated that a great genetic diversity exits among the germplasm evaluated in this research Genetic distances were used to prepare a dendogram for the 35 A. pintoi accessions, which separated them in four distinct groups. Callus induction was achieved on two differe nt M.S. basal protocols after 28 d of incubation. Analysis of variance demonstrated that Pr otocol 1 was superi or to Protocol 2 for both variables related with callus growth. Ca llus rating and weight values of Protocol 1 confirmed the great variability for these two variables among the accessions. Average callus rating was 3.11 and average callu s weight 0.333 g. Accessions PI 604800, 604815, 604858, 604798, 604812, 604805, 604799, and CIAT 22234 showed the highest values. In Protocol 2, average callus ra tings and weight we re significantly lower than the values obtained with Protocol 1. Average callus rating was 1.49 and averag e callus weight was 0.072 g. Accessions PI 497541 and 604857 did not produce callus growth. Accessions PI 604799, 604804, 604809, 604814, 604815, 604858, and CIAT 22256 were the ones with highest values. Shoot regeneration was ach ieved for several accessions on both media with no structures indicativ e of somatic embryogenesis being detected. Callus growth was not correlated with shoot regeneration. In Protocol 1 shoot regeneration was obtained from 15 accessions, whereas in Protocol 2, s hoot regeneration was attained from 18 accessions, but only PI 604856, 604818, 604809, and CIAT 20826 and 22234 regenerated shoots on both pr otocols. Root induction was very difficult to obtain, and invariably many shoots died during this proce ss. At the end, just 16 regenerated plants were recovered between the two protocols.

PAGE 69

55 Although differences in callus ratings and weight among protocols were observed earlier in the tissue culture process, we c onclude that based on shoot development and plant regeneration both protocol s were similar. RAPD band profiles of regenerated tissue culture plants were similar to their parent plants. However, we should point out that the number of regenerated plants was too small to conclusively affirm that genetic stability will be maintained by these two protocols. Fu rther investigations should be conduced to definitely confirm our findings.

PAGE 70

56 CHAPTER 4 MORPHOLOGICAL CHARACTERIZATION OF Arachis pintoi GERMPLASM Introduction Morphological characterization is used to assess and understand the genetic variability of germplasm collections. When wo rking with a wild species of a genus of interest, this usually is the chance to first gather basic knowledge about it. Polymorphic, highly heritable morphological traits were originally used in early scientific investigations of genetic diversit y; such as the ones pe rformed by Mendel and DeVries In general, morphological studies did not involve sophisticated equipment or laborious procedures, and these monogenic or oligogenic morphological traits were simple, rapid, and inexpensive to score (Hawkes et al., 2000). The information generated from this type of morphological characterization research can be used to identify individua l accessions based on a set of particular phenotype traits. Such data can also be us eful to estimate genetic diversity of a germplasm collection, which will possibly impact the decision to enlarge the gene pool by further collection trips. Additionally, this activity can generate information about the genetic divergence among the accessions, whic h can be used to group accessions. Such grouping of the germplasm accessions can be used to assemble core collections, especially important in large germplasm banks. Several statistical techniques may be used to assess genetic divergence. However, when a large number of accessions are present, multivariate analysis is most appropriate.

PAGE 71

57 The advantage of this technique is the ability to discriminate among accessions considering multiple variables at the same time. Among the multivariate techniques the most used in genetic studie s are Principal Component and Cluster Analysis (Hawkes et al., 2000). Principal components analysis (PCA) has been widely used in genetic resource related research. The technique can be used with several objectives: Quantification of genetic diverg ence among germplasm accessions Selection of divergent pare ntal genotypes to hybridize Variable reduction in sets of data with great number of parameters Variable exclusion based on its co ntribution to the total variance Calculation of a similarity index for the purpose of grouping accessions The goal of this research was to morphologically characterize the A. pintoi germplasm accessions stored at the USDA-N PGS germplasm bank and to cluster the accessions based on similarity indices. Materials and Methods Accessions of A. pintoi stored in the Southern Regi onal Plant Introduction Station of the National Plant Germplasm System (N PGS) located at Gr iffin, Georgia were transferred to the University of Florida in 2001 and 2002. A list of these accessions with information related to the respective PI numbe rs and sites of collection is presented in Appendix A. Morphological characterization of the above accessions was accomplished by evaluating each individual acc ession for a list of morphologica l descriptors selected from the IBPGR/ICRISAT list of mo rphological descriptors for Arachis (1990 and1992). Data from leaves, stems, flowers, pegs, pods, and seeds were collected from plants of each

PAGE 72

58 accession. The list of morphological descriptors ev aluated is presented in more detail in Table 4.1. Table 4-1. Morphological de scriptors applied to Arachis pintoi accessions Morphological descriptor Descriptor code # of structures measured Unit Equipment used Flower/inflorescence FPI 10 number Visual evaluation Flower standard width FSW 10 mm Caliper Flower standard length FSL 10 mm Caliper Flower standard color FSC 10 Color scale Visual evaluation Flower standard crescent FScr 10 Present/absent Dissect scope Flower wing width FWW 10 mm Caliper Flower wing length FWL 10 mm Caliper Flower keel length FKL 10 mm Caliper Flower hypanthium length FHL 10 mm Caliper Flower hypanthium width FHW 10 mm Caliper Flower hypanthium color FHC 10 Present/absent Dissect scope Flower hypanthium hairiness FHH 10 Present/absent Dissect scope Flower pollen size FPSi 10 mm Caliper Flower pollen shape FPS 10 IBPGR scale Microscope Stem internode length SIL 50 mm Caliper Stem internode diameter SID 50 mm Caliper Stem color SC 50 Present/absent Dissect scope Stem hairiness SH 10 Pres/abs/abun Dissect scope Stem bristles SGH 10 Present/absent Dissect scope Leaflet shape LS 10 Shape scale Visual evaluation Leaflet hairiness sup. surface LHU 10 Present/absent Dissect scope Leaflet hairiness margin LHM 10 Present/absent Dissect scope Leaflet hairiness inf. surface LHL 10 Present/absent Dissect scope Leaflet bristles sup. surface LGHU 10 Present/absent Dissect scope Leaflet bristles margin LGHM 10 Present/absent Dissect scope Leaflet bristles inf. surface LGHL 10 Present/absent Dissect scope Leaflet length LL 10 mm Caliper Leaflet width LW 10 mm Caliper Leaflet length/Leaf width LLLW 10 mm Caliper Leaf Petiole length LPL 10 mm Caliper Peg length PegL 10 mm Caliper Peg width PegW 10 mm Caliper Peg color PegC 10 Present/absent Visual evaluation Peg hairiness PegH 10 Present/absent Visual evaluation Pod weight PodWe 10 grams Scale Pod length PodL 10 mm Caliper Pod width PodW 10 mm Caliper Pod beak PodB 10 IBPG R scale Dissect scope Pod reticulation PodR 10 IBPGR scale Dissect scope Seed width SW 10 mm Caliper Seed length SL 10 mm Caliper Seed weight SWe 10 grams Scale Seed color Scolor 10 ISCC-NBS color chart Dissect scope Pod weight/Seed weight PodweSwe 10 %

PAGE 73

59 Two plots of four plants (2 m x 2 m) were established at the Forage Research Unit of the Agronomy Department of the Universi ty of Florida in Gainesville-FL. Six random plants of each accession were selected from thes e plots and used as plant-part sources. For each genotype, 10 stems per plant were collected, the terminal part of each stem (1st 3 internodes) was discarde d and then five internodes were evaluated on each stem. For the other part categories, 10 pieces were collected and evaluated. Continuous variables were measured with a 15 cm electro nic digital caliper (C hicago Brand Industrial Inc., Fremont, CA), and categorical vari ables were scored under the dissecting microscope using the standards prop osed by IBPGR/ICRISAT (1990 and 1992). The data were tabulated in a Microsoft Excel spreadsheet and the mean, standard deviation, and range were calculated for qua ntitative descriptors, and the mode was determined for qualitative ones. Genetic variability among the accessions with respect to the morphological descriptors was examined by calculating Simpsons (1949) and Shannon-Weavers (1949) diversity indices. These indices give a measure of phenotypic diversity and range from zero to one, where one represents grea t genetic diversity a nd zero the opposite or no genetic diversity. The indices correspond to th e probability that two individuals randomly selected from a group of populations will ha ve the same morphological feature. The formula to calculate both i ndices is presented below. Shannon-Wiever Diversity index: H = -SUM (pi log pi)/log pi Simpson Diversity index: D = 1-SUM (pi 2) where i= 1 to n, and p is the proportion of the total morphotypes made up of the ith morphotype.

PAGE 74

60 The data matrix was then analyzed with SAS software (SAS institute, 1989). Phenotypic correlations among morphological desc riptors were computed with the PROC CORR procedure. After that, qualitative characteristics we re transformed with PROC PRINQUAL, and a principal component anal ysis was performed with the procedure PROC PRINCOMP. Finally, a cluster analysis using the Complete linkage method was prepared. Means of quantitative traits of each group were compared by using the Newman-Keuls procedure (SAS institute, 1989). Result and Discussions Great morphological variability was ob served among the accessions for all descriptors, exceptions being, pollen size and shape, and br istles on the superior and inferior leaf surface, which showed no pol ymorphism. In Table 4-2, the mean/mode, standard deviation, variance and range of each morphological descriptor are listed. Flowers arise from inflorescences located at reproductive buds under the lea axils. According to Conagin (1959), each infloresce nce produces one to nine flowers, which will bloom in a sequence, usually with a 1-2 d interval. Numbers observed in this study revealed accessions with a mean of two flow ers per inflorescence and a maximum of four flowers per inflorescence. Although important, these differences do not seem to have great impact in terms of seed production. Arachis pintoi flowers are typical of papilion acea legumes possessing five petals. They displayed one standard petal, two wing pe tals, and a keel that is actually formed by the fusion of two petals. The other flower structures are the ca lyx and the hypanthium.

PAGE 75

61 Table 4-2. Morphological characteristics of Arachis pintoi germplasm accessions Morphological Descriptor Mean/Mode Standard Deviation Variance Range FPI 3.14 0.33 0.11 2.00 3.80 FSW 15.31 1.63 2.67 11.99 18.85 FSL 11.58 1.10 1.22 9.84 14.45 FSC Yellow FScr Present Present/Absent FWW 6.27 0.69 0.47 4.47 7.95 FWL 8.07 0.69 0.48 6.92 9.80 FKL 5.09 0.23 0.05 4.45 5.55 FHL 70.66 12.58 158.28 52.39 -104.79 FHW 0.97 0.10 0.01 0.77 1.12 FHC Absent Present/Absent FHH Present Present/Absent FPSi 4.00 0.00 0.00 FPS Round SIL 29.47 7.43 55.27 18.23 52.23 SID 2.95 0.84 0.71 2.05 7.15 SC Absent Present/Absent SH Present Present/Absent SGH Absent Present/Absent LS Obovate LHU Absent Present/Absent LHM Absent Present/Absent LHL Present Present/Absent LGHU Absent Present/Absent LGHM Present Present/Absent LGHL Absent Present/Absent LL 24.89 4.07 16.60 15.79 32.03 LW 15.14 2.79 7.80 9.45 20.78 LLLW 1.66 0.21 0.04 1.29 2.06 LPL 22.26 6.40 40.91 11.67 39.33 PegL 11.90 3.36 11.28 6.30 21.40 PegW 0.94 0.14 0.02 0.57 1.23 PegC Present Present/Absent PegH Present Present/Absent PodWe 0.13 0.07 0.01 0.07 0.25 PodL 9.33 4.52 20.40 7.55 14.25 PodW 4.70 2.22 4.91 4.53 6.91 PodB Moderate PodR Slight SW 3.89 1.86 3.45 3.55 5.46 SL 7.13 3.50 12.22 5.71 10.51 SWe 0.10 0.06 0.001 0.04 0.18 Scolor Orange yellowish PodweSwe 75.47 6.71 44.99 59.71 87.07

PAGE 76

62 Most of the accessions (91%) have yellow flowers, with the 9% exception showing a lighter shade of yellow classified as lem on yellow (Figure 4.1). This was the case of accessions PI 604814, PI 604818, and CIAT 20826. Sta ndard color is considered a very good genetic marker, and so it could be used in genetic studies with A. pintoi Our results in terms of standard petals color are different from the ones reported by Maass et al. (1993) who performed morphologi cal characterization of eight accessions of A. pintoi. They reported that lemon yellow was the standard petal color displayed by all eight germplasm lines. Qualitative traits su ch standard petals color are subject to individual interpretation and th at is probably why their results are very different them the ones reported in this research. Also different are the results of Upadhya ya (2003) who worked with the ICRISAT core collection (1704 accessions) of Arachis hypogaea. He stated that 97% of the germplasm in that collection had orange st andard petals. White, lemon yellow, yellow, and dark orange were not observed in any of his accessions. Differences in standard petal length a nd width, wing length a nd width, keel length, and hypanthium length and width were also obse rved. Those features are directly related to overall flower size. Figure 4.2 illustrates the large differences displayed by the germplasm in relation to this group of charac teristics. Also, differences in hypanthium color were found and can be observed in Figur e 4.2. The mode for this characteristic was absence of color, but some accessions presented a distinct purple color, which is probably associated with the presence of anthoc yanin. One of the re ported functions of anthocyanin is to attract inse cts: since it absorbs UV radiatio n that is very attractive to them (Mann, 1987).

PAGE 77

63 Arachis pintoi stems were hollow with repr oductive and vegetative nodes occurring along the stem length. Vegetative node s have the ability to root which allows the species to be vegetatively propagate d, a desirable agronomic characteristic. Figure 4-1. Flower st andard colors of Arachis pintoi germplasm. Figure 4-2. Flower sizes and hypa nthium colors displayed by Arachis pintoi germplasm. Large differences were observed in relati on to stem internode length and diameter, and stem color. The average internode length was 29 mm, ranging from 18 to 52 mm. Average internode diameter was 3 mm, vary ing from 2 to 7 mm. As in the case of hypanthium color, it seems that stem inter node color is determined by the presence of

PAGE 78

64 anthocyanin. However, here more variants in color were present than in the former feature, although scoring was just made in terms of presence and absence. Absence was the mode for this characteri stic with 57% of the accessi ons showing no pur ple coloration, which is translated as a light green color. Among those that displa yed stem coloration, yellowish, pinkish, light and deep pu rple were observed (Figure 4-3). A. pintoi leaves are compound with four leafle ts. Usually the basal leaflets were smaller than the distal, and exhibited an ellipt ic shape. As observed in flower and stem features, large differences we re also present in leaves. Figure 4-3. Stem characteristics of Arachis pintoi germplasm. When examining leaflet characteristics, the f eatures related to overa ll leaf size (leaflet length and width and petiole length) and shape were the ones that showed the most variability. Average leaflet length was 25 mm with a range of 16 to 32 mm. Leaflet width varied from 9 to 21 mm, with an average va lue of 15 mm. Due to le aflet length and width variation, an array of leaflet si zes arise (Figure 4-4). In terms of leaflet shape ten different types were present with obovate shape being th e mode, this agreed with the findings of

PAGE 79

65 Maass et al. (1993). However, Upadhyaya (2003) reported differe nt results among the 1704 accessions of the ICRISAT A. hypogaea core collection, which had elliptic leaf shape without exceptions. Similar to others species of the genus, A. pintoi also possesses geocarpic fruits. Flowers self-pollinate and then lose their petals as the fertilized ovary begins to enlarge. The budding ovary or "peg" grows under the ground, away from the plant, forming a small stem which contains the ovary a nd the embryo in its tip. The embryo turns horizontal to the soil surface and begins to ma ture inside a pod. In the cultivated species ( A. hypogaea ), two to four or more seeds may be formed in each pod. However, in A. pintoi single seed pods were exclusively observed. Figure 4-4. Leaflet characteristics of Arachis pintoi germplasm. Pegs, pods, and seeds exhibited an array of variability among the accessions. One of the differences is the fact th at accessions PI 604804, 604813, 604817, and CIAT 22152, 22159, 22234 did not produce any of the a bove plant structures under the conditions of these experiments. Additionall y, differences were also present in the

PAGE 80

66 accessions that produced these structures. Av erage peg length was 12 mm, ranging from 6 to 21 mm. With respect to the other peg features, average peg width was 0.90 mm, and presence of peg color and hairiness were the mode. Pod size and weight showed great di vergence among the germplasm. These characteristics were most impacted by pod length and width. Large pod size and weight were reflected in large seed size and weight. Figure 4-5 illustrates th e differences in seed size among the genotypes. Average seed we ight was 0.12 g, ranging from 0.04 to 0.18 g. Also, the average ratio of seed weigh t/pod weight was 0.75, ranging from 0.59 to 0.87. Figure 4-5. Seed characteristics of Arachis pintoi germplasm. Phenotypic correlations were calculate d among morphological descriptors, and Pearsons correlation coefficien t and significance test were also calculated (Appendix C). Pollen size and shape (FPSi and FPS) and bristles on the superior and in ferior leaf surface (LGHU and LGHL) were not included because they did not show any variability. In addition, leaf length/leaf width ratio ( LLLW) and pod weight/s eed weight ratio (PodweSwe) were left out of the correlation ma trix because they were derived from two other variables. Accessions PI 604804, 604813, 604817, and CIAT 22152, 22159, 22234 were excluded because they did not produ ce any pegs, pods, or seeds. In summary, 35

PAGE 81

67 morphological descriptors and 28 germplas m accessions were used to calculate correlation coefficients. Therefore, with 26 degrees of freedom (df = N-2 or 28-2), any correlation coefficient with an absolute value greater than 0.361 had a P value of 0.05. From 595 correlations computed, 96 were higher than 0.361. These 96 significant correlations could be divided into eight diffe rent groups (Table 4-3). Table 4-3. Correlations among mo rphological descriptors of Arachis pintoi germplasm Group Group Description 1 Correlations between flower features dimensions 2 Correlations between leaflet features dimensions 3 Correlations between flower features dimensions and leaflet dimensions 4 Correlations between stem diameter and leaflet dimensions 5 Correlations between flower features dimensions and pod and seed dimensions 6 Correlations between stem diam eter and pod and seed dimensions 7 Correlations between leaflet features dimensions and pod and seed dimensions 8 Correlations between pod features dimensions and seed dimensions Correlations in Groups 1, 2, and 8 are not very meaningful and could be explained by the fact that part proportions should be maintained within each plant organ or structure. So, flowers with large standards length will also have large standard width, wing length and width, and keel length. For th e same reason we should expect that pods with large measures will also have large seed if an adequate development had occurred, because the seeds are enclosed by the pod. Correlations in Groups 3 and 4 would have ag ronomic importance if leaflet size is translated into higher vegetative mass produc tion. Likewise leaf/stem ratio could be important assuming leaves have higher nutri tive values than stem. These are very important features in forage species. These co rrelations suggest that flower size and stem diameter could be included as selection indices in A. pintoi forage selection and breeding

PAGE 82

68 programs. The same idea could be applied to Groups 5, 6, and 7, where stem, flower, and leaflet size appear to im pact pod and seed size. Although 96 correlation coefficients were considered sign ificant (r = 0.361). Skinner et al. (1999) suggested that only corr elation coefficients with absolute values higher than 0.71 should be consid ered biologically meaningful. They explain that only in these situations is more than 50% of the vari ance of one trait is pr edicted by the other. Reexamining the correlation coeffici ent table (Appendix C) under this new criterion only 29 correlation coefficients out of 595 would be considered biologically meaningful. In this situation only Groups 1, 2, 7, and 8 remain in place. As stated before, Groups 1, 2, and 8, do not have great biolog ical importance, since they could be explained by the fact that pa rt proportions should be mainta ined within each plant organ or structure. Therefore, only correlati ons relative to Group 7 may have meaningful biological importance. Examining this inform ation more carefully, we can observe that meaningful correlations were found between leaf length (LL) and Pod weight (Podwe); leaf length (LL) and pod width (PodW); leaf length (LL) and seed weight (Swe); leaf length (LL) and seed width (SW); and finally leaf length (LL) and seed length (SL). Thus, leaf length could be used as a selecti on criterion in programs where increased seed size/weight is one of the objectives. One of the biggest problems with forage legumes species is slow establishment, which sometimes is associated with small s eed size. Having a large seed could represent more stored reserves and higher seedling vigor, which would reduce the establishment time. These suggestions based on the above co rrelations should be verified in future research to evaluate the use of these characteristics as selection indices.

PAGE 83

69 To support this hypothesis we could examin e the findings of Skinner et al. (1999) who worked with the Australian annual Medicago collection (20997 accessions) measuring 27 traits. They reported that the tr aits seeds per gram and winter and spring herbage yield were correlated (0.42 and 0.29). Th ey also stated that seeds per gram and seedling vigor were correlated (0.44). Upadhya ya (2003) also found correlations between 100 seed weight and yield (0.32) in the ICRISAT Arachis hypogaea core collection. Although we can affirm that phenotypic genetic diversity was observed among the accessions by examining the information contained in Appendix B and Table 4.2 there is a need to quantify this diversity. To achieve this goal we should make use of a genetic diversity index. Two of the most used diversity indices are Shannon-Weaver and Simpson indices. These indices are often used in ecological studies where species richness and composition of a particular community or ecosystem are evaluated. Recently, these indices have been applied to quantify genetic diversity of germplasm collections when phenotypic frequencies are co llected. The greater the index value, the greater the genetic diversity. In Table 44 the values of the indices for each morphological descriptor are presented. The to tal genetic diversity was also calculated, and it is an indication of how different the accessions are in relation to the morphological features utilized in this research. Diversity values were variable among traits, but in gene ral all morphological features expressed high genetic diversity. A ccording to Simpsons index, leaf shape (0.83), seed color (0.82), flow er standard width (0.81), and seed length (0.75) were the descriptors with greatest dive rsity. In opposition, the lowest di versity values were related

PAGE 84

70 to flower standard color ( 0.18), flower standard crescent (0.24), and flower hypanthium hairiness (0.24). The total Simpsons inde x to all morphological descriptors was 0.58. Shannon-Weavers diversity values were in general higher than Simpsons. Descriptors with higher values for the Shannon-Weaver index we re leaf hairiness inferior surface (1.00), flower hypanthium color (1.00), leaf hairiness margins, and pod reticulation (0.97). Total genetic diversity was estimated as 0.71. Shannon-Weaver diversity index values obser ved in this work are much higher than the ones observed by Upadhyaya et al. ( 2002), who applied 38 agromorphological descriptors to the whole A. hypogaea ICRISAT collection (13342 accessions). These authors found a total genetic diversity va lue of 0.50. Leaflet length (0.62) and shelling percentage (0.62) were the tr aits showing most variation. These values are also higher than those reported by Upadhyaya (2003) who evaluated a core collection prep ared using the results of the previous work. He obtained a total diversity index of 0.44 in 32 agromorphol ogical traits. He concluded that the core collection had significant variation for the mo rphological and agronomic traits evaluated. A principal components analysis (PCA ) was performed with the goal of discriminating among accessions and grouping even further. Principal components analysis (PCA) can be used in sets of data with large num ber of variables. The goal of PCA is to provide a reduced dimension mode l that would indicate measured differences among groups. It also can contri bute to a better understandin g of the set of variables by describing how much of the to tal variance is explained by each one. With this objective the PCA was performed with the matrix of morphological data generated by applying the list of descriptors presented in Table 4-1.

PAGE 85

71 Table 4-4. Simpson and Shannon-Weaver diversity indices for Arachis. pintoi morphological descriptors Morphological Descriptor Simpson Shannon-Weaver Flower/inflorescence 0.60 0.34 Flower standard width 0.81 0.89 Flower standard length 0.68 0.80 Flower standard color 0.18 0.47 Flower standard crescent 0.24 0.58 Flower wing width 0.56 0.33 Flower wing length 0.60 0.33 Flower keel length 0.54 0.74 Flower hypanthium length 0.73 0.89 Flower hypanthium width 0.73 0.46 Flower hypanthium color 0.50 1.00 Flower hypanthium hairiness 0.24 0.58 Stem internode length 0.50 0.58 Stem internode diameter 0.65 0.45 Stem color 0.47 0.96 Stem hairiness 0.63 0.96 Stem bristles 0.36 0.80 Leaf shape 0.83 0.83 Leaflet hairiness inf. surface 0.50 1.00 Leaflet hairiness margins 0.48 0.98 Leaf bristles margins 0.36 0.80 Leaf length 0.64 0.46 Leaf width 0.58 0.34 Leaf length/Leaf width 0.67 1.00 Leaf petiole length 0.74 0.84 Peg length 0.59 0.50 Peg width 0.46 0.71 Peg color 0.45 0.92 Peg hairiness 0.45 0.92 Pod weight 0.69 0.45 Pod length 0.47 0.74 Pod width 0.74 0.89 Pod beak 0.73 0.87 Pod reticulation 0.64 0.97 Seed width 0.69 0.48 Seed length 0.75 0.88 Seed weight 0.64 0.37 Seed color 0.82 0.89 Pod weight/Seed weight 0.57 0.87 Total 0.58 0.71

PAGE 86

72 Variables pollen size and shape (FPSi and FPS) and leaf bristles superior and inferior surface (LGHU and LGHL) were not included because they did not show any variability. In addition, leaf length/leaf width ratio (LLLW) and pod weight/seed weight ratio (PodweSwe) were also left out because they are derived from two other variables. After the first analysis, the leaflet hairiness superior surface variable (LHU) was also excluded from the analysis because it contri buted little to the e xploration of total variance. The first five principal co mponents (PCs) were responsible for 67.7% of the total variation (Table 4-5). Values similar to these were reported by Stalker (1990), Upadhyaya et al. (2002), Upa dhyaya (2003), and Upadhyaya et al. (2002), who worked with wild species of Arachis groundnut, and chickpea germplasm collections, respectively, to explain their results. The first PC explaine d 30.0% of the variation, the second accounted for 15.2%, the third for 10.1%, the fourth for 6.6%, and the fifth explained 5.8% of th e total variation. Examining the variable loadings of the fi rst five PCs (Table 4-5) we can clearly observe that the characteristics of pegs pods, and seeds are the ones with highest contribution to PC1. Therefore, PC1 could be termed the sexual reproduction axis. Performing the same exam to PC2 we can conclu de that the features related with flower and leaf dimensions were the ones with highes t loadings. Therefore, PC2 could be called the vegetative axis. Finally, examining the loading of PC3 we note that features related to the shape, color, and hairiness of morphol ogical structures were the ones with the most contribution. Those are all qualitative features and because of that we could call PC3 the qualitative axis.

PAGE 87

73 Therefore, we could state that the prin cipal components analysis was able to discriminate and separate the accessions in terms of these three dimensions, represented by sexual reproduction, vegetative, and qualitative axes. This is clearly observed when accessions were projected in two-dimens ional graphs formed by PC1 and PC2, PC1 and PC3, and PC2 and PC3 (Fi gures 4-6, 4-7, and 4-8). Table 4-5. Vector loadings and percentage of variation explained by the first five principal components for mor phological characteristics of Arachis pintoi Principal components Characteristics 1 2 3 4 5 Variance explained (%) 30.01 15.15 10.09 6.63 5.76 Cumulative variance explained (%) 30.01 45.16 55.25 61.88 67.65 Flower/inflorescence 0.077 0.041 0.215 0.036 -0.019 Flower standard width -0.060 0.338 -0.085 -0.092 0.027 Flower standard length -0.068 0.369 -0.028 0.070 -0.014 Flower standard color -0.060 -0.038 0.072 0.333 0.317 Flower standard crescent 0.145 -0.117 0.056 -0.132 0.248 Flower wing width -0.065 0.331 0.006 -0.099 0.122 Flower wing length -0.080 0.350 -0.017 0.025 0.085 Flower keel length -0.084 0.213 0.146 0.200 0.141 Flower hypanthium length -0.097 0.299 0.089 0.136 -0.142 Flower hypanthium width 0.017 0.218 0.273 0.235 -0.128 Flower hypanthium color 0.023 0.013 -0.164 -0.165 -0.382 Flower hypanthium hairiness -0.009 0.087 0.133 0.124 -0.422 Stem internode length -0.030 0.079 -0.304 -0.145 0.051 Stem internode diameter -0.043 -0.036 0.157 -0.194 0.338 Stem color -0.076 -0.042 0.090 0.203 -0.274 Stem hairiness 0.015 0.051 0.438 0.104 -0.115 Stem bristles -0.033 0.198 0.013 0.085 0.112 Leaflet shape -0.105 0.074 -0.275 0.231 0.115 Leaflet hairiness margin -0.056 -0.063 0.302 -0.290 0.023 Leaflet hairiness inf. surface 0.006 -0.086 0.366 -0.180 0.139 Leaflet bristles margin 0.021 0.071 -0.186 0.285 0.290 Leaflet length 0.057 0.285 0.039 -0.310 0.011 Leaflet width 0.155 0.233 0.071 -0.245 -0.101 Leaf Petiole length 0.062 0.251 -0.138 -0.248 0.028 Peg length 0.231 0.070 -0.092 0.038 -0.165 Peg width 0.256 -0.001 0.015 0.185 -0.075 Peg color 0.138 -0.090 -0.195 -0.134 -0.140 Peg hairiness 0.109 -0.002 -0.240 0.113 -0.075 Pod weight 0.273 0.099 0.016 -0.029 0.095 Pod length 0.293 0.016 0.023 0.046 0.032 Pod width 0.291 0.033 0.047 0.079 0.035 Pod beak 0.274 0.017 0.003 -0.034 0.022 Pod reticulation 0.288 -0.016 0.024 0.111 -0.005 Seed width 0.292 0.044 0.044 0.066 0.041 Seed length 0.293 0.042 0.008 0.001 0.042 Seed weight 0.268 0.101 -0.005 -0.031 0.122 Seed color 0.287 -0.028 0.003 0.099 0.012

PAGE 88

74 Figure 4-6. Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC1 and PC2. Figure 4-6 corresponds to the plane fo rmed by PC1 and PC2 dimensions. PC1 values are on the Y axis and PC2 on the X axis. Moving from the bottom of the Y axis where coefficients were negative to the top where they were positive represents moving from lower values of peg, pods, seed dimens ions, and weight. Doing the same to the X axis represented by PC2, and moving from th e left (coefficients) to the right (+ coefficients) means that we are moving from lo wer values of flower and leaf dimensions to higher values. Accessions were obviously discriminated, and two groups were formed. The group in the bottom was composed of six accessions and the othe r 29 in the top of the graph. These six accessions in the bottom are the ones with no pegs, seeds, and pods.

PAGE 89

75 Figure 4-7. Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC1 and PC3. In the same way, figure 4-7 corresponds to the plane formed by PC1 and PC3 dimensions, where PC1 represents the Y axis and PC3 represents the X axis. Here, the Y axis indicates the same tendency as in the pr evious figure. PC3 is the represented by the qualitative characteristics (shape, color, and hairiness), and a more comprehensive reading of the tendency here is difficult to achieve. However, accessions were also discriminated with two groups formed. The gr oup in the bottom was again composed of the six accessions with no pegs, seeds, and pods.

PAGE 90

76 Figure 4-8. Projection of the 35 Arachis pintoi accessions in a two-dimensional graph defined by PC2 and PC3. Finally, figure 4-8 represen ts the plane formed by PC2 and PC3 dimensions, where PC2 represents the Y axis and PC3 represents the X axis. Moving from the bottom (coefficients) to the top (+ coefficients) mean s that we are moving from smaller to bigger flowers and leaves. Accession discrimina tion was also obtained. However, group formation was more difficult to obtain, on ce accessions were spread in the plane. We could state that by using this se t of descriptors we accomplished the discrimination of the A. pintoi germplasm. All the 35 morphol ogical features used in the PCA presented high loading values at least on ce when the first five PC were analyzed, reinforcing the importance of each one as an A. pintoi descriptor.

PAGE 91

77 Since discrimination was obtained by usi ng of the genetic diversity indices and also by the principal component analysis, th e next step was to perform a grouping or cluster analysis. The first nine principal components were used to execute a cluster analysis using the complete linkage cluste rs method. The dendogram resulting from this analysis is presented in Figure 4-9. From the dendogram we can differentiate four distinct groups of accessions. Group 1 was composed of accessions PI 497541, 497574, 604800, 604810, 604811, 604812, 604815, 604856, 604857, and CIAT 22271. Group 2 was formed by PI 604799, 604801, 604808, 604809, 604814, 604818, CIAT 18745, 20826, 22260, and 22265. Group 3 was composed by PI 476132, 604798, 604803, 604805, 604807, 604858, 604859, CIAT 22150, and 22256. Finally, Group 4 was composed by PI 604804, 604813, 604817, CIAT 22152, 22159, and 22234. Morphological characteristics of each of the four groups created by the cluster analysis are presented in Table 4-6. Based on these features we could characterize the four groups. Group 1 was composed by accessions with small leaves, flowers, pods, pegs, and seeds. Therefore, we could name this group as the small type group. In Figure 410, accession PI 497541, a member of this group is presented. Group 2 was formed by accessions with intermediate size features and could be called as the intermediate type gro up. Accession PI 604814 is presented as a representative of th is group in Figure 4-11. Group 3 was formed by accessions with la rge leaves, flowers, pegs, pods, and seeds. It was named as the large type gr oup due to their large f eatures. In Figure 4-12 accession PI 604798 is displayed as a representative of this type.

PAGE 92

78 Figure 4-9. Dendogram of 35 Arachis pintoi accessions based on morphological descriptors and the first ni ne principal components. Finally, Group 4 was composed of acce ssions which did not produce any pegs, pods, and seeds. Their leaves, flowers, and stem s characteristics were relatively similar to

PAGE 93

79 the ones displayed by the members of Group 3, which presented large sizes for these structures. So, based on the fact that this group did not produced seeds, this group could be called the vegetative type group. Accessi on 604817 is presented in Figure 4-13 as an exemplar of this group. Table 4-6. Morphological characteristics of Arachis pintoi accession groups obtained by the cluster analysis Characteristics Group 1 Group 2 Group 3 Group 4 Quantitative descriptors Flower/inflorescence 3.14 a 3.03 a 3.34 a 3.00 a Flower standard width 14.17 b 15.15 ab 16.23 a 16.07 a Flower standard length 10.80 b 11.32 ab 12.40 a 12.06 a Flower wing width 5.72 c 6.13 bc 6.81 a 6.62 ab Flower wing length 7.53 b 7.85 b 8.64 a 8.49 a Flower keel length 4.98 a 5.02 a 5.21 a 5.20 a Flower hypanthium length 65.00 b 64.00 b 79.00 a 78.00 a Flower hypanthium width 0.94 b 0.91 b 1.08 a 0.94 b Stem internode length 26.40 a 32.27 a 28.13 a 31.94 a Stem internode diameter 2.72 a 2.79 a 3.14 a 3.34 a Leaflet length 20.68 b 26.61 a 27.80 a 24.69 a Leaflet width 13.01 b 16.73 a 17.23 a 12.91 b Leaf Petiole length 16.13 c 27.44 a 23.89 ab 21.42 b Peg length 9.85 b 13.66 a 12.14 ab 0.00 c Peg width 0.97 a 0.89 a 0.96 a 0.00 b Pod weight 0.11 b 0.18 a 0.17 a 0.00 c Pod length 10.23 b 12.07 a 11.53 a 0.00 c Pod width 5.26 c 5.66 b 6.12 a 0.00 d Seed width 4.17 c 4.80 b 5.15 a 0.00 d Seed length 7.33 b 9.56 a 8.96 a 0.00 c Seed weight 0.08 b 0.14 a 0.13 a 0.00 c Qualitative descriptors Flower standard color Yellow Yellow Yellow Yellow Flower standard crescent Present Present Present Absent Flower hypanthium color Absent Present Absent Absent Flower hypanthium hairiness Present Present Present Present Stem color Absent Absent Absent Present Stem hairiness Present Absent Abundant Present Stem bristles Absent Absent Absent Absent Leaflet shape Obovate Obovat e Narrow ellipitc Obovate Leaflet hairiness sup. surface Absent Absent Absent Absent Leaflet hairiness margin Absent Absent Present Present Leaflet hairiness inf. surface Pr esent Absent Present Absent Leaflet bristles margin Pres ent Present Present Present Peg color Present Present Absent Peg hairiness Present Present Absent Pod beak Moderate Slight Moderate Pod reticulation Moderate Moderate Slight Seed color Yellow brownishOrange yellowishOrange yellowish Mean Mode Differences between means of different gr oups were tested by Student Newman-Keuls test. Means followed by same letter are not different at p =0.05.

PAGE 94

80 Figure 4-10. Group 1 represen tative accession (PI 497541). Figure 4-11. Group 2 represen tative accession (PI 604814).

PAGE 95

81 Figure 4-12. Group 3 represen tative accession (PI 604798). Figure 4-13. Group 4 represen tative accession (PI 604817). PI 604798

PAGE 96

82 Summary and Conclusions Thirty-five germplasm accessions of Arachis pintoi were morphologically characterized using a list of descriptor s prepared by IBPGR/ ICRISAT (1990 and 1992) Data from stems, leaves, flowers, pegs, pods and seeds were collected and comparisons among accessions were made, based on the mean, standard deviations, and range of the quantitative features and th e mode of the qualitative characteristics. Phenotypic correlations were conducted among descriptors, and Pearsons correla tion coefficient and significance test were also calculated. Simpson and Sha nnon-Weavers diversity index were computed for each descriptor to acce ss the genetic diversity among the accessions for individual descriptors. Principal component analysis was then executed to discriminate the accessions, and finally a cluster analysis was performed to group the germplasm in accordance with its morphological similarities. The germplasm presented great morphological variability with all the descriptors, except pollen size and shape, leaf bristles superior and inferior surface, showing polymorphism. From 595 correlations computed, 96 were statistically si gnificant. These 96 significant correlations could be divided in eight different groups. However, when only the biologically meaningful correlations (r 0.71) were evaluated, the number of significant correlations dropped to 29, and only four groups were observed. These meaningful correlations were found between leaf length (LL) and pod weight (Podwe); leaf length (LL) and pod width (PodW); leaf length (LL) and seed weight (Swe); leaf length (LL) and seed width (SW); and finally leaf length (LL) and seed length (SL). Thus, leaf length could be used as a selecti on criterion in programs where increased seed size is one of the objectives.

PAGE 97

83 Diversity values were variable among traits, but in gene ral all morphological features expressed high genetic diversity. A ccording to Simpsons index, leaf shape (0.83), seed color (0.82), flow er standard width (0.81), and seed length (0.75) were the descriptors with most diversit y. In contrast, the lowest valu es were related to flower standard color (0.18), flower standard crescent (0.24), and flower hypanthium hairiness (0.24). The total Simpsons index to all morphological descriptors was 0.58. Alternatively, Shannon-Weavers values were in general higher than Simpsons, with higher values displayed by leaf hairine ss inferior surface ( 1.00), flower hypanthium color (1.00), leaf hairine ss margins (0.80), and pod reticulation (0.97). Total genetic diversity was estimated as 0.71. The first five principal co mponents explained 67.7% of the total vari ation, with PC1 explaining 30.0% of the variation, PC2 15.2%, PC3 10.1%, PC4 6.6%, and PC5 5.8% of the total variation. The principal com ponent analysis was able to discriminate and separate the accessions in terms of three dimensions represented by the sexual reproduction, vegetative, and qualitative axes. The cluster analysis based on the first nine principal components differentiated four distinct groups of accessions. Group 1 was composed by accessions with small leaves, flowers, pods, pegs, and seeds. Group 2 was fo rmed by accessions with intermediate size features, Group 3 was formed by accessions with large leaves, flowers, pegs, pods, and seeds, and finally, Group 4 was composed of accessions which did not produce any pegs, pods, and seeds.

PAGE 98

84 CHAPTER 5 AGRONOMIC EVALUATION OF Arachis pintoi GERMPLASM Introduction Agronomic evaluation is a very important step in germplasm characterization programs. Although molecular and morphologi cal characteristics ar e relevant, plant breeders and ultimately producers have their attention focused on the potential of the plant to grow well in their environment a nd to produce forage, grain, or other economic products. Thus, agronomic evaluation will always be a key component in breeding programs. When evaluating a species outside its origin al environment, it is important to assess its adaptation to the new ecosystem. Emphasis must be given to how soils, climate, and rainfall conditions will impact the growth of this new species. Along with adaptation, several agronomic characteristics can be measured. The importance of each variable will be defined by the use of the plant and by the environment where it will be cultivated. In the case of a forage crop such as Arachis pintoi growing in a subtropical environment like Florida, forage yield, forage nutritive va lue, seed production, wi nter survival, and nematode resistance are just some of the characteristics that should be evaluated. Arachis pintoi is native of and well adapted to certain tropical environments. According to Pizarro and Rincn (1994), A. pintoi was evaluated by the International Tropical Pastures Evaluation Network (RIEPT ) in Brazil, Uruguay, Bolivia, Colombia, Peru, and Venezuela under savanna and humid tropical conditions. They concluded that it

PAGE 99

85 presented a wide range of adaptation and grows best under humid tropical conditions with total annual rainfall ranging from 2000 to 4000 mm. Fisher and Cruz (1994), howev er suggested that although A. pintoi grows well under high rainfall conditions it can tolerate periods of drought. They reported that A. pintoi was able to maintain a large proportion of its aerial parts at the expense of root tissue when exposed to 8 wk of water defic it. The same authors also revealed that A. pintoi did not tolerate long periods of floodi ng. They concluded that 3 wk of flooding severally restricted plant grow th with severe leaf chlorosi s and reduction of leaf area. Pizarro and Rincn (1994) reported that plants growing in a subtropical environment in Pelotas-Brazil were exposed to severe frosts (Tempe rature < 0C) that reduced growth, but did not kill the plant stands. They recovere d after the return of warm and rainy conditions. The literature has abundant A. pintoi forage yield data collected in the tropics. In evaluations performed in Bolivia, Brazil, Ec uador, Colombia, and Peru, accession CIAT 17434 produced between 0 and 2.7 Mg ha-1 of DM during the rainy season and 0.04 to 2.8 Mg ha-1 of DM during the dry season with a growing period of 12 wk (Pizarro and Rincn, 1994). In Costa Rica, Ar gel and Valerio (1993) report ed forage yields of 7, 12 and 7 Mg ha-1 of DM for accessions CIAT 17434, 18744, and 18748, with 20 mo of growth in Guapiles and San Isidro. In Puerto Rico, forage dry matter yields of 2.1 Mg ha-1 of DM were harvested 16 wk after planting from the accession CIAT 17434 (Argel, 1994). Several publications also report forage nutritive value and seed production data of Arachis pintoi in the tropics. Crude protei n values ranging from 120 to 220 g kg-1 and in

PAGE 100

86 vitro digestibility (IVOMD) ranging from 560 to 700 g kg-1 were reported from Argel Pizarro (1992); Rincn et al (1992); Carulla et al. (1991) ; and Pizarro and Carvalho (1996). Average seed yields of A. pintoi varied from 1 to 2 Mg ha-1 when harvested at 15 to 18 mo after planting (Ferguson et al., 1992). However, in Colombia, when planted in soils with high fertility, Ferguson (1 994) reported yields of 7.3 Mg ha-1. In Australia, Cook and Franklin (1988) reporte d seed yields of 1.4 Mg ha-1, 12 mo after sowing cv. Amarillo. Cook and Lock (1993), also working w ith the cv. Amarillo in Australia, stated that seed yields of 2.8 Mg ha-1 were obtained in a commercial seed crop. As stated before, the literature has severa l examples of research work done in the past where agronomic characterization of Arachis pintoi was the primary goal. However, most of this work wase done with a single germplasm accession, that latter was released as a commercial cultivars in several differe nt countries. These studies also have in common the fact that most of them where execut ed in tropical regions Therefore, there is a lack of information about other accessions of A. pintoi stored in germplasm banks, there are little or no existing data regarding the performance of the same germplasm in subtropical conditions. The goal of this research was to evalua te the agronomic adaptation, forage yield, seed yield, forage nutritive value, and nematode resistance of several A. pintoi germplasm accessions stored at the USDA-NPGS germplasm bank. Material and Methods The Germplasm Germplasm of A. pintoi stored at the USDA-NPGS germplasm bank located in Griffin, GA was transferred to the University of Florida on three different occasions. In

PAGE 101

87 May 2001 the first set composed of 25 acces sions was sent, the second set with 15 accessions, came in February 2002, and finally th e third set with 13 accessions was sent in May 2003, for a total of 53 accessions. All accessions were received as vegetative material because seeds were not available at th e time and also to ensure the genetic purity of each germplasm accession. A list with th e identification numb er and geographical information about the site of colle ction is presented in Appendix A. Field Evaluation Adaptation and forage dry matter yield In September 2001, 4 m (2 m x 2 m) field plots of the original 25 accessions were established with four rooted cuttings in a randomized complete block design with two replications at the Agronomy Department Forage Research Unit of the University of Florida near Gainesville. The experiment was located on a Pomona sand (siliceous, Hyperthermic Typic Psammaquents Entisol) so il type, a flatwood soil characteristic of north central Florida. The soil wa s prepared by disking and 190 kg ha-1 of 0-10-20 fertilizer was applied and incorporated into the soil. Arachis glabrata cultivars Florigraze and Arbrook were used as lo cal standards and were also planted following the same scheme. During the years 2001, 2002, and 2003 plots were maintained with periodical hand weeding and applications of the herbicide Cadre (Imazapic), in accordance with the recommended dosage for cultivated peanut (100 g ha-1). Annual applications of 800 kg ha-1 of 0-10-20 were made in may 2002 and 2003. Visual evaluations were ma de following the winter 2001/2002. In these evaluations number of plants per plot, plant survival and rate of spread were assessed. Plant survival was expressed as percentage of total plants alive and rate of spread in accordance with a

PAGE 102

88 scale of 1 to 5 (1 < 10% plot coverage, 2 = 20%, 3 = 30%, 4 = 40% and 5 > 50% plot coverage). These parameters were intended as indications of adaptation and winter survival. No dry matter harvest was made in 2002. In 2003 visual evaluations of plot covera ge and plant height were made. These evaluations were used as a criteria to determ ine which plots would be sampled to estimate forage dry matter yield (FDMY). FDMY wa s estimated on three dates during 2003, with an interval of 8 wk between harvests. These dates were 13 June (Harvest 1), 13 August (Harvest 2), and 21 October (Harvest 3). To estimate FDMY a sample of 0.25 m was harvested at soil level using battery-powered hand clippers as a cutti ng tool. Pintoi peanut and weeds were manually separated and the samples were placed in a 65C forced-air drier for 72 h. Dry weight was measured us ing a digital scale. Plot borders were uniformly cut at soil level with a rotary lawn mowing tractor after each harvest. FDMY was analyzed with Proc GLM of SAS (SAS institute, 1989) using the following model: Yij = + Bi + Aj + eij Where: = mean Bi = block effect Aj = accession effect eij = experimental error LSD at 5% significance level was us ed as the mean separation test. This same model was used for forage nutritive value and seed production.

PAGE 103

89 Forage nutritive value Forage samples harvested on 13Aug 2003 were ground in a Wiley Mill, to pass through a 1 mm screen. This ground tissue was analyzed for crude protein and in vitro organic matter digestibility (IVOMD) in the Forage Evaluation Support Laboratory of the Agronomy Department, Univ ersity of Florida. The N analysis started with sample digestion using a modification of the aluminum block digestion pro cedure (Gallaher et al., 1975). In this procedure samples of 0.25 g and catalyst of 1.5 g of 9:1 K2SO4:CuSO4 were used during a 4 h digestion at 375C using 6 ml of H2SO4 and 2 ml of H2O2. Digestate was then filtrated and N was determined by semiautomated colorimetry (Hambleton, 1977). IVOMD was performed by a modified tw o-stage technique (Moore and Mott, 1974). Both crude protein and IVOMD were expressed on an organic matter basis. Seed production In February 2003 and 2004 plots were samp led to assess seed production. Samples were taken in each plot using a soil core sampler of 12.5 cm diameter and 24 cm depth (0.01227m). The soil was screened on a sieve with a 0.6 x 0.6 cm mesh to remove the pods from the soil. The pods were then dried at room temperature for 6 wk and weighed. Calculations were made to extrapolate the values in terms of production per hectare. Nematode Response Evaluation Stems of the A. pintoi germplasm accessions were cut and placed in vermiculate trays under an automated mist system (10 s ec every 30 min) for rooting. After 45 d under the mist systems, the rooted cuttings were transferred to 150 cm Conetainers filled with methyl-bromide-fumigated fine sand tops oil. After transferring, plants were allowed to establish for 2 wk and then used in this experiment. On 12November 2003 plants were

PAGE 104

90 inoculated with either M. arenaria race 1, M. javanica or M. incognita race 1. During the experimental period, plants were watered da ily and fertilized w ith 20-20-20 fertilizer every 2 wk. The green house temperature ra nged from 15 to 25C during the 12 wk that the trial lasted. Tomato plants were used to propagate the nematodes and then the Hussey and Barker (1973) method was used to extract e ggs and juveniles. In this method, roots are cleaned, split in small pieces and washed in a 0.525 % sodium hypochlorite (NaOCl) solution for 2 min. The roots are then stir red strongly and passed through a 200-mesh sieve (openings 0.149-0.074 mm). The eggs a nd juveniles are collected on a 500-mesh (openings 0.028 mm) sieve placed under the 200 -mesh one. Eggs are subsequently rinsed with H2O, pored to a beaker and water is added to bring the volume to 1000 ml. A sample is taken, placed on a slide, and the number of eggs per ml is estimated by counting under the microscope. Prior to injecting the egg suspension into the soils, the solution was diluted to 300 eggs per ml. This procedure was followed for each one of the three nematodes used in this experiment. To each container 5 ml of egg suspension was applied, which brings the total eggs per container to 1500 or 10 eggs cm3 of soil. Application was delivered with a veterinarian surgical syringe, and during th e whole process the eggs were kept in continual suspension by a magnetic stirrer. The experimental design was a randomized co mplete block, with four replications for M. arenaria and three replications for M. javanica and M. incognita A single plant constituted each replication. Arachis hypogaea cv. Florunner was used as a susceptible control to verify inoculum viability.

PAGE 105

91 Twelve weeks after inoculation plants we re removed from the containers and soil was carefully washed from the roots with tap water. Plants were then placed in a bucket with roots immersed in a 0.25% Phloxine B so lution to stain the egg masses. Roots were rated for gall index (GI), gall size (GS), and percent galled ar ea (GA) in a 1-9 scale and after that a damage index (DI) was calculated based on the same parameters (Sharma et al., 1999). DI was calculated by the following equation: DI = (GI+GS+GA)/3. GI, GS, GA and DI scales are presented in Table 5-1. Number of egg masses (EI) was rated with a 1-9 scale similar to gall index, where 1 represented no egg masses and 9 more than 100 egg masses. Accessions with EI = 1 were considered highly resistant to nematode reproduction and with EI = 9 were highly susceptible. Intermediate values followed the DI scale. Mean and standard error of the mean we re computed for EI and DI and these variables were used to cla ssify the accessions in relati on to its nematode reaction. Table 5-1. Gall Index, gall size, percent ga lled area and damage index values Scale value Gall index (GI) Gall size (GS) Percent galled area (GA) Damage Index (DI) 1 No galls No galls No galls Highly resistant 2 1-5 galls Resistant 3 6-10 galls 10% increase 110% root galled Resistant 4 11-20 galls Moderate resistant 5 21-30 galls 30% increase 11-30% ro ot galled Moderate resistant 6 31-50 galls Susceptible 7 51-70 galls 31-50% increase 31-50% root galled Susceptible 8 71-100 galls Highly susceptible 9 >100 galls >50% increase >50% root galled Highly susceptible Accessions with DI and/or EI =1, were consid ered highly resistant; DI and/or EI > 1 and 3, were classified as resist ant; DI and/or EI > 3 and 5, were classified as moderately resistant; DI and/or EI > 5 and 7, were classified as suscep tible; and finally accessions

PAGE 106

92 with DI and/or EI > 7 and 9, were classified as highl y susceptible. If there was discrepancy between DI a nd EI values, the higher value was applied. Results and Discussion Adaptation and forage dry matter yield The evaluations performed in 2002, eight, el even, and twelve months after planting, revealed that although A. pintoi presented great reduction of green tissue during the winter, most plants did not die, and in fact they fully recovered when the temperature warmed and soil moisture increased (T able 5-2). During the 2001/2002 winter, temperatures reached 0C or less on 24 different occasions (Appendix D). The average plant survival (PS%) in Evaluation 1 was 79%, which shows that A. pintoi can tolerate winters where freezing and frosting are normal occurrences. However, when rate of spread (RS) was analyzed, mo st of the accessions were covering less than 50% of the plot area. Cultivars Florigra zeand Arbrook presented 100% of plant survival, but they also covered less than 50% of the plot area. This same trend was displayed in Evaluati ons 2 and 3, where average plant survival recorded was superior to 80% and plot coverage inferior to 50%. Actuality, this is one of the biggest problems that A. pintoi and also A. glabrata display, the fact that both species required a long period to establish themselv es and cover the area where they were planted. Even with general low plot cover, we can observe differences among accessions in relation to plant survival a nd plot coverage, which is an indication of variable in adaptation to the north Florida e nvironment. Accessions PI 497574, 604798, 604800, 604803, 604807, 604814, 604817, and 604857 are the ones that appear better adapted, with values of plant survival and rate of spread superior to the others.

PAGE 107

93 Table 5-2. Winter survival evaluations of Arachis pintoi at the forage research unit in Gainesville-FL Accession Evaluation 1 (04/25/2002) Evaluation 2 (07/03/2002) Evaluation 3 (08/23/2002) PI number #PL PS (%) RS #PL PS (%) RS #PL PS (%) RS 476132 3.0 75 1.0 3.0 75 2.4 4.0 100 2.0 497541 2.0 50 0.5 2.0 50 1.0 4.0 100 1.0 497574 4.0 100 2.5 3.0 75 1.5 4.0 100 2.5 604798 3.5 87 2.0 3.0 75 3.0 4.0 100 3.0 604799 3.0 75 3.0 3.0 75 2.5 4.0 100 2.0 604800 3.5 87 2.0 3.0 75 2.0 3.5 87 2.5 604801 1.0 25 0.5 1.0 25 1.0 3.5 87 2.0 604803 3.5 87 2.0 3.5 87 2.8 4.0 100 3.0 604804 4.0 100 1.5 3.0 75 1.5 3.0 75 1.5 604805 3.0 75 2.0 3.0 75 2.0 4.0 100 1.5 604807 3.5 87 2.5 3.5 87 2.0 4.0 100 3.0 604808 4.0 100 4.0 4.0 100 4.5 4.0 100 2.0 604809 3.0 75 2.0 3.0 75 3.0 3.5 87 2.5 604810 3.0 75 2.5 2.0 50 2.0 3.5 87 1.5 604811 4.0 100 2.5 4.0 100 2.0 4.0 100 1.5 604812 2.0 50 1.0 1.5 37 1.5 4.0 100 2.0 604813 4.0 100 1.5 4.0 100 1.5 4.0 100 1.5 604814 1.5 37 0.5 3.0 75 2.4 3.5 87 4.0 604815 2.0 50 2.0 0.5 12 1.0 4.0 100 2.0 604817 4.0 100 2.5 3.8 94 3.5 4.0 100 3.5 604856 2.5 62 1.0 3.0 75 1.5 4.0 100 2.0 604857 3.5 87 2.0 3.8 94 3.2 4.0 100 3.0 604858 3.5 87 3.0 3.3 81 3.5 4.0 100 2.5 604859 1.0 25 1.0 3.8 94 3.2 1.5 37 1.5 Arbrook 4.0 100 4.0 3.3 81 3.5 4.0 100 4.5 Florigraze 4.0 100 2.5 3.0 75 3.0 4.0 100 2.5 #PL = number of plants per plot; PS (%) = plant survival; RS= Rate of spread Pizarro and Rincn (199 4) affirmed that A. pintoi is slow establishment is a limitation to adoption as producers want to see quick results. They suggested that when seeds are used as the propagation material the establishment time is shortened when compared to vegetative propagation. French et al. (1994) stated th at although adequate establishment has been achieved in Florida with rhizoma peanut ( A. glabrata ) in a single year, a typical planting demands 2 to 3 yr fo r full development. Cook et al. (1994) agreed

PAGE 108

94 with the authors previously cited. They believe that the major constraint to farmer acceptance of these species is the comp lexity and cost of establishment. Slow establishment is particularly signifi cant where weed pressure is high. This fact will require extra resources to assure an ade quate initial coverage that will extend the period of utilization of the pa sture or hay field. Research is necessary to determine the best planting method, using the different fa ctors that could result in a fast and inexpensive pasture establishment. In 2003, 21 mo after establishment, samples were harvested on three different occasions to determine FDMY. Before each harvest, evaluations were made to estimate percentage of plot covered (% cove r) and plant height (Table 5-3). At Harvest 1 (13 June 2003), average pl ot coverage was estimated as 62% among the A. pinto i germplasm. Accessions PI 497574, 604808, 604817, and 604858 showed the best plot coverage with values a bove 90%. In contrast, accessions 604801, 604804, 604812, 604814, and 604859 had plot coverage less than 25%. Plot coverage is reflected in the FDMY. In terms of height, the averag e was 6 cm with little variation. This number is very contrasts markedly with Florigraze and Arbrook that presented values of 11 and 20, respectively. It seems clear th at the two species have a very different growth habit, and while A. pintoi exhibits a more prostrate habit, A. glabrata is more erect. At Harvest 2 (13 Aug. 2003), average plot coverage was 66%, with a general increase in covered area by all accessions. The average height was similar to the previous evaluation. Harvest 3 (21 Oct. 2003) presented similar average plot coverage and plant height as Harvest 2. A general increase in covered area was also observed in this evaluation.

PAGE 109

95 Table 5-3. Plot coverage and plant height be fore forage dry matter yield evaluations of Arachis pintoi at the forage research un it in Gainesville-FL in 2003 Accession Harvest 1 Harvest 2 Harvest 3 PI number % cover Height (cm) % coverHeight (cm) % cover Height (cm) 476132 47 5.5 30 7 40 8 497541 57 3 42 6.5 47 3 497574 95 7.5 95 8 100 6.5 604798 62 6 70 8.5 75 8.5 604799 87 5 75 5.5 65 5 604800 85 7 80 8 80 7 604801 27 4 42 5 55 5.5 604803 80 14 85 15.5 77 10.5 604804 17 2.5 15 3 10 1.5 604805 62 5 65 5.5 50 6 604807 57 6.5 40 5 40 4.5 604808 92 5.5 100 6 100 4.5 604809 50 6 60 8.5 65 6 604810 62 4 75 5 75 5 604811 67 5.5 85 7 85 7 604812 22 2.5 37 5.5 47 4.5 604813 67 5 80 6.5 90 3 604814 17 4 20 3.5 25 3 604815 82 8.5 100 9 95 8 604817 95 8 100 10.5 100 7 604856 60 4 72 5 80 5 604857 80 9 75 10 82 8.5 604858 97 7.5 95 7.5 100 6.5 604859 22 6.5 37 9 55 9 Florigraze 82 11.5 100 14 95 7.5 Arbrook 100 20.5 100 22.5 100 12 Harvest 1: 13 Sep. 2003; Harvest 2: 13 Aug. 2003; Harvest 3: 21 Oct. 2003 A careful analysis of these evaluations demonstrates again that there are large differences among accessions in relation to environmental adaptation. Accessions PI 604801, 604804, 604812, 604814, and 604859 presented low va lues of plot coverage in each of the three harvest. This is very signi ficant, since the Harvest 3 was completed at 25 months after plot establishment. Another si gnificant fact that can be extracted from these evaluations is that in Harvest 3 the values of plot coverage and plant height

PAGE 110

96 displayed a significant drop in re lation to the values of the same variables for Harvest 1 and 2. Forage dry matter yield (FDMY) was a ssessed three times during summer 2003 with an 8-wk interval between harvests. Anal ysis of variance table for Harvest 1, 2, and 3 is presented below (Table 5-4). Table 5-4. Analysis of variance table of fo rage dry matter yield (FDMY) evaluations of Arachis pintoi at the forage research un it in Gainesville-FL in 2003 Harvest 1 Harvest 2 Harvest 3 Source df MS Pr>F MS Pr>F MS Pr>F Accession 25 1.19 0.00612.28 0.00331.05 0.0060 Replication 1 0.82 0.17580.13 0.68240.15 0.5253 Error 25 0.42 0.74 0.37 Total 51 Harvest 1: 13 Sep. 2003; Harvest 2: 13 Aug. 2003; Harvest 3: 21 Oct. 2003 Significant differences (P < 0.01) were present among accessions in all three FDMY harvests, indicating that accessions show variability in their environmental adaptation, an attribute that will consequen tly impact their forage production potential. Replication effects we re not significant. Forage dry matter yield at each harvest date in 2003 is presented in the Table 5-5. A large amount of variability was present among these A. pintoi accessions with respect to their FDMY. The average FDMY at Harvest 1 was 1.34 Mg ha-1, ranging from 0.06 to 2.86 Mg ha-1. More than half of the accessions pr esented superior performance producing above the average, but accessions PI 604801, 604804, 604812, and 604814 had very low yields. These low yields are a reflection of poor plot coverage for these accessions as they had less than 30% plot coverage at Harvest 1. In comparison, accessions PI 604817, 604815, 604803, 604808, 604810, 604857, and 497574 were the most productive, showing no differences in FDMY to the rhizoma peanut cultivars.

PAGE 111

97 Table 5-5. Forage dry matter yield of Arachis pintoi germplasm at the Forage Research Unit near Gainesville, FL in 2003 Accession Harvest 1 Harvest 2 Harvest 3 ------------------------------------Mg ha-1-----------------------------------476132 0.79 0.75 0.98 497541 1.18 0.94 0.44 497574 1.85 3.00 2.50 604798 1.51 1.52 1.38 604799 1.35 1.74 1.60 604800 1.39 1.98 1.33 604801 0.34 1.10 0.93 604803 1.99 1.93 1.74 604804 0.06 0 0 604805 1.58 1.48 0.88 604807 1.38 0.78 0.56 604808 1.98 2.62 1.99 604809 1.47 0.38 1.14 604810 1.96 2.54 2.14 604811 1.06 1.45 1.63 604812 0.19 0.6 0.82 604813 0.98 1.72 1.01 604814 0.10 0 0 604815 2.32 2.12 2.30 604817 2.86 3.45 2.79 604856 1.12 1.90 1.56 604857 1.85 2.26 2.24 604858 1.59 2.25 1.87 604859 1.38 2.29 1.63 Arbrook 3.16 4.75 1.76 Florigraze 2.14 1.89 0.95 LSD 0.79 0.75 0.98 Harvest 1: 13 Sep. 2003; Harvest 2: 13 Aug. 2003; Harvest 3: 21 Oct. 2003 At Harvest 2 the average FDMY of A. pintoi germplasm showed a 20% increase reaching 1.62 Mg ha-1. Accessions PI 604804 and 604814 showed no production, which reflects low plot coverage and poor adaptati on to the north Florid a environment. Here once more, half of the accessions presented superior performance with FMY above the average. Accessions PI 604817 and 497574 we re the most productive, showing no difference to Arbrook, which produced 4.75 Mg ha-1. Florigraze however, was intermediate in FDMY a nd similar to a group of A. pintoi accessions. This is somewhat

PAGE 112

98 surprising, since it is regarded by many as the most adapted and productive of the rhizoma peanut cultivars. Harvest 3 presented mean FDMY of 1.39 Mg ha-1, which corresponded to a drop of about 20% in relation to th e preceding harvest. Arbrook and Florigraze exhibited a significant reduction in FDMY of ca. 50% compared to previous period. A. pintoi germplasm also exhibited this reduction, howev er with less decrease (Figure 5.1). This reduction in FDMY could represent a change in nutrient partitioni ng as a process to overcome restriction in growth due to the shortening days in autumn. Accessions PI 604817, 497574, 604815, and 604857 had the highest FDMY and were significantly higher than Florigraze and Arbrook rhizoma peanut. 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 JuneAugOct Year 2003kg.ha-1 A.pintoi Arbrook Florigraze Figure 5-1. Forage dry ma tter yield (FDMY) of Arachis pintoi and A. glabrata cultivars at the forage research unit in Gainesville-FL in 2003.

PAGE 113

99 Adding the FDMY of the three individual harvests allows the analysis of the annual forage production of the germplasm. The an alysis of variance table for total FDMY shows that there were significant (P < 0.01) differences among accessions. Annual average FDMY for A. pintoi accessions was 4.36 Mg ha-1 (Table 5-7). Arbrook produced 9.67 Mg ha-1 and was the most productive. Accessions PI 604817, 497574, 604815, 604810, 604808, and 604857 were the most productive among the A. pintoi germplasm and were not different from Arbrook. Table 5-6. Analysis of variance table of th e annual forage dry matter yield (FDMY) of Arachis pintoi at the forage research un it in Gainesville-FL in 2003 Total FDMY Source df MS Pr>F Accession 25 11.48 0.0014 Replication 1 2.72 0.3726 Error 25 3.30 Total 51 Total FDMY = Harvest 1 + Harvest 2 + Harvest 3 Several A. pintoi accessions yielded more than 5.00 Mg ha-1 which can be considered a good legume production in north cen tral Florida. In fact this number is similar to yields displayed by other summer le gumes used in Florida, and even could be compared to A. pintoi or others legume species yields obtained in tropical regions. Kretschmer et al. (1988) repor ted 3-yr average FDMY for Desmodium heterocarpon cv. Florida, Macroptilium atropurpureum cv. Siratro, and Arachis Kretschmeri Pantanal of the or der of 2.97, 2.39, and 1.16 Mg ha-1 respectively. Annual yields of Stylosanthes guianensis cv. Savanna, Alysicarpus vaginalis and Indigofera hirsuta cv. Flamingo ranged from 5 to7 Mg ha-1, in plantings at Br ooksville, FL during years 1989 and 1990 (Williams et al., 1993). Misl evy and Martin (2001) reported similar annual FDMY (5.70 Mg ha-1) for Aeschynomene evenia at Ona, FL.

PAGE 114

100 Table 5-7. Total forage dry matter yield of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 Accession Total FDMY -------------------Mg ha-1------------------476132 2.52 497541 2.56 497574 7.36 604798 4.42 604799 4.69 604800 4.71 604801 2.37 604803 5.67 604804 0.06 604805 3.93 604807 2.73 604808 6.59 604809 2.99 604810 6.63 604811 4.14 604812 1.61 604813 3.71 604814 0.10 604815 6.74 604817 9.10 604856 4.58 604857 6.35 604858 5.71 604859 5.30 Arbrook 9.67 Florigraze 4.98 LSD 3.74 Total FDMY = Harvest 1 + Harvest 2 + Harvest 3 In tropical areas, comparable yiel ds were presented by Souza et al (1992) who evaluated Stylosanthes and Centrosema germplasm in central Brazil. Average FDMY of 1.50 Mg ha-1 were harvested in a 12-wk growi ng period during the summers of 1989 and 1990. Pizarro et al. (1996c) reported annual FDMY of Calopogonium mucunoides of the order of 1.72 Mg ha-1 in research conducted at severa l different locations in South America.

PAGE 115

101 Data from diverse agronomic characterization of A. pintoi are also similar to those presented in this work. Argel and Pizarro ( 1992) stated that cv. Amarillo produced 2.30 Mg ha-1 in Planaltina, Brazil, during the rainy season. In Australia, Cook et al. (1990) stated that Amarillo yielded 6.50 and 7.30 Mg ha-1 yr-1 in unirrigated and irrigated conditions respectively. Also in Australia, Cook et al. (1994) pres ented average annual yields of 5.80 Mg ha-1 for Amarillo in two years of evaluation. Although the yields obtained in this research were similar to those of other summer legumes in Florida, these data are relative to just 1 yr of evaluation. Supplementary information must be obtained if broader applic ation of the data are to be made, because persistence is one of the majo r issues with forage legumes. Forage nutritive value The analysis of variance table for CP a nd IVOMD demonstrated that significant variation (P < 0.01) was observed among accessi ons for both variables (Table 5-8). Table 5-8. Analysis of variance table of cr ude protein (CP) and in vitro organic matter digestion (IVOMD) of 8-wk regrowth of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 CP IVOMD Source df MS Pr>F MS Pr>F Accession 25 2381.7 0.0002 301.9 0.0045 Replication 1 470.2 0.0001 392.4 0.0568 Error 25 495.2 97.5 Total 51 CP= Crude Protein and IVOMD= In vitro organic matter digestibility Crude protein and IVOMD values of A. pintoi accessions from this research confirm those reported in the literature, and suppo rt the information that this species produces high quality forage. The results endorse the claim of some that A. pintoi could be considered a tropical alfalfa (Table 5-9).

PAGE 116

102 Average crude protein concentration across A. pintoi accessions was 180 g kg-1 ranging from 139 (PI 604800) to 225 g kg-1 (PI 604858). Florigraze CP was 172 g kg-1 while that of Arbrook was 153 g kg-1. When CP of the A. pintoi germplasm was compared to the rhizoma peanut cultivars, the majority of the accessions demonstrated higher values than either rhizoma peanut cultivars. Average IVOMD of the A. pintoi germplasm was 670 g kg-1, and ranged from 600 to 730 g kg-1. Accession PI 604812 had the lowest value, while PI 604801 was the highest. Florigraze presented the highest IVOMD value (740 g kg-1), however several A. pintoi accessions were not different (P 0.05) from it. CP and IVOMD displayed in this res earch could be considered high and comparable to those reported for A. pintoi in other investigations Average values of CP in the leaves of accession CIAT 17434 varied from 122 to 218 g kg-1 in Colombia, during dry and rainy seasons, respectively. In stem s the same accession presented values of 93 and 135 g kg-1 during the same seasons (Argel and Pizarro, 1992). In Brazil, Purcino and Viana (1994) reported total-herbag e CP values of 183, 157 and 161 g kg-1, for accessions BRA-013251 (PI 338447), BRA-015253 (PI 604859) and BRA-015598 (PI 604815), respectively. Rincon et al. (1992) reported CP values for the whole plant of 130 and 180 g kg-1, during the dry and rainy season, respec tively. Average values of IVOMD during the same periods were 67 and 62 g kg-1. Average IVOMD of 168 d growth forage from three A. pintoi accessions in Brazil was 610 g kg-1 (Pizarro and Carvalho, 1992). In Australia, Amarillo IVDMD was 730 g kg-1 in 2 yr of evaluation (Cook et al., 1994).

PAGE 117

103 Table 5-9. Crude protein (CP) and in vitro organic matter di gestibility (IVOMD) of 8-wk regrowth of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 Accession CP (g kg-1) IVOMD (g kg-1) 476132 178 690 497541 147 630 497574 184 640 604798 170 680 604799 189 700 604800 139 700 604801 188 730 604803 195 670 604805 151 690 604807 156 680 604808 189 620 604809 148 620 604810 187 610 604811 189 690 604812 183 600 604813 166 660 604815 194 690 604817 212 670 604856 180 710 604857 222 720 604858 225 720 604859 173 710 Arbrook 153 690 Florigraze 172 740 + LSD CP (0.05) = 30.3 g kg-1, LSD IVOMD (0.05) =65 g kg-1 Seed production Another important factor for which geneti c variability should be investigated is seed production. In February 2003 and 2004, re spectively, 18 and 30 mo after planting, samples were collected in each plot to assess this trait. The analysis of variance table for seed pr oduction (Table 5-10) showed significant variation (P < 0.01) among accession s for both the 2003 and 2004 years. In 2003, the average seed production was 0.32 Mg ha-1, ranging from zero to 2 Mg ha-1 (Table 5-11). Only 60% of the A. pintoi germplasm yielded seeds, however, if we

PAGE 118

104 consider only those accessions with production above 0.1 Mg ha-1 this number falls below 50%. Accession PI 604857 was the most productive yielding 2 Mg ha-1. Table 5-10. Analysis of varian ce table of seed production of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 and 2004 Seed Production 2003 Seed Production 2004 Source df MS Pr>F MS Pr>F Accession 25 707.75 0.0001 929.12 0.0062 Replication 1 706.52 0.0023 850.83 0.1504 Error 25 47.46 40.120 Total 51 pod weight In 2004, with the exception of three acce ssions, all accession produced seeds. The average seed production was higher than the previous year reaching 0.43 Mg ha-1, ranging from zero to 1.58 Mg ha-1 (Table 5-11). In general, yields of most accessions increased when compared to the 2003 evaluation. Exceptions were PI 604803, 604809, 604815, and 604857. Curiously, these accessions we re among the most productive in 2003. Although 87% of the accessions produced some seeds, six of then yielded less than 0.10 Mg ha-1, amount that would not be useful fo r commercial harvest. PI 604799 had the highest yield (1.58 Mg.ha-1) in 2004 although it was intermedia te in seed yield in 2003. Seed production obtained in this rese arch for some of the accessions was comparable to that obtained by other authors working with A. pintoi in tropical conditions. For legumes this trait has particul ar importance because it can be related to persistence. A species with high seed production capacity may have an advantage over one lacking this characterist ic, because in theory, pastur e establishment by seeds is simpler and cheaper than when vegetative mate rial is used. Also, plants established by seeds will grow much faster, and conseque ntly will have higher ground covered area. A.

PAGE 119

105 pintoi has also a notable characteristic that diffe rentiates it from other legumes in relation to seed production, in that the seeds are located below ground. Table 5-11. Seed production of Arachis pintoi at the Forage Research Unit near Gainesville, FL in 2003 and 2004 Accession Seed Production 2003 Seed Production 2004 ---------------------------Mg ha-1---------------------------476132 0.01 0.07 497541 0.00 0.03 497574 0.00 0.36 604798 0.00 0.69 604799 0.75 1.58 604800 0.03 0.16 604801 0.00 1.10 604803 0.76 0.20 604804 0.00 0.00 604805 0.59 1.21 604807 0.29 0.36 604808 0.30 1.01 604809 0.91 0.43 604810 0.07 0.06 604811 0.00 0.02 604812 1.30 1.32 604813 0.00 0.00 604814 0.00 0.10 604815 0.47 0.17 604817 0.00 0.00 604856 0.08 0.05 604857 2.00 0.62 604858 0.06 0.27 604859 0.00 0.63 Arbrook 0.00 0.00 Florigraze 0.00 0.00 LSD 0.48 0.90 pod weight As previously stated, yields in this resear ch were comparable to others found in the literature. Cook and Franklin (1988) reported mechanized harvest yield for Amarillo in Australia of 1.40 Mg ha-1 12 mo after planting. In other work, seed yield of Amarillo in Australia reached an averag e production of 2.80 Mg ha-1 (Cook and Loch, 1993).

PAGE 120

106 In a Costa Rica location with 4260 mm annua l rainfall, Diulgheroff et al. (1990) obtained seed yields of 1.95 Mg ha-1, 12 mo after sowing the cv. Man Forrajero perenne. Also in Costa Rica, Argel and Vale rio (1993) reported av erage yields of 0.59, 0.54, 0.50, and 0.52 Mg ha-1 over the accessions CIAT 17434 18744, and 18748, at 8, 12, 16, and 20 mo after sowing, respectively. Rincon et al. (1992) stated that seed yields of 2 Mg ha-1 were obtained for cv. Man Forraj ero perenne in mixed pasture with Brachiaria under grazing. This supports the idea of excellent persistence potential for A. pintoi due to seedling recruitment. Nematode response evaluation Reaction to Meloidogyne arenaria M. javanica and M. incognita was established in accordance with the methodology proposed by Sharma et al. (1999). The analysis of variance of A. pintoi reaction to M. arenaria showed significant (P < 0.01) differences among the accessions (Table 5-12). M. arenaria reaction of A. pintoi germplasm is presented in Table 5-13. Large genetic variability was observed among the acces sions with respect to this characteristic. Among the 44 accessions evaluated, 12 were clas sified as highly resistant, 14 were classified as resistant, 15 were considered moderately resistant, 2 were considered susceptible, and one was considered highl y susceptible. Overall 93 % of the accessions presented some level of resistance and onl y 7% were classified as susceptible. The A. pintoi accessions also demonstrated sign ificant variatio n (P < 0.01) in response to infestation with M. javanica (Table 5-14). Although, si gnificant variation was presented in M. javanica reaction, all 39 accessions evaluate d were classified as highly resistant or resistant (Table 5-15).

PAGE 121

107 Table 5-12. Analysis of variance table of Arachis pintoi germplasm reaction to M. arenaria DI EM Source df MS Pr>F MS Pr>F Accession 44 8.61 0.0001 8.44 0.0001 Replication 3 0.99 0.5515 0.44 0.7428 Error 78 1.09 1.05 Total 125 Damage index; Egg mass In the case of M. incognita reaction significant differe nces were observed among accessions only for DI (Table 5-16). All except two accessions showed no galling or egg mass production (Table 5-17). Other repor ts have shown that in general Arachis pintoi have near immunity to M. incognita In fact, A. hypogaea is used as a non-host differential for M. incognita in the standard test to char acterize populations of root-knot nematodes in to major species and races. It is however generally susceptible to M. arenaria Nematode resistance is a valuable attribute for any species that wi ll be incorporated into agriculture systems. It is more important with perennial plants that will have longterm exposure to soil borne problems. For a forage crop, nematode susceptibility can impact the ability to persist over a long period in the pasture. In the case of A. pintoi, which is known as multiple use legume, this characteristic coul d improve its utilization as ground cover and in crop rotations with cult ures that are susceptible to root-knot nematodes. This is the case of the common peanut planted in the southeastern USA, which requires a crop rotation with bahiagra ss (Paspalum notatum). The introduction of A. pintoi in bahiagrass pasture coul d improve nematode control and additionally improve the nutritive value of the pasture.

PAGE 122

108 Table 5-13. Reaction of Arachis pintoi germplasm to M. arenaria race 1 Accession Egg Mass (EM) Damage Index (DI) CIAT / PI mean SE mean SE Highly resistant 20826 1 0 1 0 22175 1 0 1 0 22232 1 0 1 0 22233 1 0 1 0 22238 1 0 1 0 22241 1 0 1 0 22259 1 0 1 0 22268 1 0 1 0 604799 1 0 1 0 604813 1 0 1 0 604815 1 0 1 0.3 604858 1 0 1 0 Resistant 22151 1 0 1.6 0.5 22152 1 0 1.6 0.5 22159 1 0 2.7 0 22234 1 0 2.7 0 22289 1 0 2.3 0.4 22324 1 0 2.7 0 22339 1 0 2.1 0.5 476132 1 0 1.6 0.5 497574 1 0 1.8 0.6 604798 1 0 2.1 0.5 604810 1 0 2.7 0 604814 1 0 2.7 0 604817 1 0 2.1 0.5 604856 1 0 2.7 0 Moderately resistant 22150 2.3 0.7 4.6 0.9 22159 2 0.5 4.2 0.7 22174 2 0.6 2.5 0.1 22236 2 0 2.3 0.4 22256 2.3 0.2 2.7 0 22260 2 0 3.5 0.6 22271 3.3 1.4 4.8 1.8 22325 1.8 0.4 1.8 0.5 604800 2 0.6 3.7 0.5 604803 2 0 3.7 0.6 604805 1.3 0.3 2.1 0.5 604809 1.5 0.3 1 0 604811 1 0 4.3 0.9 604812 2.3 0.7 4.3 1 604859 1 0 3.8 0.5 Susceptible 22265 4 0 7.2 1.3 22154 6 1.9 5.2 1.1 Highly Susceptible 604808 9 0 7 0.5 Florruner 8.7 0.3 7.3 0.8

PAGE 123

109 Table 5-14. Analysis of variance table of Arachis pintoi germplasm reaction to M. javanica DI EM Source df MS Pr>F MS Pr>F Accession 39 0.203 0.0054 0.011 0.0002 Replication 2 0.241 0.0905 0.033 0.3986 Error 55 0.096 0.012 Total 96 Damage index; Egg mass In the case of A. pintoi nematode resistance is rema rkably important to permit a wide use of the species as forage crop or even as a cover crop. Also it is important due to the fact that the species could be consider ed a useful source of genes for its relative A. hypogaea which is worldwide cultivated. Since direct crossing among the two species is not possible, some authors include A. pintoi in the tertiary gene pool of A. hypogaea However, with the recent progress of molecu lar biology tools, direct transfer could be achieved even for non-related species of the ge nus, which makes this source of resistance potentially important. Even though, knowledge about sources of nematode resistance is extremely important to the general use of the species and for its us e in breeding programs of A. hypogaea little was know about A. pintoi germplasm accessions response to root-knot nematodes. Information available is usually re stricted to one or a few accessions. Sharma et al. (1999) studied M. javanica race 3 reaction of 161 accessions of wild Arachis species. They reported that of the nine accessions of A. pintoi evaluated, eight were considered susceptible or highly susceptib le, but a single accession was classified as moderately resistant. By contrast, all A. pintoi accessions were highly resistant to the M. javanica population used in this res earch (Not classified as a race).

PAGE 124

110 Table 5-15. Reaction of Arachis pintoi germplasm to M. javanica Accession Egg Mass (EM) Damage Index (DI) PI / CIAT mean SE mean SE Highly resistant 20826 1.0 0.0 1.0 0.0 22150 1.0 0.0 1.0 0.0 22151 1.0 0.0 1.0 0.0 22152 1.0 0.0 1.0 0.0 22154 1.0 0.0 1.0 0.0 22159 1.0 0.0 1.0 0.0 22174 1.0 0.0 1.0 0.0 22175 1.0 0.0 1.0 0.0 22232 1.0 0.0 1.0 0.0 22233 1.0 0.0 1.0 0.0 22234 1.0 0.0 1.0 0.0 22236 1.0 0.0 1.0 0.0 22238 1.0 0.0 1.0 0.0 22241 1.0 0.0 1.0 0.0 22256 1.0 0.0 1.0 0.0 22259 1.0 0.0 1.0 0.0 22265 1.0 0.0 1.0 0.0 22268 1.0 0.0 1.0 0.0 22271 1.0 0.0 1.0 0.0 22289 1.0 0.0 1.0 0.0 22324 1.0 0.0 1.0 0.0 22325 1.0 0.0 1.0 0.0 476132 1.0 0.0 1.0 0.0 497574 1.0 0.0 1.0 0.0 604798 1.0 0.0 1.0 0.0 604799 1.0 0.0 1.0 0.0 604800 1.0 0.0 1.0 0.0 604803 1.0 0.0 1.0 0.0 604805 1.0 0.0 0.7 0.3 604809 1.0 0.0 1.0 0.0 604810 1.0 0.0 1.0 0.0 604812 1.0 0.0 1.0 0.0 604813 1.0 0.0 1.0 0.0 604815 1.0 0.0 1.0 0.0 604818 1.0 0.0 1.0 0.0 604858 1.0 0.0 1.0 0.0 604859 1.0 0.0 1.0 0.0 Resistant 22339 1.7 0.3 2.1 0.6 604808 1.0 0.0 1.6 0.6 Florunner 1.0 0.0 2.1 0.6 Queneherve et al. (2002) examined A. pintoi reaction to Radopholus similis Pratylenchus coffeae Hoplolaimus seinhorsti Meloidogyne incognita and M. mayaguensis Forty-five days after inoculation R. similis H. seinhorsti and P. coffeae multiplied in the roots. A. pintoi did not allow the multiplication of M. incognita and M.

PAGE 125

111 mayaguensis indicating the inability of A. pintoi to act as a host to these two root-knot nematodes. Table 5-16. Analysis of variance table of Arachis pintoi germplasm reaction to M. incognita DI* EM** Source df MS Pr>F MS Pr>F Accession 39 0.0955 0.0015 0.0000 Replication 2 0.0364 0.3969 0.0000 Error 48 0.0386 0.0000 Total 89 Damage index; ** Egg mass Santiago et al. (2002) investigated the A. pintoi reaction to M. paranaensis and M. incognita races 1, 2, 3, and 4. They reporte d that no root penetration by M. incognita and M. paranaensis juveniles had occurred, and henc e there was no gall or egg mass formation. They concluded that in general A. pintoi accessions had an antagonistic effect on the nematodes, suggesting that they could be used as an interccrop or cover crop to reduce M. paranaensis and M. incognita populations. This res earch supports this conclusion and includes populations of M. arenaria since many accessions presented resistance to this species. Information published and available seems to support the results obtained in this research with respect to nematode reaction of A. pintoi The great majority of the accessions evaluated presented some level of resistance to M. arenaria race 1, M. javanica and M. incognita race 1. Once more, the source of resistance of these accessions could be used in breeding programs of A. hypogaea and more important qualify A. pintoi as potential forage, at least by this criteria, in environments where nematode infestation is a factor. Another positive outcome of this resu lt is the ability of th e species to suppress

PAGE 126

112 the multiplication of nematodes, and then be an important cover crop to species with nematode susceptibility problems. Table 5-17. Reaction of Arachis pintoi germplasm to M .incognita Accession Egg Mass (EM) Damage Index (DI) PI / CIAT mean SE mean SE Highly resistant 20826 1.0 0.0 1.0 0.0 22150 1.0 0.0 1.0 0.0 22151 1.0 0.0 1.0 0.0 22152 1.0 0.0 1.0 0.0 22154 1.0 0.0 1.0 0.0 22159 1.0 0.0 1.0 0.0 22174 1.0 0.0 1.0 0.0 22175 1.0 0.0 1.0 0.0 22232 1.0 0.0 1.6 0.6 22233 1.0 0.0 1.0 0.0 22234 1.0 0.0 1.0 0.0 22236 1.0 0.0 1.0 0.0 22238 1.0 0.0 1.0 0.0 22241 1.0 0.0 1.0 0.0 22256 1.0 0.0 1.0 0.0 22259 1.0 0.0 1.0 0.0 22265 1.0 0.0 1.0 0.0 22268 1.0 0.0 1.0 0.0 22289 1.0 0.0 1.0 0.0 22290 1.0 0.0 1.0 0.0 22324 1.0 0.0 1.0 0.0 22325 1.0 0.0 1.0 0.0 22339 1.0 0.0 1.0 0.0 476132 1.0 0.0 1.0 0.0 497574 1.0 0.0 1.0 0.0 604798 1.0 0.0 1.0 0.0 604800 1.0 0.0 1.0 0.0 604803 1.0 0.0 1.0 0.0 604804 1.0 0.0 1.0 0.0 604805 1.0 0.0 1.0 0.0 604808 1.0 0.0 1.0 0.0 604811 1.0 0.0 1.0 0.0 604812 1.0 0.0 1.0 0.0 604813 1.0 0.0 1.0 0.0 604815 1.0 0.0 1.0 0.0 604817 1.0 0.0 1.0 0.0 604818 1.0 0.0 1.0 0.0 604859 1.0 0.0 1.0 0.0 Resistant 22271 1.0 0.0 2.7 0.0 Florunner 1.0 0.0 1.0 0.0

PAGE 127

113 Summary and Conclusions Arachis pintoi germplasm displayed great variabi lity with respect to adaptation, dry matter yield, nutritive value, seed production, and nematode reaction. Average FDMY during the year 2003 was 4.35 Mg ha-1, and ranged from zero to 9.10 Mg ha-1. Crude protein and IVOMD were hi gh and confirmed the fact that A. pintoi has excellent nutritive value. Average CP was 180 g kg-1 of DM and ranged from 130 to 220 g kg-1 of DM. IVOMD averaged 670 g kg-1 of DM and ranged from 600 to 730 g kg-1 of DM. Some accessions produced high seed yields reaching values above 1.00 Mg ha-1, 18 and 30 mo after sowing. The average seed production was 0.32 Mg ha-1 in 2003, and 0.43 Mg ha-1 in 2004. Overall, accessions PI 604817, 497574, 604815, 604810, 604808, and 604857 were the best adapted to the north centr al Florida environment and with the best agronomic characteristics. Arachis pintoi germplasm presented high levels of resistance to M. arenaria M. javanica and M. incognita Overall, 93% of the accessions were classified as resistant to M. arenaria and all of them presented resistance to M. javanica and M. incognita. Although it sounds obvious, this res earch demonstrates again, that general conclusions about the ad aptation and agronomic value of a species to a particular environment should not be based on a single or a few germplasm lines.

PAGE 128

114 CHAPTER 6 CONCLUSIONS Germplasm accessions of Arachis pintoi Krap. and Greg. stor ed at the Southern Regional Plant Introduction Sta tion of the National Plant Germplasm System (NPGS) located at Griffin, GA were transferred to the University of Florida, where they was characterized and evaluated at the molecula r, morphological and agronomic levels during the period 2001 to 2004. The three different levels of germplasm accession characterization employed in this research we re effective in demonstrating differences among the accessions, and furthered the ut ilization of the germplasm based on the information gathered at these three levels. Molecular characterization was achieved with RAPD molecular markers, which proved to be very informative and efficien t to characterize the genetic diversity and relationship among germplasm accessions of the species. DNA amplifications were obtained with eight primers (A4, B4, B5, C 2, D4, D13, E4, and G5), which amplified 100 different bands with ninety eight sh owing polymorphism. Individual germplasm accessions were discriminated based on th eir RAPD band profile. Ten bands were accession specific and with further work (sequencing) they can be used as markers to identify these individual germplasm accessions. The molecular parameters used to assess the genetic diversity of the germplasm indicat ed that a large amount of genetic diversity exists among the germplasm evaluated in this research. These same parameters allowed a construction of a dendogram that separated the accessions in four distinct groups.

PAGE 129

115 Accessions from different groups could be sel ected as parents in breeding program based on their genetic divergence. Two tissue culture protocols were used to assess their organo genesis ability and variations on RAPD profile of regenerate d plants were examin ed. Callus induction was achieved by both protocols 28 d after incubati on, and Protocol 1 (Rey et al., 2000) was superior to Protocol 2 (Ngo and Quesenbe rry, 2000) for callus rating and weight. The germplasm presented large variability for these two variables, and the accession by protocol interaction was signi ficant. In Protocol 1, shoot regeneration was obtained for accessions PI 604856, 604857, 604805, 604811, 604809, 604814, 604818, and CIAT 22234, 20826, 22152, and 22265. In Protocol 2, shoot regeneration was attained by accessions PI 604856, 604805, 604799, 604804, 604818, 604809, 604810, 604800, 604813, 604857, and CIAT 22256, 22234, 20826, 22152, and 22265. Although shoot regeneration was achieved for several accessi ons, root induction was very difficult to obtain, and many shoots died during this proce ss. At the end, just 16 regenerated plants were recovered from both protocols. Although differences in callus rating and we ight were observed we can conclude that both protocols were equivalent in A. pintoi plant regeneration. Even though differences between protocol s were not observed, large differences in individual accessions plant regeneration were present. In vitro regeneration is one of the requirements for genetic transformation, so one or more of the 10 accessions that showed effective regeneration could be used to develop a system to A. pintoi transformation. RAPD band profiles of regenerated tissue cu lture plants were identical to their parent plant. Although our sample was small, the results suggest that the germplasm

PAGE 130

116 could be stored in vitro thus reducing the space to store large numbers of accessions if further tests confirmed that no somaclonal va riation was generated in the tissue culture process. This should lower the costs when compared to growing and organizing plants annually in the field This A. pintoi germplasm presented great morphol ogical variability with all the descriptors, except pollen size and shape, leaf bristles superior and inferior surface, showing polymorphism. Meaningful correlati ons were found between leaf length (LL) and pod weight; leaf length and pod width; l eaf length and seed we ight; leaf length and seed width; and finally leaf length and seed length. Thus, le af length could be used as a selection criterion in programs where incr eased seed size is one of the objectives. The large amount of genetic diversity that exits among the germplasm accessions evaluated in this research, suggests for most traits, breeding and selection for specific attributes should be possible. Principal component analysis of the morphological data proved to be a useful technique to work with a large number of accessions and variables. The principal components analysis was able to discriminate and separate the accessions in terms of three dimensions, represented by sexual repr oduction, vegetative, and qualitative axis. The cluster analysis based on the first nine principal component s differentiated four distinct groups of accessions. It is important to comment that the molecular and the morphological dendograms were different. Arachis pintoi germplasm displayed large variabi lity with respect to its adaptation, dry matter yield, nutritive value, seed produc tion, and nematode reaction. Annual average dry matter yield for some A. pintoi accessions was close to the forage production

PAGE 131

117 presented by both rhizoma peanut cultivar s. Crude protein and IVOMD were high and confirmed the fact that A. pintoi has superior nutritive va lue. Some accessions produce elevated seed yields reaching values superior to 1 Mg ha-1 18 and 30 mo after sowing, which would give a pintoi peanut cultivar a great advantage over a rhizoma peanut cultivar, in terms of pasture establishment and seedling recruitment, if similar forage yields and persistence were present. Arachis pintoi germplasm showed somewhat variable but generally high levels of resistance in response to M. arenaria M. javanica and M. incognita Overall 93% of the accessions were classified as resistant to M. arenaria and all were resistant to M. javanica and M. incognita. Susceptibility to rootknot nematodes, especially M. arenaria, is one of the major problems that groundnut cultivars face in the southeastern USA. A. pintoi is considered by some authors as representative of the tertiary or quaternary gene pool of A. hypogea and so the germplasm could be a source of resistance to M. arenaria if genetic barriers could be overcome. The great genetic diversity presented at molecular, morphological, and agronomic levels by the germplasm evaluated shows that e ffort to collect germplasm in the wild are in the right direction. The information genera ted by this research reaffirms once more for plant selectors and plant breed ers that conclusions about a species gene tic diversity, adaptation to environments, and agronomic ch aracteristics must not be based on a single or small number of germplasm accessions. This information will also be useful for genetic resources management, germplasm se lection and plant breeding programs, and genetic studies in general. Finally we believe that a great contribu tion for the utilization

PAGE 132

118 of the genetic resources of A. pintoi stored in various germpl asm banks was made by this work.

PAGE 133

119 APPENDIX A LIST OF A. pintoi GERMPLASM EVALUATED AT UF Table A-1. List of A. pintoi germplasm evaluated at UF UF number PI number Lat. (South) Long. (West) Altitude (meter) 01 604856 16 53 42 07 360 02 604857 13 23 44 05 450 03 604858 15 26 47 21 700 04 604798 16 18 46 58 630 05 604803 14 25 44 22 510 06 604805 16 59 45 57 570 07 604812 14 28 46 29 500 08 604810 13 06 46 45 600 09 604811 13 51 46 52 490 10 604799 16 19 46 51 580 11 604800 16 41 46 29 540 12 604809 13 02 46 45 610 13 604817 18 38 44 04 630 16 604815 15 49 47 58 1080 17 604814 15 52 39 08 50 18 497541 18 38 44 04 640 20 604813 14 27 47 00 480 21 604801 16 42 46 25 560 22 604804 14 20 44 25 560 23 604808 13 18 46 42 500 24 604807 13 18 46 48 510 26 604818 29 338447 15 52 39 08 50 31 476132 16 08 47 12 690 32 497574 13 23 44 05 450 33 604859 17 03 42 21 360

PAGE 134

120 Table A-1. Continued. UF number CIAT number Lat. (South) Long. (West) Altitude (meter) 34 22256 16 10 46 01 580 35 22234 13 14 46 44 463 36 20826 37 22159 15 17 47 23 650 38 22152 16 52 46 35 550 39 22150 15 07 44 08 510 40 22265 41 22260 14 04 47 18 720 42 22271 15 47 47 56 1040 43 18745 16 05 42 05 280 44 22339 700 45 22259 14 29 47 09 510 47 22233 49 22238 15 11 39 29 256 50 22174 16 12 46 54 580 53 22241 12 40 39 05 53 54 22231 13 18 46 24 500 55 22175 16 12 46 54 600 56 22290 15 06 39 16 120 58 22325 15 06 39 16 120 59 22289 16 09 46 10 560 60 22232 61 22236 12 49 38 24 37 62 22151 16 18 46 50 590 63 22268 66 22324 16 09 46 12 550 67 22154 16 44 46 12 580

PAGE 135

121 APPENDIX B CTAB DNA EXTRACTION PROTOCOL Solutions and Reagents 2x CTAB Extraction Buffer Prepare 100ml 2% CTAB 2g 1.4 M NaCl 28ml 5M NaCl Stock 100 mM Tris-Cl (pH 8.0) 10ml 1M Tris-Cl Stock 20 mM EDTA (pH 8.0) 4ml 0.5M EDTA Stock 1% PVP 10ml 10% PVP Stock Complete with ddH2O to 100ml 2% (v/v) -Mercaptoethonol 20l/ml (Add just before use) CTAB Precipitation Buffer (Add just before use) Prepare 100ml 1% CTAB 1g 50 mM Tris-Cl (pH 8.0) 5ml 1M Tris-Cl Stock 10 mM EDTA (pH 8.0) 2ml 0.5M EDTA Stock Complete with ddH2O to 100ml High Salt TE Buffer (Add just before use) Prepare 100ml 10 mM Tris-Cl (pH 8.0) 1ml 1M Tris-Cl Stock 1 mM EDTA (pH 8.0) 0.2ml 0.5M EDTA Stock 1 M NaCl 20ml 5M NaCl Stock Complete with ddH2O to 100ml 10% CTAB Prepare 100ml 10% CTAB 10g 0.7M M NaCl 14ml 5M NaCl Stock Complete with ddH2O to 100ml (Modification of Rogers, S.O. and Bendich, J. 198 5. Extraction of DNA from milligram amounts of fresh, herbarium and mummified plant tissue. Plant Molecular Biology, 5:69-76)

PAGE 136

122 Procedure 0.1 g grinded peanut leaf tissue Add 400 l of pre-warmed (65C) 2x CTAB precipitation buffer Incubate in a 65 C water bath fo r 60 minutes and occasionally mix Add an equal volume of chloroform:isoamyl alcohol (24:1), mix gently but thoroughly Centrifuge @ 10,000 rpm for 5 minutes at 4 C Transfer supernatant to new tube a nd add 1/10 volume of 65 C 10% CTAB Repeat steps 4 and 5 Transfer supernatant to new tube and a dd an equal volume of CTAB precipitation buffer and mix well by inversion Centrifuge @ 10,000 rpm for 5 minutes at 4 C Remove the supernatant and rehydrate pellet in 50-100 l of high salt TE buffer by heating to 65 C for 10 minutes Re-precipitate the DNA with 0.6 volumes of isopropanol Centrifuge @ 10,000 rpm fo r 15 minutes at 4 C Wash pellet with 80% ethanol and dry Resuspend DNA in a small volume (10 50 l) of DDW

PAGE 137

APPENDIX C MORPHOLOGICAL DESCRIPTORS CORRELATION TABLE

PAGE 138

124Table C-1. Morphological desc riptors correlation table FPI FSW FSL FSC FSCr FWW FWL FKL FHL FHW FHC FHH SIL SID SC SH SGH LS FPI 0.0143* 0.9421 0.1313 0.5052 0.0370 0.8517 -0.037 0.8486 -0.034 0.8635 0.0300 0.8793 0.0576 0.7709 0.1088 0.3104 0.2248 0.2501 -0.185 0.3456 -0.204 0.2960 -0.068 0.7279 0.3517 0.0664 0.2129 0.2766 0.4547 0.0150 0.1109 0.5739 -0.135 0.4913 FSW 0.7815 0.0001 -0.119 0.5457 -0.092 0.6406 0.8180 0.0001 0.7402 0.0001 0.1992 0.3095 0.5253 0.0041 0.2200 0.2605 0.2164 0.2687 0.0457 0.8172 0.3348 0.0815 0.1129 0.5673 -0.294 0.1287 -0.111 0.5732 0.2431 0.2125 -0.076 0.7003 FSL -0.115 0.5586 -0.318 0.0989 0.6398 0.0002 0.7495 0.0001 0.3867 0.0421 0.7251 0.0001 0.5685 0.0016 0.0902 0.6480 0.1311 0.5060 0.3092 0.1093 0.1762 0.3696 -0.237 0.2246 0.1649 0.4016 0.3667 0.0549 -0.072 0.7149 FSC 0.1885 0.3366 0.1117 0.5714 0.1477 0.4530 0.3146 0.1029 -0.084 0.6675 0.1367 0.4878 -0.346 0.0709 -0.120 0.5430 -0.147 0.4525 -0.040 0.8377 0.2786 0.1510 0.0351 0.8590 -0.066 0.7361 -0.235 0.2275 FSCr 0.1457 0.4592 0.0545 0.7829 0.0164 0.9337 -0.102 0.6051 -0.215 0.2719 0.2041 0.2975 -0.141 0.4729 -0.193 0.3230 0.0792 0.6886 0.1194 0.5450 -0.103 0.5997 -0.235 0.2273 -0.265 0.1714 FWW 0.8531 0.0001 0.2884 0.1366 0.4839 0.0091 0.2156 0.2705 0.1016 0.6067 0.0600 0.7614 0.2756 0.1557 0.2632 0.1775 -0.235 0.2268 0.0336 0.8650 0.1799 0.3594 -0.200 0.3065 FWL 0.4417 0.0186 0.6665 0.0001 0.5087 0.0057 0.1016 0.6067 0.0608 0.7614 0.2756 0.1557 0.2623 0.1775 -0.235 0.2268 0.0336 0.8650 0.1799 0.3594 -0.200 0.3065 FKL 0.2849 0.1417 0.4173 0.0271 -0.175 0.3703 0.0497 0.8017 -0.161 0.4123 0.2028 0.3005 -0.040 0.8383 0.2290 0.2409 0.1392 0.4798 -0.170 0.386 FHL 0.5460 0.0026 -0.077 0.6940 0.1320 0.5030 0.2150 0.2718 0.2699 0.1648 -0.144 0.4642 0.3555 0.0633 0.4061 0.0320 -0.231 0.2364 FHW -0.110 0.5763 0.3542 0.0644 0.0114 0.9540 0.3544 0.0642 -0.082 0.6767 0.5758 0.0013 0.1454 0.4604 -0.126 0.5211 FHC -0.115 0.5585 0.3486 0.0690 -0.158 0.4201 -0.073 0.7115 -0406 0.0320 -0.247 0.2043 0.1474 0.4539 FHH 0.0470 0.8121 -0.055 0.7804 0.0422 0.8311 0.3634 0.0573 -0.066 0.7361 0.0028 0.9886 SIL 0.0352 0.8587 -0.126 0.5219 -0.333 0.0827 -0.017 0.9285 0.0449 0.8203 SID -0.124 0.5270 0.3883 0.0412 0.1753 0.3723 -0.089 0.6495 SC 0.1262 0.5222 0.0422 0.8311 0.0772 0.6958 SH Pearson Correlation Coefficients (N = 28) 0.2931 0.1301 -0.152 0.4393 SGH P rob > |r| under H0: Rho=0 -0.241 0.2116

PAGE 139

125Table C-1. Continued LHL LHM LGHM LL LW LPL PegL Pegw PegC Pe gH Podwe PodL PodW Po dB PodR SW SL Swe FPI 0.2564 0.1877 0.1952 0.3194 -0102 0.6028 0.1894 0.3343 0.1369 0.4873 0.0030 0.9878 -0.170 0.3852 0.0653 0.7404 0.0350 0.8595 -0.002 0.9914 0.3400 0.0767 0.3115 0.1066 0.4076 0.0313 -0.125 0.5244 0.0034 0.9862 0.3653 0.0559 0.2432 0.2122 0.3225 0.0942 FSW 0.0075 0.9695 0.0362 0.8548 -0.004 0.9800 0.3841 0.0432 0.3035 0.1163 0.3140 0.1037 0.0795 0.6874 -0.144 0.4637 -0.034 0.8605 0.0703 0.7219 0.3768 0.0481 0.1551 0.4304 0.4408 0.0189 0.2867 0.1391 -0.342 0.0743 0.4135 0.0287 0.2892 0.1355 0.3333 0.0831 FSL -0.029 0.8806 -0.041 0.8331 0.1160 0.5563 0.3525 0.0657 0.3811 0.0990 0.2725 0.1606 0.0686 0.7286 -0.261 0.1796 -0.235 0.2268 0.1616 0.4112 0.2514 0.1968 0.0937 0.6351 0.3961 0.0369 0.1224 0.5349 -0.223 0.2527 0.3917 0.0392 0.2357 0.2271 0.2360 0.2266 FSC 0.1405 0.4755 -0.166 0.3966 0.3333 0.0830 -0.263 0.1775 -0.369 0.0527 -0.429 0.0224 -0.219 0.2628 0.3192 0.0977 -0.258 0.1846 -0.258 0.1846 0.0105 0.9575 -0.050 0.7970 0.1213 0.5385 -0.170 0.3850 -0.088 0.6529 0.1082 0.5835 -0.198 0.3104 0.0134 0.9460 FSCr 0.4385 0.0196 0.3535 0.0649 -0.235 0.2273 -0.101 0.6071 0.0125 0.9495 -0.177 0.3656 -0.082 0.6754 0.1659 0.3987 0.1217 0.5372 -0.091 0.6441 -0.109 0.5802 -0.050 0.7970 0.1213 0.5385 -0.170 0.3850 -0.088 0.6529 0.1082 0.5835 -0.198 0.3104 0.0134 0.9460 FWW 0.1723 0.3805 0.1883 0.3371 -0.086 0.6629 0.5312 0.0036 0.3290 0.0873 0.2401 0.2183 0.1427 0.4687 -0.153 0.4348 -0.143 0.4653 -0.041 0.8333 0.4399 0.0191 0.1940 0.3223 0.5754 0.0014 0.2270 0.2454 -0.204 0.2962 0.5303 0.0037 0.3372 0.0793 0.4224 0.0251 FWL 0.0547 0.7821 0.0565 0.7751 0.0764 0.6989 0.4038 0.0331 0.3128 0.1050 0.1827 0.3520 0.1462 0.4578 -0.049 0.8015 -0.256 0.1876 -0.061 0.7546 0.3385 0.0780 0.1447 0.4625 0.4675 0.0121 0.2780 0.1519 -0.320 0.0960 0.4674 0.0121 0.2715 0.1622 0.3561 0.0628 FKL 0.0396 0.8411 0.0870 0.6597 0.3039 0.1159 0.2430 0.2127 0.1902 0.3321 0.2732 0.1595 -0.030 0.8777 -0.044 0.8233 -0.518 0.0047 0.0219 0.9117 0.1655 0.4000 -0.036 0.8534 0.1217 0.5372 -0.117 0.5506 0.0950 0.6305 0.2372 0.2242 0.0823 0.6770 0.1635 0.4057 FHL 0.0106 0.9572 -0.118 0.5467 -0.048 0.8067 0.1806 0.3575 0.2292 0.2407 0.0987 0.6172 0.1541 0.4335 -0.070 0.7232 -0.378 0.0468 -0.074 0.7073 0.0178 0.9280 -0.234 0.2307 0.2345 0.2296 0.1626 0.4082 -0.130 0.5072 0.1619 0.4103 -0.051 0.7965 0.0154 0.9377 FHW -0.028 0.8843 0.1356 0.4913 0.0683 0.7297 0.1472 0.4545 0.2272 0.2449 0.0188 0.9242 -0.083 0.6718 0.1908 0.3307 -0.418 0.0265 -0.245 0.2085 0.1026 0.6032 0.1141 0.5629 0.3248 0.0917 -0.057 0.7728 -0.049 0.8019 0.3388 0.0778 0.1030 0.6019 0.1003 0.6112 FHC -0.358 0.0614 -0.144 0.4637 -0.082 0.6765 0.0458 0.8167 0.3060 0.1132 0.2828 0.1448 0.2688 0.1666 -0.071 0.7172 0.4472 0.0170 0.2981 0.1233 0.1037 0.5993 0.1133 0.5657 -0.133 0.4986 0.2762 0.1547 -0.384 0.0431 0.0084 0.9659 0.1828 0.3517 0.0622 0.7531 FHH -0.090 0.6453 0.3000 0.1209 0.0666 0.7361 0.0590 0.7655 -0.036 0.8552 0.0292 0.8827 0.1762 0.3698 0.0090 0.9621 -0.017 0.9307 -0.258 0.1846 -0.231 0.2368 -0.259 0.1816 -0.078 0.6925 0.0394 0.8422 -0.088 0.6529 -0.106 0.5908 -0.227 0.2451 -0.268 0.1679 SIL -0.452 0.0157 -0.336 0.0797 0.0876 0.6573 0.2531 0.1936 0.0366 0.8532 0.1508 0.4436 0.3853 0.0428 -0.160 0.4141 0.2207 0.2590 -0.082 0.6754 0.2671 0.1694 0.1160 0.5566 0.2562 0.1881 -0.047 0.8084 -0.179 0.3596 0.2503 0.1988 0.2443 0.2101 0.2624 0.1772 SID 0.4422 0.0184 0.3568 0.0623 -0.122 0.5363 0.5379 0.0031 0.5611 0.0019 0.1336 0.4977 0.0087 0.9647 0.0357 0.8567 -0.216 0.2692 -0.385 0.0430 0.4273 0.0233 0.3511 0.0669 0.5263 0.0040 0.2264 0.2465 0.1685 0.3914 0.4716 0.0113 0.4563 0.0147 0.4199 0.0261 SC 0.0157 0.9368 -0.105 0.5930 -0.042 0.8311 -0.158 0.4200 -0.134 0.4963 -0.224 0.2513 0.2703 0.1642 0.1361 0.4898 -0.163 0.4057 -0.010 0.9561 -0.236 0.2259 -0.325 0.0909 -0.188 0.3366 -0.257 0.1852 -0.253 0.1934 -0.028 0.8867 -0.315 0.1018 -0.147 0.455 SH 0.2981 0.1234 0.4250 0.0242 -0.058 0.7670 0.2096 0.2842 0.2472 0.2047 0.0649 0.7428 -0.072 0.7156 0.1936 0.3236 -0.348 0.0695 -0.348 0.0695 -0.020 0.9174 -0.062 0.7530 0.2225 0.2549 -0.265 0.1719 0.1856 0.3443 0.2056 0.2938 -0.075 0.7029 -0.033 0.8671 SGH 0.0413 0.8345 0.0000 1.0000 0.3333 0.0830 0.2816 0.1465 0.1272 0.5188 0.2230 0.2540 0.0707 0.7207 0.1600 0.4158 -0.086 0.6632 0.0860 0.6632 0.0239 0.9038 -0.106 0.5912 0.2303 0.2384 0.0093 0.9622 0.0000 1.0000 0.1956 0.3184 0.0062 0.9748 0.0552 0.7801 LS -0.082 0.6755 0.1170 0.5529 -0.354 0.0629 0.0272 0.8905 0.1057 0.5922 0.0917 0.6423 -0.059 0.7622 -0.306 0.1128 -0.087 0.6563 0.1429 0.4682 -0.083 0.6737 0.0663 0.7372 -0.106 0.5885 -0.131 0.5050 -0.073 0.7108 -0.075 0.7013 -0.014 0.9430 -0.098 0.6195

PAGE 140

126Table C-1. Continued LHL LHM LGHM LL LW LPL PegL Pegw PegC Pe gH Podwe PodL PodW Po dB PodR SW SL Swe LHL 0.5168 0.0049 -0206 0.2912 0.1238 0.5300 0.1638 0.4048 -0.152 0.4392 -0.304 0.1155 0.0691 0.7268 0.0533 0.7873 -0.245 0.2079 -0.014 0.9430 -0.002 0.9908 0.1114 0.5723 0.2036 0.2986 -0.110 0.5765 0.1343 0.4956 0.0184 0.9259 -0.002 0.9915 LHM -0.333 0.0830 0.2748 0.1570 0.1428 0.4684 -0.017 0.9278 -0.287 0.1374 -0.156 0.4273 0.0430 0.8279 -0.107 0.5858 -0.048 0.8083 0.0794 0.6878 0.1050 0.5949 0.0328 0.8683 -0.069 0.7255 0.0674 0.7332 0.0149 0.9398 -0.099 0.6139 LGHM -0.114 0.5614 -0.167 0.3936 0.0470 0.8123 0.0154 0.9379 0.3608 0.0592 0.0860 0.6632 0.0860 0.6632 0.0208 0.9160 0.0171 0.9309 0.0317 0.8726 0.0656 0.7399 0.0000 1.0000 -0.006 0.9739 -0.071 0.7192 0.0385 0.8457 LL 0.7944 0.0001 0.7660 0.0001 0.4667 0.0123 -0.111 0.5715 -0.071 0.7183 -0.205 0.2947 0.7168 0.0001 0.5248 0.0041 0.7082 0.0001 0.1951 0.3196 -0.011 0.9548 0.7382 0.0001 0.7255 0.0001 0.7098 0.0001 LW 0.7746 0.0001 0.3786 0.0469 0.0180 0.9274 -0.155 0.4285 -0.111 0.5721 0.5963 0.0008 0.4659 0.0125 0.5482 0.0025 0.3316 0.0847 -0.003 0.9843 0.6248 0.0004 0.6532 0.0002 0.6062 0.0006 LPL 0.5087 0.0057 0.0044 0.9822 -0.031 0.8723 0.0057 0.9767 0.5699 0.0015 0.3719 0.0513 0.4228 0.0250 0.1943 0.3216 -0.034 0.8597 0.4906 0.0080 0.5636 0.0018 0.5485 0.0025 PegL 0.0280 0.8875 -0.004 0.9832 -0.148 0.4523 0.2357 0.2271 -0.011 0.9524 0.1156 0.5577 0.2483 0.2025 -0.255 0.1887 0.1716 0.3824 0.1789 0.3622 0.2446 0.2096 PegW 0.0446 0.8214 -0.273 0.1592 0.1271 0.5191 0.1792 0.3614 0.1496 0.4473 0.1614 0.4119 -0.134 0.4949 0.2339 0.2309 0.0408 0.8364 0.1874 0.3395 PegC 0.0666 0.7361 -0.023 0.9067 0.0961 0.6263 -0.179 0.619 0.1865 0.3419 -0.258 0.1846 -0.152 0.4381 0.0777 0.6941 -0.084 0.6694 PegH -0.141 0.4739 0.0090 0.9634 -0.244 0.2108 0.1526 0.4382 0.1434 0.4665 -0.207 0.2892 -0.088 0.6550 -0.097 0.6203 PodWe 0.8475 0.0001 0.8549 0.0001 0.1641 0.4039 0.0380 0.8475 0.8744 0.0001 0.9218 0.0001 0.9701 0.0001 PodL 0.7082 0.0001 0.2147 0.2725 0.0849 0.6675 0.7120 0.0001 0.8892 0.0001 0.8367 0.0001 PodW 0.0939 0.6344 0.0070 0.9718 0.8915 0.0001 0.7340 0.0001 0.8375 0.0001 PodB -0.306 0.1127 0.0204 0.9177 0.1888 0.3359 0.1419 0.4713 PodR -0.060 0.7604 0.0786 0.6906 0.0760 0.7006 SW 0.8162 0.0001 0.9042 0.0001 SL 0.9204 0.0001

PAGE 141

127 APPENDIX D CLIMATOLOGICAL DATA AT THE FORAGE RESEARCH UNIT IN GAINESVILLE-FL, DURING THE PERIOD OF THE AGRONOMIC EVALUATION FIELD TRIAL. Table D-1. Climatological data at the forage research unit in Gainesville, FL, during the period of the agronomic evaluation field trial Temperature C N of days below temperature Year Month Min. Max. Mean Rainfall (mm) N days drought 0 C 4 C 12 C 2001 Sep 23.22 10.1834.34 292.2 13 0 0 1 Oct 19.46 -0.1131.48 2.9 29 1 4 11 Nov 17.51 -0.2228.88 27.1 26 1 3 14 Dec 15.21 -5.4428.24 38.4 27 6 9 18 2002 Jan 12.77 -7.3129.11 133.0 19 6 14 19 Feb 12.41 -8.0928.8 27.7 25 6 16 24 Mar 17.45 -6.3731.65 83.1 22 4 8 19 Apr 21.92 3.6433.15 9.5 23 0 1 4 May 23.11 6.4235.57 39.5 26 0 0 7 Jun 24.89 15.0837.47 104.7 10 0 0 0 Jul 25.71 18.0536.84 138.3 13 0 0 0 Aug 25.05 11.9435.61 265.3 13 0 0 1 Sep 25.66 17.1333.99 128.2 19 0 0 0 Oct 22.14 6.8234.88 41.6 21 0 0 4 Nov 14.07 -4.3130.08 133.1 22 7 13 23 Dec 10.96 -3.0126.44 183.1 21 11 18 27 2003 Jan 8.22 -8.3324.17 4.5 27 17 23 31 Feb 13.42 -0.8228.42 172.5 17 4 12 23 Mar 18.69 2.0330.64 194.4 19 0 1 14 Apr 19.22 -0.2630.99 41.4 26 1 2 15 May 24.05 11.6434.81 50.9 24 0 0 1 Jun 25.27 16.4434.49 297.8 12 0 0 0 Jul 25.92 19.9136.04 121.4 11 0 0 0 Aug 25.53 19.5534.04 206.2 9 0 0 0 Sep 23.87 13.6233.17 211.0 17 0 0 0 Oct 20.66 7.6131.34 103.7 25 0 0 7 Nov 17.39 -2.6331.62 34.1 25 2 4 15 Dec 10.28 -6.4625.16 18.9 24 12 21 31 2004 Jan 10.98 -6.1627.29 33.9 24 13 17 28 Feb 12.35 -1.7627.82 143.6 15 2 9 24 Mar 16.54 -1.3529.00 25.5 27 1 6 25 Apr 17.95 1.6931.09 40.3 25 0 6 23

PAGE 142

128 Table D-1. Continued Temperature C N of days below temperature Year Month Min. Max. Mean Rainfall (mm) N days drought 0 C 4 C 12 C 2004 May 23.92 5.41 36.13 16.4 28 0 0 4 Jun 25.77 18.03 36.15 174.3 16 0 0 0 Jul 25.99 18.22 35.28 254.8 13 0 0 0 Aug 25.69 20 34.94 150.7 12 0 0 0

PAGE 143

129 LIST OF REFERENCES Akasaka, Y., H. Daimon, and M. Mii. 2000. Improved plant regeneration from cultured leaf segments in peanut ( Arachis hypogaea L.) by limited exposure to thiadiazuron. Plant Science 156: 169-175. Argel, P.J. 1994. Regional experience with forage Arachis in Central America. p. 134143. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Argel, P.J. and E.A. Pizarro. 1992. Germplasm case study: Arachis pintoi . p. 57-76. In Pastures for the tropical lowlands CIAT 's Contribuition. CIAT, Cali, Colombia. Argel, P.J. and A. Valerio. 1993. Effect of crop age on seed yield of Ar achis pintoi at two sites in Costa Rica, Central America. p. 1696-1698. In Proc. Int. Grassl. Congr., 17 th., Palmerston North, New Zealand, 8-21 Feb, New Zealand Grassland Association. Bertozo, M.R. and J.F.M. Valls. 1996. Ge netic variability in populations of Arachis pintoi Krap. & Greg. and A. repens Handro (Leguminosae). In Proc. of Congresso Nacional de Gentica. Caxamb, Brazil. SB G, Revista Brasileira de Gentica. Bertozo, M.R. and J.F.M. Valls. 1997a. Vari ation of RAPD profile in accessions of Arachis pintoi and Arachis repens section Caulorrhizae (L eguminosae). In Proc. do Simpsio Latino-Americano de Recursos Ge nticos Vegetais. Campinas, Brazil, IAC. Bertozo, M.R. and J.F.M. Valls. 1997b. Utilizat ion of storage proteins, isoenzymes and RAPD in studies of a hybrid population of Ar achis pintoi (Leguminosae). In Proc. do Simpsio Latino-Americano de Recursos Ge nticos Vegetais. Campinas, Brazil, IAC. Bretting, P.K. and M.P. Widrlechner. 1995. Ge netic markers and Plant Genetic Resource Management. p. 11-75. In J. Janick (ed.). Plant Br eeding Reviews, John Wiley & Sons, Inc. New York. Burtnik, O.J. and L.A. Mr oginski. 1985. Regeneration of Arachis pintoi plants (Leguminosae) by in vitro leaf tissue culture. Oleagineux 40:609-612. Cameron, D.G. 1983. To breed or not to breed. p. 237-250. In J.G. McIvor and R.A. Bray (ed.). Genetic resources of fora ge plants, CSIRO, Melbourne. Cameron, D.G., Jones, R.M., Wilson, G.P.M. L., Bishop, H.G., Cook, B.G., Lee, G.R., and Lowe, H.F. 1989. Legumes for heavy graz ing in coastal sub tropical Australia. Tropical Grassland 23:153-161.

PAGE 144

130 Carulla, J.E., C.E. Lascano, and J.K. Ward. 1991. Selectivity of resistent and oesophageal fistulated steers grazing Arachis pintoi and Brachiaria dictyoneura in Llanos of Colombia. Tropical Grassland 25: 317-324. Carvalho, S., Monato, L., Valls, J.F.M., and Lopes, C.R. 1998. Evaluation of the genetic diverisity of Arachis pintoi using molecular and morphological markers. In Proc. do Congresso Brasileiro de Genetica, 44 th., guas de Lindia, SBG. Chang, T.W., T.J. Yiu, and F.S. Thesng. 1999. Genetic diversity of peanut varieties (lines) released in Taiwan. Journal of Agriculture and Forestry 48: 41-53. Chengalrayan, K., V.B. Mhaske, and S. Harza. 1995. In vitro regulation of morphogenesis in peanut ( Arachis hypogaea L.). Plant Science 110: 259-268. Conagin, C.H.T.M.. 1959. Fruit development in wild species of peanut ( Arachis spp.). Bragantia 18: 51-70. Cook, B.G. and I.C. Crosthwaite. 1994. Utili zation of Arachis species as forage. p. 624663. In J. Smart (ed.). The groudnut crop: a scientific basis for improvement, Chapmam & Hall, London. Cook, B.G. and T.G. Franklin. 1988. Crop management and seed harvesting of Arachis pintoi Krap. et Greg. nom. nud. Journal of Applied Seed Production 6: 26-30. Cook, B.G., R.M. Jones, and R.J. Willia ms. 1994. Regional experience with forage Arachis in Australia. p. 158-168. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Cook, B.G. and D.S. Loch.1993. Commercialisation of Arachis pintoi cv. Amarillo in northern Australia. In Proc. Int. Grassl. Congr., 17 th., Palmerston North, New Zealand, 8-21 Feb, New Zeal and Grassland Association. Cook, B.G., R.J. Williams, and G.P.M. W ilson. 1990. Register of Australian plant cultivars. B. Legumes. 21. Arachis (a) Arachis pintoi Krap. & Greg. nom. nud. (Pinto peanut) cv. Amarillo. Australian Journal of Experimental Agriculture 30: 445-446. Diulgheroff, S., E.A. Pizarro, J.E. Ferguson, and P.J. Argel. 1990. Seed multiplication of tropical forage species in Costa Ri ca. Pasturas Tropicales 12: 15-23. FAO. 1996. Global plan of action for the conser vation and sustainable utilization of plant genetic resources for food and agri culture The Leipzig declaration. In International Technical Conference on Plant Genetic Resources. 1996. Leipzig, Germany, FAO. FAO. 1998. The state of the world's plant ge netic resources for food and agriculture. FAO, Rome.

PAGE 145

131 Ferguson, J.E. 1994. Seed biology and systems for Arachis pintoi p. 159-168. In P.C. Kerridge and B. Hardy (ed.). Biol ogy and Agronomy of Forage Arachis CIAT, Cali, Colombia. Ferguson, J.E., C.I. Cardozo, and M.S. Snc hez. 1992. Advances and perspectives of Arachis pintoi seed production. Pastur as Tropicales 14: 14-22. Fernandes, A.T.F., Fernandes, C.D., Euclid es, V.P.B., and Grof, B. 1992. Evaluation of Paspalum spp. accessions in consortiation with Arachis pintoi in low fertility lowlands. In Proc.Reunin SabanasRed Internacional de Evaluacin de Pastos Tropicales (RIEPT). Braslia, Brazi,: CIAT. Fernndez, A. and A. Krapovickas. 1994. Chromosomes and evolution of Arachis (Leguminosae). Bonplandia 8: 187-220. Ferreira, M.E. and D. Grattapaglia. 1998. In troduction to molecular marker is genetic analysis. Embrapa-Cenargen Documentos, Brasilia. Fisher, M.J. and P. Cruz. 1994. So me ecophysiological aspects of Arachis pintoi p. 5370. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Flick, C. E., D. A. Evans, et al. 1983. Basic techniques of plant cell culture: Organogenesis. p.13-81. In D. A. Evans, W. R. Sharp, et al (Ed.). Handbook of plant cell culture: Tecniques for propaga tion and breeding. New York, v.1, 1983. Ford-Lloyd, B. and M. Jackson. 1986. Plant Gene tic Resources: an introduction to their conservation and use. J.W. Arrowsmith Ltd., Bristol. Ford-Lloyd, B.V. 2001. Genotyping in pl ant genetic resources. p. 59-82. In R.J. Henry (ed.). Plant genotyping: the DNA fingerprinting of plants. CA BI Publishing: Lismore, Autralia. Frankel, O.H. and A.H.D. Brown. 1984. Plan t genetic resources today: A critical reappraisal. p. 249-257. in J.H.W. Holden and J.T. Williams (ed.). Crop Genetic Resources: conservation & evalua tion. Allen and Unwin, London. French, E.C., Prine, G. M., Ocumpaugh, W. R., and Rice, R. W. 1994. Regional experience with forage Arachis in the United States. p. 169-186. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Galgaro, M.L., Gimenes, M.A., Valls, J. F. M., and Kochert, G. 1998. Genetic variation between several species of sections Extr anervosae, Caulorrhizae, Heteranthae, and Triseminatae (genus Arachis) esti mated by DNA polymorphism. Genome 41: 445454. Gallaher, R.N., C.O. Weldon, and J.G. Futr al. 1975. An Aluminum block digester for plant and soil analysis. Soil Sc i. Soc. Amer. Proc. 39: 803-806.

PAGE 146

132 Gimenes, M.A., Lopes, C.A., Galgaro, M.L., Valls, J. F. M., and Kochert, G. 2000. Genetic variation and phylogenetic relations hips based on RAPD analysis in section Caulorrhizae, genus Arachis (Leguminosae). Euphytica 116: 187-195. Gorbet, D.W., R.L.J. Stanley, and D.A. Kn auft. 1994. Forage potential of cultivated peanut ( Arachis hypogaea L.). Peanut science 21: 112-115. Gregory, W.C., A. Krapovickas, and M.P. Gr egory. 1980. Structure, variation, evolution, and classification in Arachis p. 469-481. In R.J. Summerfield and A.H. Bunting (ed.). Advances in Legume Science. R oyal Botanic Gardens, Kew, England. Grof, B. 1985. Forage attributes of pereni al groundnut Arachis pi ntoi in a tropical savanna environment in Colombia. In Proc. Int. Grassl. Co ngr., 15th., Kyoto, Japan, 24-31 Aug., The Japanese Societ y of Grassland Science. Hambleton, L.G.. 1977. Semiautomated method for simultaneous determination of phosphorous, calcium and crude protein in animal feeds. J.A.O.A.C. 60: 845-852. Hawkes, J.G., N. Maxted, and B.V. Ford-Ll oyd. 2000. The Ex Situ Conservation of Plant Genetic Resources. Kluwer Academic Publishers, Dordrecht, The Netherlands. Hilu, K.W. and H.T. Stalker. 1995. Genetic relantionships between peanut and wild species of Arachis sect. (Fabaceae): evidence from RAPDs. Plant Systematics and Evolution 198: 167-178. Holbrook, C.C., W.F. Anderson, and R.N. P ittman. 1993. Selection of a core collection from the U.S. germplasm collection of peanut. Crop Science 33: 849-861. Hulli, D.M. 1987. Molecular versus morphologica l approaches to systematics. Ann. Ver. Econ. Syst. 18: 23-42. Hussey, R.S. and K.R. Baker. 1973. A comparis on of methods of collecting inocula for Meloidogyne spp. including a new technique Plant Dis. Rep. 57: 1025-1028. IBPGR/ICRISAT. 1990. Report of a work shop on genetic resources of wild Arachis species: including preliminary descriptors for Arachis In International Crop Network series 2. IBPGR/ICRISAT:, Rome. IBPGR/ICRISAT. 1992. Descriptors fo r groundnut. IBPGR/ICRISAT, Rome. Karp, A., Kresovich, S., Bhat, K.V., Aya d, W.G., and Hodgkin, T. 1997. Molecular tools in plant genetic resources conservation: a guide to technologies. IPGRI Technical Bulletin 47. IPGRI, Rome. Kerridge, P.C. 1994. Future prospects for utilization and research in forage Arachis p. 199-206. In P.C. Kerridge and B. Hardy (ed.) Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia.

PAGE 147

133 Keystone Center. 1991. International Dial ogue Series on Plant Genetic Resources, In Final Consensus report of the Keystone In ternational Dialogue Series on Plant Genetic Resources (3rd Plenary Session, Os lo, Norway), Keystone Center, Boulder, Co. Krapovickas, A. and W.C. Gre gory. 1994. Taxonomy of the genus Arachis (Leguminosae). Bonplandia 8: 1-186. Kretschmer, A.E., Brolmann, J., Coleman, S. W., and Snyder, G. H. 1979. Florida Carpon Desmodium Perennial Tropical Fo rage Legume for Use in South Florida. In Florida Agricultural Experiment Station Circular University of Florid a, Ona, Florida. Kretschmer, A.E. and T.C. W ilson. 1988. A new seed production Arachis sp. with potential as a forage in Florida. So il Crop Sci. Soc. Fla. Proc. 47:229-233. Larbi, A., Dung, D. D., Olorunju, P. E., Smit h, J. W., Tanko, R. J., Muhammad, I. R., and Adekunle, I. O. 1999. Groundnut ( Arachis hypogaea ) for food and fodder in croplivestock systems: forage and seed yi elds, chemical composition and rumen degradation of leaf and stem fractions of 38 cultivars. Animal Feed Science and Technology 77: 33-47. Lowe, A.J., Thorpe, W., Teale, A., and Hanson, J. 2003.Characterization of germplasm accessions of Napier grass ( Pennisetum purpureum and P. purpureum x P. glaucum hybrids) and comparison with farm clones using RAPD. Genetic Resources and Crop Evolution 50: 121-137. Mallikarjuna, N. and D.C. Sastri. 2002. Mor phological, cytological and disease resistance studies of the inters ectional hybryd between Arachis hypogaea L. and A. glabrata Benth. Euphytica 126: 161-167. Mann, J. 1987. Secondary Metabolism. Oxford Univ. Press., London. Marcon, G. 1988. Estrutura gentica de populaes de Stylosanthes humilis H.B.K. (Leguminosae) de trs regie s ecogeogrficas do Estado de Pernambuco. M.S. thesis. Escola Superior de Agricultura "Lui z de Queirz, Universidade de So Paulo,Piracicaba. Martins, P.S. 1984. Biological aspects of Brazilian herbaceous legumes. p. 173-184. In M.L.R. Aguiar-Perecin, P.S. Martins, and G. Bandel (ed.). Topicos de Citogenetica e evoluo de plantas. FEALQ, So Paulo, Brazil. Maass, B.L., A.M. Torres, and C.H. Ocampo. 1993. Morphological and isozyme characterization of Arachis pintoi Krap. et Greg. Nom. Nud. germplasm. Euphytica 70: 43-52. McKently, A.H., G.A. Moore, and F.P. Gardner. 1990. Regeneration of peanut and perennial peanut from cultured l eaf tissue. Crop Science 31: 833-837.

PAGE 148

134 Mislevy, P. and E.G. Martin. 2001. Harvest management of Aeschynomene evenia In Proc. Soil and Crop Science Society of Florida, Stuart, Florida, Soil and Crop Science Society of Florida. Monato, L. 1997. Morphological de scriptors applied to germpl asm of species of section Caulorrhizae. In Proc. do Simpsio Latino-Amer icano de Recursos Genticos Vegetais. Campinas, Brazil, IAC. Moore, J.E. and G.O. Mott. 1974. Recovery of residual organic matter from in vitro digestion of forages. J. Dairy Sci. 57: 1258-1259. Muir, J.P.. 2002. Hand-plucked forage yield a nd quality and seed pr oduction from annual and short-lived perennial warm-season legum es fertilized with composted manure. Crop Science 42: 897-904. Nei, M. 1972. Genetic distance between populat ions. The American Naturalist 106: 283292. Nei, M. 1973. Analysis of gene diversity in subdivided populations. Proc. Nat. Acad. Sci. 70: 3321-3323. Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University Press., New York. Ngo, H.L. and K.H. Quesenberry. 2000. Day length and media effects on Arachis pintoi regeneration in vitro Soil Crop Sci. Soc. Fla. Proc. 59: 90-93. Nbile, P.M., Gimenes, M.A., Valls, J. F. M., and Lopes, C.R. 2004. Genetic variation within and among species of genus Arachis section Rhizomatosae. Genetic Resources and Crop evolution 51: 299-307. Norden, A.J., R.W. Lipscomb, and W.A. Carver. 1969. Registration of 'Florunner' peanuts. Crop Science 9: 850. Oliveira, M.A.P. and J.F.M. Valls. 1999. Vari ability of the morphol ogical characters and inheritance of flower color in wild peanut A. pintoi and A. repens species. In Proc.do Congresso Nacional de Gen tica. Gramado, RS, SBG. Omokanye, A.T., Onifade, O. S., Olorunju, P. E., Adamu, A. M.,Tanko, R. J., and Balogun, R. O. 2001 The evaluati on of dual-purpose groundnut ( Arachis hypogaea ) varieties for fodder and seed production at Shika, Nigeria. Journal of Agricultural Science 136: 75-79. Paganella, M.B. and J.F.M. Valls. 2002. Morp hological characteriza tion of cultivars and selected accessions of Arachis pintoi Krapov. & Gregory. Pasturas Tropicales 24: 2229. Paterniani, E. 1988. Genetic diversity of cultivated plants. In Proc. do Encontro sobre Recursos Geneticos. 1988. Jaboticabal, FCAV.

PAGE 149

135 Pizarro, E.A. and M.A. Carvalho. 1996. A lternative forages for the tropics: Arachis and Paspalum In Proc.of Symposium of the Crop Science Society of America. Seattle. WA, CSSA. Pizarro, E.A., Carvalho, M.A., Valls J. F. M., and Maciel, D.1992. Arachis spp.: Evaluacion Agronomica en areas bajas del Cerrado. In Proc.Reunin SabanasRed Internacional de Evaluacin de Pastos Tropi cales (RIEPT). Braslia, Brazil,: CIAT. Pizarro, E.A., Ramos, A. K. B., Ayarza, M. A., Carvalho, M. A., and Costa, P. H. 1996b. Agronomic evaluation of forage legumes in consortium with B. decumbens in Uberlndia-MG. In Proc. da Reunio Anual da Soc. Bras. Zootc. 33th., Fortaleza, Brazil, SBZ. Pizarro, E.A., A.K.B. Ramos, and M.A. Carvalho. 1996c. Forage potential and seed production of accessions of selected Calopogonium mucunoides in the Brazilian Cerrados. Pasturas Tropicales 18: 25-31. Pizarro, E.A. and A.C. Rincn. 1994. Regional experience with forage Arachis in South America. p. 144-157. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Pizarro, E.A., Valls, J. F. M., Carvalho, M.A., Charchar, M. J.D. 1993. Arachis spp.: introduction and evaluation of new accecions in seasonally flooded land in the Brazilian Cerrado. In Proc. Int. Grassl. Congr., 17 th., Palmerston North, New Zealand, 8-21 Feb, New Zeal and Grassland Association. Pizarro, E.A., Valls, J.F.M., Ramos,A.K.B., Godoy, I.J., Carvalho, M.A., and Singh, A.K. 1996a. Forage potential of Arachis hypog aea in the Brazilian Cerrados. Pasturas Tropicales 18: 17-24. Plucknett, D.L., Smith, N.J.H., Williams, J.T., and Anishetty, N.M. 1987. Gene banks and the world's food. Priceton University Press, New Jersey. Prine, G., Dunavin, L.S., Moore, J.E.,and R oush, R.D. 1986. Registration of 'Florigraze' rhizoma peanut. Crop Science 26 : 1084-1085. Prine, G.M., Dunavin, L.S., Glennon, R.J ., and Roush, R.D. 1990. Registration of 'Arbrook' rhizoma peanut. Crop Science 30: 743-744. Purcino, H.M.A. and M.C.M. Viana. 1994. Preliminary evaluations of Arachis pintoi in lowland areas. In Proc. da Reunio Anual da Soc. Bras. Zootc. 31st., Maring, PR, Brazil, EDUEM. Queneherve, P., Y. Bertin, and C. Chabrier. 2002. Arachis pintoi : a cover crop for bananas: Advantages and di sadvantages as regards nematology. Infomusa 11: 28-30.

PAGE 150

136 Rao, I.M. and P.C. Kerridge. 1994. Mineral nutrition of forage Arachis p. 71-83 In P.C. Kerridge and B. Hardy (ed.). Biol ogy and Agronomy of Forage Arachis CIAT, Cali, Colombia. Reiter, R. 2001. PCR-based marker systems. p. 9-30. In R.L. Phillips and I.K. Vasil (ed.). DNA-based markers in plants. Kluwer Academic Publishers: Dordrecht. Renu, M.S.K. 2003. Characterization of Pisum germplasm based on molecular markers. Indian Journal of Pulses Research 16: 84-91. Rey, H.Y., Scocchi, A. M., Gonzalez, A. M., and Mroginski, L. A.2000. Plant regeneration in Arachis pintoi (Leguminosae) through leaf culture. Plant Cell Reports 19: 856-862. Rincn, A.C., Cuesta, M.P.A., Prez, B.R., Lascano, C.E., and Fergunson, J. 1992. Man forrajero perenne ( Arachis pintoi Krapovickas et Gregory) : Una alternativa para ganaderos y agricultores. In Boletn Tcnico 219. IAC/CIAT, Cali, Colombia. Rogers, S.O. and J. Bendich. 1985. Extracti on of DNA from milligram amounts of fresh, herbarium and mummified plant tissue. Plant Molecular Biology 5: 69-76. Santiago, D.C., Homechin, M., Krzyzanowski, A. A., Carvalho, S., and Fonseca, I.C. 2002. Antagonist effect of Arachis pintoi on Meloidogyne paranaensis and M. incognita Nematologia-Mediterranea 30: 147-152. SAS Institute Inc. 1989. SAS/STAT User's Guide, Version 6, 4th ed. SAS Inst. Inc., Cary, NC. Shannon, C.E. and W. Weaver. 1949. The ma thematical theory of communication. University of Illinois Press, Urbana, IL. Sharma, S.B., Ansari, M.A., Varapras ad, K.S., Singh, A.K., and Reddy, L.J. 1999. Resistance to Meloidogyne javanica in wild Arachis species. Genetic Resources and Crop evolution 46: 557-568. Simpson, C.E. and J.L. Starr. 2001. Registra tion of 'COAN' peanut. Crop Science 41: 918. Simpson, E.H. 1949. Measurement of diversity. Nature 163: 688. Simpson, M.J.A. and L.A. Withers. 1986. Char acterization of plant genetic resources using isozyme electrophoresis: a guide to the literature. FAO/IBPGR, Rome. Singh, A.K., H.T. Stalker, and J.P. Moss. 1990. Cy togenetics and use of alien variation in groundnut improvement. In T. Tsuchiya and P.K. Gupta (ed.). Chromosome engineering in plants: Genetics, breed ing, evolution. Elsevier, Amsterdan.

PAGE 151

137 Singh, U. and B. Singh. 1992. Tropical grain legumes as important human foods. Economic Botanic 46: 310-321. Skinner, D.Z., G.R. Bauchan, G. Auricht, and S. Hughes. 1999. A Method for the efficient management and utilization of large germplasm collections. Crop Science 39:1237-1242. Smith, R.D. and S. Linington.1977. The mana gement of the Kew Seed Bank for the conservation of arid land and U.K. wild species. Bocconea 15:237-280. Solbrig, O.T. 1980. Demography and natural se lection, in Demography and evolution in plant population. Blackwell, Oxford. Sousa, M.A., Pizarro, E.A., Carvalho, M.A., Grof, B., and Schulz, A. L. 1992. Agronomic evaluation of forage grasses and legumes in Planaltina, Distrito Federal, Braslia. In Proc.Reunin SabanasRed Intern acional de Evaluacin de Pastos Tropicales (RIEPT). Braslia, Brazil,: CIAT. Stalker, H.T. 1990. A morphological appr asal of wild species in Section Arachis of peanuts. Peanut Science 17:117-122. Starr, J.L., G.L. Schuster, and C.E. Simps on. 1990. Characterization of the resistance to Meloidogyne arenaria in an interspecific Arachis spp. Hybrid. Peanut Science 17:106-108. Str, W.W. and J. Ndikumana. 1994. Regional experience with forage Arachis in other tropical areas: Asia, Africa, and the Pacific. p. 187-198. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Thomas, R.J. 1994. Rhizobium requirements, nitrogen fixation, and nutrient cycling in forage Arachis. p. 84-94 In P.C. Kerridge and B. Har dy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Toledo, J.M. 1982. Manual para la evaluaci n agronmica. Red Internacional de Evaluacin de Pastos Tropicales (Serie CIAT 075G-1-82). CIAT, Cali, Colombia. Upadhyaya, H.D. 2003a. Phenot ypic diversity in groudnut ( Arachis hypogaea L.) core collection assessed by morphological a nd agronomical evaluations. Genetic Resources and Crop evolution:539-550. Upadhyaya, H.D. 2003b. Geographical patter ns of variation for morphological and agronomic characteristics in the chic kpea germplasm collection. Euphytica 123:343352. Upadhyaya, H.D., P.J. Bramel, R. Ortiz, and S. Singh. 2002. Geographical patterns of diversity for morphological and agronomic traits in the groundnut germplasm collection. Euphytica:191-204.

PAGE 152

138 Valente, S.E.S., C.R. Lopes, and J.F.M.Valls 1998. Restricted isoenzimatic variation in acessions of Arachis pintoi (Leguminosae). In Proc. International Plant and Animal Genome Conference. 6th., San Diego, CA. Valentim, J.F. 1994. Adapation, productivity an d seasonal distribution of forage herbage mass in Arachis sp. Germplasm in Acre. In Proc. da Reunio Anual da Soc. Bras. Zootc. 31st., Maring, PR, Brazil, EDUEM. Valles, B., E. Castillo, and T. Hernndez. 1992. Seasonal production of forage legumes in Veracruz, Mxico. Pastur as Tropicales 14: 32-36. Valls, J.F.M. 1988. Morphological characteri zation, reproductive and biochemical of vegetative germplasm. In Proc. do Encontro sobre Recursos Genticos. Jaboticabal, SP, Brazil, FCAV. Valls, J.F.M. 1992. Origin of the germplasm of Arachis pintoi available in Brazil. In Proc.Reunin SabanasRe d Internacional de Evalua cin de Pastos Tropicales (RIEPT). Braslia, Brazil,: CIAT. Valls, J.F.M. and E.A. Pizarro. 1994. Collection of wild Arachis germplasm. p. 19-27. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Valls, J.F.M. and C.E. Simpson. 1994. The taxon omy, natural distribution and attributes of Arachis p. 1-18. In P.C. Kerridge and B. Hardy (ed.). Biology and Agronomy of Forage Arachis CIAT, Cali, Colombia. Watson, J.D. and F.H.C. Crick. 1953. Molecular structure of nucleic acids. Nature 171: 737-738. Westman, A.L. and S. Kresovich. 1997. Us e of molecular marker techniques. In J.A. Callow, B.V. Ford-lloyd, and H.J. New bury (ed.).Biotechnology and plant genetic resources: conservation and use. CAB International, Wallingford.. Williams, J.G., A.R. Kubelik, K.J. Livak, J.A. Rafalski, and S.V. Tingey. 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res 18:6531-5. Williams, M.J., C.G. Chambliss, and J.B. Br olmann. 1993. Potential of 'Savanna' stylo as a stockpiled forage for the subtropical US A. Journal of Production Agriculture 6:553556. Wynne, J.C., and T. Halward. 1989. Cytogeneti cs and genetics of Arachis. Crit. Rev. Plant Sci. 8:189-220.

PAGE 153

139 BIOGRAPHICAL SKETCH Marcelo Ayres Carvalho, the older son of Milton de Sousa Carvalho and Magaly Ayres Carvalho, was born in Brasilia, DF, Brazil, on May 4th 1965. When Marcelo was 12, his father died in a car accident, and hi s mother moved to her parents house, where he and his sister Patricia grew up under thei r mother and grandparent s guardianship. He inherited his fathers interest in agronomy on several visits during school holidays to his grandfather Policarpos farm in the north of Minas Gerais state. In 1983, Marcelo began his college studies in Agronomy at the Univ ersity of Brasilia, where he received his Agronomist degree in 1988. During his time in college he spent 1 year in EMBRAPA doing an internship in the forage department. In 1990, Marcelo was hired as a Research Assistant at the co llaborative Project EMBRAPA\CIAT\IICA, where he worked under the supervision of Dr. Esteban Pizarro conducting research with tropical grasses and legumes. The time spent with Dr. Pizarro was extremely important in establishing Marcelos passion for agronomic research. Encouraged by Dr. Pizarro, Marcelo received his M.S. degree in Agronomy from the University of Brasilia in 1996, working unde r the guidance of Dr. Jose F.M. Valls. In 1995, he was hired by EMBRAPA to wo rk at the Cerrados National Research Center located in Brasilia, where he is res ponsible for research projects with genetic resources and evaluation of tropica l forage legumes and grasses. In 2001, after being awarded with schola rships from EMBRAPA and CNPq, he started his Ph.D. program at the Agronomy depart ment of the University of Florida. He is

PAGE 154

140 studying evaluation of ge netic resources of Arachis pintoi under the guidance of Dr. Kenneth H. Quesenberry. After the conclusion of his program he will return to Brazil, where he will reassume his position in EMBRAPA. Marcelo married his lovely wife Aline in September of 1998. Aline gave him her entire support and dedication during the time spent at the University of Florida.