Citation
The North American Plums (Prunus Spp.) and Their Use as Germplasm Resources

Material Information

Title:
The North American Plums (Prunus Spp.) and Their Use as Germplasm Resources From Population to Phylogenetic Studies - a Breeder's Perspective -
Creator:
Chavez Velasquez, Dario Javier
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (464 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Horticultural Sciences
Committee Chair:
CHAPARRO,JOSE XAVIER
Committee Co-Chair:
OLMSTEAD,JAMES W
Committee Members:
SOLTIS,PAMELA S
KIRST,MATIAS
Graduation Date:
12/13/2013

Subjects

Subjects / Keywords:
Breeding ( jstor )
Chills ( jstor )
Diploidy ( jstor )
Haplotypes ( jstor )
Majority rule ( jstor )
Maximum likelihood estimations ( jstor )
Parsimony ( jstor )
Peaches ( jstor )
Plums ( jstor )
Species ( jstor )
Horticultural Sciences -- Dissertations, Academic -- UF
breeding -- cpdna -- endodormancy -- its
City of Gainesville ( local )
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
Horticultural Sciences thesis, Ph.D.

Notes

Abstract:
NorthAmerica is a center of diversity for Prunusspecies. Tree architecture, endodormancy requirement, heat requirement,fruit development period, fruit size, fruit texture, fruit flesh, diseaseresistance, and adaptive changes to multiple environmental conditions, are afew examples of the tremendous genetic variability available in the plum germplasm. Wild native Prunus species constitutes an importantsource for genetic diversity for stone fruit breeding and selection. The studyof genetic variability within the subgenus Prunussection Prunocerasus was theprimary objective of this research. Analyses recovered the major relationshipswithin section Prunocerasus. Highvariability was found across the major groups. In addition, possible candidatesof selection for adaptation were identified that could be transferred to otherstone fruits, such as peaches, apricots, and cherries. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
General Note:
Description based on online resource; title from PDF title page.
General Note:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
General Note:
Adviser: CHAPARRO,JOSE XAVIER.
General Note:
Co-adviser: OLMSTEAD,JAMES W.
General Note:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-12-31
Statement of Responsibility:
by Dario Javier Chavez Velasquez.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2016
Resource Identifier:
907646428 ( OCLC )
Classification:
LD1780 2013 ( lcc )

Downloads

This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EB8U189DU_P5PHHF INGEST_TIME 2014-05-13T16:59:54Z PACKAGE UFE0045247_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

THE NORTH AMERICAN PLUMS ( Prunus spp.) AND THEIR USE AS GERMPLASM RESOURCES: FROM POPULATION TO PHYLOGEN ETIC STUDIES A BREEDERS PERSPECTIVE By DARIO JAVIER CHAVEZ VELASQUEZ A DISSERTATION PRESENTED TO THE GRADUATE SCH OOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013 1

PAGE 2

2013 Daro Javier Chvez Velsquez 2

PAGE 3

To my savi o r Jesus Christ my love Rachel and my family 3

PAGE 4

ACKNOWLEDGMENTS I thank Rachel, my parents and grandparents for their love and support in my life. I thank Mr. and Mrs. Itle for their friendship and kindness I thank my advisor, Dr. Jos X. Chaparro, for his support on my research project and for his honest and true excitement of working with students and doing what he loves the most : plant breeding. I thank him, Dr. Paul Lyrene and Dr. Wayne Sherman, for teaching me to love plant breeding. I express gratitude to my c ommittee members, Dr. James Olmstead, Dr. Matias Kirst, and Dr. Pamela Soltis I thank Mark Gal John Thomas, Cecil Shine Rachel Odom and Kamille Chaparro, for their support in hard days of work Finally, I am grateful to My Lord Jesus Christ for giving me strength and companionship during this stage of my life. New goals and challenges appear at every turn but just remember that God will always be with you no matter what. I thank the Prunus Crop Germplasm Committee for their financial support for my ger mplasm collection, USDA ARS project No. 530621000 01800D. I thank Dr. Thomas Beckman, Dr. William Okie, Dr. John Preece, Dr. Thomas Gradziel, Dr. Cameron Peace and Mr. Kent Perkins for their logistic support. I thank Dr. Beatriz Pace Aldana from the Tig er Creek Preservation, Dr. Carl Weekley from the Archbold Biological Station, Dr. Kristen Kneifl from the Lake Wales Ridge NWRs, Ms. Nicole Ranalli from the Lake Wales Ridge Wildlife and Environmental Area Carter Creek North, Ms. Jennifer Navarra from t he Lake Wales Ridge State Forest, Dr. Cheryl Peterson from the Bok Tower Gardens, Ms. Jackie Rolly, Dr. Travis Marisco, Dr. Theo Witsell, Mr. Ray Erickson, Dr. George Johnson, Dr. Eric Sundell, Dr. Daniel Potter, Dr. Craig Ledbetter, Dr. David Ramming, Dr. Sternberg, Dr. Andrew Hipp, Dr. Kunso Kim, Mr. Rick Phillippe, Dr. Eric Grimm, Dr. Mark Widrlechner, Dr. Michael Dosmann from 4

PAGE 5

the Arnold Arboretum, Ms. Kathryn Richardson, Dr. Art Gilman, Dr. Brian Connolly, Dr. James Rassman, Dr. Mike Penskar, Ms. Alicia Selden, Dr. Glenn Palmgren, Dr. Sean Lynch from the Chrysler Herbarium, Dr. Timothy A. Block, Dr. Andrew St. John, Dr. Ann Rhoads, Dr. Isaac Bonnie, Mr. Marshall Enquist, Mr. Roger McCoy, Dr. Eugene Wofford, Mr. David Bamberger, Dr. Will McClatchey, Dr. A manda Neill, Dr. Joseph Rohrer, and Dr. Tom Wendt for their collaboration and knowledge provided for and during my collection trips 5

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 9 LIST OF FIGURES ........................................................................................................ 12 ABSTRACT ................................................................................................................... 25 CHAPTER 1 LITERATURE REVIEW .......................................................................................... 26 Introduction ............................................................................................................. 26 Taxonomic Treatment ............................................................................................. 27 Prunus Phylogenetic Studies .................................................................................. 29 Research Significance ............................................................................................ 33 2 GENETIC DIVERSITY IN PEACH [ Prunus persica (L.) BATSCH] AT THE UNIVERSITY OF FLORIDA: PAST, PRESENT AND FUTURE .............................. 43 Introduction ............................................................................................................. 43 Material and Methods ............................................................................................. 47 Plant Material ................................................................................................... 47 DNA Isolation ................................................................................................... 48 SSR Fingerprinting ........................................................................................... 49 Data Analyses .................................................................................................. 50 Genetic and Cluster Analyses .................................................................... 50 Population Structure .................................................................................. 50 Results .................................................................................................................... 51 Genetic and Cluster Analyses .......................................................................... 51 Population Structure ......................................................................................... 54 Discussion .............................................................................................................. 57 Conclusions ............................................................................................................ 59 3 THE NORTH AMERICAN PLUMS ( Prunus spp.) PHYLOGENETIC SIGNAL ........ 67 Introduction ............................................................................................................. 67 Material and Methods ............................................................................................. 71 Plant Material ................................................................................................... 71 DNA Isolation ................................................................................................... 72 Genomic Regions ............................................................................................. 73 SSR Markers .............................................................................................. 73 Chloroplast DNA ........................................................................................ 74 Internal Transcribed Spacer ....................................................................... 74 6

PAGE 7

Nuclear Genes ........................................................................................... 75 Sequencing ...................................................................................................... 76 Phylogenetic Analyses ..................................................................................... 76 Results .................................................................................................................... 79 SSRs ................................................................................................................ 79 Chloroplast DNA ............................................................................................... 81 Internal Transcribed Spacer ............................................................................. 82 Nuclear Genes ................................................................................................. 83 Isozymes .................................................................................................... 83 Branching ................................................................................................... 85 Flowering ................................................................................................... 89 Tot al Evidence .................................................................................................. 94 Discussion .............................................................................................................. 95 Conclusions .......................................................................................................... 100 4 THE PHYLOGENY OF THE NORTH AMERICAN PLUMS ( Prunus spp. L.) ........ 120 Introduction ........................................................................................................... 120 Materials and Methods .......................................................................................... 124 Collection ........................................................................................................ 124 Phylogenetic Study Material ........................................................................... 125 DNA Isolation ................................................................................................. 125 Molecular Data ............................................................................................... 126 Chloroplast DNA ...................................................................................... 127 Nuclear Genes ......................................................................................... 127 Sequencing .............................................................................................. 128 Phylogenetic Analyses ................................................................................... 128 Results .................................................................................................................. 131 Phylogenetic Analyses ................................................................................... 133 Discussion ............................................................................................................ 143 Conclusions .......................................................................................................... 150 5 GENETIC DIVERSITY AND POPULATION STRUCTURE OF PRUNUS UMBELLATA ELLIOT IN FLORIDA ...................................................................... 179 Introduction ........................................................................................................... 179 Material and Methods ........................................................................................... 181 Plant Material ................................................................................................. 181 DNA Isolation ................................................................................................. 182 SSR Markers .................................................................................................. 183 Data Analyses ................................................................................................ 184 Genetic and Cluster Analyses .................................................................. 184 Population Structure ................................................................................ 185 Detection of Loci Under Selection ............................................................ 185 Results .................................................................................................................. 186 Genetic and Cluster Analyses ........................................................................ 186 Population Structure ....................................................................................... 191 7

PAGE 8

Detection of Loci Under Selection .................................................................. 193 Discussion ............................................................................................................ 194 Conclusions .......................................................................................................... 200 6 CONCLUSIONS A BREEDERS PERSPECTIVE ........................................... 215 APPENDIX A COMPLEMENTARY TABLES AND FIGURES ..................................................... 221 LIST OF REFERENCES ............................................................................................. 452 BIOGRAPHICAL SKETCH .......................................................................................... 464 8

PAGE 9

LIST OF TABLES Table page 1 1 Cultivated and indigenous plums in North America by group, area of origin and cultivation. .................................................................................................... 34 1 2 Summary of Prunus phylogenetic studies. ......................................................... 36 2 1 Peach germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present. ................................................ 60 2 2 Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earl ygold peach (T E) reference map. ........................................... 61 2 3 Summary statistics of 36 SSR markers for 168 peach germplasm representatives of the UF varieties and advanced materials at the University of Florida stone fruit breeding and genetics program. ........................................ 62 2 4 Average statistics of 36 SSR markers for 170 germplasm representatives of the UF founder cultivars, varieties, and advanced materials at the U niversity of Florida stone fruit breeding and genetics program. ........................................ 63 3 1 List of the North American Plums (section Prunocerasus ) and other Prunus species collected and used in this project. ....................................................... 102 3 2 Summary statistics of 41 SSR markers of a subset of 11 Prunus species. ...... 103 3 3 Summary statistics of cpDNA regions sequenced for 11 Prunus species. ....... 104 3 4 Summary statistics of ITS and candidate gene regions sequenced for 11 Prunus species. ................................................................................................ 105 4 1 Li st of specimens used for the study of the phylogeny of the North American Plums ( Prunus spp.)z. ....................................................................................... 153 4 2 Summary statistics of trnH psbA PGI MAX4 PHYE, and VRN1 using haplotype phased sequences. .......................................................................... 157 5 1 List of the native plums (section Prunocerasus ) collected in the southeastern US and used for this study. .............................................................................. 202 5 2 S ummary statistics of 33 SSR markers for 101 genotypes of the North American Plums germplasmz available in the Stone Fruit Genetics and Breeding Program at University of Florida, Gainesville, FL. ............................. 205 9

PAGE 10

5 3 Average statistics of 33 SSR markers by species in the southeastern US and core collection. .................................................................................................. 206 5 4 Outlier locus detection analyses for pair wise comparison of phenotypic character istics classes of 33 SSR markers for 99 Prunus specimens representatives of the southestern US species. ............................................... 207 A 1 Phenotypic characteristics of the peach germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program. ...................................................... 221 A 2 Summary statistics of 36 SSR markers for 195 peach germplasm represen tatives of the genetic pools utilized in breeding and selection at the University of Florida stone fruit breeding and genetics program. ...................... 229 A 3 Summary of the population stratification simulat ion results of 36 SSR markers for 195 germplasm representatives of University of Floridas genetic pools. ................................................................................................................ 230 A 4 Analysis of the population structure results for 195 germplasm representatives of the University of Florida genetic pool using the Evanno method implemented in the Structure Harvester software. ............................... 231 A 5 Summary of the population stratification simulation results of 36 SS R markers for 168 peach UF varieties and advanced materials. .......................... 232 A 6 Analysis of the population structure results for 168 peach UF varieties and advanced materials using the Evanno method impl emented in the Structure Harvester software. .......................................................................................... 233 A 7 Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earlygold peach (T E) reference map. ......................................... 234 A 8 Primer sequences of cpDNA as described by Shaw et al. (2005, 2007) and Morris et al. (2008). .......................................................................................... 236 A 9 Primers for candidate genes associated with flowering time response, branching, plant architecture and isozymes in peach. ...................................... 237 A 10 Internal primers used to improve sequence coverage, quality and basecalling accuracy of gene regions associated with axillary meristem formation, dormancy response and isozyme expression. .................................................. 240 A 11 Summary statistics of candidate gene regions based on exon and intron boundaries for 11 Prunu s species. ................................................................... 241 A 12 Summary results of the phylogenetic analyses for diploid sequence data. ....... 244 10

PAGE 11

A 13 Summary results of the phyl ogenetic analyses for phased haplotype sequence data. ................................................................................................. 247 A 14 Models of sequence evolution of cpDNA sequence data using JModelTest for 11 Prunus species. ........................................................................................... 250 A 15 Models of sequence evolution of diploid ITS and candidate gene regions sequence data using JModelTest for 11 Prunus species. ................................ 251 A 16. Models of sequence ev olution of cpDNA sequence data (classified by haplotypes) using JModelTest for 11 Prunus species. ..................................... 259 A 17 Models of sequence evolution of phased haplotype ITS and candidate gene regions sequence data using JModelTest for 11 Prunus species. .................... 260 A 18 List of Prunus accessions collected and available at the Florida Museum of Natural History, Gainesville, FL. ....................................................................... 268 A 19 List of the Prunus accessions submitted to the USDA National Clonal Germplasm Repository in Davis, CA. ............................................................... 283 A 20 Summary statistics of PGI, MAX4 PHYE, and VRN1 using introns haplotype phased sequence information. .......................................................................... 285 A 21 Summary statistics of PGI, MAX4 PHYE, and VRN1 using exon haplotype phased sequence information. .......................................................................... 286 A 22 Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earlygold peach (T E) reference map. ......................................... 287 A 23 Phenot ypic information available for the southeastern US plums specimens. .. 289 A 24 Summary statistics of 33 SSR markers for 106 genotypes of the North American Plums germplasmz available in the Ston e Fruit Genetics and Breeding Program at University of Florida, Gainesville, FL. ............................. 292 A 25 Summary of the population stratification simulation results of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. .......................................................... 293 A 26. Analysis of the population structure results for 106 Prunus specimens representatives of the Prunus core c ollection and southeastern US species using the Evanno method implemented in the Structure Harvester software. .. 294 11

PAGE 12

LIST OF FIGURES Figure page 2 1 UPGMA cluster analysis based on Neis genetic distance and population structure results for k=2 to k=10 of 36 SSR markers for 195 g ermplasm representatives of the University of Florida stone fruit breeding program. .......... 64 2 2 Analysis of the population structure results for 195 germplasm representatives of the University of Florida genetic pool using the Evanno method implemented by the Structure Harvester software .. ............................... 66 3 1 Representatives of the North American plums ( Prunus spp.), subgenus Prunus section Prunocerasus and outgroup species. ...................................... 110 3 2 NJ cluster analysis based on Neis genetic distance for 11 Prunus species using SSR markers. Cladogram rooted with P. fasciculata (PfasCA131) at the base. ........................................................................................................... 112 3 3 Sequence percent variability, %= [(Subst+Indels)/L] 1 00, comparisons for Section Prunocerasus vs. outgroup species and e xons vs. introns gene regions. ............................................................................................................. 113 3 4 Phylogenetic analyses of combined cpDNA sequence data (3584 bp). Majority rule consensus trees using maximum parsimony maximum parsimony (including gaps) and m aximum likelihood. ..................................... 114 3 5 A non standarized multi locus combined c pDNA sequence data network ....... 115 3 6 Phylogenetic analyses of combined nuclear genes diploid sequence data (23429 bp). Majority rule consensus t rees using maximum parsimony, maximum parsimony (including gaps) and m aximum likelihood. ..................... 116 3 7 A non standarized multi locus combined nuclear genes sequence data network. ............................................................................................................ 117 3 8 Phylogenetic analyses of combined nuclear + cpDNA + ITS diploid sequence data (27623 bp). Majority rule consensus t rees using maximum parsimony, maximum parsimony (including gaps) and m aximum likelihood. ..................... 118 3 9 A non standarized multi locus combined nuclear + cpDNA + ITS sequence data network. .................................................................................................... 119 4 1 Maximum likelihood tree using RAxML for trnH psbA sequenc e data (553 bp) (lnL= 1243.68) ................................................................................................ 158 4 2 Maximum likelihood tree using RAxML for PGI diploid sequence data (457 bp) (lnL= 1294.72). .......................................................................................... 159 12

PAGE 13

4 3 Maximum likelihood tree using RAxML for MAX4 diploid sequenc e data (329 bp) (lnL= 1133.54) .......................................................................................... 160 4 4 Maximum likelihood tree using RAxML for PHYE diploid sequence data (553 bp) (lnL= 1294.97). .......................................................................................... 161 4 5 Maximum likelihood tree using RAxML for VRN1 diploid sequence data (560 bp) (lnL= 1247.39). .......................................................................................... 162 4 6 Maximum likelihood tree using RAxML for combined nuclear genes diploid sequence data (1859 bp) (lnL= 5267.71) ....................................................... 163 4 7 Maximum likelihood tree using RAxML for combined diploid sequence data total evidence approach (2452 bp) (lnL= 6204.00). ....................................... 164 4 8 Best maximum likelihood tree using RAxML for combined diploid sequence data total evidence approach (2452 bp) (lnL= 6204.00). ............................ 165 4 9 Maximum likelihood tree using RAxML for trnH psbA sequence classification with haplotype number prefix (553 bp) (lnL= 1243.68). ................................... 166 4 10 Haplotype network for trnH psbA sequence data (553 bp) considering gaps as a 5th character ............................................................................................. 167 4 11 Maximum likelihood tree using RAxML for PGI haplotype phased sequence data (459 bp) after removing minor frequency haplotypes (lnL= 1742.83). .... 168 4 12 Maximum likelihood tree using RAxML for MAX4 haplotype phased sequence data (364 bp) after removing minor frequency haplotypes (lnL= 1640.83).. ......................................................................................................... 170 4 13 Maximum likelihood tree using RAxML for PHYE haplotype phased sequence data (553 bp) after removing minor frequency haplotypes (lnL= 1536.13).. ......................................................................................................... 173 4 14 M aximum likelihood tree using RAxML for VRN1 haplotype phased sequence data (562 bp) after removing minor frequency haplotypes (lnL= 1976.81).. .... 176 5 1 Floridas map of distribution for Prunus americana Marsh., P. angustifolia Marsh., P. umbellata Elliot, and P. geniculata Harper.. .................................... 208 5 2 Specimes geographical distribution for population structure and genetic diversity studies of P. americana (orange), P. angustifolia (blue), P. geniculata (red), P. umbellata (yellow), and P. umbellatalike (purple). ............ 209 5 3 UPGMA cluster analysis based on Neis genetic distance and population structure results for k=2 to k=10 of 33 SSR markers for 106 Prunus specimen representatives of the Prunus core collect ion and southeastern US species .. 210 13

PAGE 14

5 4 Analysis of the populati on structure results for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species using the Evanno method implemented by the Structure Harvester software .. 212 5 5 A non standarized neighbor net network of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collect ion and southeastern US species ...................................................................................................... 213 A 1 UPGMA cluster analysis based on CavalliSforza chord distance of 36 SSR markers for 195 germplasm representatives of the University of Florida stone fruit breeding and genetics program. ................................................................ 295 A 2 NJ cluster anal ysis based on CavalliSforza chord distance of 36 SSR markers for 195 germplasm representatives of the University of Florida stone fruit breeding and genetics program. ................................................................ 296 A 3 NJ cluster analy sis based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida stone fruit breeding and genetics program. ....................................................................... 297 A 4 Peach (green) and nectar ine (black) fruit types genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ................................... 298 A 5 White (green) and yellow (black) fruit flesh color genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ................................... 299 A 6 Normal (green) and highlighter/reduced anthocyanin (black) genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ......... 300 A 7 Round (green) and peento (black) fruit shape genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ................................... 301 A 8 Reniform (turquoise), globose (green), and eglandular (black) leaf glands genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida stone fruit breeding and genetics program. .......................................... 302 A 9 Showy (green) and nonshowy (black) flower type genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ................................... 303 14

PAGE 15

A 10 Low chill <400 hours (turquoise), medium chill 400700 hours (green), and high chill >700 hours (black) genotypes traced over the UPGMA cluster based on Neis genetic distance of 36 SSR markers for 195 germplasm representatives of the University of Florida. ..................................................... 304 A 11 Genotypes with 0 150 chill hours (dark blue), 151250 chill hours (powder blue), 251350 chill hours (turquoise), 351450 chill hours (sea foam green), 451550 chill hours (green), 751850 chill hours (yellow), 851950 chill hours (orange), and 9511050 chill hours (blac k) traced over the UPGMA cluster based on Neis genetic distance for 195 germplasm representatives of the University of Florida. ......................................................................................... 305 A 12 Introduction and/or genotypes selection decades of 1890s (dark blue), 1940s (powder blue), 1950s (turquoise), 1960s (light blue), 1970s (sea foam green), 1980s (green), 1990s (yellow), 2000s (orange), and 2010s (black) traced over the UPGMA cluster based on Neis genetic distance for 195 germplasm representat ives of the University of Florida. ..................................................... 306 A 13 Analysis of the population structure results for 168 peach germplasm representatives of the UF varieties and advanced materials using the Evanno met hod implemented by the Structure Harvester software .............................. 307 A 14 UPGMA cluster analysis based on Neis genetic distance and population structure results for k=2 to k=10 of 36 SSR markers for 168 peach UF varieties and advanced materials. .................................................................... 308 A 15 Majority rule consensus trees for 3'trnV ndhC cpDNA region using: maximum parsimony, maximum pa rsimony (including gaps), and maximum likelih ood. .. 310 A 16 Majority rule consensus trees for trnL trnF and trnL intron cpDNA regions using : maximum parsimony and maximum likelihood. ...................................... 311 A 17 Majority rule consensus trees for trnQ 5rps16 cpDNA region using: maximum parsimony, maximum par simony (including gaps), and maximum likelihood. ......................................................................................................... 312 A 18 Majority rule consensus trees for trnH psbA cpDNA region using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood .. 313 A 19 Majority rule consensus trees for ndhF rpL32 cpDNA region u sing: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. .. 314 A 20 Majority rule consensus trees for atpB rbcL cpDNA region using: maximum parsimony, maximum parsimo ny (including gaps), and maximum likelihood. .. 315 A 21 Majority rule consensus trees for ITS diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 316 15

PAGE 16

A 22 Majority rule consensus trees for PGDH diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood ......................................................................................... 317 A 23 Majority rule consensus trees for PGI diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 318 A 24 Majority rule consensus trees for S6PDH diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 319 A 25 Majority rule consensus trees for AXR1 diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 320 A 26 Majority rule consensus t rees for BRC1 diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 321 A 27 Majority rule consensus trees for BRC2 diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 322 A 28 Majority rule consensus trees for CUC1A diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 323 A 29 Majority rule consensus trees for CUC1B diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood ......................................................................................... 324 A 30 Majority rule consensus trees for CUC2 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 325 A 31 Majority rule consensus trees for CUC3 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 326 A 32 Majority rule consensus trees for LAS diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 327 A 33 Majority rule consensus trees for MAX1A diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 328 A 34 Majority rule consensus trees for MAX1B diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 329 16

PAGE 17

A 35 Majority rule consensus trees for MAX2 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 330 A 36 Majority rule consensus trees for MAX3 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 331 A 37 Majority rule consensus trees for MAX4 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 332 A 38 Majority rule consensus trees for PIN diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 333 A 39 Majority rule consensus trees for RAX1 diploid and haplotypic sequence data using: maximum parsimony, maximum pars imony (including gaps), and maximum likelihood. ......................................................................................... 334 A 40 Majority rule consensus trees for RAX2RAX3 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ....................................................................... 335 A 41 Majority rule consensus trees for REV diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum li kelihood. ......................................................................................... 336 A 42 Majority rule consensus trees for SPS diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 337 A 43 Majority rule consensus trees for AGL24 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 338 A 44 Majority rule consensus trees for AGL20SOC1 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ....................................................................... 339 A 45 Majority rule consensus trees for BFT diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 340 A 46 Majority rule consensus trees for BRM diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 341 17

PAGE 18

A 47 Majority rule cons ensus trees for CO diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 342 A 48 Majority rule consensus trees for CRY1 di ploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 343 A 49 Majority rule consensus trees for CRY2 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 344 A 50 Majority rule consensus trees for ELF6 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 345 A 51 Majority rule consensus trees for FD diploid and haplotypic sequence data using: maximum parsimony, maxi mum parsimony (including gaps), and maximum likelihood. ......................................................................................... 346 A 52 Majority rule consensus trees for FD1 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (includi ng gaps), and maximum likelihood. ......................................................................................... 347 A 53 Majority rule consensus trees for FG diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum l ikelihood. ......................................................................................... 348 A 54 Majority rule consensus trees for FLC FLF diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 349 A 55 Majority rule consensus trees for FPF1 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 350 A 56 Majority rule consensus trees for FRIGIDA diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 351 A 57 Majority rule consensus trees for FT TSF diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 352 A 58 Majority rule consensus trees for GI FB diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 353 18

PAGE 19

A 59 Majority rul e consensus trees for HOS1 diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 354 A 60 Majority rule consensus trees for LFY diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 355 A 61 Majority rule consensus trees for MAF1MAF3 AGL31 diploid and haplotypic sequence data using : maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ....................................................................... 356 A 62 Majority rule consensus trees for MAF2A diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 357 A 63 Majority rule consensus trees for MAF2B diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony ( including gaps), and maximum likelihood. ......................................................................................... 358 A 64 Majority rule consensus trees for MAF4 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 359 A 65 Majority rule consensus trees for MAF5 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 360 A 66 Majority rule consensus trees for MFT diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 361 A 67 Majority rule consensus trees for PHYA diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 362 A 68 Majority rule consensus trees for PHYB PHYD diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ....................................................................... 363 A 69 Majority rule consensus trees for PHYE diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 364 A 70 Majority rule consensus trees for RGA RGA1 di ploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ............. 365 A 71 Majority rule consensus trees for RGL1RGL2 RGL3 diploid and haplotypic sequence data using: maximu m parsimony, maximum parsimony (including gaps), and maximum likelihood. ....................................................................... 366 19

PAGE 20

A 72 Majority rule consensus trees for SPY diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 367 A 73 Majority rule consensus trees for TFL1ATC diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (includi ng gaps), and maximum likelihood. ......................................................................................... 368 A 74 Majority rule consensus trees for TFL2 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 369 A 75 Majority rule consensus trees for VRN1 diploid and haplotypic sequence data using: maximum parsimony and maximum likelihood. ...................................... 370 A 76 Majority rule consensus trees for VRN2 diploid and haplotypic sequence data using: maximum parsimony, maximum parsimony (including gaps), and maximum likelihood. ......................................................................................... 371 A 77 Phylogenetic analyses of combined isozymes diploid sequence data (1399 bp). Majority rule consensus trees using maximum parsimony maximum parsimony (including gaps) and m aximum likelihood. .................................... 372 A 78 A non standarized multi locus combined isozymes genes sequence data network. ............................................................................................................ 373 A 79 Phylogenetic analyses of combined branching genes diploid sequence data (6659 bp). Majorit y rule consensus t rees using maximum parsimony, maximum parsimony (incl uding gaps), and m aximum likelihood. ..................... 374 A 80 A non standarized multi locus combined branching genes sequence data networ k. ............................................................................................................ 375 A 81 Phylogenetic analyses of combined dormancy related genes diploid sequence data (15371 bp). Majority rule consensus t rees using maximum parsimony, maximum parsimony (including gaps) and m aximum likelihood. .. 376 A 82 A non standarized multi locus combined dormancy related genes sequence data network. .................................................................................................... 377 A 83 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data (885 characters) including gaps. ............................................................................... 378 A 84 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data (553 bp) without gaps. .................................................................................................... 379 20

PAGE 21

A 85 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI diploid sequence data (525 characters) including gaps. ............................................................................... 380 A 86 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI diploid sequence data (457 bp) without gaps. .................................................................................................... 381 A 87 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid p arsimony analysis of MAX4 diploid sequence data (393 characters) including gaps. ............................................................................... 382 A 88 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analys is of MAX4 diploid sequence data (329 bp) without gaps. .................................................................................................... 383 A 89 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE diploid sequence data (659 characters) including gaps. ............................................................................... 384 A 90 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE diploid sequence data (553 bp) without gaps. .................................................................................................... 385 A 91 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 diploid sequence data (564 characters) including gaps. ............................................................................... 386 A 92 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 diploid sequence data (560 bp) without gaps. .................................................................................................... 387 A 93 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined nuclear genes diploid sequence data (2101 characters) including gaps ............................................ 388 A 94 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined nuclear genes diploid sequence data (1859 bp) without gaps. ............................................................ 389 A 95 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined diploid sequence data total evidence approach (3026 characters) including gaps ............................ 390 A 96 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined diploid sequence data total evidence approach (2452 bp) without gaps. ........................................... 391 21

PAGE 22

A 97 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data with haplotype number prefix (885 characters) including gaps. ............................... 392 A 98 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA haplotype phased sequence data (553 bp) without gaps. .............................................................. 393 A 99 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (550 characters) including gaps. ....................................................................... 394 A 100 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PGI haplotype phased sequence data (549 characters) after removing minor frequency haplotypes and including gaps. .............................. 397 A 101 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PGI haplotype phased sequence data (459 bp) without gaps. ........ 399 A 102 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (459 characters) after removing minor frequency haplotypes. .......................... 402 A 103 Maximum likelihood tree using RAxML for PGI haplotype phased sequence data (459 bp) (lnL= 1967.55). .......................................................................... 404 A 104 Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 haplotype phased sequence data (466 characters) including gaps ............................................................... 406 A 105 Majority rule consensus tree of t he MPTs using the parsimony ratchet method of MAX4 haplotype phased sequence data (466 characters) after removing minor frequency haplotypes and including gaps. .............................. 409 A 106 Majority rule consens us tree of the MPTs using the parsimony ratchet method of MAX4 haplotype phased sequence data (364 bp) without gaps. ..... 411 A 107 Majority rule consensus tree of the MPTs using the parsimony ratc het meth od of MAX4 haplotype phased sequence data (364 bp) after removing minor frequency haplotypes. ............................................................................. 414 A 108 Maximum likelihood tree using RAxML for MAX4 haplotype phased sequenc e data (364 bp) (lnL= 1902.66) ......................................................... 417 A 109 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PHYE haplotype phased sequence data (661 characters) including gaps. .................................................................................................. 420 22

PAGE 23

A 110 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PHYE haplotype phased sequence data (648 characters) after removing minor frequency haplotypes and including gaps. .............................. 422 A 111 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PHYE haplotype phased sequence data (553 bp) without gaps. ..... 424 A 112 Majority rule consensus tree of the MPTs using the parsimony ratchet method of PHYE haplotype phased sequence data (553 bp) after removing minor frequency haplotypes. ............................................................................. 426 A 113 Maximum likelihood tree using RAxML for PHYE haplotype phased sequence data (553 bp) (lnL= 1550.06). ......................................................... 428 A 114 Majority rule consensus tree of the MPTs using the parsimony ratc het method of VRN1 haplotype phased sequence data (570 characters) including gaps. ................................................................................................................. 431 A 115 Majority rule consensus tree of the MPTs using the parsimony ratchet method of VRN1 haplotype phased sequence data (570 characters) after removing minor frequency haplotypes and including gaps. .............................. 433 A 116 Majority rule consensus tree of the MPTs using the parsimony ratchet method of VRN1 haplotype phased sequence data (562 bp) without gaps. ..... 435 A 117 Majority rule consensus tree of the MPTs using the parsimony ratchet method of VRN1 haplotype phased sequence data (562 bp) after r emoving minor frequency haplotypes. ............................................................................. 437 A 118 Maximum likelihood tree using RAxML for VRN1 haplotype phased sequence data (562 bp) (lnL= 2094.19). .......................................................................... 439 A 119 UPGMA cluster analysis based on Neis genetic distance of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. ................................................................................. 442 A 120 NJ cluster analysis based on Neis genetic distance of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. ................................................................................. 443 A 121 Prunus americana (dark blue), P. angustifolia (powder blue), P. geniculata (turquoise), P. umbellatalike (sea foam green), P. umbellata (yellow), and other Prunus species (black) traced over the UPGMA cluster. ......................... 444 23

PAGE 24

A 122 Low chill <400 hours (dark blue), medium chill 400700 hours (green), medium high chill 7001110 hours (yellow), and high chill >1110 hours (black) genotypes traced over the UPGMA cluster based on Neis genetic dist ance of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. ................................ 445 A 123 Genotypes with 201300 chill hours (dark blue), 301400 chi ll hours (powder blue), 401500 chill hours (light blue), 501600 chill hours (turquoise), 601700 chill hours (sea foam green), 701800 chill hours (green), 801900 chill hours (yellow), 11001200 chill hours (orange), and >1400 chill hours (black) traced over the UPGMA cluster based on Neis genetic distance of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. .......................................................... 446 A 124 A non st andarized neighbor net network of 33 SSR markers for 99 Prunus specimens representatives of the southestern US species. Colors represent population structure results with k=5. ............................................................... 447 A 125 Population structure result for k=5 and k=7 separated by species of 33 SSR markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. .......................................................... 449 A 126 Fsto utlier detection using the Beaumont and Nicholss modified frequentist method of 33 SSR markers for 69 Prunus umbellata specimens. .................... 450 A 127 Posterior probability calculations to selection usi ng a Bayesian method and a reversiblejump MCMC approach of 33 SSR markers for 69 Prunus umbellata specimens. ....................................................................................... 451 24

PAGE 25

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Pa rtial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE NORTH AMERICAN PLUMS ( Prunus spp.) AND THEIR USE AS GERMPLASM RESOURCES: FROM POPULATION TO PHYLOGENETIC STUDIES A BREEDERS PERSPECTIVE By Dar o Javier Ch vezVel squez December 2013 Chair: Jos X. Chaparro Major: Horticultural Sciences North America is a center of diversity for Prunus species. Tree architecture, chilling requirement, heat requirement, fruit development period, fruit size, fruit texture, disease resis tance, and adaptive changes to multiple environmental conditions, are a few examples of the traits of which tremendous genetic variability is available in the native plum species. Wild native Prunus species constitute an important potential source of genet ic diversity for stone fruit breeding and selection. The study of genetic variability within the subgenus Prunus section Prunocerasus was the primary objective of this research. Analyses recovered the primary species relationships within section Prunoceras us High levels of genetic variation were found across the major groups. In addition, possible candidates of selection for soil and climate adaptation were identified that are potentially useful in rootstock and scion breeding programs for peaches, apricot s, and plums 25

PAGE 26

CHAPTER 1 LITERATURE REVIEW Introduction The genus Prunus L. belongs to the subfamily Amygdaloideae (=Prunoideae) of the Rosaceae. It has worldwide distribution with approximately 200 species. Edible species are mostly distributed in the northern hemisphere (Bortiri et al., 2001; Hedrick, 1911; Rehder, 1940; Sargent, 1905; Wight, 1915). The genus contains species that are important in the production of fruit, nuts, and lumber. Plums, cherries, almonds, apricots and peaches are the most comm o nl y known fruit and nuts in this genus. The worlds net production of almonds, apricots, cherries, peaches, nectarines, plums and sloes in 2010, was approximately 40.8 million tons. Peach and nectarine production was the largest in the world with 20.5 mill ion tons. US peach and nectarine production was approximately 1.3 million tons, with a farm gate value of ~683 million dollars (FAOSTAT, 2010). North America is an important center of diversity for plum species adapted (native) to widely divergent climates and soils represent ing an important potential source of genes for plant breeding Layne and Bassi (2008) reported high levels of variation in the Prunus germplasm for tree size, growth habit, flower size and color, chill hour requirement, fruit size, fles h texture, flesh color, resistance to diseases, and adaptability to diverse climatic and geographic regions. Plums are the stonefruit with the greatest diversity of flavor, aroma, texture, color, form and size (Hedrick, 1911; Waugh, 1901). Stone fruit breeders have used this tremendous genetic variability through inter specific hybridizations (in particular with species in the subgenus Prunus or 26

PAGE 27

Prunophora) for the improvement of Prunus scion and rootstock cultivars (Layne and Sherman, 1986). Among those, native North American plum species have been identified as a source of resistance to blossom blight and brown rot ( Monilinia fructicola ), bacterial spot ( Xanthomonas campestris pv. pruni ), bacterial canker ( Pseudomonas syringae pv. syringae), plum leaf sca ld ( Xylella fastidiosa ), peach tree short life (PTSL), root knot nematode ( Meloidogyne spp.), lesion nematode ( Pratylenchus spp.), clitocybe root rot ( Armillaria tabescens ), and others (Beckman and Okie, 1994; Beckman et al., 1998; Layne and Sherman, 1986; Okie and Weinberger, 1996). Resistance to bacterial leaf spot and bacterial canker was identified in a cultivar derived from P. salicina Lindl. P. cerasifera Ehrh. P. angustifolia Marshall P. americana Marshall and P. munsoniana W. Wight & Hedrick P runus hortulana L.H. Bailey was found resistant to root knot nematode, and has been recommended as a rootstock for European plums. Improved tolerance for PTSL was found in hybrids from P. americana, P. hortulana, P. angustifolia and/or P. umbellata Elliot P otential uses of the native North American plum species as breeding parents, scions and/or rootstocks were summarized by Beckman and Okie (1994), and Okie and Weinberger (1996). Taxonomic Treatment Waugh (1901) described the genus Prunus as trees or s h rubs, mostly with edible fruit ; flowers, white or pink, with spreading petals. Stamens 1530, distinct, with filiform filaments. Style, terminal; stigma, usually truncate. The fruit with a fleshy exterior, glabrous, and containing a hard bony pit, which contains the seed. Inconsistencies in the taxonomy of Prunus were recognized by Waugh (1901) and Hedrick (1911). Bortiri et al. (2001) summarized the classification discrepancies in 27

PAGE 28

Prunus as follows: 1) four different genera: Amygdalus Cerasus Prunus and Padus (Linnaeus, 1737), and later two: Amygdalus and Prunus (Linnaeus, 1754); 2) Five genera: Amygdalus Persica Prunus Armeniaca, and Cerasus (including Padus and Laurocerasus ) (De Candolle, 1825); 3) Prunus as a single genus divided in seven section s: Amygdalus Armeniaca, Prunus Cerasus Laurocerasus Ceraseidos and Amygdalopsis (Bentham and Hooker, 1865); 4) Prunus with previous seven sections as subgenera (Koehne, 1893); 5) Prunus classified into five subgenera: Prunophora ( Prunus ), Amygdalus C erasus Padus and Laurocerasus and with subgenus Prunus divided in three sections: Euprunus Prunocerasus and Armeniaca (Rehder, 1940); and 6) Prunus divided into three genera: Padus Laurocerasus and Prunus (Hutchinson, 1964). Recently, the concept o f Prunus as single genus has become widely accepted but subgenera classification is still undistinguished as new phylogenetic relationships within Prunus come to light. The USDA GRIN (2010) germplasm collection organizes the genus Prunus into subgenus Amy gdalus Cerasus Emplectocladus and Prunus Subgenus Cerasus was divided into sections Cerasus and Laurocerasus and subgenus Prunus into sections Armeniaca, Microcerasus (including some plums) Penarmeniaca, Prunocerasus (the North American plums), and P runus Waugh (1901) recognized the difficulty in classifying the North American plums and stated plums grow pretty much as they please, and the botanist has to take them as he finds them The distribution, cultivation, hybridization, and breeding value o f native plums has been extensively studied (Britton and Shafer, 1908; Hedrick, 1911; Mason, 1913; Sargent, 1905; Waugh, 1901; Wight, 1915). 28

PAGE 29

Waugh (1901) classified the cultivated and indigenous Prunus of North America into groups. These groups were clust ered into seven series: Americana, Chickasaw, Hortulana, Maritima, Sand Cherry, Choke Cherry, and Black Cherry (Waugh, 1899; Table 11 ). The Americana series included: the Americana group (including P. americana var. lanata) and the Nigra group ( Prunus nigra Aiton) The Chickasaw series included : the Chickasaw and the Sand plum groups. The Hortulana series, categorized as hybrids, included the Wildgoose group, the Wayland group, and the Miner group. The Maritima series: the Beach plum group, the Southern sloe group [ including P. umbellata Elliot var. injuncunda (Small) Sarg.] the Oklahoma plum group, and P. glandulosa Thunb. (ungrouped). The Sand Cherry series were equivalent to the Dwarf cherries group. The Choke Cherry and the Black Cherry series conser ve d their name as groups (Waugh, 1899, 1901; Table 11 ). Wight (1915) separated the genus Prunus in plums, cherries and dwarf cherries. Waughs (1899, 1901) taxonomic treatment included cherries as part of plums. Wights (1915) groups/series were: Americana, Subcordata, Hortulana, Angustifolia, Maritima, and Gracilis. The Angustifolia group agreed with Waughs (1899) Chickasaw series. Waugh (1899) did not include P. mexicana S. Watson (Americana group), P. munsoniana (Angustifolia group), P. subcordata Ben th. (Subcordata group), P. alleghaniensis Porter (Maritima group), and P. umbellata (Maritima group), as part of his groups/series. Prunus Phylogenetic Studies Phylogeny and systematics in the genus Prunus was reported by Mowrey and Werner (1990).They employed isozymes to study the phylogenetic relationships in Prunus Section Prunocerasus was found to be polyphyletic, with a clade formed by P. 29

PAGE 30

americana, P. munsoniana, P. hortulana, P. subcordata, and P. angustifolia and a clade formed by P. maritima Marshall and P. umbellata. Chloroplast DNA is an alternative source of genetic variation and is maternally inherited in Prunus Chloroplast DNA is highly conserved and in relative abundance in the cell as compared with the nuclear DNA. Kaneko et al. (1986) an d Uematsu et al. (1991) used cpDNA to classify cherries, apricots, and wild and cultivated peaches in Japan. In 1995, Badenes and Parfitt reported a phylogeny similar to Mowrey and Werner (1990). All the Prunus species were grouped as in conventional subgenus classifications (Rehder, 1940). Prunus persica L. P. dulcis (Mill.) D.A. Webb P. domestica L. P. salicina Lindl. and P. cerasus L. P.fruticosa Pall were monophyletic. Lee and Wens (2001) phylogenetic analysis of the genus Prunus using ITS sequences recognized two major groups: the Amygdalus Prunus group, and the Cerasus Laurocerasus Padus group. The results were not congruent with Rehders (1940) taxonomic treatment In Bortiri et al. (2001) the p hylogeny and systematics of Prunus based on ITS and chloroplast trnL trnF spacer DNA sequences identified t wo major clades: subgenera Padus Laurocerasus Cerasus and subgenera Prunus Amygdalus Emplectocladus Cerasus (sect. Microcerasus ) sect. Penarmeniaca [ similar to Mowrey and Werner (1990), Lee and Wen (2001) and Bortiri et al. (2001) ] The ir results indicated that plums of northeastern North America were closely related, and that P. mexicana belonged to a sister clade. Bortiri et al. (2002) used the nuclear gene s6pdh, which encodes NADP+ dependent sorbitol 6 phosphate dehydrogenase, to assess the lack of support for deep 30

PAGE 31

nodes in the clade subgenera Prunus Amygdalus Emplectocladus (as reported in previous data). The phylogenies based on ITS, cpDNA trnL trnF and s6pdh sequences were compared and combined. Phylogenetic analysis of the combined data supported two major clades: subgenera Cerasus Laurocerasus Padus and subgenera Amygdalus Emplectocladus Prunus Section Microcerasus (subgenera Cerasus ) was reported nested within subgenus Prunus Prunus subg. Prunu s sect. Prunocerasus was reported to be monophyletic by Shaw and Small (2004). The phylogenetic analysis was based on seven cpDNA regions: rpS16 rpL16, trnL trnG trnL trnF trnS trnG and trnH psbA Three clades were strongly supported in sect. Prunocer asus : the American Clade, the Chickasaw Clade, and the Beach Clade (names based on Waughs (1901) classification). The American clade included P. americana Marshall var. americana Sudw. P. americana Marshall var. lanata P. mexicana, P. rivularis Sc heele, P. hortulana, P. umbellata var. injucunda; the Chickasaw clade included P. angustifolia, P. munsoniana, P. gracilis Engelm. & A. Gray P. nigra, P. umbellata Elliot var. umbellata, P. alleghaniensis Porter var. alleghanienses and P. alleghaniensis Porter var. davisii (W. Wight) Sarg. ; and the Beach clade included P. geniculata Harper P. maritima Marshall var. maritima and P. maritima Marshall var. gravesii (Small) G.J. Anderson Similarly, a survey of cpDNA haplotypes available within sect ion Prun ocerasus was reported by Shaw and Small (2005). The cpDNA rpL16 region was sequenced for 207 accessions representatives of 17 North American Plums, including P. texana D. Dietr. (as described before). More than one of the three primary cpDNA haplotypes w as found in many of the taxa. 31

PAGE 32

Bortiri et al. (2006) studied the evolution of vegetative and morphological characters of 37 species of Prunus and other genera of Rosaceae. Morphological characters were combined with ITS, trn L trn F, and trn S trn G data from pr evious studies (Bortiri et al., 2001; Bortiri et al., 2002). The addition of the morphological data with trn S trn G supported some nodes that were found in ITS and trn L trn F studies. Three clades were reported: Clade A with subgenera Padus and Laurocerasus; Clade B with subgenera Amygdalus Emplectocladus and Prunus ; and Clade C with subgenera Cerasus Clade B was characterized by the production of three axillary buds. Padus and Laurocerasus were not supported as monophyletic (high homoplasy). Genet ic diversity within Prunus cerasifera (cherry plum) was studied using morphological characters, cytometry, cpDNA, and SSR markers (Horvath et al., 2008). Morphological characters showed differences between clones. Analysis of cpDNA reported fifteen haploty pes clustered in three groups. Considerable diversity among accessions was reported based on these studies. Endocarp and leaf morphometrics combined with AFLP markers were used to study the morphologic al and genetic variation of five European members of section Prunus : P. cerasifera, P. cocomilia Ten., P. domestica P. insititia L. P. spinosa L. and P. fruticans (Depypere et al., 2009). Three clusters were reported: a first cluster P. cerasiferaP. cocomilia a second P. domesticaP. insistia and a thi rd P. spinosa and P. fruticans Phylogenetic analysis based on four singlecopy cpDNA regions ( atp B rbc L, mat K, rpl 16, and trn L trn F) of Eurasian plums, Prunus section Prunus confirmed this section to be monophyletic. Four well supported clades were rep orted : Clade A with P. 32

PAGE 33

salicina P. sogdiana, and P. ussuriensis ; Clade B with P. cocomilia ; Clade C with P. brigantina, P. ramburii and P. spinosa; and Clade D with subclade D1 P. domesticaP. insititia P. divaricataP. ursine and subclade D2 P. cerasifera (Reales et al., 2010). Previous studies demonstrated the value of morphology, cytometry, nuclear DNA and cpDNA as data for phylogenetic studies in Prunus. Most of the previous phylogenetic research used Mason (1913) and Rehders (1940) taxonomic classification. Hereafter, in our study the North American plums will be referred to as sect. Prunocerasus (Bortiri et al., 2001; Shaw and Small, 2004). A complete summary of Prunus phylogenetic research is summarized i n Table 12. Research Significance U rban sprawl and agriculture have impacted the survival of several of the North American Prunus species In addition, climate change threatens some species habitat and growing environment. The study of the subgenus Prunus (in particular section Prunocerasus) constituted the main objective of this research, including collection and identification of the plant specimens from the wild to clarify their phylogenetic framework (relatedness, evolution, and character/trait diversification). Our objective was to col lect, identify, and archive vouchers for the collected North American plum species, thereby avoiding any misidentification of plant specimens/collections (vouchers and/or germplasm collections) that could undermine our research. The study of the subgenus P runus sect. Prunocerasus allowed us to identify, to support, and to preserve these species as important genetic resources (gene pool) of unique traits that could be used in the near future. 33

PAGE 34

Table 11 Cultivated and indigenous pl ums in North America by group, area of origin and cultivation (Waugh, 1901). Group Species Origin Cultivation Cutivated Domestica plums Prunus domestica Eastern Europe and west central Asia Nova Scotia, central New England, New York, southern Ontario and Michigan, and the Pacific coast states. Damsons Prunus domestica Europe Myrobalan plums Prunus cerasifera Europe and US used as rootstock Simon plums Prunus simonii China New York, California Japanese plums Prunus triflora China, Japan Maine, V ermont, Ontario, and southern Iowa Indigenous Americana group Prunus americana USA (Ohio, Texas, northward to Minnesota and Montana). Prince Edward Island, Manitoba, and Vancouver, to Florida, Louisiana, and Texas. Nigra group Prunus americana nigra CA N (Newfoundland west to Rainy and Assiniboine rivers), USA (New England states) Prince Edward Island, Manitoba, and Vancouver, to Florida, Louisiana, and Texas. Miner Group Prunus hortulana mineri USA (Standing between P. americana and the Wildgoose grou p) No t cultivation Wayland group Prunus rivularis z Prunus hortulana y USA (Colorado, Guadalupe and the Leona) North of Burlington, Vermont, and Iowa. Wildgoose group Prunus hortulana USA (Mississippi valley) From Texas to Massachusetts. Chickasaws P runus angustifolia USA (Southern range to Delaware and Kentucky, including southern Atlantic and Gulf states) Iowa, Vermont, New York, and Massachusetts. Sand plum Prunus angustifolia watsonii USA (South and southeast Nebraska and central and western Kan sas) Cultivated by settlers in Kansas, and Maryland Beach plum Prunus maritima USA (Sea beaches, New Brunswick to Virginia, Georgia, Alabama, and Connecticut) No t cultivation 34

PAGE 35

zClassified as Prunus rivularis but with doubts. yPrunus hortulana consider as part of the Wayland and the Wildgoose group. Table 1 1. Continued. Group Species Origin Cultivation Indigenous Pacific plum Prunus subcordata USA (Pacific coast) Sierra regions of California, and southern Oregon Oklahoma plum Prunus gracilis USA (Southern Kansas to Texas and Tennessee) No t cultivation Alleghany plum Prunus alleghaniensis USA (Alleghany mountains in Pe nnsylvania) No t cultivation Southern sloe Prunus umbellata USA (Seashore from South Carolina to Florida, and westward to Mississippi, Louisiana and Arkansas) No t cultivation Dwarf cherries Prunus pumila Prunus pumila besseyi Prunus cuneata P. pumila in USA (coasts of northern states), P. pumila besseyi (from Manitoba to Kansas, westward to California and Utah), and P. cuneata in USA (New Hampshire to Minnesota and southward to North Carolina) Nebraska eastward. Choke cherry Prunus virginiana CAN (N ewfoundland to Manitoba and British Columbia) to USA (Georgia, Texas and Colorado) No t cultivation Black cherry Prunus serotina CAN (Quebec) to USA (Kansas and southward, New Mexico, and Mexico) No t cultivation 35

PAGE 36

Table 12. Summary of Prunus phylogenetic studies Paper z Kaneko et al. (1986) Mowrey and Werner (1990) Badenes and Parfitt (1995) Phylogenetic analysis Molecular Mole cular Molecular Analytical methods Phenetics Percent differential restriction fragments and Engel's genetic distance Phenetics Principal components MP Metrics (Analysis) cpDNA using BamH I, Hind III, and Sma I Isozyme cpDNA cutting with 21 3.2kb and 10 2.1kb endonucleases Taxa (no.) / subgenus (sect.) / genus 11 species / 3 subgenus: Cerasus Padus Armeniaca (Rehder, 1940) / genus Prunus 34 species / 4 subgenus: Prunus (sect.: Prunus Prunocerasus Armeniaca) Amygdalus Cerasus (sect.: Sargentiell a Microcalymma Magniculpula Phyllomahaleb), and Lithocerasus (sect.: Microcerasus Armeniacocerasus ) (Krussmann, 1986) 9 species / 5 subgenus: Prunus Amygdalus and Cerasus. Outgroups Fragaria vesca Trees (no.) 2 2 (average 30 principal component s) 10 Characters or bp (no.) 23 Informative characters (no.) Indels (no.) Substitutions (no.) Inversions (no.) PIC Percent variability Phylogeny in classification Support for subgenus Prunus. Subgenus Lithocerasus was identified as an artificial grouping of species Support for subgenus Prunus Cerasus and Amygdalus Relative small number of taxa used in the study. Subgenus Cerasus suggested to be more extensively evolved than either Prunus or Amygdalus Notes Lithocerasus fo rmed part of Cerasus in Rehder's (1940) classification. 36

PAGE 37

Table 1 2 Continued Paper Lee and Wen (2001) Bortiri et al. (2001) Phylogenetic analysis Molecular Molecular Analytical methods MP, NJ, ML MP Metrics (Analysis) ITS nuclear ribosomal DNA I TS nuclear ribosomal DNA and chloroplast trn L trn F spacer DNA Taxa (no.) / subgenus (sect.) 40 species (represented by 52 accessions) / 5 subgenus : Prunus (sect. : Prunus Prunocerasus, Armeniaca) Amygdalus Cerasus (sect.: Microcerasus Pseudocerasus Mahaleb, Phyllomahaleb), Padus and Laurocerasus (Rehder, 1940) 48 species / 5 subgenus : Prunus (sect. : Prunus, Prunocerasus, Armeniaca) Amygdalus Cerasus (sect.: Microcerasus, Pseudocerasus, Mahaleb, Phyllomahaleb), Padus and Laurocerasus (Rehder 1940) Outgroups Exochorda giraldii Madenia hypoleuca Oemleria cerasiformis Prinsepia sinensis, Prinsepia uniflora, Lyonothamnus floribundus Exochorda racemosa Oemleria cerasiformis Prinsepia sinensis Physocarpus capitatus Sorbaria sorbifolia, a nd Spiraea cantoniensis Trees (no.) MP = 15000 MPT (L=630, CI=0.632, RC=0.510). Consensus tree 16383 MPTs (L=630, CI=0.632, RI=0.808). ML tree log likelihood = 3641.3155 trn L trnF sequence MP = 76 MPT (L=187, CI= 0.733, RI= 0.834). ITS sequence MP = stopped at 30000 MPT (L=678, CI=0.567, RI=0.714). Combined data set consensus tree 8318 MPT (L=876, CI=0.695, RI=0.727). Characters (no.) 662 bp aligned (ITS1 = 223 242 bp, 5.8s = 154 bp, and ITS2 = 201 219 bp) trn L trn F = 563 bp, ITS = 759 bp Informa tive characters (no.) 218 bp aligned (ITS1 = 114 bp, 5.8s = 12 bp, and ITS2 = 92 bp) trn L trn F = 26 bp (excluding outgroups), ITS = 76 bp (excluding outgroups = among Prunus species) Indels (no.) 29 indels (>3 bp) aligned (ITS1 = 13 bp, ITS2= 16 bp) trn L trn F = 9 indels (> 2bp), ITS = 2 indels (> 2 bp) Substitutions (no.) Inversions (no.) PIC 218 bp aligned (ITS1= 114 bp, 5.8s = 12 bp, ITS2 = 92 bp) (not including indels) trn L trn F = 26 bp (excluding outgroups), ITS = 76 bp (excluding outgroups = am ong Prunus species) (not including indels) Percent variability 32.9% aligned (ITS1 = 47.1%, 5.8s = 7.79%, ITS2 = 42.0%) trn L trn F = 4.62%, ITS = 10.01% Phylogeny in classification Genus Prunus was monophyletic. Support for Madenia nested within genus P runus Within genus Prunus two major groups were recognizable: Amygdalus Prunus group and Cerasus Laurocerasus Padus group. Genus Prunus was monophyletic. Exochorda Oemleria and Prinsepia were not supported as sister groups with Prunus Genus Prunus was divided in two clades: subgenera Amygdalus Prunus Cerasus (sect. Microcerasus ) Emplectocladus group and subgenera Cerasus Laurocerasus Padus group. Subgenus Prunus sect. Prunus was monophyletic. Notes Number of parsimony informative characters included outgroups. The % variability cannot be directly compared with studies that excluded the outgroups for the number of PICs First time that P. fasciculata (sect. Emplectocladus ) was used in a study 37

PAGE 38

Table 1 2 Continued Paper Bortiri et al. (2002) Phyl ogenetic analysis Molecular Analytical methods MP, ML Metrics (Analysis) Nuclear gene sorbitol 6 phosphate dehydrogenase ( s6pdh ), and data from previous study ITS and trn L trn F (Bortiri et al., 2001). Taxa (no.) / subgenus (sect.) 22 species (represent ing all the major clades found in previous study)/ 5 subgenus : Prunus (sect. : Prunus, Prunocerasus, Armeniaca ) Amygdalus Cerasus (sect.: Microcerasus, Pseudocerasus, Mahaleb, Phyllomahaleb), Padus and Laurocerasus (Rehder, 1940) Outgroups Exochord a racemosa Oemleria cerasiformis Sorbaria sorbifolia,Spiraea cantoniensi, Holodiscus microphyllus Chamaebatiaria millefolium Kageneckia oblonga, Vauquelinia californica, Gillenia stipulata, Pyrus caucasica, Sorbus sp., Amelanchier alnifolia, Aruncus di oicus Neilla sinensis and Spiraea betulifolia Trees (no.) s6pdh sequence MP = 273 MPT (L=1198, CI= 0.58, RI= 0.81). s6pdh sequence ML tree log likelihood = 7720.96. For combined data set MP = 9 MPT (L=1592, CI= 0.58, RI= 0.61). For combined data set ML tree log likelihood = 12056.56. Characters or bp (no.) s6pdh = 1387 bp. Combined data set = 2760 bp ( s6pdh, trn L trn F, and ITS) Informative characters (no.) s6pdh = 234 bp (excluding outgroups = among Prunus species). Combined data set = 226 b p ( s6pdh = 148, trn L trn F = 18, and ITS = 60) Indels (no.) Substitutions (no.) Inversions (no.) PIC s6pdh = 234 bp (excluding outgroups = among Prunus species). Combined data set = 226 bp (s6pdh= 148, trnL trnF = 18, and ITS = 60) Percent variabili ty s6pdh = 16.87%. For combined data set = 8.18% [ s6pdh = 10.67%, trn L trn F = 3.19%, and ITS = 7.9% = calculated with characters from Bortiri et al.(2001)] Phylogeny in classification Genus Prunus was monophyletic. In the combined data set, the genus Prunu s was formed by two groups: subgenera Cerasus Laurocerasus Padus and subgenera Amygdalus Emplectocladus Prunus Cerasus (sect. Microcerasus ). Notes Includes P. fasciculata sect. Emplectocladus 38

PAGE 39

Table 1 2 Continued Paper Shaw and Small (2004) Phyl ogenetic analysis Molecular Analytical methods MP, BI Metrics (Analysis) Seven non coding chloroplast DNA regions: trnL UAA rpS 16, rpL16 and trnG UUC introns; trnS GCU trn G UUC trnL UUA trnF GAA and trnH GUG psbA intergeneric spacers. Taxa (no.) / subgenus (sect.) 43 species / 5 subgenus : Prunus [sect. : Prunus Prunocerasus (17 taxa), Armeniaca ] Amygdalus Cerasus (sect.: Microcerasus Pseudocerasus Mahaleb Phyllomahaleb ), Padus and Laurocerasus (Rehder, 1940) Outgroups Physocarpus opulifolius Tr ees (no.) Combined data set MP= 25171 MPT (L=422, CI= 0.92, RI= 0.94) Characters or bp (no.) PRUNOCERASUS ANALYSIS Introns: trnL UAA = 522 bp, rpS 16 = 683 bp, rpL16 = 996 bp, and trnG UUC = 711 bp. Intergeneric spacers: trnSGCUtrn GUUC = 703 bp, trnLUUAtrnFGAA= 397 bp, and trnHGUGpsbA = 363 bp. Combined data = 4375 bp. PRUNUS ANALYSIS trnHGUGpsbA = 516 bp, rpL16 = 1105 bp, trnSGCUtrnGUUC = 903 bp, trnGUUC = 746 bp. Combined data = 3270 bp. Informative characters (no.) Indels (no.) PRUNOCERASUS ANA LYSIS Introns: trnL UAA = 0 bp, rpS 16 = 2 bp, rpL16 = 7 bp, and trnG UUC = 0 bp. Intergeneric spacers: trnS GCU trn GUUC = 2 bp, trnLUUAtrnFGAA= 0 bp, and trnHGUGpsbA = 3 bp. Combined data = 14 bp. PRUNUS ANALYSIS trnHGUGpsbA = 13 bp, rpL16 = 10 bp, trnSGCUtrnGUUC = 14 bp, trnGUUC = 4 bp. Combined data = 41 bp. Substitutions (no.) PRUNOCERASUS ANALYSIS Introns: trnL UAA = 1 bp, rpS 16 = 4 bp, rpL16 = 6 bp, and trnG UUC = 4 bp. Intergeneric spacers: trnS GCU trn GUUC = 4 bp, trnLUUAtrnFGAA= 3 bp, and trnHGUGps bA = 1 bp. Combined data = 23 bp. PRUNUS ANALYSIS trnHGUGpsbA = 11 bp, rpL16 = 21 bp, trnSGCUtrnGUUC = 28 bp, trnGUUC = 32 bp. Combined data = 92 bp. Inversions (no.) PRUNOCERASUS ANALYSIS Introns: trnL UAA = 0 bp, rpS 16 = 0 bp, rpL16 = 0 bp, and trnG UUC = 0 bp. Intergeneric spacers: trnS GCU trn GUUC = 0 bp, trnLUUAtrnFGAA= 0 bp, and trnHGUGpsbA = 0 bp. Combined data = 0 bp. PRUNUS ANALYSIS trnHGUGpsbA = 0 bp, rpL16 = 0 bp, trnSGCUtrnGUUC = 1 bp, trnGUUC = 0 bp. Combined data = 1 bp. PIC PRUNOCERASUS ANALYSIS Introns: trnL UAA = 1 bp, rpS 16 = 6 bp, rpL16 = 13 bp, and trnG UUC = 4 bp. Intergeneric spacers: trnSGCUtrn GUUC = 6 bp, trnLUUAtrnFGAA= 3 bp, and trnHGUGpsbA = 4 bp. Combined data = 37 bp. PRUNUS ANALYSIS trnHGUGpsbA = 24 bp, rpL16 = 31 bp, trn SGCUtrnGUUC = 43 bp, trnGUUC = 36 bp. Combined data = 134 bp. Percent variability PRUNOCERASUS ANALYSIS Introns: trnL UAA = 0.19%, rpS 16 = 0.88%, rpL16 = 1.31%, and trnG UUC = 0.56%. Intergeneric spacers: trnSGCUtrn GUUC = 0.85%, trnLUUAtrnFGAA= 0.76%, and trnHGUGpsbA = 1.10%. Combined data = 37 bp. PRUNUS ANALYSIS trnHGUGpsbA = 4.65%, rpL16 = 2.80%, trnS GCU trnG UUC = 4.76%, trnG UUC = 4.80%. Combined data = 4.09%. Phylogeny in classification Genus Prunus was monophyletic. Subgenus Prunus sect. Prunocera sus and sect. Prunus were monophyletic. The genus Prunus was formed by two groups: subgenera Laurocerasus Padus and subgenera Amygdalus Emplectocladus Prunus Cerasus (sect. Microcerasus ). Prunus texana and P. subcordata were included in sect. Prunocerasus Within sect. Prunocerasus three groups were identified: the American, the Chickasaw, and the Beach clades. Notes Prunus texana was first used in this study. Prunus texana and P. fasciculata were not recognized by Waugh (1901), Wight (1915), and Rehder (19 40) 39

PAGE 40

Table 1 2 Continued Paper Rohrer et al. (2004) Shaw and Small (2005) Katayama and Uematsu (2005) Phylogenetic analysis Molecular Molecular Molecular Analytical methods UPGMA MP UPGMA Metrics (Analysis) Fifteen microsatellites primer pairs rpL 16 intron CpDNA analysis based on five restriction enzymes ( Sal I, XhoI, Bam HI, Sac I, and Pst I) by RFLP Taxa (no.) / subgenus (sect.) / genus 18 species / subgenus Prunus sect. Prunocerasus (13 and 3 undetermined hybrids), subgenus Prunus ( P. cerasifera), and subgenus Armeniaca ( P. armeniaca ). A total of 207 accessions = 18 species (subgenus Prunus sect. Prunocerasus ). A total of 18 accessions = 14 Prunus species and 1 interspecific hybrid Outgroups Pyrus ussuriensis var. hondoensis Trees (no.) Stric t consensus = 3 MPT (L=34, CI= 0.97, RI= 0.99) Strict consensus = 8 MPT (L=68, CI= 0.93, RI= 0.64) Characters or bp (no.) A total of 186 putative alleles with a mean value of 12.4 per locus rpL16 intron = 797 bp. Informative characters (no.) rpL16 i ntron = 23 bp. Indels (no.) Substitutions (no.) Inversions (no.) PIC rpL16 intron = 23 bp. Percent variability rpL16 intron = 2.88%. Phylogeny in classification No clear phylogenetic relationships were determined. The microsatellites are evolving too rapidly in North American plums to be truly useful at resolving species relationships. Twenty two unique haplotypes were identified in sect. Prunocerasus Ten different haplotypes were associated with the American clade, two haplotypes with the Beach clade, and seven haplotypes with the Chickasaw clade. Additionally, one Texana haplotype, one Subcordata haplotype, and one peculiar Umbellata haplotype. Eleven genome types. The UPGMA tree consisted of two major groups: genome types A I (subgenus Amygdalus Prunus and Cerasus sect. Microcerasus ) and other with genomes J K (subgenus Laurocerasus and Padus ). Notes The congeneric relationship of plums to peach and cherry allowed the successful use of these primers in section Prunocerasus Micros atellites are evolving too rapidly to be truly useful at resolving species phylogeny. The common practice of choosing one specimen to represent a taxon can be misleading in closely related groups. Choosing different genotypes could have resulted in a different result than previous studies. The 9.1 kb region between psb A and atp A genes would be useful tool to study the cpDNA evolution in Prunus. 40

PAGE 41

Table 1 2 Continued Paper Bortiri et al. (2006) Wen et al. (2008) Phylogenetic analysis Morphology and molecu lar Molecular Analytical methods MP, ML, and BI. MP and BI Metrics (Analysis) ITS nuclear ribosomal gene, trn L trn F spacer, trn S trn G spacer, trn G intron, and 25 morphological characters. Chloroplast ndh F region and ITS nuclear ribosomal gene. Taxa (no. ) / subgenus (sect.) / genus 37 species / 5 subgenus : Prunus (sect. : Prunus Prunocerasus Armeniaca) Amygdalus Cerasus (sect.: Microcerasus Pseudocerasus Mahaleb, Phyllomahaleb), Padus and Laurocerasus (Rehder, 1940) A total of 59 ( ndh F) or 51 ( ITS) accessions of Prunus / 5 subgenus : Prunus (sect. : Prunus, Prunocerasus, Armeniaca) Amygdalus Cerasus (sect.: Microcerasus, Pseudocerasus, Mahaleb, Phyllomahaleb), Padus and Laurocerasus (Rehder, 1940). In addition, Madenia hypoleuca and the Pyg eum group. Outgroups Oemleria cerasiformis Sorbaria sorbifolia, Spiraea cantoniensi, Gillenia stipulata, Lyonothamnus floribundus Maddenia hypoleuca, Physocarpus capitatus Physocarpus opulifolius and Rhodotypos scandens. Oemleria cerasiformis Prins epia uniflora Physocarpus monogynus Lyonothamnus floribundus and Holodiscus discolor Trees (no.) Morphological data set MP=50000 MPT (L=110, CI= 0.36, RI= 0.73). Molecular data results from Bortiri et al. (2001) and Bortiri et al. (2002). Combined data set MP=20 MPT (L=1741, CI= 0.49, RI= 0.65). Combined data set ML tree log likelihood = 12499.63 ndh F sequence MP = 196200 MPT (L=815, CI=0.71 COI=056, RI=0.86 ). ITS sequence MP = 49200 MPT (L=791, CI=0.56, COI=0.45, RI=0.70). Characters or bp (no.) Combined data set = 771 bp. Informative characters (no.) ITS = 178 bp, trn L trn F = 50 bp, and trn S trn G = 142 bp. Indels (no.) Combined data set = 3 Substitutions (no.) Inversions (no.) PIC ITS = 178 bp, trn L trn F = 50 bp, and trn S t rn G = 142 bp. Percent variability Phylogeny in classification Three clades were reported: Clade A with subgenera Padus and Laurocerasus ; Clade B with subgenera Amygdalus Emplectocladus and Prunus ; and Clade C with subgenera Cerasus Clade B was characterized by the production of three axillary buds. Padus and Laurocerasus were not supported as monophyletic (highly homoplasy). Both data set identified genus Prunus as a monophyletic group. Both data sets were incongruent at the species level in Prunus The ndhF data supported two major groups: subgenera Laurocerasus (including Pygeum ) and Padus and subgenera Amygdalus Cerasus and Prunus The ITS data supported a clade composed of subgenera Amygdalus Prunus and Cerasus sect. Microcerasus and the paraphyletic clade of Padus and Laurocerasus 41

PAGE 42

Table 1 2 Continued Paper Depypere et al. (2009) Phylogenetic analysis Morphology and Molecular Analytical methods UPGMA, PCo, and BI Metrics (Analysis) Leaf and endocarp morphometrics, and AFLP p rimers Taxa (no.) / subgenus (sect.) / genus A total of 82 accessions / 5 species: P. cerasifera P. domestica P. insititia P. spinosa, and P. x fruticans Outgroups Trees (no.) Characters or bp (no.) Informative characters (no.) Indels (no.) Substitutions (no.) Inversions (no.) PIC Percent variability Phylogeny in classification PCoA and AFLP of three distinct clusters. A first cluster consists of all P. cerasifera samples and the sole P. cocomilia A second cluster includes all individuals of P. domestica and P. insititia A third cluster comprises all P. spinosa and P. x fruticans samples. Notes Low number of Prunus species for sampling. zPIC = total indels + nucleotide substitutions + inversions. Percent variability = PIC/charac ters or bp. PIC = potentially informative character. 42

PAGE 43

CHAPTER 2 GENETIC DIVERSITY IN PEACH [ Prunus persica (L.) BATSCH] AT THE UNIVERSITY OF FLORIDA: PAST, PRESENT AND FUTURE Introduction Prunus L. belongs to the subfamily Amygdaloideae (=Prunoideae) of t he Rosaceae. The genus Prunus is an economi cally important source of fruit nuts and timber It includes plums, cherries, almonds, apricots and peaches. It is distributed around the world, with approximately 200 species (Bortiri et al., 2001; Hedrick, 1911 ; Rehder, 1940; Sargent, 1905; Wight, 1915). FAOSTAT (2010) reported that the world s net production of almonds, apricots, cherries, peaches, nectarines, plums and sloes in 2010 was approximately 40.8 million tons Peach and nectarine world production was the largest with 20.5 million tons US peach and nectarine production was approximately 1.3 million tons, with a farm gate value of ~ 683 million dollars The peach and almonds origin is believed to be eas tern and southwestern Asia; apricot to be northern China; cherries from the Eurasian region, wher e Europe and Asia meet s; and plums to be Asia, Europe and America (Hedrick, 1911). Peach cultivation d ates back to approximately 4000 years in China with three groups being recognized (Scorza and Okie, 1991; W ang, 1985) T he southern group of peaches originated in a climate similar to the sout h easter n US with mild winters and hot wet summers along Yangtze River in the provinces of Jiangsu, Zhejiang, Jiangxi, Hubei, Hunan, and Sichuan. Th e n orthern group is fou nd in a climate with cold winters, and hot and dry summers along the Yellow River in Shandong, Hebei, Henan, Shanxi, Shaanxi, and Gansu provinces. The third group is found in the arid northwest China. T he peaches 43

PAGE 44

spread from China along the silk trade rou tes through Persia and the R omans distributed peach es throughout Europe (Scorza and Okie, 1991; Wang, 1985) The first peaches in North and South America were brought by the early Spanish explorer s through St. Augustine, FL and Mexico ( Scorza and Okie, 1991 ) Native Americans spread the early yellow nonmelting peaches across the American continents In the 1800s, French and English white melting peach cultivars were brought into US At this time t he new world peaches lacked fruit quality and commercial attributes (Scorza and Okie, 1991; Sharpe et al., 1954 ; Sherman et al., 1996 ; Wang, 1985) Peach c ultivar s released in the US during late 1800s and early 1900s originated from superior chance seedlings found by farmers and growers from OP (open pollinated) se ed (Floyd, 1920; Layne and Bassi, 2008) H ybridization among desired parents was later implemented for breeding and selection by state and federal peach breeding programs. A cultivar named Chinese Cling, an introduction from China, changed US i ndustry and cultivation in 1850 Georgia Belle and Elberta, OP seedlings from Chinese Cling became important cultivars and were used as breeding parents (Scorza and Okie, 1991). Elberta was widely used in the US peach breeding programs to obtain peaches with higher fruit quality and firmness adapted to our climate and growing conditions. Elberta can be found in the pedigree of most commercial peaches developed in the US (Layne and Bassi, 2008; Scorza and Okie, 1991; Wang, 1985). US p each breeding programs w ere and are classified based on their geographic origin and chilling requirement : north (highchill), southcentral (mid chill), south (low chill) and west ( mid to high chill) (Layne and Bassi, 2008) Breeding peaches adapted to 44

PAGE 45

low chill areas of the sout hern US began with the use of germplasm from the southern peento Chinese group due to its superior adaptation to the subtropics In the US the University of California Riverside started breeding low chill peaches in 1907 followed by University of Florida in 1952 (Layne and Bassi, 2008; Scorza and Okie, 1991 ; Sharpe et al., 1954; Sharpe, 1961, 1969 ; Sherman et al., 1984). The stone fruit breeding and genetics program at the University of Florida was established to develop early ripening stone fr uit cultiva rs with high fruit quality, adapt ed to subtropical tropical climates with low chilling requirements, and the ability to withstand high disease pressure ( Andersen et al., 2001; Sharpe et al., 1954; Sherman et al 1984; Sherman et al., 1996) The University of Floridas peach germplasm pool has been selected since its creation to accommodate changes in cultivation practices disease pressure, fruit quality standards and commercialization. These genetic modifications can be traced back to cultivars or selecti ons that have had an important impact on the breeding pool UFs l ow chill germplasm founders originated from three different seed importations from south China The first seed lot came through Charleston, SC in 1870, from which cultivars Waldo and Jewel originated The second seed lot came through Hawaii, from which the cultivar Hawaiian was selected. A third seed lot was imported directly from Okinawa in 1953, from which the cultivar Okinawa was obtained ( Floyd, 1920; Sharpe et al., 1954; Sharpe, 1957, 1961 1969). Superior fruit quality and firmness can be traced to the high quality higher chill cultivars Southland and Springtime (Floyd, 1920; Sherman et al., 1996). The n ectarine trait was introduced in 1956 with Panamint and in 1958 with NJN21 and NJ 45

PAGE 46

5107397 (Sharpe and Aitken, 1971). Additional introductions of superior nectarine materials were made using New Jersey and California varieties (Sherman et al., 1996). The nonmelting flesh trait was introduced in 1969 from medium chill fer al peaches from Zacatecas, Mexico (Sherman et al., 1990; Sherman et al., 1996). Additional nonmelting flesh germplasm was introduced using the US cultivar Springcrest in 1972 and Diamante from southern Brazil in 1982 ( Sherman et al., 1990; Sherman et al., 1996). N ovel genotypes with deep yellow or light orange skin and flesh (highlighter) were also obtained from the Mexican cling population (Sherman et al., 1996; Sherman and Lyrene, 2003). Different types of m olecular markers are available in peach suc h as isozymes (Durham et al., 1987; Messeguer et al., 1987; Mowrey et al., 1990a, 1990b) random amplified polymorphic DNAs (RAPDs) and restriction f ragment length polymorphic DNAs (RFLP) (Ars et al., 1994; Chaparro et al., 1994; Dirlewanger and Bodo, 199 4; Quarta et al., 1998; Rajapakse et al., 1995), simple sequence repeats (SSRs) ( Aranzana et al., 2002, 2003a, 2003b; Cipriani et al., 1999; Dirlewanger et al., 2002; Sosinki et al., 2000; Testolin et al., 2000; Wang et al., 2002) and single nucleotide polymorphisms (SNPs) ( Eduardo et al., 2012; Martnez Garca et al., 2012; Verde et al., 2012) These molecular markers have been used to study g enetic diversity and genome structure changes (Wnsch and Hormaza, 2002). Germplasm restrictions, bottlenecks and population shifts can be associated with changes on the germplasm genetic diversity US peach breeding programs are mostly based on few founders from Europe and Chinese Cling material, and they have been used extensively in the breeding of commercial var ieties grown around the world. 46

PAGE 47

Aranzana et al. (2010) reported that peach cultivars from Europe and North America showed low genetic diversity when screened with 50 SSRs distributed across the peach genome. They found that differences in population structure and molecular variation were associated to important fruit characters : melting flesh peaches, melting flesh nectarines, and nonmelting flesh varieties. This structure was largely due to the genetic bottleneck s present in modern breeding programs US pe ach breeding programs are mostly based on a few founders from Europe and Chinese Cling germplasm T he objective of this research was to study the genetic diversity and population structur e of the University of Florida stone fruit breeding and genetics pr ogram from its creation until present A retrospective look at breeding (selection) trends and its effect on genetic diversity was performed using legacy cultivars and more recent cultivars and selections. This information is for determining the level of diversity in the Stone Fruit Breeding Program, detecting trends, and searching for possible marker trait associations Material and Methods Plant Material A total of 19 5 peach genotypes from five major breeding pools of the University of Florida stone fruit breeding program were used for this study. Genotypes w ere classified based on their origin : UF varieties and advanced materials (20002010 selections ) (n=168), UF UGA USDA varieties and selections (n=1 3 ) landrace cultivars (n= 4 ) high chill NCSU cultivar s (n=5) and other Prunus (n=5) species material ( Table 2 1). UF varieties and advanced materials landrace cultivars, and other Prunus species were obtained from UFs germplasm collection in Gainesville, FL. UF UG A USDA varieties were provided by Dr. T hom as G. Beckman (Southeastern Fruit and Tree Nut 47

PAGE 48

Research Lab, Byron, GA and Attapulgus, GA) NCS U cultivars were provided by Dr. Dennis J. Werner (North Carolina State University, Raleigh, NC). Each genotype was characterized for different phenotypic trait s: peach vs. nectarine fruit; melting, nonmelting and crispy flesh; white vs. yellow flesh; normal vs. highlighter (reduced anthocyanin) fruit ; round vs. peento fruit shape; leaf glands ( reniform, globose and eglandular ) ; showy vs. nonshowy flower, chill ing requirement ; and decade of genotype selection (See additional Table A 1). DNA I solation DNA was extracted from leaf tissue using a modified CTAB method as described by B laker ( 2010), and Chavez and Chaparro (2011) Leaf tissue (30 mg fresh or 10mg dry ) of each sample was added to 2 mL Eppendorf microcentrifuge tubes with three 5 mm stainless steel beads, 750 L of CTAB buffer (2% CTAB, 100 mM Tris pH 8.0, 1.4 M NaCl, 0.5M EDTA, 1% PVP) previously mixed with Mercaptoethanol (1 L /mL), and 8 L of RNAse (10 mg/mL). Samples were grounded two/three times at 30 Hz in a Tissue Lyser (QIAGEN Inc., Valencia, CA, USA) for 1.5 min until tissue clumps were not visible. Samples were vortexed and incubated in a 65C water bath for 6 min. Then, tubes were vortexed and a volume of 750 L of chloroform:isoamyl (24:1) was added. Tubes were vortexed, incubated at 20C for 6 min, and then centrifuged at 12000 rcf for 10 min. The aqueous phase was transferred to a new 2 mL centrifuge tube, and 500 L of cold isopropanol were added. Tubes were gently mixed, incubated at 20C for 6 min and then centrifuged at 16100 rcf for 10 min. Supernatant was removed, and pellet was washed with 500 L of cold 70% EtOH (by inverting the tubes carefully). Tubes were incubated at 20C for 5 min., and then centrifuged at 16100 rcf for 5 min. Supernatant was removed, and pellet was washed with 500 L of cold 90% EtOH 48

PAGE 49

(repeating mixing, incubation and centrifugation, as described before). Ethanol was poured off and the pellet was dried at bench top for approx. 3045 min (room temperature). The pellet DNA was resuspended in 50 L TE buffer (10 mM Tris HCl, 0.1mM EDTA) and 50 L of DNA grade water DNA concentration was quantified in a UV10 Spectrophotometer (Thermo Scientific, Waltham, MA, USA ). DNA concentration for all the samples was standardized to 20 ng /uL SSR F ingerprinting A total of 36 SSR markers distributed across the peach genome ( ~ 12.8 cM between markers) were selected from the Prunus T exas almond E arlygold peach (T E) refe rence map ( Dirlewanger et al., 2004; Jung et al., 2008) Forward primers were modified using a 5 fluorophore label 6FAM (standard) [6 FAM] or HEX [5HEX] (Eurofins MWG Operon, Huntsville, AL, USA ) for multiplex product fragment analysis ( Table 22). Two h aploid genotypes were used as controls to test product amplification for each SSR marker. PCR products were amplified in a 1 6 L volume reaction containing 2 L of 20 ng/L DNA template, 2.25 L 10X ThermoPol Reaction Buffer (10mM KCl, 10mM (NH4)2SO4, 20mM Tris HCl, 2mM MgSO4, 0.1% Triton X 100, pH 8.8 @ 25C), 1 L 2.5 mM dNTPs, 0.2 L Taq DNA Polymerase, 6.55 L DNA grade water and 4 L 5M ( 2 L forward and 2 L reverse) primers PCR parameters were: 3 min at 94C followed by 40 cycles of 30 s denaturing at 94C, 30 s at primers specific annealing temperature [ Ta (C)] ( Ta ble 22) and 1 min of elongation at 72C, ending with 7 min at 72C. PCR products were separated on 3 % (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a di gital gel documentation system. Gel images were used to determine PCR dilution ratios for fragment analysis on a n ABI3730 sequencer 49

PAGE 50

(Applied Biosystems Grand Island, NY USA ) at the Interdisciplinary Center for Biotechnology Research (ICBR), University o f Florida, Gainesville, FL Fragment analysis of chromatographs were visualized using GeneMarker v.1.6 software (SoftGenetics, LLC, State College, PA USA ) using 600 LIZ size standard (Applied Biosystems, Grand Island, NY, USA) Data A nalyses Genetic an d C luster A nalyses G enetic variation and cluster analyses of fingerprinting data were performed using Powermarker v.3.25 (Liu and Muse, 2005) and GenAlEx v. 6.5 software (Peakall and Smouse, 2006 2012). Genetic variation was characterized for each locus by the number of observed alleles ( A ), effective number of alleles ( Ae ), observed heterozygosity ( Ho ), expected heterozygosity ( He ), Wrights fixation inde x [ F=(He Ho)/He=1 (Ho/ He) ], and polymorphism information content ( PIC ) Cavalli Sforza chord distance ( CavalliSforza and Edwards, 1967) and Neis genetic distance (Nei and Takezaki, 1983) were calculated for the whole dataset. Unweighted Pair Group Method wi th Arithmetic Mean (UPGMA) (Sokal and Michener, 1958) and Neighbor Joining (NJ) (Saitou and Nei, 198 7) cluster analyses were obtained. Phenotypic traits (Table A 1) were traced over the UPGMA and NJ clusters using Mesquite v.2.73 software (Maddison and Maddison, 2011) to identify associations between peach phenotypic traits and cluster results. The genet ic distance and cluster analys is combination that represented best the known pedigree information w as chosen. Population S tructure Population structure analys e s w ere performed for all the genotypes, and separately for the UF varieties and advanced material s, using Structure v.2 software 50

PAGE 51

(Pritchard et al., 2000). Structure simulation parameters were ru n under the admixture model assumption with correlated alleles using four reps per run for 2 to 14 K subgroups with 105 interactions after a burnin of 104 int eractions Structure Harvester software (Earl and VonHoldt, 2012) was used to implement the Evanno method to analyze the population structure results (Evanno et al., 2005). Distruct v.1.1 software (Rosenberg, 2004) was used to visualize and modify bar plot structure results Results Genetic and C luster A nalyses A total of 36 SSR markers successfully amplified 423 alleles for 195 peach genotypes representing five breeding pools of the University of Florida stone f ruit b reeding p rogram An average of 18 genot ypes were amplified per marker an average number of alleles ( A ) of 11.43, an average effective number of alleles ( Ae ) of 2.58, observed heterozygosity ( Ho ) of 0.4, expected heterezogosity ( He ) of 0.52, Wrights fixation index ( F ) of 0.25 and an average polymorphism information content ( PIC ) of 0.48 (Table A 2 ) The number of genotypes amplified per marker ranged from 37 for BPPCT008 to 3 for EPDCU3117 and CPSCT004. The number of alleles varied from 22 for CPDCT038 to 2 for EPDCU3117 and CPSCT004. Similarly the effective number of alleles ranged from 5.94 to 1.07 for BPPCT008 and CPSCT004, respectively. EPDC3832 had the highest observed heterozygosity of 0.88 and CPSCT008 had the lowest heterozygosity of 0.03. The value of expected heterozygosity varied fro m 0.83 to 0.06 for BPPCT008 and CPSCT004, respect ively. Wrights fixation index ( F ) ranged from 0.94 for CPSCT008 to 0.65 for EPDC3832. BPPCT008 had the highest polymorphism information content of 0.81 compared to CPSCT004 of 0.06. 51

PAGE 52

A total of 237 alleles were amplified for the University of Florida varieties and advanced materials (20002010 selections) with average values of 11.97 genotypes amplified per marker, A of 6.41, Ae of 2.37, Ho of 0.41, He of 0.49, F of 0.17 and PIC of 0.44 (Table 23). These values were lower than the whole dataset averages because the founding genotypes and other germplasm (including other Prunus species) were removed from this analysis. The number of amplified genotypes per marker ranged from 30 for BPPCT008 to 2 for CPSCT004. Value for A ranged from 12 for BPPCT008 and CPPCT022, to 2 for CPSCT004 and EPDCU3117. BPPCT008 had the highest Ae of 5.34 compared to CPDCT027 of 1.05. Observed heterozygosity ( Ho ) ranged from 0.92 for EPDC3832 to 0.02 for CPSCT008 and CPDCT027. Similarl y, He ranged from 0.81 to 0.05 for BPPCT008 and CPDCT027, respectively. Values of F varied from 0.95 for CPSCT008 to 0.82 for EPDC3832. BPPCT008 had the highest PIC of 0.79 compared to CPDCT027 of 0.05. The University of Florida varieties and advanced mat erials genetic diversity results were compared to Aranzanas et al. (2010) results for 224 peach/nectarine cultivars from around the world. UFs germplasm genetic diversity was higher than Aranzanas results based on 16 SSR primers common to both studies with an average Ho of 0.50, He of 0.57 and F of 0.12 compared to an average Ho of 0.34, He of 0.48 and F of 0.28, respectively. T he genetic diversity of the UFs germplasm material including founder cultivars did not change across the decades since the programs creation in 1952 until present ( Table 24). L ower averages of F were reported in the early decades of the programs creation due to higher heterozygosity levels created by the introgressions from founder 52

PAGE 53

genotypes A n increase in the F average for the 80s decade was due to selection for superior genotypes with desired commercial attributes A decrease in F during the 90s was produced by the shift in the programs breeding focus from melting to nonmelting flesh varieties with the introduction of non melting genotypes from Mexico, Brazil and the US (Sherman et al., 1990; Sherman et al., 1996). A stable F was detected during the 00 s to 0 s decades that can be attributed to the introgression of higher chill UF UGA USDA germplasm Genetic distance using CavalliSforza chord distance (Cavalli Sforza and Edwards, 1967) and Neis genetic distance (Nei and Takezaki, 1983) were calculated in combination with Unweighted Pair Group Method wi th Arithmetic Mean (UPGMA) (Sokal and Michener, 1958) and Neighbor Joining (NJ) (Saitou and Nei, 1987) cluster analyses ( Figures A 1, A 2 A 3 ) The UPGMA cluster analysi s based on Neis genetic distance represented best the known pedigree information for the germplasm pools (Figure 2 1). The 36 SSR markers identified 193 genotypes (out of 195). The genotype pairs Flordaking and Flordastar, and a neuploid nectarine and AP0518ws haploid, had identical genotypes, respectively About 9799 % of the SSR markers identified single loci in the OK PK and 0201c_hap haploids with the detection of two loci each for CPSCT034 and EPDC3832. Two major cluster groups were detected in the germplasm corresponding to melting and nonmelting flesh varieties. Peaches and nectarines did not cluster into two major groups. Additional clad es were associated and named accordingly to other phenotypic traits being analyzed ( Figure 21 ). 53

PAGE 54

Outgroups (other Prunus species) and landrace germplasm had the highest diversity. The ornamental nectarine selections and varieties and the high chill nonsh owy NCSU cultivars group were monophyletic respectively Nectarine and peach fruit type s w ere found throughout the cluster, similar to other phenotypic characteristics (Figure A 4 to A 12) Breeding for s elections of non melting, low chill, highlighter, a nd/or white flesh color has increased in the last 10 years Population S tructure Population stratification for all genotypes representatives of the University of Florida genetic pool (n=195) were analyzed for two to ten K subgroups ( Table A 3) with K=3 an d K=5 having the highest delta K ( ) values using the Evanno method (Evanno et al., 2005) as implemented in the Structure Harvester software (Earl and VonHoldt, 2012) ( Table A 4 Figure 22 ). Population stratification for K=3 showed clear differences between: 1) outgroups (other Prunus sp p.) and landrace germplasm [blue bars] 2) melting varieties [orange bars] and 3) non melting varieties [ yellow bars ] Transition zones between melting and nonmelting varieties (orange bars among yellow bars and vice versa) corresponding to transition se lections from the 80 s and 90 s decades were identified ( Figure 21) Population stratification for K=5 represented differences between: 1) outgroups, landrace germplasm and UFs ornamental nectarine selections [blue bars]; 2) high chill nonshowy NCSU me lting peach cultivars, UFs melting peach /nectarine varieties from the 70s and 80s and UFs globose leaf gland melting peach varieties [purple bars]; 3) UFs melting peach /nectarine varieties from the 70 s and 80 s, melting nectarine selections ( showy a nd nonshowy series ) and nonmelting peach/nectarine transition selections from the 90s [orange bars]; 4) 1st generation nonmelting peach varieties 54

PAGE 55

(Diamante cling), nonmelting peach/nec tarine transition selections 90s decade, peento nonmelting non showy selections from the 0 s and 10 s, highlighter nonmelting nectarine selections from the 00s and 0 s, white flesh non melting selections from the 0 s, and white flesh non melting selections from the 00 s and 10 s [yellow bars]; and 5) nonmelting peach/nec tarine transition selections 90s decade, non melting nectarine selections 10s decade, nonmelting peach selections, yellow nonmelting peach selections 00s decade, and white flesh non melting peach/nectarine selections from the 00s and 10 s [g reen bars] ( Figure 21 ). Other K analyses showed additional subgroups/clades relationships associated with additional phenotypic characteristics ( Figure 21 ) Stratification K=2 differentiated 1) outgroups, landrace germplasm and ornamental nectarine selec tions [blue bars] from 2) UFs selections and advanced materials, UF USDA UGA varieties, and NCSU high chill cultivars [orange bars]. Stratification with K=4 showed similar major subgroups as with K=3. Highlighter nonmelting nectarine selections peento non showy nonmelting selections, and white flesh nonmelting selections all from the 00s and 10s formed a subgroup [yellow bars]. Yellow flesh nonmelting varieties and white flesh low chill nonmelting selections from the 0 s decade constituted another subgroup [green bars]. Population stratification with K=6 separated major subgroups as with K=3. Additional observed subgroups were: orn amental nectarine selections from the 00s [peach bars]; high chill NCSU cultivars and UF UGA USDA selections [purple bars ]; UFs melting peach/nectarine varieties from the 70 s and 80 s and nectarine melting selections (showy and nonshowy series) [orange bars]; and UFs melting peach/nectarine varieties from the 70s and 80s [yellow bars]. Similarly, m ajor 55

PAGE 56

subgroups w ere recovered with K=7 to 10. T he highlighter nonmelting nectarine selections and peento nonshowy nonmelting selections both from the 00s and 10s, formed a subgroup starting from K=7 [brown bars]. Similarly landrace germplasm s were differentiated in a subgroup starting from K=9 [dark red bars]. Structure analyses of the UF varieties and advanced materials (n=168) provided similar results to the analyses with all the germplasm (n=195) Population stratification with K=5 had the highest value using the Evanno method (Evanno et al., 2005) (Tables A 5, A 6 ; Figure s A 13 and A 14). Population stratification for K=5 provided clear differences between: 1) U Fs ornamental nectarines [blue bars]; 2) melting flesh selections [orange bars]; 3) peento nonmelt ing nonshowy selections highlighter non melting nectarine selections and white flesh nonmelting selections all from the 00s and 10s [yellow bars]; 4) nonmelting nectarine selections from the 0 s decade and white flesh nonmelting low chill selections from the 00 s decade [purple bars]; and 5) yellow flesh nonmelting peach/nectarine selections [green bars]. No additional major subgroups were observed with other K values. Population structure results supported major subgroups: melting and nonmeltin g flesh varieties/ selections in all analyses (Figures 2 1, A 14). No clear separation was observed between peach and nectarine selections. The stone fruit breeding program at the Univers ity of Florida has not maintained a separate breeding pool for the dev elopment of nectarine cultivars and t he majority of the peach selections are believed to be heteroz ygous at the peach/nectarine locus (nectarine recessive to peach). Similar ly most white peach selections are believed to be heterozygous for yellow flesh. No additional structure was detected for other phenotypic characteristics. 56

PAGE 57

Discussion A total of 195 peach genotypes representatives of five major breeding pools of the University of Florida stone fruit breeding and genetics program were fingerprinted with 3 6 SSR markers to determine the programs genetic diversity, population structure and the effect of selection since its creation in 1952 until present The screening of SSR primers allowed the identification of primers with high PIC that can be used for fin gerprinting and patent variety protection. Fingerprinting profiles of v arieties and selections constitute a searchable database of unique alleles that could later be used to find alleles associated with important traits in conjunction with their pedigree i nformation (identity by descent analyses). The reduction in genetic variation observed when the entire dataset (n=195) and the dataset consisting exclusively of UF germplasm (n=168) was due to the removal of founder genotypes, landrace germplasm highchill material (UF UGA USDA varieties, NCSU cultivars), and other Prunus species from the analysis ( Tables A 2 2 5 ). The total number of alleles (423 vs. 273), A e (11.43 vs. 6.41), He (0.52 vs. 0.49), F (0.25 vs. 0.17) and PIC (0.48 vs. 0.44) reported a reduc tion when comparing both datasets, respectively. Only Ho (0.4 vs. 0.41) w as maintained across analyses. The use of high chill selections (primarily UF UGA USDA varieties) and other Prunus species ( such as P kansuensis and P. dulcis ) are an important sourc e of novel traits and genetic diversity. The use of these resources will i ncrease and maintain the genetic diversity of UFs breeding pool in the near future A total of 237 alleles were amplified for 168 peach and nectarine cultivars and selections of the University of Florida using 36 SSR markers ( Table 23). Aranzana et al. (2010) reported that a total of 318 alleles were amplified for 224 peach and nectarine 57

PAGE 58

cultivars from around the world using 50 SSR markers. Although, the total number of alleles iden tified was lower in the UF stone fruit breeding pool the values for A (6.41 vs. 6.36), Ae (2.37 vs. 2.08), Ho (0.41 vs. 0.34), and He (0.49 vs. 0.46) were higher respectively Similarly, the value of F (0.17 vs. 0.26) was lower, respectively. These trends were maintained (as reported before) when comparing 16 SSR markers common to both studies University of Floridas genetic diversity had the higher diversity compared to Aranzanas et al. (2010) study, which is the most comprehensive study of peach and n ectarine germplasm from across the world reported to date. The genetic diversity of the University of Florida stone fruit breeding germplasm does not appear to have changed since its initiation in 1952 ( Table 24). The source of fruit quality from high chi ll germplasm and introgression of new traits acro ss generations (UF UGA USDA breeding pool and other Prunus species) has maintained the genetic diversity of the breeding population. The UPGMA cluster analysis based on Neis genetic distance represented bes t the known pedigree relationships within the germplasm pools ( Figure 21). Two major groups were identified: melting and nonmelting flesh varieties Similar results were reported by Aranzana et al. (2003a, 2010) No additional groups were found associate d with other phenotypic characteristics. The highest genetic diversity using the cluster analysis was identified in founder genotypes, landrace germplasm, highchill material (UF UGA USDA varieties, NCSU cultivars), and other Prunus species. Population str ucture results were consistent with the cluster analysis. Three main groups were identified across simulations (K=2 to 10, with K=3 being optimal cluster ): outgroups and landrance germplasm, melting varieties, and nonmelting varieties 58

PAGE 59

( Figure 21). Aranzana et al. (2003a, 2010) identified three main groups: melting flesh peaches, melting flesh nectarines and nonmelting varieties. No nectarine and peach structure was visible. Nectarines and peaches were found throughout the clusters ( Figure A 4). Breeding and selectio n of peaches and nectarines is intermixed at UF It is believed that the majority of UFs peach varieties and selections are heterozygous in the peach/nectarine locus. Other minor population subgroups were identified (k=2 to 10): outgroups (other Prunus species), landrace germplasm, ornamental nectarine selections, highchill NCU cultivars and UF UGA USDA selections ( Figure 21). Analysis with all germplasm (n=195) and only UF varieties and selections (n=168) provided similar results. No additio nal structures were identi fied associated with other phenotypic characteristics. Conclusions The University of Florida genetic diversity has been maintained through breeding and introgression of superior germplasm sources since its creation in 1952 (UF UGA USDA NCSU, and others) UFs genetic diversity was high compared to other genetic diversity studies based on cultivars from around the world (Aranzana et al., 2010). Similarly, a p opulation structure was detected between melting and nonmelting flesh var ieties. Several loci closely located to the genome regions where different phenotypic traits have been previously mapped w ere detected under selection. This study constitutes the baseline for genetic diversity studies of low chill breeding programs around the world since UFs germplasm pool has been used as their founder material. Future studies will include quantitative traits and the use of pedigree information. 59

PAGE 60

Table 21 Peach germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program sinc e its creation in 1952 until present Peach germplasm representatives UF varieties and advanced materials Aztec Gold, EarliGrande, Early Amber, Flordabelle, Flordabest, Fl ordacrest, Flordadawn, Flordaglo, Flordagold, Flordaguard, Flordaking, Flordaprince, Flordastar, Martha Jane, Oro A, Rayon, Sunbest, Sunblaze, Suncoast, Sunhome, Sunlite, Sunmist, Sunraycer, Sunred, Sunsplash, TropicBeauty, TropicSnow, Tropic Sweet, UF2000, UFBeauty, UFBlaze, UFFlair, UFGold, UFO, UFOne, UFSharp, UFSun, UFQueen, UFRoyal, and advanced selection s from 2000 to 2010. UF UGA USDA varieties Gulfcrest, Gulfcrimson, Gulfking, Gulfp rince, Sunfre, Attapulgus White, and advanced selections. Landrac e cultivars Late Arkansas, Okinaw a, Red Ceylon, Strickland High chill NCSU cultivars Candor, Carolina Gold, China Pearl, Contender, Legend. Other Prunus species material P. dulcis cv. Nonpareil, P. kansuensis and UF apricot ( P. mume and P. armeniaca hybrid ) selections. 60

PAGE 61

Table 22. Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earlygold peach (T E) reference map. Chromosome Position Marker Fluorophore T a ( C) z Forward sequence Reverse sequence 1 9 CPSCT008 HEX 6 2 TGGATCCAATCCAAGAGTCTG GCAGCAAGTTGTTCTTGGTTC 1 25.8 CPDCT038 HEX 62 ATCACAGGTGAAGGCTGTGG CAGATTCATTGGCCCATCTT 1 33.9 CPPCT026 HEX 55 AGACGCAGCACCCAAACTAC CATTACATCACCGCCAACAA 1 47.3 BPPCT027 HEX 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 1 77.4 BPPC T028 HEX 57 TCAAGTTAGCTGAGGATCGC GAGCTTGCCTATGAGAAGACC 2 9.6 UDP98 025 FAM 57 GGGAGGTTACTATGCCATGAAG CGCAGACATGTAGTAGGACCTC 2 25 BPPCT013 HEX 57 ACCCACAAATCAAGCATATCC AGCTTCAGCCACCAAGC 2 38 BPPCT030 HEX 57 AATTGTACTTGCCAATGCTATGA CTGCCTTCTGCTCACACC 2 4 8.6 CPSCT034 HEX 62 AGGTGGACAATAGCCGTGAT TTTCCAGACCCTGAGAAAGC 3 18 BPPCT039 FAM 57 ATTACGTACCCTAAAGCTTCTGC GATGTCATGAAGATTGGAGAGG 3 36.4 CPDCT025 HEX 62 GACCTCATCAGCATCACCAA TTCCCTAACGTCCCTGACAC 3 46.4 CPDCT027 HEX 62 TGAGGAGAGCACTGGAGGAG CAACCGATCCCTCT AGACCA 4 10.4 CPPCT005 FAM 52 CATGAACTCTACTCTCCA TGGTATGGACTCACCAAC 4 28.3 UDP96 003 FAM 57 TTGCTCAAAAGTGTCGTTGC ACACGTAGTGCAACACTGGC 4 34.1 EPDC3832 FAM 57 CTTTTGAAGGCCCAATACCA ATCACTGCTTCGCCTTCATT 4 45.4 BPPCT023 HEX 57 TGCAGCTCATTACCTTTTGC AGATGTGC TCGTAGTTCGGAC 4 52.7 EPPISF032 HEX 57 TCCCCCACAGATATTTCAGC GTCGAGGAGAGAGGGCTTTT 5 5.2 BPPCT026 FAM 57 ATACCTTTGCCACTTGCG TGAGTTGGAAGAAAACGTAACA 5 20.1 BPPCT017 HEX 57 TTAAGAGTTTGTGATGGGAACC AAGCATAATTTAGCATAACCAAGC 5 32.9 BPPCT038 FAM 57 TATATTGTTGGCTT CTTGCATG GAGCTTGCCTATGAGAAGACC 5 44 BPPCT014 HEX 57 TTGTCTGCCTCTCATCTTAACC CATCGCAGAGAACTGAGAGC 6 8.7 CPPCT008 FAM 59 GAGCTCTCACGCATTAGTTT TTTGACTGCATAACAAAACG 6 17.5 UDP96 001 HEX 57 AGTTTGATTTTCTGATGCATCC TGCCATAAGGACCGGTATGT 6 30.1 BPPCT008 FAM 57 A TGGTGTGTATGGACATGATGA CCTCAACCTAAGACACCTTCACT 6 35.8 CPPCT015 FAM 50 TGGAGTGCCAATACTATTTA CATATGCATGGTTATGGT 6 41 EPPISF002 FAM 56 CGACGTGTGACCAAAGGAC GCAACTCCATCCACATTTCTC 6 56.4 BPPCT025 FAM 57 TCCTGCGTAGAAGAAGGTAGC CGACATAAAGTCCAAATGGC 6 72 UDP98 41 2 HEX 57 AGGGAAAGTTTCTGCTGCAC GCTGAAGACGACGATGATGA 7 9.5 CPSCT004 FAM 62 GCTCTGAAGCTCTGCATTGA TTTGAAATGGCTATGGAGTACG 7 18.6 CPPCT022 FAM 50 CAATTAGCTAGAGAGAATTATTG GACAAGAAGCAAGTAGTTTG 7 29.6 BPPCT029 FAM 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 7 3 8.9 CPPCT033 HEX 50 TCAGCAAACTAGAAACAAACC TTGCAATCTGGTTGATGTT 7 47.8 PMS2 FAM 55 CACTGTCTCCCAGGTTAAACT CCTGAGCTTTTGACACATGC 8 14.1 BPPCT006 HEX 57 GCTTGTGGCATGGAAGC CCCTGTTTCTCATAGAACTCACAT 8 20.8 UDP96 019 HEX 57 TTGGTCATGAGCTAAGAAAACA TAGTGGCACAGAGCAA CACC 8 54.7 EPDCU3117 FAM 57 CAGAGGGAACAGTGTGAGCA TGTTGTTGTCGACCCTGAAA zTa = annealing temperature. 61

PAGE 62

Table 23 Summary statistics of 36 simple sequ ence repeat (SSR) markers for 168 peach germplasm representatives of the UF varieties and advanced materi als (20002010 selections) at the University of Florida stone fruit breeding and genetics program sinc e its creation in 1952 until present Marker Genotype No. A z Ae Ho He F PIC BPPCT006A 19 7 3.96 0.66 0.75 0.11 0.72 BPPCT006B 12 7 3.02 0.53 0.67 0.21 0 .6 BPPCT008 30 y 12 5.34 0.61 0.81 0.25 0.79 BPPCT013 8 6 1.48 0.25 0.32 0.22 0.3 BPPCT014 8 5 1.85 0.44 0.46 0.04 0.41 BPPCT017 15 7 3.32 0.71 0.7 0.01 0.65 BPPCT023 16 9 2.55 0.6 0.61 0.01 0.54 BPPCT025 21 9 4.84 0.68 0.79 0.14 0.76 BPPCT026 10 6 1.71 0.36 0.42 0.14 0.38 BPPCT027 9 5 2.31 0.17 0.57 0.7 0.5 BPPCT028 4 3 1.22 0.16 0.18 0.12 0.17 BPPCT029 9 7 1.47 0.28 0.32 0.14 0.31 BPPCT030 11 8 1.92 0.37 0.48 0.23 0.42 BPPCT038 11 7 2.07 0.46 0.52 0.12 0.42 BPPCT039 5 4 1.33 0.22 0.25 0.13 0. 22 CPDCT025 14 6 2.27 0.41 0.56 0.26 0.52 CPDCT027 4 3 1.05 0.02 0.05 0.66 0.05 CPDCT038 26 11 4.35 0.65 0.77 0.15 0.73 CPPCT005 20 11 3.02 0.66 0.67 0.01 0.63 CPPCT008 6 7 1.06 0.03 0.06 0.49 0.06 CPPCT015 5 3 1.95 0.47 0.49 0.04 0.38 CPPCT022 24 1 2 3.25 0.53 0.69 0.24 0.65 CPPCT026 21 10 3.71 0.66 0.73 0.09 0.69 CPPCT033 15 6 2.54 0.54 0.61 0.11 0.56 CPSCT004 2 2 1.07 0.06 0.06 0.03 0.06 CPSCT008 5 5 1.87 0.02 0.47 0.95 0.37 CPSCT034 9 6 1.8 0.35 0.44 0.21 0.41 EPDC3832 4 3 2.02 0.92 0.5 0. 82 0.38 EPDCU3117 3 2 1.38 0.18 0.28 0.36 0.24 EPPISF002 6 3 2.04 0.46 0.51 0.09 0.43 EPPISF032 7 5 1.51 0.29 0.34 0.14 0.31 PMS2 6 3 1.28 0.18 0.22 0.2 0.21 UDP96 001 4 3 1.48 0.29 0.32 0.09 0.27 UDP96 003 24 9 4.16 0.66 0.76 0.13 0.72 UDP96 019 19 9 3.79 0.55 0.74 0.25 0.69 UDP98 025 12 7 1.66 0.32 0.4 0.2 0.36 UDP98 412 19 9 2.17 0.49 0.54 0.09 0.51 Average 11.97 6.41 2.37 0.41 0.49 0.17 0.44 zA = number of observed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = polymorphism information content. yNumbers in bold represent highest and lowest values for each variable. 62

PAGE 63

Table 2 4. Average statistics of 36 simple sequence repeat (SSR) markers for 170 germplasm representatives of the UF founder cultivars, varieties, and advanced materials (20002010 selections) separated by decades of selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Decade z Plant No. Genotype No. A y Ae Ho He F PIC Founders 3 2.46 2.84 2.51 0.26 0.53 0.53 0.46 1960 s 5 2.70 2.76 2.10 0.46 0.44 0.02 0.39 1970 s 9 3.51 2.95 1.97 0.39 0.41 0.04 0.36 1980 s 12 5.08 3.84 2.50 0.40 0.51 0.20 0.46 1990 s 15 4.54 3.32 2.21 0.45 0.44 0.04 0.39 2000 s 9 5 8.49 4.89 2.19 0.40 0.46 0.17 0.42 2010 s 31 6.03 3.81 2.17 0.42 0.45 0.05 0.40 Average 4.69 3.49 2.24 0.40 0.46 0.13 0.41 zDecade = decade in which genotype was selected. yA = number of observed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = polymorphism information content. 63

PAGE 64

Figure 21 Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) and population structure results for k=2 to k=10 of 36 simple sequence repeat (SSR) markers for 195 germplasm representatives of the geneti c pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program sinc e its creation in 1952 until present Cladogram rooted with apricots at the base. Clades are described on the right hand side of the clad ogram. Branch lengths represent Neis genetic distance values. Population structure results are located on the right hand side of the cladogram with different colors representing a different subgroup. 64

PAGE 65

Outgroups Landrace germplasm Ornamental Nectarines 00s High Chill Non showy NCSU 1 st Gen Non melting / Diamante cling Globose Sun Nectarines Showy series Sun Nectarines Non showy series Melting Flesh Non melting transition 80s and 90s Peento Non showy 00s and 10s Highlighter Nectarines 00s and 10s Non melting transition 90s Nectarines 10s White Flesh 00s Yellow Flesh Low chill (150 250 h) 00s White Flesh Low chill (150 250 h ) 00s White Flesh 00s and 10s Non melting Flesh 70s and 80s Low chill (0 150 h) K=2 K=3 K=4 K=5 K=6 K=7 K=8 K=9 K=10 65

PAGE 66

F igure 22. Analysis of the population structure results for 195 germplasm r epresentatives of the University of Florida genetic pool using the Evanno method (Evanno et al., 2005) implemented by the Structure Harvester software (Earl and VonHoldt, 2012) A) Second order change in the log likelihood delta K ( ). B) Rate of change of the log l ikelihood distribution (mean). C) Absolute value of the second order change in the log l ikelihood distribution (mean). D) The average log likelihood and the standar d error of two to five reps per run A B C D 66

PAGE 67

CHAPTER 3 THE NORTH AMERICAN PLUMS ( Prunus spp.) PHYLOGENETIC SIGNAL Introduction The genus Prunus is distributed around the world, with approximately 200 species. North America is an important center of diversity for plum species adapted to divergent climates and soil s. They harbor the greatest variation of flavor, aroma, texture, color, form and size among the stone fruits (Hedrick, 1911; Waugh, 1901) T axonomic conflicts in Prunus have been studied by Waugh (1901) and Hedrick (1911). Waugh (1901) reported that plums grow pretty much as they please, and the botanist has to take them as he finds them. He recognized that sometimes the new classifications changed the previously published species groupings and classifications. Prunus classifications and discrepancies were well summarized by Bortiri et al. (2001), with Rehders (1940) classification being recognized as the most widely used and accepted. Rehders (1940) classification divided Prunus into five subgenera: Pru nophora ( Prunus ), Amygdalus Cerasus Padus and Laurocerasus The subgenus Prunus was further divided into sections: Euprunus Prunocerasus (North American plums), and Armeniaca. Recent p hylogenetic studies supported the concept of Prunus as a monophyleti c group (single genus) However, the genus Prunus contained several poorly supported subclades/terminals (subgenera/species) ( Bortiri et al., 2001, 2002, 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Lee and Wen, 2001; Shaw and Small, 2004, 2005; Wen et al., 2008). Mason (1913) and Rehders (1940) taxonomic classification have b een mostly used for all of the Prunus phylogenetic analyses. Hereafter, in our study the North 67

PAGE 68

American plums will be referred to as section Prunocerasus (Bortiri et al., 20 01; Shaw and Small, 2004). The phylogeny of the genus Prunus was studied by Mowrey and Werner (1990) using isozymes. Prunus section Prunocerasus was found to be polyphyletic with P americana, P. munsoniana, P. hortulana, P. subcordata, and P. angustifoli a forming a clade, and with P. maritima and P. umbellata in another Badenes and Parfitt (1995) reported similar results to Mowrey and Werner (1990) using chloroplast DNA (cpDNA) sequences They recovered all conventional subgenus classifications based on R ehders ( 1940) taxonomic treatment : Prunus persica P. dulcis P. domesticaP. salicina and P. cerasus P.fruticosa monophyletic clades Two major groups: the Amygdalus Prunus group, and the Cerasus Laurocerasus Padus group were recovered by Lee and Wens (2001) phylogenetic analysis using ITS (internal transcribed spacer) sequences of ribosomal DNA The se results were not congruent with Rehders (1940) taxonomic treatment Bortiri et al. (2001) supported the genus Prunus monophyly using ITS and chloroplast trnL trnF spacer DNA sequences. Subgenera Padus Laurocerasus Cerasus and subgenera Prunus Amygdalus Emplectocladus Cerasus (sect. Microcerasus ) sect. Penarmeniaca formed two major clades, respectively. The plums of northeastern North America were found cl osely related, with P. mexicana sister to the rest of this clade. Bortiri et al. (2002) used the nuclear gene sorbitol 6phosphate dehydrogenase ( s6pdh) combined to ITS and trnL trnF sequences to improve the lack of definition for deep nodes in the subgenera Prunus Amygdalus Emplectocladus clade, as previously reported ( Bortiri et a l., 2001). Phylogenetic analysis of the combined data supported two 68

PAGE 69

major clades: subgenera Cerasus Laurocerasus Padus and subgenera Amygdalus Emplectocladus Prunus Section Microcerasus (subgenera Cerasus ) was found nested within subgenus Prunus Prunus subg enus Prunus sect ion Prunocerasus was reported to be monophyletic by Shaw and Small (2004). S even cpDNA regions: rpS16 rpL16, trnL trnG trnL trnF trnS trnG and trnH psbA w ere used Three clades were strongly supported in sect. Prunocerasus based on Waughs (1901) classification: the American Clade, the Chickasaw Clade, and the Beach Clade The American clade included P. americana var. americana, P. americana var. lana ta P. mexicana, P. rivularis P. hortulana, P. umbellata var. injucunda; the Chickasaw clade included P. angustifolia, P. munsoniana, P. gracilis P. nigra P. umbellata var. umbellata, P. alleghaniensis var. alleghanienses and P. alleghaniensis var. dav isii; and the Beach clade include d P. geniculata, P. maritima var. maritima and P. maritima var. gravesii The majority of phylogenetic research in Prunus has been done using cpDNA sequences (Badenes and Parfitt, 1995; Bortiri et al., 2001, 2002, 2006; Ka neko et al., 1986; Katayama and Uematsu, 2005; Shaw and Small, 2004, 2005; Wen et al., 2008; Table 12 ). The advantages and disadvantages of using cpDNA for plant phylogenetic analyses have been well summarized by Soltis and Soltis (1998). One of the advantages includes that the chloroplast genome of plants is small. Peach ( P. persica cv. Hakuhou) cpDNA has been estimated to be about 152 kb (Katayama and Uematsu, 2005). The chloroplast genome rate of evolution is considered to be slow in comparison with nuc lear genes. The latter could be considered an advantage and disadvantage when studying closely related species and species at the population level. In addition, a 69

PAGE 70

di sadvantage could be chloroplast transfer by inter specific hybridization (Soltis and Soltis 1998). The slow rate of evolution of several cpDNA sequences and their value to infer phylogenetic relationships in closely related species for some angiosperms, including Prunus was revised by Small et al. (1998), Shaw et al. (2005), and Shaw et al. (2007). Shaw et al. (2007) main objective was to identify chloroplast regions that would provide the greatest number of characters for low level molecular phylogenetic studies. He recommended selecting the top few choices with the greatest variability to be screened in a particular lineage in order to determine which of those are the most informative for a specific group. Morphological characters and nuclear DNA sequences proved of importance for Prunus phylogenetic analyses. However, t hey have been not widel y used in Prunus phylogeny as compared with cpDNA sequences ( trnL trnF trnS trnG etc). Bortiri et al. (2006) showed the importance of ITS sequences, cpDNA regions, and morphological characters for Prunus phylogenetic reconstruction. The combined data provided support of some nodes that were previously identified in Prunus (Bortiri et al., 2001). Several synapomorphies supported large groups, proving benefic ial resolution for some clades. Similarly, Rohrer et al. (2004) reported that microsatellite markers could provide the genetic variability necessary to resolve the relationships within Prunus at the species level. The main objective of this research is to measure and to identify additional genomic regions that could provide the greatest number of charact ers, variability, and 70

PAGE 71

improved phylogenetic signal at the species level for Prunus section Prunocerasus relationships Material and Methods Plant Material A total of 11 species one accession per taxa, were collected from the wild (excluding P. persica hap loids ) by T. G. Beckman, W R. Okie D.J. Chvez and J. X. Chaparro (Table 31) A total of 13 samples (including two peach haploids) were used for this study. These species represented outgroups and major clades within subgenus Prunus section Prunocerasus a s reported by Shaw and Small (2004) The American clade was represented by P. americana, P. hortulana, and P. mexicana; the Chickas aw clade included P. angustifolia P. munsoniana and P. umbellata; and the Beach clade was represented by P. geniculata and P. maritima Outgroups included: P. pumila (subgenus Prunus section Penarmeniaca); P. persica cv. Okinawa, AP0518ws (UF haploid peach selection ), 02 01c (UF haploid peach selection) (subgenus Amygdal us ); and P. fasciculata (subgenus Emplectocladus ). Peach haploids were used for confirmation of sequencing products. Collected plant specimens were deposited as herbarium vouchers in the University of Florida Herbarium (FLAS) at the Florida Museum of Natural History Gainesville, FL, USA. Additional leaf materi al from the collected samples were freeze dried using a Labconco freezone 2.5L system (Labconco, Kansas City, MO USA ) or dried using a mixture of 2.5 kg silica gel dessicant 28200 mesh (Cat No. S157212) and 500 g silica gel Tel Tale (TM) dessicant indi cating 1018 mesh (Cat No. S161500) ( Thermo Scientific Waltham, MA USA ) Dried sample materials are stored at the Stone Fruit Breeding and Genetics Program at the University of Florida, Gainesville, FL. 71

PAGE 72

DNA I solation DNA was extracted from leaf tissue using a modified CTAB method as described by Blaker (2010), and Chavez and Chaparro (2011). Lyophilized leaf tissue ( ~10 mg) of each sample was added to 2 mL Eppendorf microcentrifuge tubes with three 5 mm stainless steel beads, 750 L of CTAB buffer (2% CTAB, 100 mM Tris pH 8.0, 1.4 M NaCl, 0.5M EDTA, 1% PVP) previously mixed with Mercaptoethanol (1 L/mL), and 8 L of RNAse (10 mg/mL). Samples were grounded two/three times at 30 Hz in a Tissue Lyser (QIAGEN Inc., Valencia, CA, USA) for 1.5 min until tiss ue clumps were not visible. Samples were vortexed and incubated in a 65C water bath for 6 min. Then, tubes were vortexed and a volume of 750 L of chloroform:isoamyl (24:1) was added. Tubes were vortexed, incubated at 20C for 6 min, and then centrifuged at 12000 rcf for 10 min. The aqueous phase was transferred to a new 2 mL centrifuge tube, and 500 L of cold isopropanol were added. Tubes were gently mixed, incubated at 20C for 6 min, and then centrifuged at 16100 rcf for 10 min. Supernatant was remov ed, and pellet was washed with 500 L of cold 70% EtOH (by inverting the tubes carefully). Tubes were incubated at 20C for 5 min., and then centrifuged at 16100 rcf for 5 min. Supernatant was removed, and pellet was washed with 500 L of cold 90% EtOH (repeating mixing, incubation and centrifugation, as described before). Ethanol was poured off and the pellet was dried at bench top for approx. 3045 min (room temperature). The pellet DNA was resuspended in 50 L TE buffer (10 mM Tris HCl, 0.1mM EDTA) and 50 L of DNA grade water DNA concentration was quantified in a UV10 Spectrophotometer (Thermo Scientific, Waltham, MA USA ). DNA concentration for all the samples was standardized to 20 ng/uL. 72

PAGE 73

Genomic Regions SSR Markers A total of 41 SSR markers distributed ac ross the peach genome (~11.3 cM between markers) were selected from the Prunus Texas almond Earlygold peach (T E) reference map (Dirlewanger et al., 2004; Jung et al., 2008). Forward markers were modified using a 5 fluorophore label 6FAM (standard) [6FAM] or HEX [5HEX] (Eurofins MWG Operon, Huntsville, AL, USA) for multiplex product fragment analysis ( Table A 7 ). Two haploid genotypes were previously used as controls to test product amplification accuracies for each SSR marker (Chapter 2) PCR products were amplified in a 16 L volume reaction containing 2 L of 20 ng/L DNA template, 2.25 L 10X ThermoPol Reaction Buffer [ 10mM KCl, 10mM (NH4)2SO4, 20mM Tris HCl, 2mM MgSO4, 0 .1% Triton X 100, pH 8.8 @ 25C] 1 L 2.5 mM dNTPs, 0.2 L Taq DN A Polymerase, 6.55 L DNA grade water, and 4 L 5M (2 L forward and 2 L reverse) primers. PCR parameters were: 3 min at 94C followed by 40 cycles of 30 s denaturing at 94C, 30 s at primers specific annealing temperature [Ta(C)] (Table A 7 ), and 1 mi n of elongation at 72C, ending with 7 min at 72C. PCR products were separated on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Gel images PCR amplification intensities were used to determine PCR dilution ratios for fragment analysis on an ABI3730 sequencer (Applied Biosystems Grand Island, NY, USA) at the Interdisciplinary Center for Biotechnology Research (ICBR), University of Florida, Gainesville, FL. Fragment analysis of chromatog raphs were visualized using GeneMarker v.1.6 software (SoftGenetics, LLC, State College, PA, USA) using 600 LIZ size standard (Applied Biosystems, Grand Island, NY, USA). 73

PAGE 74

Chloroplast DNA Several cpDNA regions and their value to infer phylogenetic relati onships in closely related species for some angiosperms, including Prunus were revised by Small et al. (1998), Shaw et al. (2005), and Shaw et al. (2007). From those studies, a total of 8 cpDNA r egions ( with product amplification less than 1000bp) were selected based on their high polymorphism rates or proportion of mutational events for Prunus Primer sequences were obtained from Shaw et al. (2005, 2007) and Morris et al. (2008) (Table A 8 ). PCR products were amplified in a 48 L volume reaction containin g 6 L of 20 ng/L DNA template, 6.75 L 10X ThermoPol Reaction Buffer [ 10mM KCl, 10mM (NH4)2SO4, 20mM Tris HCl, 2mM MgSO4, 0.1% Triton X 100, pH 8.8 @ 25C] 3 L 2.5 mM dNTPs, 0.6 L Taq DNA Polymerase, 19.65 L DNA grade water, and 12 L 5M (6 L forward and 6 L reverse) primers. A universal cpDNA PCR procedure was used to simplify amplification as reported by Shaw et al. (2007). PCR cycling conditions were: 5 min at 80C followed by 30 cycles of 1 min denaturing at 95C, followed by a ramp of 0.3C/s to 65C, and 4 min of elongation at 65C, ending with a final extension step of 5 min at 65C. PCR products were confirmed on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Only cpD NA re gions that were easily amplified for all the samples were used for further sequencing ( ndhJ trnF and trnT trnL region primers did not amplify products for several species and therefore were not used for further analyses ). Internal Transcribed Spacer The us e of ITS sequence data has proven of value to define species relationships within the genus Prunus (Bortiri et al 2001, 2002, 2006; Lee and Wen, 74

PAGE 75

2001). This region was amplified using the primers published by Wen and Zimmer (1996) : C26A, 5 GTT TCT TTT C CT CCG CT 3and N c18S10 5 AGG AGA AGT CGT AAC AAG 3 PCR products were amplified in a 48 L volume reaction as previously described for cpDNA amplification. PCR cycling conditions were: 3 min at 94C followed by 20 cycles of 1 min denaturing at 94C, 2 min at 50C, and 2 min of elongation at 72C PCR products were confirmed on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Nuclear Genes A total of 3 3 major genes associated with dormancy response, two major genes associated with branching and three genes expressing isozymes were used for this study. GenBank annotated collection database of DNA sequences was used to find the Arabidopsis Heynh. gene sequences and accessions corresponding to those genes (Benson et al., 2013). Arabidopsis gene sequences were blasted against the annotated peach genome v1.0 using BLASTTN option in Phytozome (Goodstein et al., 2012). BLAST resu l ts with the highest score and E value were used to identify t he best peach annotated transcript matching the Arabidopsis genes. Gene ancestry for each peach transcript was analyzed in Phytozome to remove possible multiple copy gene targets Peach gene seq uences and their mRNA were aligned to identify exonintron boundaries using Muscle v3.8 (Edgar, 2004) Specific primers for each region were created in Primer3Plus v2.3.5 webinterface (Untergasser et al., 2012) using consecutive exons as annealing sites t o amplify the spanning intron region (unless otherwise specified) Similarly, a n additional fourteen c andidates genes primer sets associated with axillary 75

PAGE 76

meristem formation and bran ching in peach were obtained as reported by Carrillo Mendoza (2012) ( Table A 9 ) PCR products were amplified in a 48 L volume reaction, as previously described for cpDNA and ITS regions amplification. PCR parameters were: 3 min at 94C followed by 40 cycles of 30 s denaturing at 94C, 30 s at primers specific annealing tempera ture [Ta(C)] (Table A 9 ), and 1 min of elongation at 72C, ending with 7 min at 72C. PCR products were separated on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Sequencing PCR produc ts with multiple bands during electrophoresis were separated, exised and purified using QIAquick Gel Extraction Kit (Qiagen, Valencia, CA, USA). Single band products were purified by Eurofins MWG Operon (Huntsville, AL, USA). All purified products were for ward and reverse direct sequenced by Eurofins MWG Operon (Huntsville, AL, USA) following company procedures. Geneious R6.0.3 software (Biomatters Ltd., New Zealand) was used to edit and assemble complementary DNA strands. Assembled contigs of gene sequence regions were used to create internal primers that would improve sequence coverage, quality, and base calling accuracy if necessary (Table A 1 0 ) Out of those gene regions, two or more polymorphic sites were identified for CUC2, ELF6, FT TSF, GIFB, LFY, MAF4, and s6pdh These regions were resequenced using these internal primers. Phylogenetic Analyses G enetic variation and cluster analyses for SSR fingerprinting data were performed using Powermarker v.3.25 (Liu and Muse, 2005) and GenAlEx v.6.5 software 76

PAGE 77

( Peakall and Smouse, 2006, 2012). Genetic variation was characterized for each locus by the number of observed alleles ( A ), effective number of alleles ( Ae ), observed heterozygosity ( Ho ), expected heterozygosity ( He ), Wrights fixation index [ F=(He Ho)/He=1 (Ho/He) ], and polymorphism information content ( PIC ). Neis genetic distance was calculated and used in Neighbor Joining (NJ) cluster analysis ( Nei and Takezaki, 1983 ; Saitou and Nei, 1987) Alignment s of DNA sequences were performed using Muscle v3.8 (Edgar, 2004). Manual alignment s w ere made in Geneious R6.0.3 software (Biomatters Ltd., New Zealand) to produce an alignment with the fewest number of changes. Alig n ments were visually checked for polymorphic sites that resulted in sequence slippage. P olymor phisms (indels and/or substitutions ) were manually determined using reference sequences from other spec ies with full sequence coverage ( without slippage from forward and reverse sequencing ) Alternative sequences or alleles were kept and used in further an alyses A cons ensus sequence was obtained for each genotype. All genotype sequences per region were re aligned using Muscle v3.8 (Edgar, 2004). Haplotypes for each species were calculated based on aligned amplicons using DnaSP v5 software (Librado and Rozas, 2009). The number of nucleotide substitutions ( Subst ) and indels collectively named Potentially Informative Characters ( PICs=Subs+ Indels ) between the ingroup species (American, C hickasaw and Beach clades in Prunus section Prunocerasus ) and between ingroup and outgroup species phased haplotypes were calculated using DnaSP v5 software (Librado and Rozas, 2009) The proportion of observed mutational events or percent variability for each genomic region w ere calculated as reported by 77

PAGE 78

Shaw et al. (2005, 2 007) by using the following equation: % variability = [( Subs+ Indels )/L] 100, where L = the total aligned sequence length. P hylogenetic analyses were performed by using the optimality criterion of maximum parsimony for diploid and phased haplotype sequenc es Gaps (indels) were ignored and also considered as an additio nal binary character matrix, with 0=absence and 1=presence of the indel Maximum parsimony (MP ) trees were obtained by using PAUP* v. 4.0b10 (Swofford, 2002). A heuristic search for MPTs was conducted by using tree bisectionreconnection (TBR) branch swapping and 100 0 random sequence addition replicates. Bootstrap sup port (Felsenstein, 1985) was estimated in 100 0 replications of heuristic search and simple taxon addition with at least 100000 re arrangements per replicate. Homoplasy was determined by calculating the consistency index (CI) and the retention index (RI). Model of sequence evolution for each region was determined using the Akaike information criterion ( AIC ; Akaike, 1974) and Bayesian information criteria (BIC; Schwarz, 1978) as implemented in the JModeltest v.0.1.1 software (Posada, 2008). A likelihood ratio test was calculated to choose the best model with the best lnL score between the AIC and the BIC criteria Heuristic searches for maximum likelihood trees (MLT) were conducted using GARLI 2.0 (Zwickl, 2006) in the CIPRES Science Gateway (Miller et al., 2010). A nonparametric b ootstrap was perfor med using two independent search replicates and 50 bootstrap repetitions. Best ML trees for each bootstrap re petitions and replicate were used to calculate a consensus tree with posterior probability values in Mesquite v.2.73 (Maddison and Maddison, 201 1 ). 78

PAGE 79

Maximum likelihood analyses of individual and combined sequence data were performed us ing RAxML v.7.1.0. (Stamatakis, 2006; Stamatakis et al., 2008). Analyses were calculated under a general time reversible model of nucleotide substitution and a gamma model of rate heterogeneity in the CIPRES Science Gateway (Miller et al., 2010). A halt bootstrapping 1000 replicates, was performed. Best ML trees for each bootstrap repetition were used to calculate a consensus tree with posterior probability values in Mesquite v.2.73 (Maddison and Maddison, 201 1 ). Network analy se s of haplotype and diploid sequence data were performed as described by Oatley et al. (2012). Networks of the phased haplotypes were estimated using TCS v.1.21 ( Clement et al., 2000). TCS analyses were performed treating gaps as missing data and as a 5th character state. Multi locus n etworks were built using SplitsTree v.4.11.3 (Huson and Bryant, 2006) based on a distance matrix generated in POFAD v.1.03 (Joly and Bruneau, 2006). Uncorrected pdistances from aligned sequences were calculated in PAUP* v. 4.0b10 (Swofford, 2002) and wer e used as input in POFAD. Results SSR s A total of 41 SSR primers, amplified 471 alleles for a subset of 11 species representing outgroups and the major clades within s ubgenus Prunus section Prunocerasus (Figure 3 1) An average of 9 genotypes (out of 11 samples) were amplified per marker, an average number of alleles ( A ) of 10.70, an average effective number of alleles ( Ae ) of 8.10, observed heterozygosity ( Ho ) of 0.58 expected heterozygosity ( He ) of 0.84 Wrights fixation index ( F ) of 0.31 and polymorphism information content ( PIC ) of 0.82 (Table 3 2). Fifteen SSR markers amplified the 79

PAGE 80

maximum number of genotypes (11 out of 11). CPPCT015 amplified the lowest number of genotypes of 2 The value of A varied from 17 for BPPCT025 to 3 for CPPCT015 BPPCT023, BPPCT025, and BPPCT039 had the highest Ae of 13.44. CPPCT022 had the lowest Ae of 2.35. The value of Ho varied from 1.00 to 0.09 for CPPCT006 and EPDCU3117, respectively. BPPCT025, BPPCT026 and CPDCT025 had the highest He of 0.93 compared to the lowest He of 0.63 for CPPCT015. Wrights fixation index ( F ) ranged from 0. 87 for CPPCT033 to 0. 05 for BPPCT029 Six SSRs, BPPCT025, BPPCT026, BPPCT039, CPDCT025, EPDC3832B, and EPDCU3392 had the highest PIC of 0.92 compared to CPPCT015 of 0.55. Neis genetic dista nce (Nei and Takezaki, 1983) was calculated in combination with Neighbor Joining (NJ) (Saitou and Nei, 1987) cluster analysi s for 41 SSR markers ( Figure 32 ). Subgenus Prunus section Prunocerasus w as monophyletic Relationships within the North American pl um species were not in agreement to previous results found by Shaw and Small (2004). The outgroup species clustered together as expected. The Chickasaw clade ( P. munsoniana and P. angustifolia ) was recovered. The American and the Beach clade were not obser ved. C luster analysis using six SSR markers with the highest PIC values yielded somewhat consistent results (Figure 32). BPPCT017, BPPCT025, BPPCT026, CPDCT025, PMS2, and UDP98412, were used for this analysis. Outgroups clustered together, however, P. me xicana was found as a sister species to the outgroup. The Beach clade ( P. geniculata and P. maritima ) and the Chickasaw clade ( P. angustifolia, P. munsoniana) were recovered. Prunus americana and P. umbellata were 80

PAGE 81

monophyletic. P runus hortulana was recover ed as sister species to all the section Prunocerasus clade. Chloroplast DNA A total of 7 cpDNA regions were PCR amplified and sequenced for all 11 species. Only ndhJ trnF region did not amplify overall and was not used for further analyses (Table A 8 ) A total of 3584 bp of alignable phased haplotype sequences (130 PICs, average polymorphism rate of 3.68 % ) were ob tained (Table 33). The highest amounts of informative sites were observed in 3'trnV ndhC and trnH psbA cpDNA regions for section Prunocerasus wit h 7.23% and 3.28% variability, respectively. Similar results were identified among the outgroup species with 3.17% and 1.51% variability respectively (Figure 3 3) Both regions yielded good species delimitations within section Prunocerasus and outgroup species in MP and ML analyses, similar to those reported by Shaw and Small (2004) (Figures A 15 and A 18 ; Tables A 1 2 and A 1 3 ) MP analyses with gaps as an additional character helped identifying additional species relationships in comparison to MP analyses without gaps Similarly, MP and ML analyses were consistent in all and across the cpDNA regions for sequence data (Figures A 15 to A 20). Phylogenetic analyses of trnL trnF trnL intron and trnQ 5rps16 cpDNA regions recovered internal relationships for section Prunocerasus ( Figures A 16 and A 17 respectively) Section Prunocerasus was found monophyletic for all regions in MP and ML analyses with diploid or with haplotype number prefix data In addition, P. americana and P. mexicana (American clade) were found cluster ed together for all cpDNA regions (except atpB rbcL ) Prunus geniculata was found near the base of section Prunocerasus for trnL trnF and trnL intron, trnQ 5rps16, and trnH psbA Other cpDNA 81

PAGE 82

regions did not show any additional species relati onships. Phased h aplotype MP analyses found that P. angustifolia, P. umbellata and P. munsoniana shared the same cpDNA haplotype for trnL trnF and trnL introns, trnQ 5rps16, trnH psbA ndhF rpL32, and atpB rbcL These species formed part of the Chickasaw clade as described by Shaw and Small (2004) Combined cpDNA regions phylogenetic analyses using MP and ML criteria found section Prunocerasus as a monophyletic clade (Figure 3 4 ). P runus americana and P. mexicana were found clustered together in all analys es. Prunus geniculata was identified as a basal sister species to P. americana and P. mexicana Similarly, P. hortulana, P. munsoniana, P. maritima P. umbellata, and P. angustifolia were clustered together. MP and ML trees were consistent across regions The non standarized multi locus combined cpDNA sequence data network (Figure 3 5 ) recovered similar species relationships to those produced by MP and ML analyses Additional sequence data information and models of sequence evolution are available i n Tables A 1 1 to A 1 7 Internal Transcribed Spacer A total of 610 bp of alignable haplotype sequences ( 66 PICs, average variability of 10.82 % ) were obtained (Table 34). Section Prunocerasus variability was 2.30% compared to outgroup variability of 8.52% (Figure 3 3) Section Prunocerasus was found monophyletic for both, MP and ML, analyses (Figure A 21; Tables A 1 2 and A 1 3 ). Prunus munsoniana and P. americana were clustered together in MP of diploid sequence data without gaps However, this species relationship w as lost when using gaps as an additional character in MP analyses. Other relationships among the North American plums were not recovered. These results were consistent across diploid and 82

PAGE 83

phased haplotype sequence data analyses. Additional sequence data inf ormation and models of sequence evolution are presented in Tables A 1 1 to A 1 7 Nuclear G enes Isozymes Three isozyme genes, PGI (Phosphoglucose isomerase 1), PGDH (Phosphoglycerate dehydrogenase), and s6pdh (sorbitol 6phosphate dehydrogenase), were sequenced (11 core species plus two peach haploids ). A total of 1405 bp of alignable sequence ( 117 PICs variability of 8.77% ) was analyzed (Table 34). The PGI primers used failed to amplify the Prunus americana PGI sequence R eadable sequences were obtained fo r all other samples and isozymes The highest amount of PICs within section Prunocerasus and between the outgroup taxa were obtained for PGI, followed by s6pdh and PGDH ( Table 34 ) The sequence polymorphism rate for PGI within section Prunocerasus was 7.6 9 % compared to 4.45% and 3.56% for s6pdh and PGDH, respectively. Similar trends of 4.01%, 3.02%, and 2.94% polymorphism rate, respectively, were observed when the outgroup species ( P. fasciculata, P. persica and P. pumila ) were removed (Figure 3 3) S ecti on Prunocerasus was consistently recovered as a monophyletic clade with MP and ML analyses using either diploid or haplotype datasets for all three genes MP analyses using gaps as an additional source of information for the enzyme sequences, did not yield additional resolution of speci es relationships (Figure A 22, A 23 and A 24; Tables A 1 2 and A 1 3 ) Prunus munsoniana and P. angustifolia consistently cluster ed together in all diploid and phased haplotype sequence data analyses with MP and ML. Both species integrated into the Chickasaw clade, as previously described. Other previously known species relationships such as the American clade and the Beach clade, were not 83

PAGE 84

identified. MP and ML analyses of PGI diploid and phased haplotype sequence data, recovered P. geniculata and P. hortulana cluster ed together at the base of the section Prunocerasus clade (Figure A 23) ML analysis of PGI haplotypes recovered the Chickasaw clade (including P. umbellata) Analysis of the s6pdh sequence data, produced a polytom y consisting of P. americana, P. geniculata, P. persica P. pumila and P. fasciculata (Figure A 24). MP and ML haplotype analyses for PGDH revealed that P. geniculata, P. angustifolia and P. munsoniana shared a common haplotype. In addition, P. geniculata was located basal to the section Prunocerasus species in a polytomy with P. persica and P. pumila (PbesIL15) (Figure A 22). MP and ML analyses of the combined isozymes diploid data recovered section Prunocerasus as a monophyletic group. Prunus angustifoli a and P. munsoniana were clustered together in all analyses (Chickasaw clade). MP analyses (with and without gaps) did not provide any additional resolution of species relationships. However, ML analyses of combined data recovered the Chickasaw clade as a monophyletic group ( P. umbellata, P. angustifolia and P. munsoniana ) T he American clade ( P. americana and P. mexicana) was found clustered to P. maritima as a sister species. Prunus hortulana and P. geniculata were identified as basal to species within section Prunocerasus (Figure A 77). The non standarized multi locus (phased haplotype) combined isozymes sequence data network (Figure A 78) recovered similar species relationships to those produced by MP and ML analyses. The Chickasaw and American clade w ere recovered. Prunus maritima was i dentified clustered to P. umbellata. Additional 84

PAGE 85

sequence data information and models of sequence evolution are available in Tables A 11 to A 1 7 Branching Sixteen axillary meristem formation and branching genes (AXR1, BR C1, BRC2, CUC1, CUC2, CUC3, LAS, MAX1, MAX2, MAX3, MAX4, PIN, RAX1, RAX2 RAX3, REV, SPS; Table A9 ) were sequenced in the 11 core species and two peach haploids CUC1 and MAX1 amplified two targets in all the species studied. Both amplicons were purified, sequenced independently, and identified by A and B suffixes i.e. CUC1A and CUC1B. A total of 6664 bp of sequence containing 426 PICs and having an average polymorphism rate of 6.39 % was ob tained (Table 34). T wo or more indel polymorphisms were detected i n the CUC2 amplicon Additional internal primers were designed to allow the generation of unambiguous sequence data as specified on Table A 1 0 Primers designed for all the target sequences yielded product for all species. However, we were not able to completely sequence the BRC2 amplicon from P. americana. The highest polymorphisms rate within section Prunocerasus and between outgroup were obtained for MAX4 MAX1 CUC2, MAX3, and AXR1 ( Table 34 ) Percent variability within section Prunocerasus was 5.02%, 3.36% (2.48% for MAX1B), 3.12%, 2.89%, and 2.28%, respectively (Figure 33). A s imilar trend was observed for these genes polymorphism rate when the outgroup species was used for the analyses MP and ML analyses of diploid and haplotype datasets for AXR1, BRC1 (only for MP), BRC2, CUC2, CUC3, LAS, MAX1 (MAX1A and MAX1B), MAX2, MAX3, MAX4, RAX1, RAX2 RAX3, REV, and SPS recovered section Prunocerasus as a monophyletic clade (Figure A 25 to A 42; Tables A 1 2 and A 1 3 ) 85

PAGE 86

AXR1 recovered the American clade relatio nships in diploid sequence analyses using MP and ML. Two major groups were identified in these analyses: 1) Prunus maritima P. umbellata and P. angustifolia which shared the same haplotype sequence, and 2) P runus angustifolia (additional haplotype), P. g eniculata, P. munsoniana, P. mexicana, P. americana, and P. hortulana. Both groups represented the Chickasaw and the American clade. Prunus geniculata and P. angustifolia haplotypes contained haplotypes in each clade (Figure A 25). AXR1 recovered the Ameri can and the Chickasaw clades within section Prunocerasus and constituted a good candidate for further studies in genus Prunus BRC1 BRC2, CUC1 (CUC1A and CUC1B) sequence data provided limited information for elucidating species relationships within section Prunocerasus MP analyses recovered section Prunocerasus as a monophyletic group ( except for CUC1; Figure s A 26, A 27 A 28, and A 29). The polymorphism rate of the s e genes was similar to some of the cpDNA regions. These regions could be candidates for a nalyses at the subgenus/genus level in Prunus. Analysis of CUC2 recovered section Prunocerasus as monophyletic group in MP and ML analyses using diploid and phased haplotype sequence data. Prunus americana, P. maritima and P. mexicana clustered together i n both analyses (Figure A 30) Similarly, CUC3 phylogenetic analyses differentiated section Prunocerasus from the outgroup species for MP and ML analyses. Only the ML analyses of the diploid sequence data yielded monophyletic clades that supported the Beac h clade (with P. umbellata joined to this clade), the American clade, and the Chickasaw clade within 86

PAGE 87

section Prunocerasus H aplotype sequence analyses recovered species haplotypes in different clades (Figure A 31). LAS analyses yielded a monophyletic clade for section Prunocerasus The diploid MP analyses did not identify additional relationships. Diploid ML and phased haplotype MP ML analyses presented P. americana and P. mexicana clustered together Similarly, P angustifolia, P. umbellata and P. munsoni ana (from the Chickasaw clade) were clustered together (Figure A 32). Analysis of diploid and haplotype sequence for MAX1 (MAX1A and MAX1B) also identified section Prunocerasus as a monophyletic group, but it did not provide any additional resolution to species relationships within section Prunocerasus (Figure A 33 and A 34). Similarly, MAX2, MAX3, PIN, and RAX1, did not recovered any species relationships within section Prunocerasus (Figures A 35, A 36, A 37 and A 38 ) MAX4 identified section Prunocerasus as a monophyletic clade. Two major groups were recovered within section Prunocerasus in all analyses: 1) The American clade species and P. geniculata, and 2) The Chickasaw clade species, P. pumila and P. maritima (Figure A 37). This was the region with th e highest percent variability MP and ML recovered several species clusters within section Prunocerasus MAX4 is a good candidate for a whole phylogenetic analysis to identify additional species clades in section Prunocerasus RAX2 RAX3 recovered the Chick asaw clade relationships in diploid sequence analyses using MP (with gaps) and ML. Prunus maritima P. umbellata, and P. angustifolia shared the same haplotype sequence and were clustered together (Figure 87

PAGE 88

A 40). The phylogenetic relationships identified wi th RAX2 RAX3 were similar to those recovered with AXR1. No additional species groupings were obtained. REV differentiated section Prunocerasus from the outgroup species in MP and ML analyses. The American clade species, P. americana and P. mexicana, were clustered together in MP analyses. Additional species relationships within section Prunocerasus were recovered when using phased haplotype sequences (Figure A 41 ). MP and ML analyses of all the branching genes diploid sequences recovered section Prunoceras us as a monophyletic group. Prunus angustifolia and P. munsoniana were clustered together in all analyses (Chickasaw clade). Similarly, P. americana and P. mexicana (Americana clade) and P. maritima and P. umbellata, formed monophyletic clades, respectively MP analyses (with gaps) yielded two major groups: 1) The American clade ( P. mexicana, P. americana, and P. hortulana) and P. geniculata; and 2) The Chickasaw clade ( P. umbellata, P. angustifolia and P. munsoniana) and P. maritima ML analyses of all se quence data recovered similar species groups to MP analysis with gaps. Prunus pumila was identified as a sister species to section Prunocerasus (Figure A 79 ). The non standarized multi locus (phased haplotype) combined branching genes sequence data networ k (Figure A 80) recovered similar species relationships to those produced by MP (with gaps) and ML analyses. The Chickasaw and American clades were recovered. Prunus maritima was identified clustered to P. umbellata (similar to the isozyme combined data an alyses). Additional sequence data information and models of sequence evolution are presented in Tables A 11 to A 1 7 88

PAGE 89

Flowering A total of 33 dormancy related genes (AGL24, AGL20SOC1, BFT, BFM, CO, CRY1, CRY2, ELF6, FD, FD1, FG, FLC FLF, FPF1, FRIGIDA, FT TSF, GI FB, HOS1, LFY, MAF1 MAF3 AGL31, MAF2, MAF4, MAF5, MFT, PHYA, PHYBPHYD, PHYE, RGA RGA1, RGL1 RGL2 RGL3, SPY, TFL1 ATC, TFL2, VRN1, and VRN2; Table A 9 ) were sequenced in the 11 core species and two peach haploids MAF2 yielded duplicate products for all species. Duplicate bands were exised and sequenced as previously described on the materials and methods section. Duplicate products were designed using A and B suffixes i.e. MAF2A and MAF2B. The AGL20 SOC1 amplicons for P. americana and P. munsonian a could not be sequenced using the available primers and were removed from the dataset. FRIGIDA sequences were difficult to assemble. Prunus umbellata, P. angustifolia P. mexicana, P. hortulana, and P. munsoniana were not able to be sequences for FRIGIDA Similarly, P. geniculata and P. pumila were not sequenced for MAF2A, and P. angustifolia and haploids were not sequenced for MAF2B. The problems with the MAF2 amplicon were due to absence/presence of duplicate amplicon in some species. A total of 15646 bp of alignable sequence (1206 PICs polymorphism rate of 7.71% ) was obtained (Table 34). Two or more indels were identified for the ELF6, FT TSF, GI FB, LFY, and MAF4 amplicons in one or more species. These amplicons were resequenced for all the samples using specif ic internal primers as indicated in Table A 1 0 The highest polymorphism rates were observed in MAF4, FLC FLF, LFY, MAF2A, AGL20 SOC1, ELF6 and AGL24 (Table 34). Percent variability within section Prunocerasus was 9.28 %, 7.13%, 6.67%, 5.60 %, 5. 18%, 4.05%, and 4.05%, 89

PAGE 90

respectively (Figure 33). Percent variability for the outgroup species did not show a similar trend Some genes had higher polymorphic rates among outgroup species. Section Prunocerasus was monophyletic in MP and ML analyses using d iploid and haplotype datasets for all gene regions except FPF1, PHYA, and RGA RGA1 (Figure A 43 to A 76 ; Tables A 1 2 and A 1 3 ). MP analyses with gaps recovered some of the species relationships within section Prunocerasus the American and the Chickasaw cl ade, as previously identified using isozymes and branching gene sequences. MP clustered the Chickasaw clade species and P. geniculata together using the AGL24 diploid and phased haplotype data. ML analyses did not yield similar results. Prunus maritima was located near the base of section Prunocerasus Haplotype MP phylogenetic analysis recovered the American clade and the Chickasaw clade (Figure A 43). The use of gaps in this analysis did not provide additional information. This region had some difficulties during sequencing due to high polymorphisms Additional internal primers could be created to improve sequence quality and coverage. AGL20 SOC1, BFT, BRM, CO, FD, FD1, FPF1, HOS1, MAF1 MAF3 AGL31, MAF2B, MAF4, MAF5, MFT, PHYA, PHYBPHYD, RGA RGA1, RGL1 RL G2 RLG3, TFL1 ATC, and VRN2, did not identify any novel relationships within section Prunocerasus MP and ML analysis of phased haplotype data for these genes produced more species clusters to those produced using diploid data. Prunus munsoniana and P. ang ustifolia were clustered together (in some occasions also with P. umbellata) These additional groupings were not consistent across genes (Figure A 43 to A 76 ). CRY1 identified section Prunocerasus as a monophyletic group in MP and ML analyses with diploid and phased haplotype sequence data. Prunus munsoniana and P. 90

PAGE 91

angustifolia were clustered together (Chickasaw clade) This clade was sister to the other species within section Prunocerasus Prunus maritima and P. umbellata were identified in a cluster. No additional species relationships were observed (Figure A 48). CRY2 ELF6 and FG diploid data analyses generated somewhat similar results than CRY1 (Figure A 49, A 50 and A 51). Prunus munsoniana and P. angustifolia were clustered together (Chickasaw clade). However, different species relationships were identified when using haplotype sequence data for each region. ELF6 MP a nalyses ( with gaps ) for diploid and haplotype data yielded more species clusters than their counterpart. FG phased haplotype MP and ML analyses recovered the Americana and the Chickasaw clade relationships. No addi tional clusters were identified across genes. FLC FLF recovered section Prunocerasus as a monophyletic clade. Additional species relationships within t his section were not cons istent, with MP (with and without gaps) and ML analyses producing different results (Figure A 54). This region did not improve speci es phylogenetic relationships within section Prunocerasus although this region had high polymorphism rate in comparison wit h other gene regions. FRIGIDA phylogenetic analyses were not representative of the core collection species relationships as the Prunus umbellata P. angustifolia, P. mexicana, P. hortulana and P. munsoniana amplicons were not able to be sequenced. FRIGIDA MP and ML analyses of diploid and phased haplotype sequence data did recover section Prunocerasus as a monophyletic clade (Figure A 56). FT TSF analyses supported section Prunocerasus as a monophyletic clade. MP analyses clustered P. angustifolia, P. muns oniana and P. geniculata in one clade. Similarly, P maritima and P. umbellata were identified as sister species in other clade. 91

PAGE 92

MP analyses using gaps ( as additional characters ) improved support for some clades The American clade was recovered. Analyses with phased haplotype sequence data yielded similar results No major clades representing the American and the Chickasaw clades within section Prunocerasus were found (Figure A 57). GIFB analyses supported section Prunocerasus monophyly. Two major groups were obtained in MP (with gaps) and ML analyses using diploid and haplotype sequence data: 1) T he American clade ( P. americana, P. mexicana and P. hortulana), and 2) The Chickasaw and the Beach clade species. Haplotype analyses did not show any additional groupings (Figure A 58). LFY identified section Prunocerasus as a monophyletic clade. Two major groups were recovered within section Prunocerasus for all analyses: 1) The American clade species and P. geniculata, and 2) The Chickasaw clade species and P. maritima (Figure A 60). This region had a high polymorphism rate and recovered several species relationships within section Prunocerasus as described above. LFY is a good candidate for a whole phylogenetic analysis within section Prunocerasus Similar res ults were observed for MAF2A (Figure A 61). MP analyses including gaps increased support for each clade. D iploid and phased haplotype results were similar PHYE analyses produced similar results that LFY and MAF2A phylogenetic results. One group represente d the Chickasaw and the Beach clade species. The other group represented the American clade species. No additional support was obtained by using gaps in MP. Haplotype sequence data produced similar results to those produced by diploid data (Figure A 69). 92

PAGE 93

S PY MP analyses with diploid and phased haplotype sequence data recovered P. americana and P. mexicana clustered together ( American clade). Similarly, P. angustifolia and P. munsoniana were found as sister species (Chickasaw clade). ML analyses did not support any species relationships (Figure A 72). Similarly, P. angustifolia and P. munsoniana were supported as sister species using TFL2 diploid and phased haplotype sequence data (Figure A 74). VRN1 supported section Prunocerasus monophyly. Two major groups were recovered within section Prunocerasus for all analyses: 1) The American clade species, P. umbellata, and P. maritima and 2) The Chickasaw clade species. Prunus maritima was observed in a polytomy with the Chickasaw clade (Figure A 75). Results from d iploid and haplotype sequence data were similar. No additional species clusters were identified using haplotype sequence data. VRN1 had a small polymorphism rate of 1.35% within section Prunocerasus However, VRN1 is a good candidate for a whole phylogenet ic analysis within section Prunocerasus because its few amount of potentially informative characters represented a good phylogenetic signal. MP and ML analyses of combined dormancy related genes diploid data recovered section Prunocerasus as a monophyleti c group. Prunus angustifolia and P. munsoniana were clustered together in all analyses (Chickasaw clade). Similar relationships were identified for P. americana and P. mexicana, and P. maritima and P. umbellata. MP analyses (with gaps) yielded three major groups: 1) The Chickasaw clade ( P. angustifolia and P. munsoniana) and P. geniculata ; 2) Prunus umbellata and P. maritima ; and 3 ) The American clade ( P. mexicana, P. americana, and P. hortulana) and P. geniculata. ML analyses of combined data recovered sim ilar species groups to 93

PAGE 94

those produced by MP analysis with gaps. Prunus pumila was identified as sister species to section Prunocerasus (Figure A 81 ). The non standarized multi locus (phased haplotype) combined branching genes sequence data network (Figure A 82) recovered similar species relationships to those produced by MP (with gaps) and ML analyses. Prunus maritima was clustered to P. umbellata (similar to isozymes and branching genes combined data analyses). Additional sequence data information and models of sequence evolution are available on Tables A 11 to A 1 7 Total E vidence A total of 27909 bp of alignable sequence (1945 PICs polymorphism rate of 6.97% ) was obtained for cpDNA, ITS, and nuclear genes (Table 34). G ene regions were summarized based on their polymorphism rate ( % variability ) within section Prunocerasus and the outgroup species. Sequence information represented both intron(s) and exon(s). As expected, i ntron(s) had the highest polymorphism rate across the core collection of 11 Prunus species (Figure 33 ; Table A 1 1 ). MP and ML analyses of the combined diploid sequence data for the branching genes recovered section Prunocerasus as a monophyletic group. Prunus angustifolia and P. munsoniana were clustered together in all analyses (Chickas aw clade). Similar ly, P. americana was paired with P. mexicana, and P. maritima with P. umbellata. MP analyses (with gaps) yielded two major groups: 1) The American clade ( P. mexicana P. americana, and P. hortulana) and P. geniculata; 2) The Chickasaw clade ( P. angustifolia, P. munsoniana and P. umbellata) and P. maritima ML analyses of combined data recov ered similar species groups to those recovered using MP analysis 94

PAGE 95

with gaps. Prunus pumila was identified as sister species to section Prunocerasus (Fig ure 38). The non standarized multi locus (phased haplotype) sequence data network for the combined branching genes sequence data (Figure 39) recovered similar species relationships to those produced by MP (with gaps) and ML analyses. Prunus maritima was clustered to P. umbellata (similar to isozymes, branchi ng, and dormancy related genes combined data analyses). Additional sequence data information and models of sequence evolution are available on Tables A 11 to A 1 7 Discussion Chloroplast DNA has been extensively used for phylogenetic studies in Prunus (Badenes and Parfitt, 1995; Bortiri et al., 2001; Bortiri et al., 2002; Bortiri et al., 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Shaw and Small, 2004, 2005; Wen et al., 2008). However, evidence for hybridization among Prunus species may not be detected if the analysis is based solely on cpDNA information Human disturbance, relocation of species, and other factors, could increase contact between previously isolated species and allow their hybridization. The existence of hybrids of native plum species with other Prunus species has been reported by Wight (1915). Wight (1915) reported that P. americana hybridized with P. angustifolia, P. besseyi (i.e. P. pumila ), P. hortulana, P. munsoniana, and P. simonnii Similarly, P. angustifolia hybridized freely with P. americana, P. triflora P. besseyi P. munsoniana, P. cerasifera and P. orthosepala. Shaw et al. ( 2005, 2007) identified chloroplast regions that provided the greatest number of characters for low level molecular phylogenetic studies. From those, 7 regions (out of 8) were sequenced for our collection. Our polymorphism rate estimates 95

PAGE 96

were different to those reported by Shaw et al. (2005, 2007) results. They used P. hortulana, P. nigra, and P. virginiana to determine the number of PICs Prunus hortulana and P. nigra belong to subgenus Prunus section Prunocerasus and P. virginiana to subgenus Cerasus The r egions with the highest PIC s prove d to be beneficial for the recovery of the phylogenetic relationships within section Prunocerasus (based on Shaw and Small, 2004). Our analyses showed that the highest polymorphism rate (indels and substitutions) were mostly found at the outgroup level (Table 34; Figure 33) The limited number of species with in section Prunocerasus and the broad comparison of levels across the genus Prunus was probably the reason for the higher polymorphism rates reported by Shaw et al. (2005, 2007) The highest amounts of informative sites in our study were observed in 3'trnV ndhC and trnH psbA cpDNA regions (Figure 3 3) Similar results were reported by Shaw et al. (2005, 2007). It is important to recognize that although PIC s are a good measure of how informative a region could be for a phylogenetic analysis it did not repre sent the final phylogenetic relationships obtained from that region. Both methods were combined in order to select regions that could be used across species levels to identify phylogenetic patterns. The use of these novel regions constitut ed of importance for our analyse s. The presence of common alleles between the different species could be evidence for interspecific hybridization and gene flow between species. The use of cytoplasmic and nuclear markers would allow for the tracking of maternal and paternal lineages present in a more diverse phylogenetic study 96

PAGE 97

Rohrer et al. (2004) analyzed the values of SSRs to resolve the relationships within Prunus at the species level. He concluded that the resolution of species relationships within section Prunocerasus require d the use of regions more variable than ITS and less variable than SSRs. In our study the outgroup species clustered together using 41 and 6 (highest PIC values) SSRs (Figure 3 2) Cluster analyses using information from the 41 SSRs supported P an gustifolia and P munsoniana as sister species to P. mexicana and P. geniculata. Similarly, P. americana, P. maritima and P. umbellata clustered together. No previously known relationships (such as the American clade, the Beach clade, and the Chickasaw clade) w ere obtained. Analysis of the species with the 6 SSR primers with the highest PIC values identified Prunus mexicana as sister species to all the outgroup species. The Beach clade ( P. geniculata and P. maritima ) and the Chickasaw clade ( P. angustifolia and P. munsoniana) were recovered. Prunus americana and P. umbellata were monophyletic, and P. hortulana was found basal to all the section Prunocerasus clade (Figure 3 2) The SSR data reported a high level of heterezygosity as described by Rohrer et al (2004). The use of SSRs d istributed across the genome was expected to provide a sample of different levels of polymorphisms However, i t was observed that the use of 6 SSR markers (with the highest PIC values) yielded more consistent results than the analyses with 41 SSR markers in comparison to known phylogenetic relationships in section Prunocerasus (Shaw and Small, 2004). The observed level of heterozygosity for all genotypes was high compared to UFs peach germplasm, 0.58 and 0.41, respectively (Tabl e 32) Wild Prunus species 97

PAGE 98

tipically express gametic self incompatibility making them highly heterozygous. Our results were concordant with these expectations. ITS results were consistent with Rohrers et al. (2004) remarks He was not able to define the species relationships within section Prunocerasus using this region. In our study, ITS sequence data was an important baseline for comparison with other regions (nuclear and cpDNA) A polymorphism rate of 10.82% was obtained for ITS (Table 34) This rate was intermediate between nuclear (high) and cpDNA (low) regions. The results obtained from the ITS phylogenetic results were consistent with our cpDNA results (Figure A 21). Three different groups of nuclear genes were used in this study: 1) Isozymes, 2) B ranching genes, and 3) Dormancy related genes. The isozyme genes were used due to their expected neutrality to selection and adaptation. The highest amount of PICs within section Prunocerasus and between outgroup were obtained for PGI, followed by s6pdh a nd PGDH (Table 3 4; Figure 33) Section Prunocerasus was recovered as a monophyletic clade for all analyses (F igure A 22 to A 24 ; Tables A 1 2 and A 1 3 ). These regions were more variable than cpDNA and less variable than ITS within section Prunocerasus (Fi gure 33). Similar results were also obtained for branching and dormancy related genes (Figures A 25 to A 76 ; Tables A 1 2 and A 1 3 ). The branching and dormancy related genes were used due to their expected selection and adaptation pressure. The highest amo unt of PICs within section Prunocerasus and between outgroup for branching genes were obtained for MAX4, MAX1, CUC2, MAX3, and AXR1 (Table 34). Section Prunocerasus was recovered as a monophyletic clade with MP and ML analyses using diploid and haplotype datasets 98

PAGE 99

(Figure A 25 to A 42; Tables A 1 2 and A 1 3 ). MP analyses with gaps did provided additional support (boostrap) for some clades when using different genes Similarly, t he highest amount of PICs within section Prunocerasus and between outgroup for dormancy related genes were observed in MAF4, FLC FLF, LFY, MAF2A, AGL20 SOC1, ELF6 and AGL24 (Table 34). Section Prunocerasus was monophyletic in MP and ML analyses using diploid and haplotype datasets for all gene regions except FPF1, PHYA, and RGARGA1 ( Figure A 43 to A 76 ; Tables A 1 2 and A 1 3 ). MP analyses with gaps recovered additional species grouping within section Prunocerasus MP and ML phased haplotype and the nonstandarized multi locus analyses yielded similar results as using diploid sequence d ata. This consistency across analyses and different gene regions supported the presence of a phylogenetic signal I t was observed that regions with the highest polymorphism rates within section Prunocerasus allowed identification of additional species relationships (clades) as compared to known clades in section Prunocerasus (Shaw and Small, 2004) These regions constituted good candidates for future phylogenetic relationships in Prunus MP and ML analyses of isozymes, branching, and dormancy related genes found two major groups similar to those described by Waughs (1901) classification of section Prunocerasus The American and the Chickasaw clades were identified in our analyses Our results support s ection Prunocerasus as a monophyletic group using indi vidual and combined analyses as previously reported by Bortiri et al. (2001), and Shaw and Small (2004). However, there was no evidence for P maritima and P. geniculata being sister species ( Beach clade) as reported by Shaw and Small (2004) In the majori ty of the analyses, P. maritima was supported as a member of the Chickasaw clade ( in specific 99

PAGE 100

sister species to P. umbellata) Similarly, P. geniculata was identified as a member species of the American clade (Figures A 77 to A 82 Figure s 3 4 and 35). O ur analyses provided similar results to those reported by Mowrey and Werner (1990) They identified a clade formed by P. maritima and P. umbellata within section Prunocerasus based on isozyme data In other study, Rohrer et al. (2008) used LEAFY and s6pdh sequences to examine the North American plums phy logeny. They reported that P. geniculata was clustered with P. texana In addition, P. umbellata, P. maritima P. alleghaniensis and P. gracilis formed a paraphyletic group using LEAFY The use of different sequence regions, in a core collection of section Prunocerasus and outgroup species, constituted an important study to identify genome regions with the highest amount of potentially informative characters ( PICs ) and the most valuable phylogenetic signal b ased on previously known clades within section Prunocerasus (Shaw and Small, 2004) Out of the 52 nuclear genes (3 isozymes, 16 branching and 33 dormancy related genes), 7 cpDNA regions, and ITS sequences, trnH psbA PGI, MAX4, AXR1, LFY, PHYE, and VRN1, were chosen to be used for a phylogenetic analysis with a larger number of taxa. Regions with high polymorphism rates and strong phylogenetic signal as compared with known phylogenetic relationships within section Prunocerasus were trnH psbA PGI, MAX4, AX R1, and LFY. PHYE and VRN1 were regions with low polymorphism rates However, they presented a strong phylogenetic signal (polymorphism within these regions were highly informative). Conclusions The identification and use of additional genomic regions that provided the greatest number of characters, variability, and improved the phylogenetic signal at the low level in Prunus section Prunocerasus relationships was achieved. The amount of 100

PAGE 101

PICs was highly variable across regions, with nuclear genes being the h ighest, followed by ITS, and cpDNA regions. The species relationships within section Prunocerasus and o utgroup species were recovered for the majority of regions in comparison with known clades as reported by Shaw and Small (2004) It is important to identify PIC s as a good measure of how informative a region will be for phylogenetic analysis, but always complemented with the phylogenetic analysis by itself A preliminary study should always include several species that fall in different levels of t he genus of study. Both methods should be combined in order to select regions that will provide the most value. These genomic region sequences could be used for linkage mapping in segregating populations from different species to identify QTLs associated with dor mancy response, and plant architecture and branching genes. These sequences could be also used to create hybridization probes for microarray chips to study the presence or absence of certain polymorphic sites associated with different candidate genes. 101

PAGE 102

Ta ble 31. List of the North American Plums (sect ion Prunocerasus ) and other Prunus species col lected and used in this project Voucher (No.) Collection (No.) Genus Species ID z Common name Country State Latitud e Longitud e Date of collection 12 Prunus ameri cana PameFL12 American plum USA FL 298.965'N 824.525W 4/13/2010 15 Prunus pumila P bes IL15 Sandcherry USA IL 410.972'N 959.167'W 4/14/2010 23 Prunus umbellata PumbFl23 Hog plum USA FL 301.178N 83 09.957W 4/28/2010 25 Prunus angustifolia P angGA25 Chickasaw plum USA GA 303.451'N 845.695'W 4/28/2010 85 Prunus mexicana PmexLA85 Mexican plum USA LA 6/4/2010 87 Prunus maritima PmarNY87 Beach plum USA NY 401.508N 732.862W 6/4/2010 88 Prunus munsoniana PmunTX88 Wild Goose Plum USA TX 6/4/2010 89 Prunus hortulana PhorMO89 Hortulan plum USA MO 6/4/2010 115 Prunus geniculata PgenFl115 Scrub plum USA FL 276.242N 814.564W 7/8/2010 121 Prunus persica cv. Okinawa PperJP121 Peach Japan 7/8/2010 131 Prunus fasciculata Pf asCA131 Desert almond USA CA 33.604N 11629.07W 7/30/2010 zID = fi rst letter represented the genus ( Prunus =P) next three letters represented the species ( americana=ame, etc.), and the following letters represented the state of origin and collection number (Flo rida collection 12=FL12, etc.). 102

PAGE 103

Table 32. Summary statistics of 41 simple sequence repeat (SSR) markers of a subset of 11 Prunus species. Marker Genotype No. A z Ae Ho He F PIC BPPCT006A 10 7 5.56 0.60 0.82 0.27 0.80 BPPCT006B 8 8 6.74 0.50 0 .85 0.41 0.83 BPPCT008A 8 8 5.41 0.50 0.82 0.39 0.79 BPPCT008B 8 12 9.52 0.60 0.90 0.33 0.89 BPPCT013 10 9 6.54 0.73 0.85 0.14 0.83 BPPCT014 11 14 11.52 0.55 0.91 0.40 0.91 BPPCT016 7 7 10.08 0.18 0.76 0.50 0.73 BPPCT017 10 13 2.85 0.45 0.90 0.02 0.8 9 BPPCT023 5 4 13.44 0.64 0.65 0.02 0.58 BPPCT025 11 17 13.44 0.91 0.93 0.21 0.92 BPPCT026 11 16 5.50 0.73 0.93 0.67 0.92 BPPCT027 8 9 9.68 0.27 0.82 0.39 0.80 BPPCT028 11 12 7.33 0.55 0.90 0.16 0.89 BPPCT029 10 10 4.57 0.73 0.86 0.05 0.85 BPPCT030 8 9 7.41 0.82 0.78 0.42 0.76 BPPCT038 8 12 12.74 0.50 0.87 0.21 0.85 BPPCT039 11 15 13.44 0.73 0.92 0.02 0.92 CPDCT023 7 7 8.64 0.20 0.74 0.07 0.71 CPDCT025 11 15 7.33 0.91 0.93 0.37 0.92 CPDCT027 11 13 10.52 0.82 0.88 0.10 0.87 CPDCT038 10 12 6.21 0.55 0.86 0.67 0.85 CPPCT005 10 14 2.67 0.82 0.90 0.20 0.90 CPPCT006 10 13 10.08 1.00 0.91 0.39 0.90 CPPCT008 8 9 11.00 0.27 0.84 0.30 0.82 CPPCT015 2 3 9.68 0.50 0.63 0.39 0.55 CPPCT022 11 13 2.35 0.55 0.90 0.05 0.89 CPPCT026 11 14 8.07 0.64 0.91 0. 17 0.90 CPPCT029 6 6 12.74 0.27 0.68 0.11 0.64 CPPCT033 11 13 3.06 0.55 0.90 0.87 0.89 CPSCT004 7 8 8.00 0.55 0.57 0.03 0.56 CPSCT008 3 5 6.21 0.67 0.78 0.24 0.74 CPSCT034 10 12 12.10 0.55 0.89 0.31 0.88 EPDC3832A 10 12 7.81 0.73 0.88 0.27 0.86 EPD C3832B 11 16 4.48 0.82 0.92 0.77 0.92 EPDCU3117 5 5 8.64 0.09 0.67 0.59 0.63 EPDCU3392 11 15 3.92 0.82 0.92 0.60 0.92 EPPISF002 9 10 10.52 0.90 0.88 0.20 0.86 EPPISF032 10 10 5.56 0.64 0.84 0.27 0.82 PMS2 11 15 6.74 0.64 0.92 0.41 0.91 UDP96 001 11 1 0 5.41 0.64 0.87 0.39 0.86 UDP96 003 8 8 9.52 0.18 0.78 0.33 0.76 UDP96 019 9 12 6.54 0.36 0.88 0.14 0.87 UDP98 025 7 5 11.52 0.30 0.75 0.40 0.70 UDP98 412 11 14 10.08 0.73 0.90 0.50 0.90 Average 9 10.70 8.10 0.58 0.84 0.31 0.82 zA = number of observ ed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = polymorphism information content. yNumbers in bold represent highest and lowest values for each variable. 103

PAGE 104

T able 33. Summary statistics of cpDNA regions sequenced for 11 Prunus species. cpDNA region 3'trnV ndhC trnL trnF z trnL intron trnQ 5'rps16 trnH psbA ndhF rpl32 atpB rbcL Aligned y L. (bp) 567 472 364 479 335 598 769 Indels Subst I ndels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst American x 17 19 0 0 0 2 4 3 8 2 1 0 0 0 Beach 13 20 0 0 0 2 3 4 8 1 1 0 0 1 Chickasaw 0 0 0 0 0 0 1 0 5 0 0 0 0 0 Prunocerasus w 25 16 0 0 0 3 5 4 9 2 2 0 0 1 PICs v 41 0 3 9 1 1 2 1 % variability u 7.23 0.00 0.82 1.88 3.28 0.33 0.13 Outgroup t 9 9 0 3 1 5 5 5 2 3 3 6 3 9 PICs 18 3 6 10 5 9 12 % variability 3.17 0.64 1.65 2.09 1.49 1.51 1.56 Total PICs 59 3 9 19 16 11 13 Normalized PICs 3.03 0.15 0.46 0.98 0.82 0.57 0.67 % v ariability 10.41 0.64 2.47 3.97 4.78 1.84 1.69 ztrnL trnF and trnL introns were sequenced together as they primers amplify both regions. yAligned L= total aligned sequence length. Indels = number of insertion / deletion events. Subst = substitutions. xAmer ican clade = P. americana, P. hortulana and P. mexicana. Beach clade = P. geniculata and P. maritima Chickasaw clade = P. angustifolia, P. munsoniana, and P. umbellata. wPrunocerasus = American + Beach + Chickasaw clades. vPICs = Indels + Subst u% varia bility = [(Subs t +Indels)/L] 100. tOutgroup = P. fasciculata P. persica and P. pumila 104

PAGE 105

Table 34. Summary statistics of ITS and candidate gene regions sequenced for 11 Prunus species. Regions ITS PGI PGDH s6pdh AXR1 BRC1 BRC2 Aligned z L. (bp) 610 299 477 629 263 299 298 Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst American y 6 3 4 4 1 6 9 12 0 0 0 3 0 4 Beach 5 1 6 6 1 14 3 5 0 5 0 2 0 0 Chickasaw 1 4 2 5 2 10 3 4 0 6 0 1 0 0 Prunocerasus x 7 7 10 13 2 15 12 16 0 6 0 4 0 6 PICs w 14 23 17 28 6 4 6 % variability v 2.30 7.69 3.56 4.45 2.28 1.34 2.01 Outgroup u 11 41 3 9 0 14 6 13 0 5 0 2 0 8 PICs 52 12 14 19 5 2 8 % variability 8.52 4.01 2.94 3.02 1.90 0.67 2.68 Total PICs 66 35 31 47 11 6 14 Normalize d PICs 3.39 1.80 1.59 2.62 0.57 0.31 0.72 % variability 10.82 11.71 6.50 7.47 4.18 2.01 4.70 Regions CUC1A t CUC1B CUC2 CUC3 LAS MAX1A MAX1B Aligned L. (bp) 430 305 577 506 378 327 323 American 0 2 0 1 1 6 0 5 0 5 3 2 0 2 Beach 1 4 1 3 0 8 2 4 0 4 2 3 1 4 Chickasaw 0 2 0 1 1 9 1 4 0 2 0 4 0 8 Prunocerasus 1 4 1 3 1 17 2 10 0 9 4 7 1 7 PICs 5 4 18 12 9 11 8 % variability 1.16 1.31 3.12 2.37 2.38 3.36 2.48 Outgroup 1 12 1 12 7 26 15 28 1 13 1 8 0 8 PICs 13 13 33 43 14 9 8 % variability 3.02 4.26 5. 72 8.50 3.70 2.75 2.48 Total PICs 18 17 51 55 23 20 16 Normalized PICs 0.93 0.87 2.62 2.83 1.18 1.03 0.82 % variability 4.19 5.57 8.84 10.87 6.08 6.12 4.95 105

PAGE 106

Table 3 4. Continued. Regions MAX2 MAX3 MAX4 PIN RAX1 RAX2 RAX3 REV Aligned L. (bp) 376 311 23 9 343 445 503 378 Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst American 0 2 0 4 1 6 1 1 0 1 0 1 1 5 Beach 0 4 1 6 1 9 0 0 2 2 2 2 0 2 Chickasaw 0 0 0 2 0 6 2 3 1 2 1 1 0 3 Prunocerasus 0 5 1 8 1 11 3 3 2 5 2 2 1 6 PICs 5 9 12 6 7 4 7 % variability 1.33 2.89 5.02 1.75 1.57 0.80 1.85 Outgroup 1 10 4 20 0 15 2 6 2 15 5 16 5 19 PICs 11 24 15 8 17 21 24 % variability 2.93 7.72 6.28 2.33 3.82 4.17 6.35 Total PICs 16 33 27 14 24 25 31 Normalized PICs 0.82 1.70 1.39 0.72 1.23 1.29 1.59 % variability 4.26 10.61 11.30 4.08 5.39 4.97 8.20 Regions SPS AGL24 AGL20SOC1 BFT BRM CO CRY1 Aligned L. (bp) 363 494 444 458 366 419 500 American 3 2 1 8 6 3 0 8 2 5 0 1 0 2 Beach 0 5 2 6 5 5 0 1 1 3 1 5 1 6 Chickasaw 3 3 0 9 4 7 0 3 1 3 0 3 1 5 Prunocerasus 3 7 4 16 12 11 0 10 3 9 1 8 2 9 PICs 10 20 23 10 12 9 11 % variability 2.75 4.05 5.18 2.18 3.28 2.15 2.20 Outgroup 3 12 10 28 8 15 9 10 0 11 5 20 2 18 PICs 15 38 23 19 11 25 20 % variability 4.13 7.69 5.18 4. 15 3.01 5.97 4.00 Total PICs 25 58 46 29 23 34 31 Normalized PICs 1.29 2.98 2.37 1.49 1.18 1.75 1.59 % variability 6.89 11.74 10.36 6.33 6.28 8.11 6.20 106

PAGE 107

Table 3 4. Continued. Regions CRY2 ELF6 FD FD1 FG FLC FLF FPF1 Aligned L. (bp) 450 593 536 481 394 449 329 Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst American 2 3 6 8 0 3 1 2 3 6 8 0 3 1 Beach 0 4 5 8 1 4 0 0 4 4 8 1 4 0 Chickasaw 2 6 5 5 3 5 0 2 6 5 6 3 5 0 Prunocerasus 3 8 9 15 4 7 1 3 8 12 13 4 7 1 PICs 11 24 11 5 13 32 1 % variability 2.44 4.05 2.05 1.04 3.30 7.13 0.30 Outgroup 2 26 15 27 7 22 1 2 26 13 26 7 22 1 PICs 28 42 29 13 18 4 8 % variability 6.22 7.08 5.41 2.70 4.57 0.89 2.43 Total PICs 39 66 40 18 31 36 9 Normalized PICs 2.01 3.3 9 2.06 0.93 1.59 1.85 0.46 % variability 8.67 11.13 7.46 3.74 7.87 8.02 2.74 Regions FRIGIDA FTTSF GI FB HOS1 LFY MAF1 MAF3 AGL31 MAF2A Aligned L. (bp) 630 507 577 198 615 496 268 American 1 0 5 1 5 11 0 4 6 19 0 10 5 3 Beach 3 9 6 5 3 2 0 1 6 22 0 3 1 2 Chickasaw 6 4 6 0 0 4 3 10 5 8 3 2 Prunocerasus 4 9 13 6 12 10 0 5 12 29 4 14 8 7 PICs 13 19 22 5 41 18 15 % variability 2.06 3.75 3.81 2.53 6.67 3.63 5.60 Outgroup 4 23 8 10 7 21 0 2 14 49 5 14 4 4 PICs 27 18 28 2 63 19 8 % variability 4.2 9 3.55 4.85 1.01 10.24 3.83 2.99 Total PICs 40 37 50 7 104 37 23 Normalized PICs 2.06 1.90 2.57 0.36 5.35 1.90 1.18 % variability 6.35 7.30 8.67 3.54 16.91 7.46 8.58 107

PAGE 108

Table 3 4. Continued. Regions MAF2B MAF4 MAF5 MFT PHYA PHYB PHYD PHYE Aligned L. (bp) 335 474 310 395 493 500 499 Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst Indels Subst American 1 1 6 5 0 2 0 6 0 0 0 2 1 6 Beach 2 3 5 4 1 0 0 5 0 2 1 2 0 0 Chickasaw 3 0 20 23 0 0 0 2 0 4 1 6 2 5 Prunocerasus 4 3 24 20 1 1 0 12 0 5 2 7 3 12 PICs 7 44 2 12 5 9 15 % variability 2.09 9.28 0.65 3.04 1.01 1.80 3.01 Outgroup 5 4 14 33 9 24 3 16 0 7 2 16 2 15 PICs 9 47 33 19 7 18 17 % variability 2.69 9.92 10.65 4.81 1.42 3.60 3.41 Total PICs 16 91 35 31 12 27 32 N ormalized PICs 0.82 4.68 1.80 1.59 0.62 1.39 1.65 % variability 4.78 19.20 11.29 7.85 2.43 5.40 6.41 Regions RGA RGA1 RGL1 RGL2 RGL3 SPY TFL1 ATC TFL2 VRN1 VRN2 Aligned L. (bp) 454 510 506 457 499 592 418 American 0 5 0 4 2 2 1 2 1 3 0 0 1 1 Beach 0 4 0 0 0 2 1 3 1 3 0 7 0 4 Chickasaw 0 8 0 0 2 5 0 5 1 6 0 4 1 4 Prunocerasus 0 10 0 4 2 8 2 6 0 10 0 8 2 6 PICs 10 4 10 8 10 8 8 % variability 2.20 0.78 1.98 1.75 2.00 1.35 1.91 Outgroup 0 6 1 30 3 15 9 9 7 15 0 23 9 19 PICs 6 31 18 18 22 23 28 % var iability 1.32 6.08 3.56 3.94 4.41 3.89 6.70 Total PICs 16 35 28 26 32 31 36 Normalized PICs 0.82 1.80 1.44 1.34 1.65 1.59 1.85 % variability 3.52 6.86 5.53 5.69 6.41 5.24 8.61 108

PAGE 109

zAligned L= total aligned sequence length. Indels = number of insertion / deletion events. Subst =substitutions. yAmerican clade = P. americana, P. hortulana and P. mexicana. Beach clade = P. geniculata and P. maritima Chickasaw clade = P. angustifolia, P. munsoniana, and P. umbellata. xPrunocerasus = American + Beach + Chickasaw clades. wPICs = Indels + Subst v% variability = [(Subst+Indels)/L] 100. u Outgroup = P. fasciculata P. persica and P. pumila tGene symbols followed by letter A and B represent duplicate fragments amplified and independently sequenced using the same primer set. Genes with duplicate fragments were CUC1, MAX1, and MAF2. Gene symbols are described on Table A 9 109

PAGE 110

Figure 31. Representatives of the North American plums ( Prunus spp.), subgenus Prunus section Prunocerasus and outgroup species: A) P. americana Marsh., B) P. umbellata Elliot, C) P. fasciculata (Torr.) A. Gray, D) P. geniculata Harper, E) P. angustifolia Marsh., F) P. mexicana S. Watson, G) P. hortulana L.H. Bailey, H) P. maritima Marsh., I) P. munsoniana W. Wight & Hedrick, and J) P. pumila L. 110

PAGE 111

A B C D E F G H I J 111

PAGE 112

Figure 32 Neighbor Joining (NJ) cluster analysis based on Neis genetic distance (Nei et al., 1983) for 11 Prunus species using A) 41 simple sequence repeat (SSR) markers and B) 6 SSR markers wi th the highest PIC values Cladogram rooted with P. fasciculata (PfasCA131) at the base. A B 112

PAGE 113

0.00 2.00 4.00 6.00 8.00 10.00 12.00 3'trnV ndhC trnL trnF trnL intron trnQ 5'rps16 trnH psbA ndhF rpl32 atpB rbcL ITS PGI PGDH s6pdh AXR1 BRC1 BRC2 CUC1A CUC1B CUC2 CUC3 LAS MAX1A MAX1B MAX2 MAX3 MAX4 PIN RAX1 RAX2 RAX3 REV SPS AGL24 AGL20 SOC1 BFT BRM CO CRY1 CRY2 ELF6 FD FD1 FG FLC FLF FPF1 FRIGIDA FT TSF GI FB HOS1 LFY MAF1 MAF3 AGL31 MAF2A MAF2B MAF4 MAF5 MFT PHYA PHYB PHYD PHYE RGA RGA1 RGL1 RGL2 RGL3 SPY TFL1 ATC TFL2 VRN1 VRN2 Percent variability (%) Prunocerasus Outgroup 0.00 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 AGL20 SOC1 AGL24 AXR1 BRM BRC1 BRC2 BFT CO CRY1 CRY2 CUC1A CUC1B CUC2 CUC3 ELF6 FD FD1 FG FLC FLF FT TSF FPF1 FRIGIDA GI FB HOS1 LAS LFY MAF1 MAF3 AGL31 MAF2A MAF2B MAF4 MAF5 MAX1A MAX1B MAX2 MAX3 MAX4 MFT PGI PGDH PHYA PHYB PHYD PHYE PIN VRN1 VRN2 RAX1 RAX2 RAX3 RGA RGA1 REV RGL1 RGL2 RGL3 s6pdh SPS SPY TFL1 ATC TFL2 Exons Introns Figure 33. Sequence percent variability, %= [(Subst+Indels)/L] 100, comparisons for A) Section Prunocerasus vs. outgroup species and B) Exons vs. introns gene regions. 113

PAGE 114

Figure 34 Phylogenetic analyses of c ombined cpDNA sequence data (3584 bp). Majority rule consensus trees using A) maximum parsimony ( 75 trees, 126 steps, CI=0.929, RI=0.890, RC=0.827) and B) maximum parsimony (includi ng gaps) (13 trees, 623 steps, CI=0.918, RI=0.809, RC=0.743) analyses. Bootstrap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL=5414.74). Branch lengths are described above the branches and posterior probabilities values below the branches Trees are rooted with P. fasciculata A B C 114

PAGE 115

Figure 35 A nonstandarized multi locus combined cpDNA sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B Outgroups Section Prunocerasus Outgroups Section Prunocerasus 115

PAGE 116

Figure 36 Phylogenetic analyses of combined nuclear genes diploid sequence data ( 23429 bp). Majority rule consensus trees using A) maximum parsimony ( 2 trees, 1400 steps, CI=0.913, RI=0.920, RC=0.840) and B) maximum parsimony (including gaps) (1 tree, 2535 steps, CI=0.879, RI=0.881, RC=0.775) analyses. Bootstrap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL= 41509.34). Branch lengths described above the branches and posterior probabilities values below the branches. Trees are rooted with P. fasciculata A B C 116

PAGE 117

Figure 37 A nonstandarized multi locus combined nuclear genes sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B Outgroups Sectio n Prunocerasus Outgroups Section Prunocerasus 117

PAGE 118

Figure 38 Phylogenetic analyses of combined nuclear + cpDNA + ITS diploid sequence data ( 27623 bp). Majority rule consensus trees using A) maximum parsimony ( 1 tree, 1 594 steps, CI=0.9 08 RI=0.91 0, RC=0.8 27) and B) maximum parsimony ( including gaps) (2 tree s 2732 steps, CI=0. 878, RI=0.8 75 RC=0.7 68 ) analyses. Bootstrap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL= 48496.34). Branch lengths described above the branches and posterior probabil ities values below the branches. Trees are rooted with P. fasciculata A B C 118

PAGE 119

Figure 39 A nonstandarized multi locus combined nuclear + cpDNA + ITS sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B Outgroups Section Prunocerasus Outgroups Section Prunocerasus Chickasaw clade American clade Beach clade Chickasaw clade American clade Beach clade 119

PAGE 120

CHAPTER 4 THE PHYLOGENY OF THE NORTH AMERICAN PLUMS ( P runus spp L. ) Introduction The genus Prunus L are trees or shrubs with white or pink flowers with spreading petals, mostly with edible fruit. Stamens 1530, distinct, with filiform filaments. Style, terminal; stigma, usually truncate. The fruit has a fleshy exterior, glaborous, and with a n interior hard bony pit which contains the seed (Waugh, 1901). The genus Prunus is part of the subfamily Amygd aloideae of the Rosaceae family, with a pproximat ely 200 species distributed worldwide (Bortiri et al., 2001; Hedrick, 1911; Rehder, 1940; Sargent, 1905; Wight, 1915). This genus contains commonly known fruit such as plums cherries, almonds, apricots, and peaches. Stone fruit world production in 2010 was approximately 40.8 million tons. P lums and sloes production in the US was approximately 0.5 million tons, with a farm gate value of ~294 million dollars (FAOSTAT, 2010). North American plums have the highest diversity of flavor, aroma, texture, color, fo rm and size as reported by Hedrick (1911) and Waugh (1901). Plum species are native to widely divergent climates and soils. T he study of the evolutionary history of these species could help us to identify important genetic source s for adaptation. Inter sp ecific hybridization has been widely used by breeders to introgress unique traits not available in commercial germplasm. However, section Prunocerasus genetic variability has not been widely used for the improvement of Prunus scion and /or rootstock material. In addition, t axonomic conflicts have been previously reported in Prunus (Hedrick, 1911; Waugh, 1901) Bortiri et al. (2001) summarized these conflicts and problems ( previously described in Chapter 3) with Rehders (1940) classification 120

PAGE 121

being recognized as the most widely used and accepted. Rehder (1940) reported five subgenera: Prunophora ( Prunus ), Amygdalus Cerasus Padus and Laurocerasus In addition, t he subgenus Prunus was further divided into sections: Euprunus Prunocerasus (North American plum s), and Armeniaca. G enus Prunus has been identified as a monophyletic group in recent phylogenetic studies. However, i nternal relationships, subclades at the subgenera and species level, have been poorly supported ( Bortiri et al., 2001, 2002, 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Lee and Wen, 2001; Shaw and Small, 2004, 2005; Wen et al., 2008) Among these studies, some phylogenetic analyses have tried to clarify section Prunocerasus species relationships. Mowrey and Werner (1990) studied the phylogenetic relationships in genus Prunus using isozymes. They recovered section Prunocerasus as a polyphyletic group, with : A) P. americana, P. munsoniana, P. hortulana P. subcordata, and P. angustifolia forming a clade, and B) P. maritima and P. um bellata in another Lee and Wen (2001) phylogenetic analysis identified two major groups using ITS : the Amygdalus Prunus group, and the Cerasus Laurocerasus Padus group. The Amygdalus Prunus group had one minor clade with P. americana, P. angustifolia, P. umbellata, P. nigra, and P. armeniaca. Bortiri et al. (2001) supported the genus Prunus as a monophyletic group using ITS and chloroplast trnL trnF spacer DNA sequences. Subgenera Padus Laurocerasus Cerasus and subgenera Prunus Amygdalus Emplectocladus Cer asus (sect. Microcerasus ) sect. Penarmeniaca formed two major clades, respectively. The plums of northeastern North America were found closely related ( P. americana, P. hortulana, P. 121

PAGE 122

munsoniana, P. maritima and P. mexicana). Similarly, Bortiri et al. (2002) used the nuclear gene s6pdh combined to ITS and trnLtrnF. Phylogenetic analysis of the combined data supported two major clades: subgenera Cerasus Laurocerasus Padus and subgenera Amygdalus Emplectocladus Prunus Subgenus Cerasus section Microcerasus ( currently subgenus Prunus sect. Microcerasus ) was found nested within subgenus Prunus Prunus maritima, P. mexicana, and P. subcordata were clustered together representing section Prunoc erasus Shaw and Small (2004) studied the phylogenetic relationships w ithin subgenus Prunus section Prunocerasus They supported section Prunocerasus monophyly using seven cpDNA regions. Three major clades were identified: the American Clade, the Chickasaw Clade, and the Beach Clade. The American clade included P. amer icana var. americana, P. americana var. lanata, P. mexicana, P. rivularis P. hortulana, P. umbellata var. injucunda; the Chickasaw clade included P. angustifolia, P. munsoniana, P. gracilis P. nigra P. umbellata var. umbellata, P. alleghaniensis var. al leghanienses and P. alleghaniensis var. davisii; and the Beach clade included P. geniculata, P. maritima var. maritima and P. maritima var. gravesii Most of the phylogenetic research in Prunus has been done using cpDNA sequences (Badenes and Parfitt, 1995; Bortiri et al., 2001; Bortiri et al., 2002; Bortiri et al., 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Shaw and Small, 2004, 2005; Wen et al., 2008; Table 12). However, hybridization among Prunus species could hinder the identification of phylogenetic relationships based solely on cpDNA information. Hybrids among native plum species with other Prunus species have been 122

PAGE 123

previously reported (Wight, 1915). The use of nuclear markers could provide enough information to recognize these putative hybrids. The use of nuclear markers was reported by Rohrer et al. (2004) T he genetic relationships within the North American plums (section Prunocerasus ) with 15 SSR primers yielded t wo major clades: A) P. americana P. mexicana hybrids, P. murrayana (cu rrently known as P. rivularis var. pubescens ), P. angustifolia P. mexicana hybrid, P. subcordata, P. rivularis P. americana, P. mexicana, P. hortulana, and P. munsoniana; and B) P. alleghaniensis P. gracilis P. umbellata, P. maritima P. americana, an d P. angustifolia Rohrer et al. (2004) concluded that the resolution of species relationships within section Prunocerasus required the use of regions more variable than ITS and less variable than SSRs. Similarly, t he use of SSRs primers found different re lationships among species than previous cpDNA analyses. S upport for additional phylogenetic relationships in subgenus Prunus section Prunocerasus could be provided by additional sequence information. Several amplicons representing enzyme, branching and dor mancy related genes were sequenced and studied for a core collection of section Prunocerasus species (Chapter 3). This preliminary study allowed us to identify trnH psbA PGI, MAX4, AXR1, and LFY as having the highest frequency of potentially informative c haracters and the phylogenetic signal [ability to recover previously known species relationships as reported by Shaw and Small (2004, 2005) and Bortiri et al. (2006)] across samples within subgenus Prunus section Prunocerasus The study of the subgenus Pru nus (in particular section Prunocerasus ) constitute d the main objective of this research. This project included collection and 123

PAGE 124

identification of the plant specimens from the wild to avoid any misidentified plant specimens/collections (vouchers and/or germplasm collections) that could affect our study to clarify section Prunocerasus phylogenetic framework In addition, open pollinated seed from several accessions were submitted to the USDA National Clonal Ger mplasm Repository in Davis, CA (DPRU) for support and preser vation of these species as important germplasm resourc es (gene pool) for unique traits. Materials and Methods Collection A total of 408 accessions, representing approximately 34 Prunus taxa were collected from the wild from April 2010 until Oct ober 2012 (Table A 1 8 ) Nine US states (AR, CA, FL, IL, MA, MI, NJ, TX, and PA) and their surrounding states ( AL, GA, IN, IO, MO, and VT) were identified as as primary collection sites to maximize resources during collection trips ( collect more than 67 sp ecies per trip ) These c ollection trips were funded by the Prunus Crop Germplasm Committee USDA ARS project No. 530621000018 00D and the Stone Fruit and Genetics Program at the University of Florida, Gainesville, FL Specimens were collected by D.J. Cha v ez and J.X. Chaparro. Collaboration during these collection trips was provided by T.G. Beckman, W.R. Okie, M. Enquist, J. Preece, T. Gradziel and others (acknowledgment s section). Open pollinated seed from 74 accessions, from approximately 16 species plus some natural oc c urring hybrids were submitted to the USDA National Clonal Germplasm Repository in Davis, CA (DPRU) for support and preservation of these species as important genetic resources (Table A 19) Collected plant specimens were deposited as her barium vouchers in the University of Florida Herbarium (FLAS) at the Florida Museum of Natural History, 124

PAGE 125

Gainesville, FL, USA. Additional leaf material from the collected samples were freeze dried using a Labconco freezone 2.5L system (Labconco, Kansas City MO, USA) or dried using a mixture of 2.5 kg silica gel dessicant 28200 mesh (Cat. No. S157212) and 500 g silica gel Tel Tale (TM) dessicant indicating 1018 mesh (Cat. No. S161500) ( Thermo Scientific Waltham, MA, USA). Dried sample materials are stor ed at the Stone Fruit Breeding and Genetics Program at the University of Florida, Gainesville, FL. Phylogenetic Study Material A t otal of 105 accessions (out of the 408) representing 35 taxa with approximately 23 accesions per species, were used for the study of the phylogeny of the North American plums (Table 41) Subgenus Prunus section Prunocerasus was rep resented by 14 taxa that are commonly accepted as North American Plums (Rehder, 1940; Shaw and Small, 2004, 2005; Waugh, 1901; Wight, 1915). Other g roups within genus Prunus were represented by an a dditional 21 species: sect. Armeniaca ( 2 ), sect. Penarmeniaca ( 3 ), sect. Prunus ( 1 ) sect. Microcerasus (1) in subgenus Prunus ; subgenus Emplectocladus (1); sect. Cerasus ( 5 ) and sect. Laurocerasus (3) in subgenus Cerasus ; and subgenus Amygdalus ( 5 ) (USDA ARS and GRIN taxonomic classifications) Prunus virginiana L. and P. serotina Ehrh. were used to root the different phylogenetic trees as they represent basal taxa to all of our samples based on previous st udies (Badenes and Parfitt, 1995; Bortiri et al., 2001; Bortiri et al., 2002; Bortiri et al., 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Shaw and Small, 2004, 2005; Wen et al., 2008) DNA Isolation DNA was extracted from leaf tissue using a modified CTAB method as described by Blaker (2010), and Chavez and Chaparro (2011). Lyophilized leaf tissue (~10 mg) of 125

PAGE 126

each sample was added to 2 mL Eppendorf microcentrifuge tubes with three 5 mm stainless steel beads, 750 L of CTAB buffer (2% CTAB, 100 m M Tris pH 8.0, 1.4 M NaCl, 0.5M EDTA, 1% PVP) previously mixed with Mercaptoethanol (1 L/mL), and 8 L of RNAse (10 mg/mL). Samples were grounded two/three times at 30 Hz in a Tissue Lyser (QIAGEN Inc., Valencia, CA, USA) for 1.5 min until tissue clumps w ere not visible. Samples were vortexed and incubated in a 65C water bath for 6 min. Then, tubes were vortexed and a volume of 750 L of chloroform:isoamyl (24:1) was added. Tubes were vortexed, incubated at 20C for 6 min, and then centrifuged at 12000 r cf for 10 min. The aqueous phase was transferred to a new 2 mL centrifuge tube, and 500 L of cold isopropanol were added. Tubes were gently mixed, incubated at 20C for 6 min, and then centrifuged at 16100 rcf for 10 min. Supernatant was removed, and pel let was washed with 500 L of cold 70% EtOH (by inverting the tubes carefully). Tubes were incubated at 20C for 5 min., and then centrifuged at 16100 rcf for 5 min. Supernatant was removed, and pellet was washed with 500 L of cold 90% EtOH (repeating mi xing, incubation and centrifugation, as described before). Ethanol was poured off and the pellet was dried at bench top for approx. 3045 min (room temperature). The pellet DNA was resuspended in 50 L TE buffer (10 mM Tris HCl, 0.1mM EDTA) and 50 L of D NA grade water. DNA concentration was quantified in a UV10 Spectrophotometer (Thermo Scientific, Waltham, MA, USA). DNA concentration for all the samples was standardized to 20 ng/uL. Molecular Data Amplicons with the highest amount of potentially informat ive characters ( PICs ) and the ability to recover previously known species relationships within section Prunocerasus as reported by Shaw and Small (2004, 2005) and Bortiri et al. (2006) 126

PAGE 127

were previously identified (Chapter 3, Tables 33 and 34). These regions were used to study the phylogeny of the North American plums. Chloroplast DNA The most informative sites among cpDNA regions were observed in 3'trnV ndhC and trnH psbA cpDNA regions (Chapter 3, Table 3 3). Similar results were reported by Shaw et al. (2005, 2007). Out of those two regions, trnH psbA was chosen to be us ed for a whole phylogenetic analysis. Primer sequences were sourced from Shaw et al. (2005, 2007). PCR products were amplified in a 48 L volume reaction containing 6 L of 20 ng/L DNA tem plate, 6.75 L 10X ThermoPol Reaction Buffer [10mM KCl, 10mM (NH4)2SO4, 20mM Tris HCl, 2mM MgSO4, 0.1% Triton X 100, pH 8.8 @ 25C], 3 L 2.5 mM dNTPs, 0.6 L Taq DNA Polymerase, 19.65 L DNA grade water, and 12 L 5M (6 L forward and 6 L reverse) primers PCR cycling conditions were: 5 min at 80C followed by 30 cycles of 1 min denaturing at 95C, followed by a ramp of 0.3C/s to 65C, and 4 min of elongation at 65C, ending with a final extension step of 5 min at 65C ( Shaw et al., 2007) PCR products were confirmed on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Nuclear Genes The highest amount of potentially informative sites and the ability to recover previously known species relationships within section Prunocerasus (Bortiri et al., 2006; Shaw and Small, 2004, 2005) were identified for the amplicons PGI, MAX4, PHYE and VRN1 (Chapter 3, Table 34). The MAX4 primer sets were sourced from Carrillo 127

PAGE 128

Mendoza (2012). PGI, PHYE, and VRN 1 primer sets were obtained as reported in Chapter 3 (Table A 9 ). PCR products were amplified in a 48 L volume reaction, as previously described for cpDNA and ITS regions amplification. PCR parameters were: 3 min at 94C followed by 40 cycles of 30 s denaturing at 94C, 30 s at primers specific annealing temperature [Ta(C)] (Table A 9 ), and 1 min of elongation at 72C, ending with 7 min at 72C. PCR products were separated on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel documentation system. Sequencing PCR products with multiple bands during electrophoresis were separated, exised and purified using QIAquick Gel Extraction Kit (Qiagen, Valencia, CA, USA). Single band products were purified by Eurofins MWG O peron (Huntsville, AL, USA). All purified products were forward and reverse direct sequenced by Eurofins MWG Operon (Huntsville, AL, USA) following company procedures. PGI was resequenced using PGIintF primer 5 TCAAATTAGCAGAGGTGGG3 (Chapter 3, Table A 1 0 ) Geneious R6.0.3 software (Biomatters Ltd., New Zealand) was used to edit and assemble complementary DNA strands. Phylogenetic Analyse s Sequence alignment was performed using Muscle v3.8 (Edgar, 2004). Alignments were check ed visually in Geneious R6.0 .3 software (Biomatters Ltd., New Zealand) to produce an alignment with the fewest number of changes. Alignments were checked for indels and polymorphic sites that resulted in sequence slippage. Polymorphisms (indels and/or substitutions) were manually det ermined using reference sequences from other species ( full coverage reference sequences from forward and 128

PAGE 129

reverse sequencing) as previously described on Chapter 3. Alleles were saved and used in further analyses. A consensus sequence was obtained for each accession. All the accessions sequences per region were realigned using Muscle v3.8 (Edgar, 2004) Haplotype sequences were calculated based on aligned regions using DnaSP v5 software (Librado and Rozas, 2009). Diploid sequence data was created using the consensus sequence obtained after assembly of the forward and reverse sequence products per specimen. Diploid consensus sequences included heterozygous sites and indels, which were manually determined by using reference sequences from other species. The nu mber of nucleotide substitutions ( Subst ) and indels, collectively named Potentially Informative Characters ( PICs=Subs+ Indels), were obtained using DnaSP v5 software (Librado and Rozas, 2009). The proportion of observed mutational events or percent variabil ity for each genomic region were calculated as reported by Shaw et al. (2005, 2007) using the following equation: percent variability = [( Subs+ Indels)/L] 100, where L = the total aligned sequence length. The number of mutations ( Eta ), number of haplotypes ( Hap ), haplotype diversity ( Hd ), nucleotide diversity per site ( ) (Nei, 1987) Watterson estimator of population mutation rate ( ) per sequence and per site (Watterson, 1975) and Tajimas D (Tajima, 1989) were calculated ( DnaSP v 5 ; Librado and Rozas, 2009) All these variables excluded indels f rom the ir analysis. Nuc leotide diversity w as measured using (the expected heterozygosity per nucleotide site) and (an estimate of 4Ne per site and per sequence) Tajimas D test was used to test for deviations from the neutral equilibrium model of evolution as described by Zhu et al. (2007). 129

PAGE 130

Phylogenetic analyses were performed by using the optimality criterion of maximum parsimony for diploid and phased haplotype sequences. Gaps (indels) were ignored and also considered as an additional binary character matrix, with 0=absence and 1=presence of the indel. Maximum parsimony (MP) trees were obtained using the parsimony ratchet method for rapid parsimony analysis with PAUPRat in PAUP* v. 4.0b10 ( Nixon, 1999; Sikes and Lewis, 2001; Swofford, 2002). A total of 1000 iterations (for diploid sequence) and 200 iterations (for haplotype sequence) with 15% of the characters being reweighted per iteration were performed. A modification to the pauprat script was created to be able to store branch lengths across MP trees This option allowe d us to collapse zero branch lengt hs Bootstrap values were calculated based on the best MP trees. Homoplasy was determined by calculating the consistency index (CI) and the retention index (RI). Incongruence length difference (ILD) tests (Farris et al., 1 994a, 1994b) were calculated to measure the incongruence among all gene regions. The test was implemented in PAUP* v. 4.0b10 (Swofford, 2002) by running 100 repetitions of the partition homogeneity test, with TBR, and a maximum setting of 1100 trees saved each step. Maximum likelihood analyses of individual and combined sequence data were performed using RAxML v.7.1.0. (Stamatakis, 2006; Stamatakis et al., 2008) Analyses were calculated under a general time reversible model of nucleotide substitution and a gamma model of rate heterogeneity in the CIPRES Science Gateway (Miller et al., 2010). A halt bootstrappping was performed. Best ML trees for each bootstrap repetition were used to calculate a consensus tree with posterior probability values in Mesquite v .2.73 (Maddison and Maddison, 2011 ). Networks of the phased trnH psbA haplotypes 130

PAGE 131

including gaps as a 5th character state, were estimated using TCS v.1.21 ( Clement et al., 2000). Results The germplasm collection trips across US allowed us to see at first h and the large genetic diversity available within Prunus Unique characteristics and different hybrid zones were identified ( Table A 1 8 see voucher notes) Waugh (1901) commented that plums grow pretty much as they please, and the botanist has to take the m as he finds them. The wide distribution and highly variable characteristics make the North American plums a good candidate for taxonomic conflicts. B otanists have found specimens that shar ed several morphological characteristics which made species delim itation quite difficult even when they belonged to different taxa. The objective of this study was to use nuclear sequence information in combination with cpDNA results to confirm and/or identify possible discrepancies resulting from interspecific hybridi zation, and to elucidate the North American plums phylogeny. A total of 35 taxa, represented by 94 acce ssions (plus 11 specimens from the core collection) were used in this study (Table 4 1). These samples were determined and confirmed as representatives of these species based on morphological characteristics, distribution range, habitat zones expert opinions (during collection), species descript ors plant ID keys (Rohrer, 2009), and collection records (GPS information, herbarium specimens, etc). A ny p uta tive hybrids between species were removed from our analysis. Hereafter in this chapter each sample was described using their collection ID as reported in Table 41. The genetic regions with the highest amount of informative sites and the ability to recover previously known species relationships within section Prunocerasus as reported 131

PAGE 132

by Shaw and Small (2004, 2005) and Bortiri et al. (2006) were previously identified in a core collection of 11 species (Chapter 3). These regions were used to study the phylogeny of the North American plums and included trnH psbA PGI, MAX4, PHYE, and VRN1. A total of 2 4 specimens per amplicon did not yield any product during P CR amplification or sequencing The specimens per amplicon were: PfasCA130 (PGI, MAX4), PhavTX275 (PGI MAX4, VRN1), PhavTX277 (PGI, MAX4, PHYE, VRN1), PfasCA266 (MAX4), Parm113 (PHYE), PfasCA269 (PHYE), PserrPA69 (PHYE), Pcer341 (VRN1), PfreCA271 (VRN1), and PminTX293 (VRN1). A total o f 2491 bp of alignable sequence for all regions were obtained per speci men (Table 42). The amplicons contained introns and exon regions representing 71% and 29% of the sequence, respectively (Tables A 2 0 and A 21) Subtitutions and indels were observed in all sequenced regions. A total of 106 indels were exclusively observed in introns, with only one indel identified in exons (PGI) The number of substitutions was higher than the number of indels across amplicons A total of 427 SNPs with an average of 17 SNP s every 100 nucleotides (polymorphism rate of 17.14% ) were reported across specimens and regions An average polymorphism rate of 22 % was observed when indels were included in these analyses. The highest variation across regions was always observed in introns for the number of haplotypes ( Hap ), haplotype diversity ( Hd ), n ucleotide diversity per site ( ), and Watterson estimator of population mutation rate ( ) per sequence and per site. Significant Tajimas D values were observed for PGI, PHYE, and VRN1 introns ( P <0.05) Similarly, trnH psbA and VRN1 exons reported a signi ficant Tajimas D value (P<0.05) Complete sequence analyses yielded significant Tajimas D values for all 132

PAGE 133

regions, except for MAX4 and PHYE PGI, PHYE, VRN1, MAX4, and trnH psbA had negative Tajimas D values (Tables 42, A 2 0 and A 2 1 ). These negative va lues rep resented the existence of a large number of low frequency mutations which supported the differences between lineages /species This constituted an important sign al of selection. Summary statistics for each locus are described on Table 42. Among th ese results, MAX4 had the highest percent var iability across regions PHYE had the lowest polymorphism rate VRN1 had the highest amount of substitutions and mutations in comparison to trnH psbA with the lowest amount The highest number of indels was observed for trnH psbA and the lowest for VRN1. PGI had th e highest number of haplotypes in comparison to trnH psbA with the lowest number of haplotypes The haplotype diversity estimate was similar across nuclear regions. All the regions under study were highly variable at the site and sequence level (Table 42). The chloroplast genome rate of evolution is slow in comparison with nuclear genes ( Soltis and Soltis, 1998). This trend was observed for the nuclear genes in our study which had a higher diversity at the nucleotide and sequence level in comparison to the cpDNA region trnH psbA (Table 42). The use of trnH psbA cpDNA region and the different nuclear genes allowed us to identify possible sources of incongruences between species a cross phylogenetic trees Phylogenetic Analyses S equences for all specimens consisted of 2452 bp representing five regions: trnH psbA (553 bp), PGI (457 bp), MAX4 (329 bp), PHYE (553 bp), and VRN1 (560 bp). Incongruence length difference (ILD) tests identified significant values of incongruence for all regions (including trnH psbA ) ( P <0. 05) Only comparisons between trnH psbA 133

PAGE 134

and VRN1 were not found incongruent ( P =0.06). Bortiri et al. (2002) reported that ITS, trnL trnF and S6PDH, showed incongruence among datasets for Prunus They found that the phylogenetic information using the combination of partitions supported additional clade relationships. Similarly, they advocated the use of combined datasets to maximize the phylogenetic information available as previously reported by N ixon and Carpenter (1996). A total of 248 parsimony informative characters (plus 311 multi indel bp binary characters) were identified. Specimens PseroPA72 and PvirKS253 had two copies for PGI. Prunus serotina and P. virginiana are tetraploid plant s (Dicks on et al., 1992;Dai and Walla, 2007). Both amplicon copies were used during PGI diploid and haplotype phased phylogenetic analyses (Figures 4 2, 4 11, A 85, A 86, A 99 to A 102 ) However, t he most divergent seq uence of PGI for PseroPA72 and PvirKS253 were not used during combined dataset analyses. Primers for MAX4, PHYE, and VRN1 amplicons did not produce two different products for these two species. The use of gaps as a binary character in diploid and haplotype sequences for MP analyses provided a higher support for some clades. However, gaps in general created conflicts with some species (outgroups) being identified as sister species to lower level groups (section Prunocerasus ) For example, PaviPA353 and PaviCA133 ( P. avium subgenus Cerasus section Ceras us ) were identified as sister species to PhorMD89 and PameKS254 ( P. hortulana and P. americana, subgenus Prunus section Prunocerasus ) in MAX4 diploid and haplotype MP analysis using gaps as a binary character s ( Figure s A 87, A 1 04 a nd A 105 ). In the other hand, MP analyses using MAX4 diploid and haplotype sequences without gaps recovered PaviPA353 and PaviCA133 species 134

PAGE 135

clustered together within the monophyletic clade of subgenus Cerasus (BS=100 % ) and within the monophyletic subclade of section Cerasus (BS= 100 % ) (Figures A 88, A 106 and A 107) Additional conflicting relationships for other samples were found across regions when gaps were used as binary characters MP analyses based on diploid and haplotype sequence data using gaps as binary characters were eliminated from further study but can be found among Figures A 83 to A 118 Maximum parsimony and ML analyses of diploid and phased haplotype sequences (without gaps) recovered similar taxa relationships at the subgenus level for all regions. Three major groups were identified: subgenus Cerasus su bgenus Amygdalus Emplectocladus Prunus (section Armeniaca, Penarmeniaca, and Microcerasus ), and subgenus Prunus section Prunocerasus (Figures 4 1 to 4 14, and A 84 to A 116 ). Relationships within section Prunocer asus were somewhat different for each region Some group relationships within this section matched the American and the Chickasaw clades (Shaw and Small 2004 and 2005). The MP analysis of trnH psbA using original sequences with and w ith out haplotype class ification (553 bp, 111 informative sites, 1000 trees, 136 steps, CI=0.96, and RI =0.98, Figure A 84 and A 85 ) recovered section Prunocerasus divided into two groups: A) A polytomy of P. americana, P. mexicana, and P. rivularis from the American clade; P. g eniculata and P. maritima from the Beach clade; and P. texana. B) A p olytomy of P. angustifolia, P. umbellata, P. munsoniana, P. gracilis and P. nigra from the Chickasaw clade; and P. hortulana, and P. alleghaniensis from the American clade. Prunus subcor data was sister to all section Prunocerasus species. Maximum likelihood analyses using original sequences with and wihout haplotype classification for trnH 135

PAGE 136

psbA sequence data did not identified additional relationships within section Prunocerasus (Figures 4 1 and 49) S pecimens of P. alleghaniensis P. maritima P. rivularis P. nigra, and P. munsoniana were found in both clades within section Prunocerasus We found that P. maritima genotypes PmarMA330 and PmarMA333, shared the same haplotype (Hap10) as th e majority of the P. geniculata samples Also, PmarNY87 and PmarMA334 had a haplotype (Hap 15) similar to most of the Chickasaw clade species (Figure 4 10) In addition, a unique haplotype (Hap 9) was identified in PgenFL240 and PrivrivTX299. Prunus geniculata and P. maritima species were identified as sister species using the ML (PP=0.54) (Figures 49 and A 98). Similar results were previously reported by Shaw and Small (2004, 2005), where several specimens of the same species were found spread across clades. Maximum parsimony analysis of PGI diploid ( 457 bp, 62 informative sites, 1000 trees, 1 21 steps, CI=0. 72, RI=0.9 2, Figure A 86) and haplotype (459 bp, 94 informative sites, 889 trees, 252 steps, CI=0.48, RI=0.88, Figure A 1 01) sequences recovered t wo ma jor groups within section Prunocerasus : A) Prunus americana, P. alleghaniensis P. mexicana, P. rivularis and P. hortulana from the American clade; P subcordata (PsubCA402 and PsubCA404) ; and P. geniculataP. texana clade This group formed a polytomy in the diploid analysis and a monophyletic clade in the haplotype analysis BS= 72% (ML PP=0.69) Prunus avium was observed as sister species to this group. B) Prunus angustifolia P. gracilis P. munsoniana, P. nigra, and P. umbellata from the Chickasaw clade; P subcordata PsubCA403; P. andersonii P. fremontii clade; and P. maritima This clade was recovered for both diploid and haplotype analysis. Similar results were obtained using ML analyses (Figures 42 and 411). 136

PAGE 137

Maximum parsimony analysis of MAX4 diplo id (329 bp, 56 informative sites, 1000 trees, 118 steps, CI=0.78, RI=0.93, Figure A 88) and haplotype ( 364 bp, 87 informative sites, 930 trees, 304 steps, CI=0.51, RI=0.84, Figure A 106) sequences identified t wo groups within section Prunocerasus : A) Prunu s angustifolia, P. munsoniana, and P. nigra, from the Chickasaw clade; P. americana from the American clade; and P. maritima P. pumila B) Prunus hortulanaP. rivularis P. mexicana, P. americana, and P. alleghaniensis from the American clade; P. umbellata and P. gracilis from the Chickasaw clade; P. texana and P. geniculata. These relationships were observed for both diploid and haplotype MP analysis No additional grouping s were reported. These two groups were not recovered using ML analyses (Figures 43 and 412). Maximum parsimony analysis of PHYE diploid (553 bp, 55 informative sites, 1000 trees, 82 steps, CI=0.95, RI=0.99, Figure A 90) and haplotype (553 bp, 76 informative sites, 200 trees, 114 steps, CI=0.88, RI=0.98, Figure A 1 11) sequences identifie d s ection Penarmeniaca as sister clade to sect ion Prunocerasus in subgenus Prunus Two major groups were observed within section Prunocerasus : A) Prunus angustifolia, P. gracilis P. munsoniana, P. nigra and P. umbellata from the Chickasaw clade; Prunus s ubcordata (PsubCA403); P. geniculata; P. texana ; and P. maritima This group formed a polytomy in the diploid and the haplotype analysis (BS=80% and 100%, respectively). B) Prunus americana, P. alleghaniensis P. mexicana, P. rivularis and P. hortulana fr om the American clade; and P subcordata (PsubCA402 and PsubCA404). Identical groups were observed in both diploid and haplotype MP analyses. Similar results were obtained for ML analyses (Figures 44 and 413). 137

PAGE 138

Maximum parsimony analysis of VRN1 diploid ( 560 bp, 52 informative sites, 1000 trees, 69 steps, CI=0.96, RI=0.99, Figure A 92 ) and haplotype (562 bp, 99 informative sites, 200 trees, 190 steps, CI=0.71, RI=0.93, Figure A 116) sequences f ound subgenus Prunus section Penarmeniaca and section Prunocerasus forming a clade, with subc lades of P. texana P. geniculata P. angustifolia and P. subcordata as sister taxa to the additional samples of section Prunocerasus Maximum likelihood analyses recovered similar phylogenetic relationships (Figures 4 5 and 414). A high level of homoplasy was observed for all datasets when using haplotype sequences in comparison to diploid data for MP analyses. This level of homoplasy was reduced when minor frequency haplotypes were removed from the analysis. In general, MP a nalyses of diploid and haplotype sequences (with or without minor frequency haplotypes) recovered the same major groups as previously described. Maximum parsimony and maximum likelihood analyses using phased haplotype sequence data allowed us to identify s pecimens with haplotypes found in two diverse clades ( indicating possible interspecific hybridization). No evidence for this was detected for samples representing subgenus Cerasus, subgenus Amygdalus subgenus Emplectocladus subgenus Prunus sections Penar meniaca, Microcerasus and Armeniaca. S everal samples were identified as possible hybrids by comparing their haplotype classification and their location within each amplicon phylogenetic tree. The samples classified as possible hybrids were: PsubCA403 (PGI PHYE ), Pmun343 (PGI PHYE ), PhorIL380 (PGI PHYE ), PangIL393 (PGI PHYE VRN1 ) PmexAR313 (MAX4), PrivLA84 (MAX4), PameAR322 (MAX4), PangAR311 (MAX4), PmunGA249 (MAX4, VRN1 ), PumbAL410 (MAX4), PameIL381 (MAX4), PameIL386 (MAX4), PameGA237 138

PAGE 139

(MAX4), PameKS2 55 (MAX4) and PangFL162 (VRN1) (Figures A 101, A 106, A 111 A 116). Previous reports have questioned the taxonomic validity of P. munsoniana and have hypothesized it to be a hybrid between P. angustifolia and other plum species (Rohrer et al 2004 and 2008). In our comparisons, both P. munsoniana samples, PmunGA249 and Pmun343, were classified as possible hybrids. MP analysis (without gaps) of c ombined nuclear diploid sequence data (1859 bp, 219 informative sites, 957 trees, 400 steps, CI=0.79, RI=0.94, Figure A 94) recovered similar taxa relationships at the subgenus level to those obtained for MP analyses using individual gene sets. Maximum likelihood analyses obtained the same results (Figure 46). Four major clades were observed: subgenus Cerasus clad e (BS=100% PP=0.99 ), subgenus Amygdalus Emplectocladus clade (BS=100% PP=0.97 ), subgenus Amygdalus Prunus (section ArmeniacaPenarmeniaca) clade (BS=52% PP=0.6 for subgenus Amygdalus Prunus and PP=0.84 for subgenus Prunus sect. ArmeniacaPenarmeniaca), and subgenus Prunus section Prunocerasus clade (BS=100% PP=0.54 ). All of these clades were monophyletic. Section Prunocerasus was further divided into two groups: A) Polytomy of P. geniculataP. texana clade (BS=100%), P. americanaP. alleghaniensis P. m exicana clade (BS=53%), and P. hortulanaP. rivularis clade (BS=100%). B) Polytomy of P. andersonii P. fremontii clade ( subgenus Prunus section Penarmeniaca, BS=100%), P. angustifoliaP. munsoniana clade (BS=100%), and P. nigraP. maritima P. umbellata cla de (BS=99%). Group A and group B represented in a large proportion to what w as previously known as the America n clade and the Chickasaw clade (Figure 4 6). 139

PAGE 140

Prunus subcordata was sister to all section Prunocerasus in the MP combined nuclear analysis Also, some specimens within section Prunocerasus were not clustered with any of the subgroups being resolved. These specimens were: PameKS254, Pmun343, PhorIL380, PameIO399, PmexAR313, PgraTX305, PumbAL410, PumbAR357, PrivLA84, and PangIL393. Out of these ten sa mples, six were previously hypothesized to have signs of introgression based on their phased haplotype data MP analysis as shown in Figures A 101, A 106, A 111, A 1 1 6 and were classified as hybrids based on these results The maximum parsimony analysis of combined nuclear and chloroplast datasets yielded similar phylogenetic relationships to those of the analysis with combined nuclear data (Figure A 96). The major difference observed across phylogenies was the position of P. andersonii P. fremontii (subgenus Prunus section Penarmeniaca) clade. This clade was recovered as sister species to all section Prunocerasus (previously clustered to the Chickasaw clade). These relationships within section Prunocerasus were not observed with ML analyses (Figure 47). The P. geniculata and P. texana clade was highly supported in MP and ML analyses (BS=100% PP=0.83 ) T hese species relationships contradict previous reports by Shaw and Small (2004, 2005) using chloroplast sequences. Zhang et al. (2012) summarized the possibl e causes of topological conflicts such as stochastic error (sampling errors), systematic errors (substitution rate variation and heterotachy), and biological factors (horizontal gene transfer, hybridization and introgression, lineage sorting, and existence of paralogs). They stated that once these 140

PAGE 141

different stochastic and systematic errors were ruled out the only possible causes left were biological factors. S ampling errors normally occurred from low number of informative characters and were ruled out in o ur study by the addition of four different nuclear genes and one cpDNA region. All of these regions were previously found to contain a high number of potentially informative characters (PICs) in a preliminary study with a core collection of 11 Prunus speci es (Chapter 3). The problems associated with systematic errors were ruled out as well. These problems can be broken by taxon sampling, faster evolving sequences and inference methods less sensitive to long branch attraction (Zhang et al. (2012). Our sampli ng involved 35 taxa with multiple samples per taxa, of which 14 taxa represented section Prunocerasus Sequences used in this phylogenetic study were highly variable as reported in Table 42. Finally, we decided to use ML in conjunction with MP approaches to avoid problems with long branch attraction. The only explanation remaining for incongruences between combined nuclear genes and cpDNA trnH psbA region in our MP analyses were due to biological factors (Figures A 94 and A 84 res pectively). These factors could be hybridization and introgression, and lineage sorting. Both nuclear and cpDNA MP ML analyses based on diploid and phased haplotype sequences positioned subgenera Cerasus Amygdalus Emplectocladus Prunus (sect. Armeniaca, Penarmeniaca, and Microc erasus ) basal to a monophyletic subgenus Prunus section Prunocerasus No incongruences between chloroplast and nuclear phylogenetic trees were observed among outgroup taxa. Incongruence between section Penarmeniaca and section Prunocerasus was observed in MP analyses (ML did not recover this 141

PAGE 142

relationship) Prunus andersonii P. fremontii clade (subgenus Prunus sect. Penarmeniaca) was observed within the sister clade to section Prunocerasus using trnH psbA (Figure A 84 ) The same clade was identified as sist er species to the Chickasaw clade within section Prunocerasus using nuclear genes (Figure A 94). This relationship could be due to the existence of genetic signals maintained after reticulation. Incongruences between chloroplast and nuclear phylogenies wer e observed in section Prunocerasus Maximum parsimony and ML analyses using trnH psbA recovered P. geniculata as sister species to P. maritima (Figure s 4 1 and A 83 ) T his species relationship was previously reported by Shaw and Small (2004, 2005). However MP and ML analyses using nuclear genes recovered P. geniculata as sister species to P. texana (Figure s 4 7 and A 94 ). It is possible that a hybridization event occurred between P. geniculata and P. maritima ( hypothesis further elaborated in the discussio n section). Incongruence was observed for P. rivularis and P. hortulana species relationships Chloroplast MP analysis recovered Prunus rivularis located within the American clade with P. americana, P. mexicana P. alleghaniensis and P. nigra. Similarly, cpDNA MP analysis recovered P. hortulana species located within the Chickasaw clade with P. umbellata, P. angustifolia, P. maritima P. alleghaniensis P. gracilis P. munsoniana (Figure A 83) Maximum parsimony and maximum likelihood analyses of nuclear g enes recovered a monophyletic clade of P.rivularis and P. hortulana species (Figures 4 6 and A 94). Both species were described by Wight (1915) to be closely related. We propose that these two species hybridized. 142

PAGE 143

The best ML tree for combined diploid seque nce data (nuclear plus chloroplast) represented best the species relationships found in this study (Figure 48). This phylogenetic tree had a similar topology to the majority consensus rule tree of the MP analysis for combined diploid sequence data (Figure A 96). Five major clades were observed: 1) Subgenus Cerasus clade with two subclades: section Laurocerasus and section Cerasus 2) Subgenus Amygdalus Emplectocladus clade, 3) Subgenus Amygdalus Prunus (section Microcerasus ) clade, 4) Subgenus Prunus (sect ion ArmeniacaPenarmeniaca) clade, and 5) Subgenus Prunus section Prunocerasus clade. Subgenus Prunus section Prunocerasus was divided in three clades: 5.1) the Sand clade, 5.2) the American clade, and 5.3) the Chickasaw clade. The results do not support t he existence of a Beach clade composed of P. maritima and P. geniculata. In constrast, we propose a new clade, the Sand clade containing P. texana and P. geniculata. The American clade composed of P. americana, P. alleghaniensis P. mexicana P. hortulana and P. rivularis The Chickasaw clade was formed by P. angustifolia, P. gracilis P. maritima P. munsoniana P. nigra, and P. umbellata. Discussion The amplification and sequencing of trnH psbA PGI, MAX4, PHYE and VRN1, was easily achieved and transfera ble across different Prunus species. The use of a representative core collection of species (at different evolutionary levels) proved to be a beneficial strategy for the identification of candidate regions for phylogenetic analysis as previously reported by Shaw et al. (2004, 2005). The use of the sequence data generated was successful in elucidating the phylogenetic relationships in Prunus as previously described (Table 42, Figures 41 to 414). 143

PAGE 144

Chloroplast DNA information has been used extensively in phylogenetic studies of Prunus (Badenes and Parfitt, 1995; Bortiri et al., 2001; Bortiri et al., 2002; Bortiri et al., 2006; Kaneko et al., 1986; Katayama and Uematsu, 2005; Shaw and Small, 2004, 2005; Wen et al., 2008). However, phylogenetic information sol ely based on chloroplast data may complicate the elucidation of relationships between closely related sexually compatible species Analysis of the nuclear sequences yielded a higher polymorphism rate as expected, and they provided a better insight into the relationships within section Prunocerasus Hybridization in Prunus is widely known and documented. Wight (1915) reported that several species within section Prunocerasus hybridized freely and several hybrids were identified. The use of both chloroplast and nuclear sequence information allowed the identification of putative interspecific hybrids and the detection of incongruences in the species relationships The relationships among Prunus species at the subgenus level were in agreement with previously reported information (Bortiri et al., 2001; Bortiri et al., 2002; Bortiri et al 2006; Lee and Wen, 2001; Shaw and Small, 2004 ; Wen et al., 2008) Five major clades were observed: 1) Subgenus Cerasus clade with two subclades: section Laurocerasus and section Cerasus 2) Subgenus Amygdalus Emplectocladus clade, 3) Subgenus Amygdalus Prunus (section Microcerasus ) clade, 4) Subgenus Prunus (section ArmeniacaPenarmeniaca) clade, and 5) Subgenus Prunus section Prunocerasus clade (Figures 4 8 and A 96 ). Bortiri et al. (2006) reported incongruence between cpDNA and ITS analyses for subgenus Cerasus section Cerasus This incongruence was not identified in our 144

PAGE 145

analyses as reported by Lee and Wen (2001) and Bortiri et al. (2001, 2002) Subgenus Cerasus is characterized by the presence of inflorescences with 12100 flowered racemes (lengths of central axes of inflorescences 425 times pedicels) (Rohrer, 2009). The main difference with the subgenus Amygdalus Emplectocladus clade was the presence of hairy drupes in compari son to glaborous drupes for subgenus Cerasus Subgenus Amygdalus Emplectocladus is characterized by unisexual flowers and entire sepals (Rohrer, 2009). Similarly, s ubgenus Amygdalus Emplectocladus subgenus Amygdalus Prunus (section Microcerasus), and subg enus Prunus section Armeniaca are characterized by hairy drupes. The difference between subgenus Prunus section Armeniaca and section Penarmeniaca ( P. pumila ) is that the fruit are pubescent and dry at maturity compared to glaborous fleshy fruits when ripe, respectively (Bortiri et al., 2006). Prunus cerasifera (subgenus Prunus section Prunus ) was found in a polytomy with P. subcordata. Both taxa are characterized by fleshy glaborous drupes. Prunus subcordata was recovered as sister to the rest of subgenus Prunus section Prunocerasus as reported by Shaw and Small (2004) and later confirmed by Bortiri et al. (2006). All of these clades and relationships at the subgenus level were highly supported based on MP and ML analysis (Figures 48 and A 96 ). Posi tions changed depending on the gene amplicon used for the phylogenetic analysis; however, they were very consistent across experiments. The regions used in this study recovered the phylogenetic relationships among subgenus as reported in previous studies (Bortir i et 145

PAGE 146

al., 2001; Bortiri et al., 2002; Bortiri et al 2006; Lee and Wen, 2001; Shaw and Small, 2004; Wen et al., 2008). One interesting point made by Shaw and Small (2004) and previously found by Bortiri et al. (2001) was that P. pumila (in their studies i t was classified as subgenus Cerasus section Microcerasus ) was not related to subgenus Cerasus They postulated this group was not a natural group. Prunus pumila was found associated with subgenus Prunus and subgenus Amygdalus (Bortiri et al., 2001; Lee an d Wen, 2001; Mowrey and Werner, 1990; Shaw and Small, 2004). The different phylogenetic locations of P. pumila and its relationships were recovered using the different gene amplicons (Figures 4 1 to 4 8). Watkins (1976) described the existence of a Microc erasus bridge. He described that the members of section Microcerasus (in our case adding also P. pumila ) that hybridized with both the subgenus Amygdalus Prunus group and subgenus Cerasus and that genetic transfer was possible between these groups. Hybri ds between P. pumila and species within section Prunocerasus have been previously reported (Wight, 1915). T hose hybrids represent an avenue for the transfer of genes from the North American plums into cultivated Amygdalus species (such as peach) for the development of rootstock and scion varieties. Similarly, this bridge could serve in the other direction for the development of new crops or fruit types based on species from the section Prunocerasus Subgenus Prunus section Prunocerasus was identified as a m onophyletic clade (BS=100%, PP=0.92). This result is consistent with previous research (Bortiri et al., 2006; Shaw and Small, 2004, 2005). Prunus subcordata was recovered as sister to the rest of subgenus Prunus section Prunocerasus as previously described. Prunus texana 146

PAGE 147

was included within section Prunocerasus The most interesting finding was that P. texana was closely allied to P. geniculata (Figures 4 8 and A 96 ). This relationship was highly supported in combined analysis (BS=100%, PP=0.83). Rohrer et al. (2008) recovered this relationship using LEAFY sequences. Prunus geniculata is a s hr ub, thorny, sometimes suckering, with glaborous drupes, endemic to the ridges of central Florida. It prefers dry, sunny, nutrient poor sites sandhills (Harper, 1911; Rohrer, 2009). Prunus texana is a shrub, weakly thorny, sometimes suckering, endemic to Texas, with hairy drupes. It grow s in deep sand, plains and sand hills, grasslands, and oak woods (Rohrer, 2009). These two species are morphologically similar and both g row in sandy and nutrient poor soils A clear morphological difference between these two species are hairy or glaborous drupes. The difference between hairy and glaborous environments probably arose as a response to different environmental conditions. We p roposed Sand clade as the name of the clade that they formed Germa i n Aubrey (2012) studied the phylogeography of central Florida scrub endemics. In this study, P. geniculata species were included to test the hypothesis on the geographical origin of the central Florida scrub. Two hypotheses were tested: 1) Scrubs species arrived in Florida between 5 and 2 Ma, and 2) Florida scrub arose at the end of the last glaciations, about 10000 years ago, with species migrating towards Florida in order to avoid the advancing ice sheet. B oth hypotheses were rejected based on ITS, rpl16 trnL intron, trnL F and trnS G reconstructed phylogenies using P. geniculata, P. texana, P. maritima and other additional species by Germai n Aubrey (2012). Our combined (chloroplast and nuclear) phylogenetic results confirmed the 147

PAGE 148

western origin for the P. geniculata species (Figures 4 8 and A 96 ). This western origin found P. texana and P. subcordata as sister to the Florida endemic P. geniculata. Furthermore, Germa i n Aubrey (2012) rep orted that not all P. geniculata and P. maritima accession s formed a monophyletic group as previously reported by Shaw and Small (2004). We observed the same type of incongruences when comparing cpDNA and nuclear phylogenetic analysis for P. geniculata and P. maritima positions (Figures 41 and A 83 for cpDNA, and Figures 47 and A 94 for nuclear). We postulate that this incongruence arose as a hybridization event between P. maritima and P. geniculata, in which P. maritima captured P. geniculata chloroplast s. This incongruence also fit ted to the geographical origin of P. geniculata and the inconsistencies found by Germai n Audrey (2012). In summary, the ancestral form of Prunus geniculata, as described by Germai n Audrey (2012) western hypothesis and confirmed with our results, probably arrived in Florida between 5 and 2 Ma. Then, about 2 Ma a semi arid biota extended fr om Florida into the western US (Myers and Myers, 1990). Some of this biota became isolated due to an increase of humidity in the Gulf of Mexico creating isolated xeric habitats in Florida in the mid Pliocene (Webb and Myers, 1990). This isolation provided the mechanism for the differentiation of P. geniculata. This hypothesis could be further supported by unique cpDNA haplotypes that have been found in Texas ( P. texana and P. rivularis ) and that also were found in low frequencies in Florida ( P. geniculata and P. umbellata) (Hap9, Figure 49). Similarly, Shaw and Small (2005) concluded that Texas appears to have been a refugium or point of origin f or many of the section Prunocerasus species based on rpL16. 148

PAGE 149

The proposed hybridization event between P. geniculata and P. maritima probably occurred in the last glaciation event. By, 10000 years ago, at the end of the last glaciation, P. maritima (or its a ncestral form) migrated from Northeastern US into north Florida ahead of the glacial sheet. In Florida, this species may have encounter ed P. geniculata (or its ancestral form), a llowing P runus geniculata and P. maritima to hybridize. This hypothesis w ould explain the presence of the same P. geniculata chloroplast within P. maritima and why some P. maritima chloroplasts are shared by other species (different than P. geniculata chloroplasts ) In our analysis, P. maritima shared nuclear and cpDNA haplotypes wi th species of the Chickasaw clade (in specific with P. umbellata) (Figures 49 to 4 14). Prunus maritima was determined to be a sister species of P. umbellata ( Figure 46, PP= 0.58; Figure A 94, BB=99%) This phylogenetic relationship was previously support ed by Rohrer et al. (2004) using SSR markers, Rohrer et al. (2008) using LEAFY sequences and by analysis of 41 SSR markers and 51 nuclear genes (Chapter 3). Subgenus Prunus section Prunocerasus clade was divided in three clades: A) the Sand clade, B) the American clade, and C) the Chickasaw clade. T he Sand clade was formed by P. texana and P. geniculata. The American clade was composed by P. americana, P. alleghaniensis P. mexicana P. hortulana, and P. rivularis The Chickasaw clade was formed by P. angu stifolia P. gracilis P. maritima P. munsoniana P. nigra and P. umbellata. The best ML tree for combined diploid sequence data (nuclear plus chloroplast) best represented the species relationships found in this study (Figure 4 8). These relationships w ere supported by the MP analysis for combined diploid sequence data (Figure A 96). 149

PAGE 150

Incongruence between P. rivularis and P. hortulana species was observed between cpDNA and combined nuclear phylogenetic analyses (Figures A 83 and 46). Ro hrer et al. (2004, 2008) described similar phylogenetic relationships among these two taxa using SSRs markers and LEAFY sequences They hypothesized that Prunus rivularis may be a hybrid with P. angustifolia as one parent. They suggested that hybridization between P. rivula ris and P. hortulana most likely happened. Shaw and Small (2004) described that P. angustifolia, P. munsoniana, P. hortulana and P. rivularis could be easily confused. Our results support the conclusion for hybridization and gene transfer between P. rivul aris and P. hortulana. Conclusions The use of nuclear markers proved beneficial to study and to clarify the North American plum species relationships. Three major clades were identified within s ection Prunocerasus : the Sand clade, the American clade, and t he Chickasaw clade (Figure 4 8) The Sand clade was constituted by P. texana and P. geniculata The American clade was composed by P. americana, P. alleghaniensis P. mexicana, P. hortulana, and P. rivularis The Chickasaw clade was formed by P. angustifol ia P. gracilis P. maritima P. munsoniana, P. nigra, and P. umbellata It was concluded that Prunus geniculata had a western origin created by isolation of semi arid biota that extended from Florida into western US about 2 Ma (Myers and Myers, 1990). This hypothesis was based on the high support for the monophyletic Sand clade, formed by P. texana and P. geniculata, sister to the remaining species in section Prunocerasus The disagreement between cpDNA analyses (Shaw and Small, 2004, 2005) and our resul ts for P. geniculata and P. maritima relationships were attributed to 150

PAGE 151

hybridization. It was hypothesized that during the last glacial event, about 10000 years ago, the two species were located in the same area and hybridized. This hypothesis w ould explain the presence of the same chloroplast haplotype within P. geniculata and P. maritima and the presence of P. maritima haplotypes shared by other species (different than P. geniculata haplotypes). In general, the use of cpDNA and nuclear gene sequences allowed us to identify the major phylogenetic relationships in section Prunocerasus It also indicated several sources of incongruence, with hybridization being the most probable source of this incongruence due to the known naturally occurring hybrids in Prunus (Wight, 1915). Additional analysis based on these results could be performed to further confirm these hybridization events. M orphological similarities among species within section Prunocerasus could be of importance to improve some clades and to identify s ynapomorphies for each clade Shaw and Small (2004) reported that phylogenetic relationships using cpDNA sequences were, in general, in disagreement with the previous morphological classifications (Waugh, 1899; Wight, 1915). Similarly, Bortiri et al. (2006) reported that morphological characters used in conjunction with molecular data (ITS and trn L trn F) improved species relationships, but proved to be homoplastic. We believe that future studies could involve adding morphological characteristics to our anal yse s. The species relationships in our study seem to be in more agreement with previous morphological classifications (Waugh, 1899; Wight, 1915). The use of fossils to obtain a time estimation of the reticulation event in Prunus and in section Prunocerasus could be of importance to further understand the 151

PAGE 152

relationships among species. Further investigations will include: 1) adding morphological characters to this phylogenetic analysis, 2) study the phylogenetic relationships among species within section Prunocerasus after removing some of the species involved in hybridization events, and 3) time divergence estimation using fossils. From a breeding perspective, this phylogenetic study provided us with a framework for our future breeding efforts to introgress tr aits from wild to cultivated species. For example, Watkins (1976) described the existence of a Microcerasus bridge. He described that the members of section Microcerasus (in our case adding also P. pumila ) hybridized with both the subgenus Amygdalus Prun us group and subgenus Cerasus and that genetic transfer was possible between these groups. Hybrids between P. pumila and species within section Prunocerasus have been previously reported (Wight, 1915). Those hybrids represent an avenue for the transfer of genes from the North American plums into cultivated Amygdalus species (such as peach) for the development of rootstock and scion varieties. Similarly, this bridge could serve in the other direction for the development of new crops or fruit types based on species from the section Prunocerasus 152

PAGE 153

Table 41. List of specimens used for the study of the phylogeny of the North American Plums ( Prunus spp.)z. Voucher (No.) Collection (No.) Genus Species IDy Rank Infraspecific Subgenusx Section Genbank accession (No.) trnH psbA PGI MAX4 PHYE VRN1 350 Prunus alleghaniensis PallePA350 Prunus Prunocerasus 352 Prunus alleghaniensis PallePA352 Prunus Prunocerasus 369 Prunus alleghaniensis PalleMI369 Prunus Prunocerasus 322 Prunus ame ricana PameAR322 Prunus Prunocerasus 237 Prunus americana PameGA237 Prunus Prunocerasus 381 Prunus americana PameIL381 Prunus Prunocerasus 386 Prunus americana PameIL386 Prunus Prunocerasus 399 Prunus americana PameIO399 Prunus Prunocerasus 254 Prunus americana PameKS254 Prunus Prunocerasus 397 Prunus americana PameMO397 var. lanata Prunus Prunocerasus 12 Prunus americana PameFL12 Prunus Prunocerasus 261 Prunus andersonii PandCA261 Prun us Penarmeniaca 262 Prunus andersonii PandCA262 Prunus Penarmeniaca 263 Prunus andersonii PandCA263 Prunus Penarmeniaca 154 Prunus angustifolia PangGA154 Prunus Prunocerasus 162 Prunus angustifolia PangFl162 Prunus Prunocerasus 256 Prunus angustifolia PangKS256 Prunus Prunocerasus 302 Prunus angustifolia PangTX302 Prunus Prunocerasus 311 Prunus angustifolia PangAR311 Prunus Prunocerasus 393 Prunus angustifolia PangIL393 Prunus Prunocerasus 25 Prunus angustifolia PangGA25 Prunus Prunocerasus 113 Prunus armeniaca cv. Canino Parm113 Prunus Armeniaca 133 Prunus avium PaviCA133 Cerasus Cerasus 353 Prunus avium PaviPA353 Cerasus Cerasus 341 Prunus cer asifera Pcer341 Prunus Prunus 141 Prunus eremophila PereCA141 Amygdalus 130 Prunus fasciculata PfasCA130 var. fasciculata Emplectocladus 153

PAGE 154

Table 4 1. Continued. Voucher (No.) Collection (No.) Genus Species IDy Rank Infraspecific S ubgenusx Section Genbank accession (No.) trnH psbA PGI MAX4 PHYE VRN1 264 Prunus fasciculata PfasCA264 var. punctata Emplectocladus 266 Prunus fasciculata PfasCA266 var. punctata Emplectocladus 269 Prunus fasciculata PfasCA269 v ar. fasciculata Emplectocladus 131 Prunus fasciculata PfasCA131 var. fasciculata Emplectocladus 271 Prunus fremontii PfreCA271 Prunus Penarmeniaca 272 Prunus fremontii PfreCA272 Prunus Penarmeniaca 274 Prunus fremontii Pfr eCA274 Prunus Penarmeniaca 174 Prunus geniculata PgenFl174 Prunus Prunocerasus 187 Prunus geniculata PgenFl187 Prunus Prunocerasus 210 Prunus geniculata PgenFL210 Prunus Prunocerasus 236 Prunus geniculata PgenFL236 Pru nus Prunocerasus 240 Prunus geniculata PgenFL240 Prunus Prunocerasus 246 Prunus geniculata PgenFL246 Prunus Prunocerasus 115 Prunus geniculata PgenFl115 Prunus Prunocerasus 139 Prunus glandulosa PglaCA139 Prunus Microcer asus 303 Prunus gracilis PgraTX303 Prunus Prunocerasus 304 Prunus gracilis PgraTX304 Prunus Prunocerasus 305 Prunus gracilis PgraTX305 Prunus Prunocerasus 275 Prunus havardii PhavTX275 Amygdalus 277 Prunus havardii PhavTX277 Amygdalus 319 Prunus hortulana PhorAR319 Prunus Prunocerasus 380 Prunus hortulana PhorIL380 Prunus Prunocerasus 394 Prunus hortulana PhorMO394 Prunus Prunocerasus 89 Prunus hortulana PhorMO89 Prunus Prunocerasus 134 Prunus ilicifolia PiliCA134 var. ilicifolia Cerasus Laurocerasus 140 Prunus ilicifolia PiliCA140 var. integrifolia Cerasus Laurocerasus 330 Prunus maritima PmarMA330 var. maritima Prunus Prunocerasus 154

PAGE 155

Table 4 1. Contin ued. Voucher (No.) Collection (No.) Genus Species IDy Rank Infraspecific Subgenusx Section Genbank accession (No.) trnH psbA PGI MAX4 PHYE VRN1 333 Prunus maritima PmarMA333 var. maritima Prunus Prunocerasus 340 Prunus maritima PmarMA340 var. maritima Prunus Prunocerasus 87 Prunus maritima PmarNY87 var. maritima Prunus Prunocerasus 255 Prunus mexicana PmexKS255 Prunus Prunocerasus 313 Prunus mexicana PmexAR313 Prunus Prunocerasus 367 Prunus mexicana PmexT X367 Prunus Prunocerasus 85 Prunus mexicana PmexLA85 Prunus Prunocerasus 291 Prunus minutiflora PminTX291 Amygdalus 292 Prunus minutiflora PminTX292 Amygdalus 293 Prunus minutiflora PminTX293 Amygdalus 117 Pr unus mume PmumTW117 Prunus Armeniaca 249 Prunus munsoniana PmunGA249 Prunus Prunocerasus 343 Prunus munsoniana Pmun343 Prunus Prunocerasus 88 Prunus munsoniana PmunTX88 Prunus Prunocerasus 335 Prunus nigra Pnig335 Prun us Prunocerasus 345 Prunus nigra PnigVT345 Prunus Prunocerasus 121 Prunus persica cv. Okinawa PperJP121 Amygdalus 324 Prunus pumila PpumsusMA324 var. susquehanue Prunus Penarmeniaca 326 Prunus pumila PpumdepMA326 var. depressa Prunus Penarmeniaca 329 Prunus pumila PpumsusMA329 var. susquehanue Prunus Penarmeniaca 371 Prunus pumila PpumMI371 var. susquehanue Prunus Penarmeniaca 97 Prunus pumila PpumPA97 var. depressa Prunus Penarmeniaca 15 Prunus p umila PpumIL15 Prunus Penarmeniaca 84 Prunus rivularis PrivLA84 Prunus Prunocerasus 278 Prunus rivularis PrivpubTX278 var. pubescens Prunus Prunocerasus 283 Prunus rivularis PrivpubTX283 var. pubescens Prunus Prunocerasus 287 Prunus rivularis PrivrivTX287 var. rivularis Prunus Prunocerasus 155

PAGE 156

Table 4 1. Continued. Voucher (No.) Collection (No.) Genus Species IDy Rank Infraspecific Subgenusx Section Genbank accession (No.) trnH psbA PGI MAX4 PHYE VRN1 290 Prunus rivularis PrivrivTX290 var. rivularis Prunus Prunocerasus 299 Prunus rivularis PrivrivTX299 var. rivularis Prunus Prunocerasus 301 Prunus rivularis PrivpubTX301 var. pubescens Prunus Prunocerasus 66 Prunus sargentii PsarPA66 Cerasus Cerasus 72 Prunus serotina PseroPA72 Cerasus Laurocerasus 69 Prunus serrulata PserrPA69 Cerasus Cerasus 402 Prunus subcordata PsubCA402 Prunus Prunocerasus 403 Prunus subcordata PsubCA403 Prunus Prunocerasus 404 Prunus subcordata PsubCA404 Prunus Prunocerasus 68 Prunus subhirtella PsubtPA68 Cerasus Cerasus 294 Prunus texana PtexTX294 Prunus Prunocerasus 297 Prunus texana PtexTX297 Prunus Prunocerasus 361 Prunus texana PtexTX361 Prunus Prunocerasus 93 Prunus triloba PtriCH93 Amygdalus 14 Prunus umbellata PumbGA14 Prunus Prunocerasus 17 Prunus umbellata PumbFl17 Prunus Prunocerasus 36 Prunus umbellata PumbFl36 Prunus Prunocerasus 104 P runus umbellata PumbFl104 Prunus Prunocerasus 164 Prunus umbellata PumbFL164 Prunus Prunocerasus 357 Prunus umbellata PumbAR357 Prunus Prunocerasus 410 Prunus umbellata PumbAL410 Prunus Prunocerasus 23 Prunus umbellata P umbFl23 Prunus Prunocerasus 253 Prunus virginiana PvirKS253 var. virginiana Cerasus Laurocerasus 70 Prunus x. yedoensis PxyedPA70 Cerasus Cerasus zAdditional samples used in some analyses: PdulTNP5= P. dulcis cv. Tardy Nonpareil and PkanPK6= P. kansuensis zID = first letter represented the genus ( Prunus =P), next three letters represented the species ( americana=ame, etc.), and the following letters represented the state of origin and collection number (Florida collection 12=FL12, etc .). ySubgenus and section classifications based on USDA GRIN. 156

PAGE 157

Table 42. Summary statistics of trnH psbA PGI MAX4 PHYE and VRN1 using haplotype phased sequences. trnH psbA PGI MAX4 PHYE VRN1 Aligned L. (bp) z 553 459 362 553 562 Substitutions (no.) 62 85 89 78 113 Indels (no.) 46 25 15 15 6 PICs y (no.) 108 110 104 93 119 % Variability x 19.53 23.97 28.73 16.82 21.17 Eta w (no.) 66 96 101 85 134 G+C content 0.25 0.41 0.35 0.37 0.39 Hap v (no.) 15 96 90 48 75 Hd u 0.791 0.961 0.959 0.941 0.891 ( per site) 0.016 0.015 0.030 0.015 0.015 (per sequence) from Eta 11.145 15.224 17.125 14.017 22.556 0.050 0.041 0.066 0.032 0.041 D r 2.040 1.871 1.702 1.559 1.955 p value <0.05 <0.05 >0.1 >0.05 <0.05 zComparisons using haplotype phased sequence data. yPICs = potentially informative characters. PICs = substitutions + indels. x% variability = [(Substitution+Indels)/L] 100. wEta = number of mutations. vHap = number of haplotypes (Nei, 1987). Gaps not considered. uHd = haplot ype diversity (Nei, 1987). Gaps not considered. t = nucleotide diversity (Nei, 1987). s = Watterson estimator of population mutation rate (Watterson, 1975). rD = Tajimas D (Tajima, 1989). 157

PAGE 158

Figure 41. Ma ximum likelihood tree using RAxML for trnH psbA sequence data (553 bp) (lnL= 1243.68). Maximum likelihood posterior probabilities values are located below the branches Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. E mplectocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunocerasus 158

PAGE 159

Figure 42. Maximum likelihood tree using RAxML for PGI diploid sequence data (457 bp) (lnL= 1294.72). Maximum likelihood posterior probabilities values are located below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus 159

PAGE 160

Figure 43. Maximum likelihood tree using RAxML for MAX4 diploid sequence data (3 29 bp) (lnL= 1133.54). Maximum likelihood posterior probabilities values are locate d below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus 160

PAGE 161

Figure 44. Maximum likelihood tree using RAxML for PHYE diploid sequence data (553 bp) (lnL= 1294.97). Maximum likelihood posterior probabilities values are located below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus 161

PAGE 162

Figure 45. Maximum likelihood tree using RAxML for VRN1 diploid sequence data (560 bp) (lnL= 1247.39). Maximum likelihood posterior probabilities values are located below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplec tocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunocerasus 162

PAGE 163

Figure 46. Maximum likeli hood tree using RAxML for combined nuclear genes diploid sequence data (1859 bp) (lnL= 5267.71). Maximum likelihood posterior probabilities values are located below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amy gdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus Subg. Prunus sect. Penarmeniaca 163

PAGE 164

Figure 47. Maximum likelihood tree using RAxM L for combined diploid sequence data total evidence approach ( 2452 bp) (lnL= 6204.00). Maximum likelihood posterior probabilities values are located below the branches. Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amygdalus Subg. Prunus sect. Mi crocerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus 164

PAGE 165

Figure 48. Best m aximum likelihood tree using RAxML for combined diploid sequence data total evidence approach (2452 bp) (lnL= 6204.00). Subg. Cerasus sect. Laurocerasus Subg. Cerasus sect. Cerasus Subg. Amygdalus Subg. Emplectocladus Subg. Amygdalus Subg. Prunus sect. Microcerasus Subg. Prunus sect. Armeniaca Subg. Prunus sect. Penarmeniaca Subg. Prunus sect. Prunus Subg. Prunus sect. Prunocerasus Chickasaw clade Americana clade Sand clade 165

PAGE 166

Figure 49. Maximum likelihood tree using RAxML for trnH psbA sequence classification with haplotype number prefix (553 bp) (lnL= 1243.68). Maximum likelihood posterior probabilities values are located below the branches. Colors represent different haplotypes. 166

PAGE 167

Figure 410. Haplotype network for trnH psbA sequence data (553 bp) considering gaps as a 5th character. Colors and haplotype numbers correspond results on Figure 49. Subgenus Prunus section Prunocerasus 167

PAGE 168

Figure 411. Maximum likelihood tree using RAxML for PGI haplotype phased sequence data ( 459 bp) after removing minor frequency haplotypes (lnL= 1742.83). Maximum likelihood posterior probabilities values are located below the branches. Colors represent different cpDNA haplotypes as reported on Figure 4 9 168

PAGE 169

A A 169

PAGE 170

Figure 412. Maximum likelihood tree using RAxML for MAX4 haplotype phased sequence data (364 bp) after removing minor frequency haplotypes (lnL= 1640.83). Maximum likelihood posterior probabilities values are located below the branches. Colors represent different cpDNA haplotypes as reported on Figure 49. 170

PAGE 171

A 171

PAGE 172

A 172

PAGE 173

Figure 413. Maximum likelihood tree using RAxML for PHYE haplotype phased sequence data (553 bp) after removing minor frequency haplotypes (lnL= 1536.13). Maximum likelihood posterior probabilities values are located below the branches. Col ors represent different cpDNA haplotypes as reported on Figure 49. 173

PAGE 174

A 174

PAGE 175

A 175

PAGE 176

Figure 414. Maximum likelihood tree using RAxML for VRN1 haplotype phased sequence data (562 bp) after removing minor freque ncy haplotypes (lnL= 1976.81). Maximum likelihood posterior probabilities values are located below the branches. Colors represent different cpDNA haplotypes as reported on Figure 49. 176

PAGE 177

A 177

PAGE 178

A 178

PAGE 179

CHAPTER 5 G ENETIC DIVERSITY AND POPULATION STRUCTURE OF PRUNUS UMBELLATA ELLIOT IN FLORIDA Introduction The North American plums are found in diverse climate and soil conditions. They have the highest diversity of flavor, aroma, texture, color, form and size (Hedrick 1911; Waugh, 1901). Layne and Bassi (2008) reported that the Prunus germplasm contains high variability for tree size, growth habit, flower size and color, chill hour requirement, fruit size, flesh texture, flesh color, resistance to diseases, and adaptability to diverse climatic and geographic regions. Floridas native plum species are P. americana, P. angustifolia, P. umbellata, and the endangered P. geniculata. The y are distributed from north Florida to south central Florida ( Figure 51 ) These species overlap in some areas of their distribution, but their habitat preferences are somewhat different. Prunus americana populations grow in creek or river banks, in deep, rich, comparatively moist soil, sometimes with partial shade, but also in the open sun (Sargent, 1905). Nevertheless, they can grow in other conditions (Waugh, 1901). Prunus angustifolia can grow in dry sandy, loose soil, in the open sun or partly shade. Populations grow in open woodlands, woodland edges, forest openings, savannahs, prairies plains, meadows, pastures, fence rows, roadsides. Prunus geniculata is a sh rub, endemic, that is known to occur in areas along the ridges of central Florida. It prefers dry, sunny, nutrient poor fire prone sites (Harper, 1911). Prunus umbellata is usual ly found in river swamps and hammocks. It can grow in semi shade or no shade. The existence of hybrids of Floridas native plum species with other Prunus species has been reported by Wight (1915). Wight (1915) reported that P. americana 179

PAGE 180

hybridized with P. angustifolia P. besseyi P. hortulana, P. munsoniana, and P. simonnii Similarly, P. angustifolia hybridized freely with P. americana, P. triflora P. besseyi (currently known as P. pumila ) P. munsoniana P. cerasifera, and P. orthosepala Prunus genicul ata has not been reported to hybridize freely with other species, and it was not included in Wights (1915) description. However, inter specific hybrids have been recovered of openpollinated P. geniculata plants in a field containing several Prunus specie s at the Stone Fruit Breeding and Genetics Program (Chaparro and Chavez personal communication) Similarly, Sherman (personal communication) described that P. umbellata hybridized freely with P. americana. It is clear, as described above, that hybridizati on could occur among the Floridas native plums. Inter specific hybridization has been widely used by breeders to introgress unique traits not available in commercial germplasm S ection Prunocerasus contains tremendous genetic variability that has not been widely used for the improvement of Prunus scion and rootstock material. N ative North American plum species have been identified as a source of resistance to blossom blight and brown rot ( Monilinia fructicola ), bacterial spot ( Xanthomonas campestris pv. pr uni ), bacterial canker ( Pseudomonas syringae pv. syringae), plum leaf scald ( Xylella fastidiosa ), peach tree short life (PTSL), root knot nematode ( Meloidogyne spp.), lesion nematode ( Pratylenchus spp.), clitocybe root rot ( Armillaria tabescens ), and other s (Beckman and Okie, 1994; Beckman et al., 1998; Layne and Sherman, 1986; Okie and Weinberger, 1996). Resistance to bacterial leaf spot and bacterial canker was identified in a cultivar derived from P. salicina P. cerasifera P. angustifolia P. americana, and P. 180

PAGE 181

munsoniana. Similarly, P hortulana were found to be resistant to root knot nematode, and this species has been used as a rootstock for European plums. Hybrids of P. americana, P. hortulana, P. angustifolia, and/or P. umbellata have demonstrated t olerance for PSTL. Other potential uses of the native North American plum species as breeding parents, scions and/or rootstocks were summarized by Beckman and Okie (1994), and Okie and Weinberger (1996). P lum species native to Florida harbor multiple desir able traits unique to our environment (relative to North America) They have evolved under high disease preassure due to our monsoonal climate and are adapted to a climate with low chilling requirements. The existence of several wild plum species adapted t o Florida creates a unique opportunity for the study of allelic variation associated with important economic traits that can then be targeted and rapidly transferred to domesticated plum using molecular markers. The main objective of this research is to analyze the genetic diversity and population structure of P. umbellata in Florida, and to generate a preliminary measure of the genetic diversity of the Florida plum native species, in specific of P. umbellata due to its wide geographic distribution range in Florida Prunus umbellata constituted a good model to identify adaptive differences associated with population genetic diversity and adaptation. Material and Methods Plant Material A total of 96 specimen s of P. americana (4), P. angustifolia (4), P. genic ulata (9 ), P. umbellata (67 ), P. umbellata x angustifolia (1), and P. umbellatalike (11) were collected from the wild by D.J. Cha vez and J.X. Chaparro (Table 5 1) These samples were chosen as they were broadly distributed across the southeastern US (Figure 5 2). 181

PAGE 182

Additional species data from the plum core collection (of 11 species) were included in the analyses ( Chapter 3, Table 31) Collected plant specimens were deposited as herbarium vouchers in the University of Florida Herbarium (FLAS) at the Florida Museum of Natural History, Gainesville, FL, USA. Additional leaf material from the collected samples were freeze dried using a Labconco freezone 2.5L system (Labconco, Kansas City, MO, USA) or dried using a mixture of 2.5 kg silica gel dessicant 28200 m esh (Cat. No. S157212) and 500 g silica gel Tel Tale (TM) dessicant indicating 1018 mesh (Cat. No. S161500) ( Thermo Scientific Waltham, MA, USA). Dried sample materials are stored at the Stone Fruit Breeding and Genetics Program at the University of Fl orida, Gainesville, FL. DNA Isolation DNA was extracted from leaf tissue using a modified CTAB method as described by Blaker (2010), and Chavez and Chaparro (2011). Lyophilized leaf tissue (~10 mg) of each sample was added to 2 mL Eppendorf microcentrifug e tubes with three 5 mm stainless steel beads, 750 L of CTAB buffer (2% CTAB, 100 mM Tris pH 8.0, 1.4 M NaCl, 0.5M EDTA, 1% PVP) previously mixed with Mercaptoethanol (1 L/mL), and 8 L of RNAse (10 mg/mL). Samples were grounded two/three times at 30 Hz in a Tissue Lyser (QIAGEN Inc., Valencia, CA, USA) for 1.5 min until tissue clumps were not visible. Samples were vortexed and incubated in a 65C water bath for 6 min. Then, tubes were vortexed and a volume of 750 L of chloroform:isoamyl (24:1) was added. Tubes were vortexed, incubated at 20C for 6 min, and then centrifuged at 12000 rcf for 10 min. The aqueous phase was transferred to a new 2 mL centrifuge tube, and 500 L of cold isopropanol were added. Tubes were gently mixed, incubated at 20C for 6 min, and then centrifuged at 16100 rcf for 10 min. Supernatant was removed, and pellet 182

PAGE 183

was washed with 500 L of cold 70% EtOH (by inverting the tubes carefully). Tubes were incubated at 20C for 5 min., and then centrifuged at 16100 rcf for 5 min. Super natant was removed, and pellet was washed with 500 L of cold 90% EtOH (repeating mixing, incubation and centrifugation, as described before). Ethanol was poured off and the pellet was dried at bench top for approx. 3045 min (room temperature). The pellet DNA was re suspended in 50 L TE buffer (10 mM Tris HCl, 0.1mM EDTA) and 50 L of DNA grade water. DNA concentration was quantified in a UV10 Spectrophotometer (Thermo Scientific, Waltham, MA, USA). DNA concentration for all the samples was standardized t o 20 ng/uL. SSR Markers A total of 44 SSR markers distributed across the peach genome (~ 10.5 cM between markers) were selected from the Prunus Texas almond Earlygold peach (T E) reference map (Dirlewanger et al., 2004; Jung et al., 2008). Forward m arkers were modified using a 5 fluorophore label 6FAM (standard) [6 FAM] or HEX [5HEX] (Eurofins MWG Operon, Huntsville, AL, USA) for multiplex product fragment analysis (Table A 2 2 ). Two haploid genotypes were used as controls to test product amplificat ion accuracies for each SSR marker ( as previously described on Chapter 2). PCR products were amplified in a 16 L volume reaction containing 2 L of 20 ng/L DNA template, 2.25 L 10X ThermoPol Reaction Buffer [10mM KCl, 10mM (NH4)2SO4, 20mM Tris HCl, 2mM MgSO4, 0.1% Triton X 100, pH 8.8 @ 25C], 1 L 2.5 mM dNTPs, 0.2 L Taq DNA Polymerase, 6.55 L DNA grade water, and 4 L 5M (2 L forward and 2 L reverse) primers. PCR parameters were: 3 min at 94C followed by 40 cycles of 30 s denaturing at 94C, 30 s at primers specific annealing temperature [Ta(C)] (Table A 2 2 ), and 1 min of elongation at 72C, ending with 7 min at 72C. PCR 183

PAGE 184

products were separated on 3% (m/v) agarose gel. Gels were stained with ethidium bromide and recorded using a digital gel doc umentation system. Gel images PCR amplification intensities were used to determine PCR dilution ratios for fragment analysis on an ABI3730 sequencer (Applied Biosystems Grand Island, NY, USA) at the Interdisciplinary Center for Biotechnology Research (IC BR), University of Florida, Gainesville, FL. Fragment analysis of chromatographs were visualized using GeneMarker v.1.6 software (SoftGenetics, LLC, State College, PA, USA) using 600 LIZ size standard (Applied Biosystems, Grand Island, NY, USA). Data An alyses Genetic and Cluster Analyses G enetic variation and cluster analyses of the fingerprinting data were performed using Powermarker v.3.25 (Liu and Muse, 2005) and GenAlEx v.6.5 software (Peakall and Smouse, 2006, 2012). Genetic variation was characteri zed for each locus by the number of observed alleles ( A ), effective number of alleles ( Ae ), observed heterozygosity ( Ho ), expected heterozygosity ( He ), Wrights fixation index [ F=(He Ho)/He=1 (Ho/He) ], and polymorphism information content ( PIC ). Neis gene tic distance (Nei and Takezaki, 1983) was calculated. Unweighted Pair Group Method with Arithmetic Mean (UPGMA) (Sokal and Michener, 1958) and Neighbor Joining (NJ) (Saitou and Nei, 1987) cluster analyses were performed. Specimens phenotypic information [ species, collection site (US state), and chilling requirement estimation (based on peach standards) ] were used in conjuction with the cluster analysis in Mesquite v.2.73 software (Maddison and Maddison, 2011) This comparison was made to identify possible associations between genotype information and cluster results. 184

PAGE 185

Neighbor net networks were built to measure the genetic relationships among genotypes and species using SplitsTree v.4.11.3 (Huson and Bryant, 2006) based on a distance matrix generated in POFA D v.1.03 (Joly and Bruneau, 2006). Neis genetic distance matrix generated in Powermarker v.3.25 software was used as input in POFAD. Population Structure Population structure analyses were performed for all the specimens (plus core collection) and separat ely for southeastern US specimens using Structure v.2 software (Pritchard et al., 2000). Structure simulation parameters were run under the admixture model assumption with correlated alleles using four reps per run, for one to ten K subgroups for all speci mens with 105 interactions after a burnin of 104 interactions. Structure Harvester software (Earl and VonHoldt, 2012) was used to implement the Evanno method to analyze the population structure results (Evanno et al., 2005). Distruct v.1.1 software (Rosenberg, 2004) was used to visualize and modify bar plot structure results. Detection of Loci Under Selection Beaumont and Nicholss modified frequentist method (Beaumont and Nichols, 1996), and Foll and Gaggiottis genomescan bayesian method (Foll and Gaggi otti, 2008), were used to estimate loci that undergo possible selective effects. LOSITAN (LOoking for Selection In a TANgled dataset) workbench software (Antao et al., 2008) and BayeScan v.2.1 software (Fischer et al., 2011; Foll and Gaggiotti, 2008; Foll et al., 2010), were used to implement each method, respectively. LOSITAN workbench parameters were: 50000 simulations, a neutral mean Fst calculation (50000 simulations) by removing all loci outside the desired confidence 185

PAGE 186

interval (95%), a force mean Fst calculation by running a bisection algorithm over repeated simulations, confidence interval of 95%, false discovery rate of 1%, and an infinite allele mutation (IAM) model. BayeScan software calculates the posterior probability for each locus using a Ba yesian method and a reversiblejump MCMC approach to selection. Parameters of the chain were: a sample size of 5000, thinning interval of 10, pilot runs of 20, pilot run length of 5000, and additional burnin of 50000. Parameters of the model were prior odds for the neutral model of 1 (neutral and selection models equally likely), and 0.1 (10 times more likely selection than neutral model). Loci with a probability value over 0.95 were reported as outliers. LOSITAN and BayeScan calculations used P. umbellata specimens SSR data ( as based on structure results). Analyses performed pair wise comparisons between genotype groups with similar class categories for chilling requirement estimation using peach known standards (Table A 2 3 ). Results Genetic and Cluster An alyses A total of 44 SSR primers were used for this study Only 33 SSRs yielded consistent results across specimens. These markers were readily transferable across species in Prunus as has been previously reported (Dirlewanger et al., 2004; Jung et al., 20 08). CPPCT015, PS1H3, CPPCT017 and CPPCT006 did not amplifify products for all samples BPPCT013, BPPCT023, BPPCT038, EPDC3832, CPDCT038, CPSCT008, and CPPCT029 yielded multiple products (multiple alleles ) and were not used in further analyses. 186

PAGE 187

A total of 762 alleles were observed across the specimens for all the primers An average of 48.17 genotypes were amplified per primer average number of alleles ( A ) of 21.77, an average effective number of alleles ( Ae ) of 10.58, observed heterozygosity ( Ho ) of 0.63, expected heterozygosity ( He ) of 0.80, Wrights fixation index ( F ) of 0.23 and an average potentially informative characters ( PIC ) of 0.87 (Table 52). BPPCT039 identified the highest number of genotypes compared to CPPCT019 with the lowest number, 77 and 3 respectively. The number of alleles per locus varied from 38 for CPPCT019 to 2 for BPPCT016 Similarly, the effective number of alleles ranged from 20.68 to 1.12 for BPPCT0 26 and BPPCT016 respectively. UDP98 412 had the highest observed heterezygosity of 0.91 and BPPCT016 had the lowest Ho of 0.01 Expected heterozygosity varied from 0.95 for BPPCT026, BPPCT039, CPPCT019, CPPCT022, to 0.11 for BPPCT016 respectively. Wrights fixation index ( F ) ranged from 0.90 for BPPCT016 to 0.01 for BPPCT030 BPPCT02 6 BPPCT039, and CPPCT019, had the highest PIC of 0.95 compared to BPPCT016 of 0.10. A total of 1826 alleles were observed across all the specimens (southeast ern US and core collection data) An average of 52.17 genotypes were amplified per primer an aver age number of alleles o f 24.57, an average effective number of alleles of 11.00, observed heterozygosity of 0.63, expected heterozygosity of 0.81, Wrights fixation index F of 0.25 and an average potentially informative characters PIC of 0.80 (Table A 2 4 ). An increment in the number of alleles was mainly due to the highly diverse plum core collection data. Summary statistics for the plum core collection using a total of 41 SSR primers yielded 471 alleles with an average of 9 genotypes, A of 10.70, Ae of 8.1 0, 187

PAGE 188

Ho of 0.58, He of 0.84, F of 0.31 and PIC of 0.82 (as previously reported on Chapter 3, Table 32). Genetic diversity estimates were similar among species with a n average of 8.57 for A 5.64 for Ae 0.61 for Ho 0.70 for He and 0.13 for F (Table 53). The majority of the species in the genus Prunus are outcrossing in nature, with most plums being self incompatible or partially self compatible (Keulemans, 1994). Self incompatibility encourages outbreeding in flowering plants (Newbigin et al. 1993). Res ults of Ho ranged between 0.54 for P. umbellatalike to 0.69 for P. americana. Results from species with small sample sizes ( P. americana, P. angustifolia, P. geniculata, and P. umbellatalike) were comparable to results for P. umbellata specimens. Prunus umbellatalike specimens yielded the highest level of F in comparison with the other species. This result was attributed to the reduced genetic base of the P. umbellatalike specimens a s they were collected in the same location. Prunus geniculata is a threatened, endemic, scrub species, native to the areas along the ridges of central Florida (Harper, 1911). Estimates of genetic diversity for P geniculata were intermediate to those of P. umbellata and P. umbellatalike specimens (Table 53). Prunus geniculata populations in Florida are currently located in limited numbers (small populations). This population size reduction could result on a genetic bot t leneck reducing its genetic diversity. The effect of inbreeding depression (produced by a genetic bottlenec k) was not shown in our results This probably is due to the fact that P geniculata plants are long lived and the effect of inbreeding depression has not been transfer to next generations. Prunus geniculata specimens showed the existence of a moderate gen etic variability. 188

PAGE 189

Neis genetic distance (Nei and Takezaki, 1983) was calculated in combination with Unweighted Pair Group Method with Arithmetic Mean (UPGMA) (Sokal and Michener, 1958) and Neighbor Joining (NJ) (Saitou and Nei, 1987) cluster analyses (Fig ures A 119 and A 1 20). The UPGMA cluster analysis based on Neis genetic distance represented best the grouping relationships among specimens and species groupings (Figure 5 3). UPGMA and NJ analysis differentiated all the genotypes Major clusters represented P. geniculata, P. americana, P. angustifolia, and P. umbellatalike species groups (Figures 5 3 and A 1 21 ). Core collection species, P. persica P. pumila P. fasciculata P. hortulana and P. mexicana, had the highest genetic diversity in comparison w ith the southeastern US specimens Prunus umbellatalike specimens were geographically closely located t o P. umbellata and P. geniculata samples in Clermont, FL They were separated from each other for approx. 2 km It was hypothesized that the P. umbellat a like specimens resulted from introgression between P. umbellata and P. geniculata species because of their unique phenotype (somewhat intermediate between both species ). However, P. umbellatalike specimens were clustered within the major grouping of P. umbellata (Figure 5 3). Structure results did not provide any evidence for introgression between these two groups. However, this lack of evidence of introgression could be due to the fact that the P. umbellatalike specimens were all collected in the same location, having a higher chance of being in Hardy Weinberg desiquilibrium (as shown in the structure result) 189

PAGE 190

A small group of P. umbellata specimens were clustered with the P.angustifolia group (Figure 5 3) These specimens were believed to have resulted from hybridization and introgression of genomic fragments from P. angustifolia. GPS data indicates that P. umbellata (Pumb) specimen s, PumbFL59 and PumbFL47 were closely located (about 10m and 1km, respectively ) to P. angustifolia (Pang) specimen PangFL60. Similarly, PumbFL32 and PumbFL33 specimens were collected about 4 km from PangFL30 and PangFL29. Voucher collection notes indicated that the specimens have an intermediate phenotype between P. umbellata and P. angustifolia species (collection descript or s and voucher information available at the Florida Museum of Natural History Gainesville, FL ). Associations between clade groups and traits of interest were studied using the available phenotypic information ( Table A 2 3 Figure A 122). Chilling range 1 (C hill1) comparisons illustrated the presence of two major groups within P. umbellata specimens: medium chill ( 400700 hours ) and medium high chill ( 700 1100 hours ) Similar r esults were obtained when comparing chilling range 2 (Chill2) genotype groups. Mant el test using g eographic and genetic distance for chill range 1 groups found possible signs of selection indicating the role of geographic isolation towards different chilling requirements (R2=0.19, P value=0.001) P run u s umbellata specimens constituted ex cellent candidates to be used in association studies to identify loci under selection. Neighbor net analysis using southeastern US and core collection specimens found similar gro ups to the cluster analyses (Figure 55). Prunus geniculata P. americana, P. angustifolia, and P. umbellatalike specimens formed major groups. 190

PAGE 191

Prunus munsoniana and P. maritima genotypes were found within P. angustifolia and P. umbellata groups, respectively. Similar results were previously reported using SSRs (Figure 5 3) and seq uence data (Chapter 3 and 4). Prunus umbellatalike specimens were group ed within P. umbellata (Figure 5 5). The neighbor net relationships for P. umbellatalike specimens showed a common base a s they were collected at the same location. These results were consistent with the level of F in comparison with other species (Table 53). Neighbor net relationships recovered a small group of P. umbellata specimens splitting up from the P.angustifolia group (Figure 55). These specimens were the same as the ones b elieved to have some type of introgression from P. angustifolia in the UPGMA c luster ( as previously described, Figure 53) Similar results were obtained when the core collection species were removed from the neighbor net analyses (Figure A 124). UPGMA and neighbor net results were high ly consistent across analyses. Population Structure Population stratification for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species were analyzed for two to ten K subgroups (Table A 2 5 ) with K=5 having the highest delta K ( ) value followed by K =7. These results were analyzed using the Evanno method (Evanno et al., 2005) as implemented in the Structure Harvester software (Earl and VonHoldt, 2012) (Table A 2 6 Figure 54). Population structure results recovered major species groups (Figure s 5 3 and A 1 25) These results were similar to cluster and neighbor net analyses. Population stratification for K =5 differentiated the following groups: 1) Core collection ( P. persica P. pumila P. fasciculata P. hortulana, P. mexicana, a nd P. maritima ), P. americana specimens, and few P. umbellata genotypes [blue bars]; 2) Prunus geniculata 191

PAGE 192

specimens [yellow bars], 3) Prunus angustifolia and some P. umbellata genotypes (believed introgressions with P. angustifolia) [purple bars]; 4) Prun us umbellata specimens [green bars]; and 5) Prunus umbellatalike specimens [orange bars]. Structure analysis with K=5 reported some individuals with i ntermixed composition between purple bars ( P. angustifolia) and blue bars ( P. umbellata) These P. umbell ata genotypes were the same that clustered with the P. angustifolia group as reported in the UPGMA and neighbor net analyses (Figure s 5 3, 5 5 A 124 and A 1 25) These r esults confirmed an introgression of P. umbellata and P. angustifolia species Other P umbellata genotypes had intermixed composition between green and blue bars (both representing P.umbellata) These genotypes were not classified as introgressions because previous analyses did not cluster them together with other southeastern US Prunus species. T hey represented the transition zone between two different P. umbellata groups as shown in structure analyses K=6 and K=7 (Figure 53) The UPGMA cluster analyses found association between genotypes chilling requirement (chill range 1) and the two different P. umbellata groups as identified on the structure analyses with K=6 and K=7 (Figure s 5 3 and A 122). These two P. umbellata groups could be used to study loci under selection associated with different chilling requirement categories P opulation structure analysis using additional K values recovered major species groups (similar results than K=5, Figure 53 ). Additional structure separation was identified for the core collection as observed with K=8 to K=10. The t wo major groups within P. umbellata specimens were identified with K=6 to K=10. Major structure results 192

PAGE 193

were consistent across analyses. No additional structure groups were observed within or between species groups. Detection of Loci Under Selection Analyses of pair wise comparisons withi n P. umbellata specimens for chilling requirement categories, chill range 1 and chill range 2, were performed to identify loci under selection associated with these phenotypic classes. Additional southeastern US and core collection species were not used for these comparisons due to the limited number of specimens. In addition structure results supported a major group for P. umbellata species (Figure 53). Mantel test using geographic and genetic distance for chill range 1 groups found possible signs of sel ection indicating the role of geographic isolation towards different chilling requirements (R2=0.19, P value=0.001). These results supported P. umbellata as an excellent candidate to be used to identify loci under selection associated with chilling requirement. Results from detection of loci under selection were complex (Table 5 4; Figures A 126 and A 127 ). Similar findings were observed when detecting loci under selection for Euroasian cattle Bos taurus (Li et al., 2010), oaks Quercus petraea and Q. robur (Goicoechea et al., 2012), and Japanese conifer Cryptomeria japonica (Tsumura et al., 2012). Bayesian calculations of loci under selection changed as the odds for the selection model increased (prior odds 10, 1 and 0.1). No loci under selection were detect ed when the prior odds was 10 (10 times more likely neutral than selection model ). The number of loci detected under selection increased as the prior of the odds changed from 1 to 0.1 ( Figure A 1 27, Table 5 4). The number of false positives increased with the prior of the odds changing in the same direction. 193

PAGE 194

Two detection methods for loci under selection were used to reduce the number of false positives. Loci that were identified across methods were reported under selection and compared to previously publis h ed genomic locations linked to chilling requirement in Prunus These comparisons were made to validate our results across tests. Comparisons of chilling requirement estimates using chill range 1 (Chill1) and chill range 2 (Chill2) categories (Table A 2 3 ) detected CPPCT026 (LG 1, 33.9 cM), CPPCT005 (LG 4, 10.4 cM), and BPPCT029 (LG 7, 29.6 cM) to be under selection. Several SSR primers were detected under balancing selection using the Fstoutlier detection method (Table 54; Figures A 126 and A 127 ). QTLs a ssociated with chilling requirement and flowering time (blooming time) have been previously reported in Prunus in LGs 1, 4, 5, 7, and 8 (Dirlewanger et al., 2012; Fan et al., 2009; Snchez Prez et al., 2012). Similarly, Blaker (2010) found two QTLs for radicle petrusion (RP) and rosette formation in peach in LG 4 (CPPCT005) and LG 6 (CPPCT008), respectively. Discussion North America is the center of diversity for Prunus species in section Prunocerasus These species constitute important resources for Prunus breeding and selection and can be used as parents, scions, and/or rootstocks (Beckman and Okie, 1994, Okie and Weinberger, 1996). Florida plum native species include d P. americana, P. angustifolia, P. geniculata and P. umbellata. These species harbor sev eral unique traits as they have grown adapted to Florida harsh environmental and disease pressure conditions. 194

PAGE 195

The main objective of this research was to study the genetic diversity of Florida plum native species P lant material consisted mainly of P. umbel lata due to its wider distribution range in Florida. Additional samples included P. americana, P. angustifolia, and P. geniculata. Core collection data was added to these analyses as a comparable measurement of the genetic diversity available in North Amer ica. The screening and transferability of several SSR primers from the Prunus Texas almond Earlygold peach (T E) reference map was possible for the southeastern US and core collection Prunus species due to the high level of synteny among Prunus gen omes (Dirlewanger et al., 2004; Jung et al., 2008). A total of 33 SSR primers (out of 44) were reproducible and consistent across specimens (Tables 52 and A 2 2 ). Several SSR primers had high PIC that can be used for fingerprinting and hybrid identificatio n. Southeastern US and core collection genetic diversity estimates per species ranged from 3.83 to 18.23 number of alleles per locus 0.54 to 0.69 observed heterozygosity, and 0.56 to 0.81 expected heterozygosity using 33 SSR primers (Table 5 3). Genetic d iversity estimates were similar between P. americana, P. angustifolia, P. geniculata, P. umbellata and P. umbellatalike specimens Similar results were reported in other Prunus species. Hovarth et al. (2008) studied P. cerasifera genetic diversity. They o bserved that 29 clones of P. cerasifera using seven SSR primers had an average number of alleles of 10.57, expected heterozygosity of 0.75, and observed heterozygosity of 0.65. Pairon et al. (2010) studied the genetic diversity of P. serotina in Europe and in North America A total of 442 and 321 genotypes respectively, were analyzed using eight SSR primers. An average of 7.74 alleles, observed heterozygosity of 0.72, and expected 195

PAGE 196

heterozygosity of 0.76 was reported for the North American populations. An average of 6.36 alleles, observed heterozygosity of 0.69, and expected heterozygosity of 0.70 was observed for the European populations. Similarly, P. davidiana genetic diversity in China was measured in more than 30 accessions using 23 SSR markers (Cheng et al., 2013). They reported an average of 3.72 alleles per locus, observed heterozygosity of 0.58 and expected heterozygosity of 0.47. These moderate genetic diversity estimates across diverse Prunus species are attributed to the presence of self incompat ibility or partially self compatibility systems (Keulemans, 1994 ; Sutherland et al., 2004). It has been previously described that self incompatibility encourages outbreeding in flowering plants (Newbigin et al. 1993). The use of the core collection data i n our comparisons showed an increase on the observed number of alleles (Tables 52 and A 2 4 ). This increment was due to the existence of rare alleles (58%) within the core collection species in comparison with the southeastern US species These estimates s howed the importance of the large genetic variation that could be introgressed through breeding and selection from these species into cultivated forms However, i t is important to mention that these estimates were obtained from a subset of the North Americ an Prunus species (only seven) Prunus geniculata genetic variability estimates were moderate (Table 53). Weekley et al. (2010) reported the existence of seedling recruitment failure in P. geniculata populations due to possible habitat fragmentation and f ire suppression. T he lost of P. geniculata original habitat and the existence of small populations comprising of fewer than 10 plants could result in high inbreeding depression (Cox et al., 2004; Turner et al., 2006; Weekley et al., 2010). Our result s observed the existence of moderate 196

PAGE 197

diversity within one P. geniculata population with similar values to other Prunus species in Florida. This genetic variability could be exploited through off site breeding nurseries (grafting genotypes representatives of diff erent populations) in order to increase seedling recruitment and pollinization. P runus umbellatalike specimens were collected nearby P. umbellata and P. geniculata genotypes in Clermont, FL. It was hypothesized that P. umbellata and P. geniculata genotypes hybridized resulting in the P. umbellatalike specimens. Inter specific hybrids have been recovered of open pollinated P. geniculata plants with other Prunus species at the Stone Fruit Breeding and Genetics Program (Chaparro and Chavez, personal communic ation). Human disturbance and other factors could have produced that these species hybridized freely. In addition, both populations are located in an urbanization development area that has changed throu ghout the years. UPGMA analyses recovered P. umbellatalike specimens clustered within the major group of P. umbellata (Figure 5 3). Signal s of i ntrogression from P. geniculata were not identified. Additional analyses using cpDNA regions could help to determine if these samples were indeed P. umbellata s pecim ens UPGMA represented the major species clusters (Figure 53). Prunus geniculata, P. americana, P. angustifolia and P. umbellatalike species were separated from each other. Cor e collection specimens were used as outgroup species for comparisons. Prunus munsoniana specimen was clustered within P. angustifolia groups (Figure 5 3) Rohrer et al. (2008) reported that P. munsoniana appeared to be a mixture of hybrids having P. angustifolia as one of the parents using LEAFY and s6pdh genes sequences. Similar r esults were obtained using SSRs, and dormancy and plant architecture related 197

PAGE 198

genes sequences for Prunus core collection species and whole phylogeny analyses ( previously reported on Chapter 3 and 4). Prunus maritima specimen was found clustered within P. u mbellata group (Figure 5 3). Analyses with SSR primers and s everal dormancy and plant architecture related genes recovered same relationships for Prunus core collection species and whole phylogeny analyses (previously reported on Chapter 3 and 4). Mowrey a nd Werner (1990) reported P. umbellata and P. maritima forming a clade using isozymes. Similarly, Shaw and Small (2005) found several P. maritima genotypes clustered together with several P. umbellata specimens using rpL16 intron. Prunus geniculata P. ame ricana, P. angustifolia, and P. umbellatalike specimens formed major groups using neighbor net analysis (Figure 55). These results were consistent with UPGMA cluster analysis. In addition, P munsoniana and P. maritima specimens were clustered within P. a ngustifolia and P. umbellata species groups. Neighbor net relationships and UPGMA analysis recovered a small group of P. umbellata specimens splitting up from the P.angustifolia group (Figures 53 and 55). These specimens were found to be introgressions from P. angustifolia and P. umbellata. Collection data supported this hypothesis as these genotypes were located between P. umbellata specimens and P. angustifolia specimens. Population structure results were consistent to the cluster and neighbor net anal yse s. Prunus geniculata, P. americana, P. angustifolia, and P. umbellatalike specimens formed major groups across simulations (K=5 to K=10) (Figures 53, 5 4, and A 1 25). Introgression of P. angustifolia and P. umbellata species were identified 198

PAGE 199

using stru cture analyses. These genotypes were the same as the ones clustered together to P. angustifolia groups using UPGMA and neighbor net analyses (Figures 53 and 55). In addition, P. umbellata specimens, PumbFL59 and PumbFL47, were closely located (about 10m and 1km, respectively) to P. angustifolia specimen PangFL60. Similarly, PumbFL32 and PumbFL33 specimens were collected about 4 km from PangFL30 and PangFL29. Both specimens had intermediate phenotype between P. umbellata and P. angustifolia species (Florid a Museum of Natural History, Gainesville, FL). Hybrids of Floridas native plum species with other Prunus species have been previously reported by Wight (1915). Wight (1915) reported that P. americana hybridized with P. angustifolia P. besseyi (currently known as P. pumila ) P. hortulana P. munsoniana, and P. simonnii Similarly, P. angustifolia hybridized freely with P. americana, P. triflora P. besseyi P. munsoniana, P. cerasifera, and P. orthosepala These structure results including UPGMA and neighbor net analyses, confirmed the presence of natural hybrids between P. angustifolia and P. umbellata species. Two major groups were identified within P. umbellata species using structure analyses (K=6 and K=7). These groups were closely associated with two groups of chilling requirement estimates (chill range 1) based on peach standards traced over the UPGMA tree (Figures 5 3 and A 122 ). Mantel test using geographic and genetic distance for chill range 1 groups found possible signs of selection indicating t he role of geographic isolation towards different chilling requirements (R2=0.19, P value=0.00 1). Similar associations between structure and UPGMA cluster results were observed in the 199

PAGE 200

University of Florida P. persica germplasm (Chapter 2). Two major peach g ermplasm groups were identified for melting and nonmelting flesh varieties (Figure 21). Pair wise comparisons between genotype groups of similar class categories for chill range 1 (Chill 1) and chill range 2 (Chill 2) requirements were used for detecti on of loci under selection in P. umbellata specimens based on UPGMA, neighbor net and structure analysis (Table 54). Resul ts for detection of outlier locus produced complex results as previously reported These results were probably due to the differences in the underline model used for each test (Table 54, Figures A 126 to A 127 ). Loci that were detected under selection for more than two methods were compared to previously known map genetic locations for chilling requirement in Prunus Synteny across sp ecies was previously described (Dirlewanger et al., 2004; Jung et al., 2008). Several loci were detected under selection with some of them closely linked to previously known genetic location. Comparisons of chilling requirement estimates using chill range 1 (Chill1) and chill range 2 (Chill2) categories (Table A 2 3 ) detected CPPCT026 (LG 1, 33.9 cM), CPPCT005 (LG 4, 10.4 cM), and BPPCT029 (LG 7, 29.6 cM) to be under selection (Table 54; Figures A 126 to A 127). QTLs associated with chilling requirement and flowering time (blooming time) have been previously reported in Prunus in LGs 1, 4, 5, 7, and 8 (Dirlewanger et al., 2012; Fan et al., 2009; Snchez Prez et al., 2012). Blaker (2010) found two QTLs for radicle petrusion (RP) and rosette formation in peac h in LG 4 (CPPCT005) and LG 6 (CPPCT008), respectively. Conclusions The use of SSR primers to measure the genetic diversity and population structure of the southeastern US species was possible due to the high level of synteny 200

PAGE 201

across Prunus species (Dirlewa nger et al., 2004; Jung et al., 2008) Major species groups were identified using UPGMA, neighbor net, and structure analyses. Hybrids between P. angustifolia and P. umbellata species were identified. Prunus umbellata specimens constituted a good candidate for association studies as demonstrated when comparing different chilling requirement estimates. These loci have been already reported based on QTL analyses for different Prunus species. Future studies will include the use of other molecular techniques su ch as genotypeby sequencing (GBS) to generate high density marker information across the genome of these species that could help us identify regions associated with other phenotypes of interest. 201

PAGE 202

Table 51. List of the native plums (section Prunocerasus ) col lected in the southeastern US and used for this study. Vou z Col y Genus Species ID x State Latitud e Longitud e 26 Prunus americana PameGA26 GA 30 44.310 N 84 27.565 W 76 Prunus americana PameGA76 GA 32 39.366 N 83 42.657 W 80 Prunus americana Pam eGA80 GA 32 29.186 N 83 35.194 W 237 Prunus americana PameGA237 GA 30 44.260 N 84 27.512 W 7 Prunus angustifolia PangFl07 FL 29 38.882 N 82 29.756 W 25 Prunus angustifolia PangGA25 GA 30 43.451 N 84 25.695 W 90 Prunus angustifolia PangGA90 GA 162 Prunus angustifolia PangFl162 FL 30 3.947 N 82 37.687 W 147 Prunus geniculata PgenFl147 FL 28 32.703 N 81 41.869 W 150 Prunus geniculata PgenFl150 FL 28 32.706 N 81 41.888 W 182 Prunus geniculata PgenFl182 FL 28 32.712 N 81 41.859 W 183 Pr unus geniculata PgenFl183 FL 28 32.716 N 81 41.863 W 184 Prunus geniculata PgenFl184 FL 28 32.719 N 81 41.875 W 185 Prunus geniculata PgenFl185 FL 28 32.714 N 81 41.889 W 186 Prunus geniculata PgenFl186 FL 28 32.707 N 81 41.896 W 187 Prunus genicul ata PgenFl187 FL 28 32.716 N 81 41.850 W 234 Prunus geniculata PgenFL234 FL 27 55.686 N 81 34.067 W 1 Prunus umbellata PumbFl01 FL 29 36.690 N 82 21.271 W 2 Prunus umbellata PumbFl02 FL 29 36.666 N 82 21.257 W 3 Prunus umbellata PumbFl03 FL 29 38.082 N 82 21.321 W 4 Prunus umbellata PumbFl04 FL 29 38.071 N 82 21.316 W 5 Prunus umbellata PumbFl05 FL 29 37.104 N 82 20.426 W 6 Prunus umbellata PumbFl06 FL 29 31.245 N 82 18.005 W 8 Prunus umbellata PumbFl08 FL 29 31.191 N 82 18.739 W 9 Prunus umbellata PumbFl09 FL 29 31.183 N 82 18.817 W 10 Prunus umbellata PumbFl10 FL 29 31.020 N 82 18.624 W 11 Prunus umbellata PumbFl11 FL 29 31.266 N 82 18.595 W 13 Prunus umbellata PumbGA13 GA 30 52.768 N 84 33.876 W 14 Prunus umb ellata PumbGA14 GA 30 52.768 N 84 33.876 W 17 Prunus umbellata PumbFl17 FL 29 01.771 N 82 09.549 W 18 Prunus umbellata PumbFl18 FL 28 45.237 N 82 05.840 W 19 Prunus umbellata PumbFl19 FL 28 39.404 N 82 03.279 W 20 Prunus umbellata PumbFl20 FL 28 39.415 N 82 03.280 W 21 Prunus umbellata PumbFl21 FL 28 18.163 N 82 08.330 W 23 Prunus umbellata PumbFl23 FL 30 21.178 N 83 09.957 W 24 Prunus umbellata PumbFl24 FL 30 36.013 N 84 25.068 W 27 Prunus umbellata PumbFl27 FL 30 34.705 N 84 21.526 W 28 Prunus umbellata PumbFl28 FL 30 29.296 N 84 19.133 W 32 Prunus umbellata PumbFl32 FL 30 29.330 N 83 30.021 W 33 Prunus umbellata PumbFl33 FL 30 28.156 N 83 24.614 W 34 Prunus umbellata PumbFl34 FL 30 23.505 N 83 11.592 W 3 5 Prunus umbellata PumbFl35 FL 30 23.505 N 83 11.592 W 36 Prunus umbellata PumbFl36 FL 27 59.255 N 82 19.553 W 37 Prunus umbellata PumbFl37 FL 28 04.565 N 82 21.177 W 202

PAGE 203

Table 5 1. Continued. Vou z Col x Genus Species ID x State Latitude Longitude 38 Prunus umbellata PumbFl38 FL 28 25.191 N 82 16.707 W 39 Prunus umbellata PumbFl39 FL 28 24.809 N 82 14.515 W 40 Prunus umbellata PumbFl40 FL 28 23.413 N 82 13.255 W 41 Prunus umbellata PumbFl41 FL 28 28.530 N 82 15.417 W 42 Prunus umbell ata PumbFl42 FL 28 35.077 N 82 12.621 W 43 Prunus umbellata PumbFl43 FL 28 54.114 N 82 06.243 W 44 Prunus umbellata PumbFl44 FL 28 59.051 N 82 08.241 W 45 Prunus umbellata PumbFl45 FL 29 00.482 N 82 09.095 W 46 Prunus umbellata PumbFl46 FL 29 06.315 N 82 11.070 W 47 Prunus umbellata PumbFl47 FL 30 22.203 N 83 14.816 W 49 Prunus umbellata PumbGA49 GA 30 45.922 N 84 28.905 W 50 Prunus umbellata PumbGA50 GA 30 45.748 N 84 28.717 W 51 Prunus umbellata PumbGA51 GA 30 45.729 N 84 28.656 W 52 Prunus umbellata PumbGA52 GA 30 45.972 N 84 28.913 W 53 Prunus umbellata PumbGA53 GA 30 45.981 N 84 28.922 W 54 Prunus umbellata PumbGA54 GA 30 45.986 N 84 28.926 W 55 Prunus umbellata PumbGA55 GA 30 45.740 N 84 28.804 W 58 P runus umbellata PumbFl58 FL 30 33.039 N 84 22.883 W 59 Prunus umbellata PumbFl59 FL 30 22.566 N 83 16.991 W 74 Prunus umbellata PumbFl74 FL 30 23.222 N 82 52.401 W 75 Prunus umbellata PumbFl75 FL 30 29.414 N 83 1.123 W 94 Prunus umbellata P umbFl94 FL 27 30.024 N 81 25.513 W 95 Prunus umbellata PumbFl95 FL 30 28.428 N 84 17.143 W 96 Prunus umbellata PumbGA96 GA 98 Prunus umbellata PumbFl98 FL 30 24.572 N 83 58.010 W 104 Prunus umbellata PumbFl104 FL 30 3.173 N 83 8.949 W 1 05 Prunus umbellata PumbFl105 FL 29 59.598 N 83 0.648 W 106 Prunus umbellata PumbFl106 FL 29 57.136 N 82 52.952 W 107 Prunus umbellata PumbFl107 FL 29 55.615 N 82 43.269 W 111 Prunus umbellata PumbFl111 FL 29 33.417 N 82 29.262 W 124 Prunus umbellata PumbFl124 FL 29 37.435 N 81 55.040 W 125 Prunus umbellata PumbFl125 FL 29 36.575 N 82 0.687 W 126 Prunus umbellata PumbFL126 FL 29 35.924 N 82 7.054 W 127 Prunus umbellata PumbFl127 FL 29 37.290 N 82 15.408 W 142 Prunus umbellata PumbFl142 FL 29 38.395 N 82 25.372 W 143 Prunus umbellata PumbFl143 FL 29 50.147 N 82 15.809 W 144 Prunus umbellata PumbFl144 FL 28 45.292 N 82 6.134 W 152 Prunus umbellata PumbFl152 FL 28 37.523 N 81 44.640 W 247 Prunus umbellata PumbGA247 GA 30 45.748 N 84 28.717 W 109 Prunus umbellata x angustifolia PumbxangFL109 FL 29 37.939 N 82 36.312 W 163 Prunus umbellata like P run FL163 FL 28 32.842 N 81 43.265 W 164 Prunus umbellata like P run FL164 FL 28 32.842 N 81 43.265 W 165 Prunus umbel lata like P run FL165 FL 28 32.842 N 81 43.256 W 166 Prunus umbellata like P run FL166 FL 28 32.842 N 81 43.242 W 167 Prunus umbellata like P run FL167 FL 28 32.839 N 81 43.227 W 203

PAGE 204

Table 5 1. Continued. Vou z Col y Genus Species ID x State Latitude Longitude 168 Prunus umbellata like P run FL168 FL 28 32.834 N 81 43.225 W 169 Prunus umbellata like P run bFL169 FL 28 32.831 N 81 43.234 W 170 Prunus umbellata like P run FL170 FL 28 32.828 N 81 43.244 W 171 Prunus umbellata like P run FL171 FL 28 32.828 N 81 43.254 W 172 Prunus umbellata like P run FL172 FL 28 32.829 N 81 43.257 W 173 Prunus umbellata like P run FL173 FL 28 32.827 N 81 43.267 W zVou = voucher number and source. yCol = collection number. xID = first letter represented the genus ( Prunus =P), next thre e letters represented the species ( americana=ame, etc.), and the following letters represented the state of origin and collection number (Florida collection 12=FL12, etc.). 204

PAGE 205

Table 52. Summary statistics of 33 simple sequence repeat (SSR) markers for 101 g enotypes of the North American Plums germplasmz available in the Stone Fruit Genetics and Breeding Program at University of Florida, Gainesville, FL. Marker Genotype No. A y Ae Ho He F PIC BPPCT006A 51 18 11.91 0.68 0.92 0.25 0.91 BPPCT006B 60 30 16.66 0. 64 0.94 0.32 0.94 BPPCT008A 18 10 2.35 0.36 0.58 0.37 0.55 BPPCT008B 59 24 13.09 0.73 0.92 0.21 0.92 BPPCT014 69 31 15.07 0.74 0.93 0.21 0.93 BPPCT016 3 x 2 1.12 0.01 0.11 0.90 0.10 BPPCT017 71 24 15.22 0.81 0.93 0.14 0.93 BPPCT025A 65 32 12.51 0.75 0 .92 0.19 0.92 BPPCT026 75 32 20.68 0.77 0.95 0.19 0.95 BPPCT027 14 10 1.99 0.23 0.50 0.53 0.47 BPPCT028 64 22 15.38 0.74 0.93 0.20 0.93 BPPCT029 17 9 2.92 0.39 0.66 0.41 0.61 BPPCT030 7 6 1.27 0.21 0.21 0.01 0.21 BPPCT039 77 35 20.22 0.81 0.95 0.15 0 .95 CPDCT023 72 31 16.00 0.71 0.94 0.24 0.93 CPDCT025 73 29 16.85 0.82 0.94 0.13 0.94 CPDCT027 72 36 17.60 0.83 0.94 0.12 0.94 CPPCT005 71 33 14.78 0.80 0.93 0.14 0.93 CPPCT008 43 25 7.88 0.41 0.87 0.53 0.86 CPPCT019 72 38 19.51 0.77 0.95 0.19 0.95 CPPCT022 72 37 18.38 0.76 0.95 0.20 0.94 CPPCT026 70 29 15.26 0.82 0.93 0.12 0.93 CPPCT033 57 23 11.19 0.78 0.91 0.15 0.90 CPSCT004 16 8 3.22 0.63 0.69 0.09 0.65 CPSCT034 28 12 4.18 0.68 0.76 0.11 0.73 EPDCU3117 21 11 4.67 0.47 0.79 0.40 0.76 EPDCU33 92 73 27 14.41 0.85 0.93 0.09 0.93 EPPISF002 49 18 9.80 0.76 0.90 0.16 0.89 EPPISF032 38 18 4.85 0.69 0.79 0.13 0.78 PMS2 64 29 14.91 0.86 0.93 0.08 0.93 UDP96 001 18 11 2.93 0.55 0.66 0.17 0.62 UDP96 003 9 7 1.43 0.17 0.30 0.44 0.30 UDP96 019 15 8 3 .00 0.48 0.67 0.27 0.63 UDP98 025 27 15 3.71 0.53 0.73 0.28 0.69 UDP98 412 76 32 15.25 0.91 0.93 0.03 0.93 Average 48.17 21.77 10.58 0.63 0.80 0.23 0. 79 zGermplasm available include d P. americana, P. angustifolia, P. geniculata, and P. umbellata specim ens. yA = number of observed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = polymorphism information content. xNumbers in bold represent highest and lowest values for each variable. 205

PAGE 206

Table 53. Average statistics of 33 simple sequence repeat (SSR) markers by species in the southeastern US and core collection. Subset Sample No. A z Ae Ho He F P. americana 5 5.40 4.34 0.69 0.69 0.02 P. ang ustifolia 4 3.83 3.17 0.59 0.56 0.06 P. geniculata 10 7.60 4.90 0.56 0.72 0.26 P. umbellata like 11 9.63 6.97 0.54 0.81 0.36 P. umbellata y 69 18.23 9.42 0.64 0.77 0.16 Others x 7 6.74 5.06 0.66 0.67 0.02 All 106 8.57 5.64 0.61 0.70 0.13 zA = number o f observed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, and F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )]. yOne specimen of P. umbellata x angustifolia was included in this group. yPrunus hor tulana, P. mari tima P. mexicana, P. munsoniana, P. persica P. pumila and P. fasciculata species (one genotype per specimen). 206

PAGE 207

Table 54. Outlier locus detection analyses for pair wise comparison of phenotypic characteristics classes of 33 simple sequenc e repeat (SSR) markers for 99 Prunus specimens representatives of the southestern US species. Chromosome Location (cM) Marker LOSITANz BayeScan y Prior = 10 Prior = 1 Prior = 0.1 1 33.9 CPPCT026 x,w Chill1 bs Chill2 out Chill2 out Chill1 out Chill2 out 1 47.3 BPPCT027 1 55.2 BPPCT016 Chill1 out Chill2 out 1 77.4 BPPCT028 Chill1 bs Chill2 bs Chill1 out 2 9.6 UDP98 025 Chill1 bs Chill2 bs Chill2 out 2 38 BPPCT030 2 48.6 CPSCT034 Chill1 bs Chill2 bs Chill1 out Chill2 out 3 18 BPPCT039 Chill1 bs Chill2 bs 3 36.4 CPDCT025 Chill1 bs Chill2 bs Chill2 out 3 46.4 CPDCT027 Chill1 bs 4 10.4 CPPCT005 Chill2 out Chill2 out 4 28.3 UDP96 003 Chill2 out 4 52.7 EPPISF032 Chill1 out Chill2 out 5 5.2 BPPCT026 Chill1 out Chill2 out 5 20.1 BP PCT017 Chill1 out Chill2 out 5 44 BPPCT014 Chill1 out Chill2 out 6 8.7 CPPCT008 Chill2 out Chill1 out Chill2 out 6 17.5 UDP96 001 Chill1 bs Chill2 bs 6 30.1 BPPCT008 A 6 30.1 BPPCT008B Chill1 bs 6 41 EPPISF002 Chill1 out Chill2 out 6 56 .4 BPPCT025 A Chill1 bs Chill2 bs Chill2 out Chill1 out Chill2 out 6 72 UDP98 412 Chill1 bs Chill2 bs Chill1 out 7 9.5 CPSCT004 Chill2 out 7 18.6 CPPCT022 Chill2 bs 7 29.6 BPPCT029 Chill1 out Chill2 out 7 38.9 CPPCT033 Chill1 bs Chill2 bs Chill1 out Chill2 out 7 47.8 PMS2 Chill2 bs Chill1 out Chill2 out 7 64.7 EPDCU3392 Chill1 bs Chill2 bs Chill2 out 8 7.8 CPPCT019 Chill1 bs Chill2 bs Chill2 out Chill1 out Chill2 out 8 14.1 BPPCT006 A Chill1 bs 8 14.1 BPPCT006B Chill1 bs Chill2 bs Chill1 out 8 20 .8 UDP96 019 Chill2 out 8 42.6 CPDCT023 Chill1 bs Chill2 bs 8 54.7 EPDCU3117 Chill1 out Chill2 out zLOSITAN workbench, 50000 simulations, a neutral mean Fst calculation, a force mean Fst calculation, confidence interval 95%, false discovery ra te of 1%, infinite allele mutation (IAM) model. yBayeScan settings: sample size of 5000, thinning interval of 10, pilot runs of 20, pilot run length of 5000, and additional burnin of 50000. Prior of the odds for the neutral model: Prior 10 = 10 times more likely neutral model than selection, Prior 1 = neutral and selection models equally likely, and Prior 0.1 = 10 times more likely selection than neutral model. xChill1 = range chilling hour requirement: 1) Low <400 chill hours, 2) Medium 400700 chill hour s 3) Medium high 7001100, and 4) High >1100 Chill2 = range chilling hour requirement: 201300 chill hours, 301400 chill hours, 401500 chill hours, 501600 chill hours, 601700 chill hours, 701800 chill hours, 801900 chill hours, and 11001200 chill hours wNames in bold represent concordant results across at least two different outlier locus detection methods. out= outlier bs= balancing selection. 207

PAGE 208

Figure 5 1. Floridas map of distribution for A) Prunus americana Marsh., B) P. angustifolia Marsh., C) P. umbellata Elliot, and D) P. geniculata Harper. Green areas represent reported areas where these species had been found (USDA PLANTS, 2010). A B C D 208

PAGE 209

Figure 52. Specimes geographical distribution for population structure and genetic diversity studi es of P. americana (orange), P. angustifolia (blue), P. geniculata (red), P. umbellata (yellow), and P. umbellatalike (purple). 209

PAGE 210

Figure 53. Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) and population structure results for k=2 to k=10 of 33 simple sequence repeat (SSR) markers for 106 Prunus specimen representatives of the Prunus core collection and southeastern US species Species clades are described on the right hand s ide of the cladogram. Population structure results are located on the right hand side of the cladogram with different colors representing a different subgroup. 210

PAGE 211

Core P. geniculata P. americana P. angustifolia Cor e P. umbellata Core P. umbellata P. umbellata like P. umbellata K=2 K=3 K=4 K=5 K=6 K=7 K=8 K=9 K=10 211

PAGE 212

Figure 54. Analysis of the populati on structure results for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species using the Evanno method (Evanno et al., 2005) implemented by the Structure Harvester software (Earl and VonHoldt, 2012). A) Second order change in the log likelihood delta K ( ). B) Rate of change of the log likelihood distribution (mean). C) Absolute value of the second order change in the log likelihood distribution (mean). D) The average log likelihood and the standard error of four re ps per run. A B C D 212

PAGE 213

Figure 55. A nonstandarized neighbor net network of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. Colors represent pop ulation structure results wit h K =5. 213

PAGE 214

P. americana P. geniculata Core collection P. angustifolia Core collection P. umbellata like P. umbellata 214

PAGE 215

CHAPTER 6 CONCLUSIONS A BREEDER S PERSPECTIVE The study of the subgenus Prunus (in particular section Prunocerasus ) constitutes a unique and important research tool to support and preserve these species as importan t genetic resources (gene pool) of unique traits It could be used in the breeding of improved stone fruit cultivars The c ollection and identification of the plant specimens from the wild to confirm their species relationships (relatedness, evolution, and character/trait diversification) archive tissue samples in the herbarium and protect living accessions in the USDA Prunus repository in Davis, California, constituted important objectives. The stone fruit breeding and genetics program at the University of Florida was initiated in 1953. The objective was to develop early ripening stone fruit cultivars with high quality, adaptation to a subtropical tropical climate low chilling requirements, and the ability to withstand high disease pressure (Sherman et al 1984). The stone fruit breeding and genetics program has mainly focused on breeding peaches and has dedicated a limited amount of effort toward development of plums apricots and cherries. This study constitutes the baseline for evaluating the use of native wild species in breeding and selection to introgress novel traits into commercial stone fruit germplasm The four collections trips that were undertaken provided first hand knowledge of the climate and soil types where the species typically grew. The information on soil type, soil drainage, and pH is invaluable information for the breeding of rootstocks adapted to specific growing conditions. Species adapted to extremely low chilling environments such as P. geniculata and the southern most populations of P. umbellata provide sources of low dormancy requirement genes for breeding subtropical 215

PAGE 216

rootstocks and scions. Species adapted to low rainfall locations and growing under alkaline vs. acidic soil conditions were also observed, again prodiving raw materi al for the breeding of improved rootstocks. It is one of our long term goals to preserve and breed plum germplasm adapted to different growing conditions and to use this material to develop rootstocks for use in home and commercial production. The use of w ild Prunus species as a genetic resource to increase UFs peach genetic diversity and to introgress traits of importance that are not available within the peach gene pool will be a long term goal. The phylogenetic framework generated by this research will be used to identify species that may serve as bridging species for the transfer of novel traits such as brown rot resistance from native plums to peach. Other traits that would have immediate economic impact include flooding tolerance and resistance to peach tree borer. First generation hybrids between several of the species have been generated. In this context, control cross pollinations between two plums species of interest proved to be difficult when hand emasculation was performed ( unpublished results ) The small size of the flowers makes emasculation difficult and the manipulation tend s to result in flower abortion after emmasculation. However, open pollinations from bees and flies pro duced several fruit. Markers allowed us to identify and classify diff erent hybrids from open pollinated seed derived from different species/genotypes around the area based on the female parent genotypic information using SSR markers The value of this SSR database was proven beneficial when identifying hybrids and their par ent al genotypes. A large collection of North American plum ( Prunus spp.) germplasm was collected across the US (Table A 1 8 ). S tates that harbored multiple species were 216

PAGE 217

selected to maximize collection efficiency. Out of those, approximately 74 accessions of open pollinated seeds were submitted to the USDA National Clonal Germplasm Repository in Davis, CA (DPRU), for preservation and future availability of these species as genetic resources (Table A 19 ). Both living and herbarium collections w ere created as a unique database that could be accessed by scientist s to identify unique characteristics and phenotypes of interest In addition, genomic information can be obtained using DNA from several of thes e specimens representing a subset of the variation available within the North American plums. It was one of our objectives to make available different levels of information for each accession that will be beneficial for future research and conservation. The identification and use of additional genomic regions provi ded the greatest number of characters, variability, and improved the phylogenetic signal at the low level in Prunus section Prunocerasus relationships Identification of the number of potentially informative characters ( PICs ) proved to provide a good measure of how informative a region will be, if complemented with the phylogenetic analysis for that region. The use of novel regions permitted the identification of common alleles between species. Furthermore, t hese sequences could be used for linkage mapping in segregating populations between species to identify QTLs as sociated with dormancy response and plant architecture. Similarly, the se genomic regions were used to understand the phylogenetic relationships within section Prunocerasus This provided us with a framework for our future breeding efforts to introgress traits from wild to cultivated species The phylogenetic relationships within the North American plums ( Prunus spp.) help ed us 217

PAGE 218

identify interspecific hybrids. The ability to hybridize is not always representative of species relationships (biological species versus phylogenetic species concept). For example, Watkins (1976) described the existence of a Microcerasus bridge. He described that the members of section Microcerasus (in our case adding als o P. pumila ) hybridized with both the subgenus Amygdalus Prunus group and subgenus Cerasus and that genetic transfer was possible between these groups. Hybrids between P. pumila and species within section Prunocerasus have been previously reported (Wight, 1915). Those hybrids represent an avenue for the transfer of genes from the North American plums into cultivated Amygdalus species (such as peach) for the development of rootstock and scion varieties. Similarly, this bridge could serve in the other direct ion for the development of new crops or fruit types based on species from the section Prunocerasus The use of nuclear genome regions allowed us to identify incongruences that appear to represent ancestral hybridization events that obscured the phylogeneti c relationships. One of the main incongruences was between P. geniculata, P. maritima and P. texana. The ancestral form of Prunus geniculata, as described by Germain Aubrey (2012) western hypothesis and confirmed with our results, probably arrived in Flor ida between 5 and 2 Ma. Then, about 2 Ma a semi arid biota extended from Florida into the western US (Myers and Myers, 1990). Some of this biota became isolated due to an increase of humidity in the Gulf of Mexico creating isolated xeric habitats in Florida in the mid Pliocene (Webb and Myers, 1990). This isolation provided the mechanism for the differentiation of P. geniculata and P. texana (sister species). In addition, the presence of common chloroplast haplotype appears to indicate that hybridization 218

PAGE 219

oc curred between P. geniculata and P. maritima This probably occurred i n the last glaciation event. By 10 000 years ago, at the end of the last glaciation, P. maritima (or its ancestral form) migrated from Northeastern US into north Florida ahead of the glacial sheet. In Florida, this species may have encountered P. geniculata (or its ancestral form), allowing Prunus geniculata and P. maritima to hybridize. In summary, the use of nuclear markers proved beneficial to the study and clarification of the North A merican plum species relationships. We propose the existence of t hree major clades within section Prunocerasus : the Sand clade, the American clade, and the Chickasaw clade (Figure 48). The Sand clade was constituted by P. texana and P. geniculata The Ame rican clade was composed by P. americana, P. alleghaniensis P. mexicana, P. hortulana, and P. rivularis The Chickasaw clade was formed by P. angustifolia, P. gracilis P. maritima P. munsoniana P. nigra, and P. umbellata. The species that are best adap ted to a monsoonal climate and high disease and pest pressure are the best species to start a plum breeding program in Florida. Prunus umbellata, P. americana, P. geniculata, and P. angustifolia are species native to these conditions. However, only P. umbellata has a wide enough geographic distribution range to allow selection for the different chilling requirement zones of the state. Chilling requirement constitutes the single most important selection criteria for adaptation in our breeding program. The ex istence of several wild plum species adapted to Florida creates a unique opportunity for the study of allelic variation associated with important economic traits that can then be targeted and rapidly transferred to domesticated plum using molecular markers 219

PAGE 220

Major species groups were identified using UPGMA, neighbor net, and structure analyses using SSR markers for the Florida species Hybrids between P. angustifolia and P. umbellata species were identified. Prunus umbellata specimens constituted a good candidate for association studies as demonstrated when comparing different chilling requirement estimates P runus umbellata is a good candidate for plum breeding and selection as shown by this species adaptation for several limiting traits resistance to oak r oot rot ( Armillaria spp.), brown rot ( Monilinia laxa ) and other diseases, that could limit plum production in the southeastern US. 220

PAGE 221

APPENDIX A COMPLEMENTARY TABLES AND FIGURES Table A 1. Phenotypic characteristics of the peach germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program. Phenotypic data obtained from the University of Florida evaluation archives and/or Okie (1998). SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu NormPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden Aneuploid_nect 1 2 2 2 1 1 1 2 300 1 3 2000s AP0008w 2 1 1 1 1 1 1 1 300 1 3 2000s AP0030wbs 2 1 1 1 1 1 1 1 450 2 4 2000s AP03 22 2 1 1 2 1 1 1 1 400 ? 4 2000s AP0411 2 1 2 2 1 1 1 1 350 1 3 2000s AP0415 2 1 1 2 1 1 1 1 350 1 3 2000s AP05 18ws_Hap 1 1 2 2 1 1 1 1 ? ? ? 2000s AP06 10n 2 2 1 2 1 1 1 1 450 ? 4 2000s Attapulgus_white 2 1 1 1 1 1 1 1 450 2 4 ? Aztec_Gold 1 1 2 2 2 1 1 2 275 1 3 1980s Candor 4 1 1 2 1 1 1 2 950 3 7 1960s Carolina_Gold 4 1 1 2 1 1 1 2 1050 3 8 1990s China_Pearl 4 1 1 1 1 1 1 1 1110 3 9 1980s Contender 4 1 1 2 1 1 1 2 1050 3 8 1980s Earligrande 1 1 1 2 1 1 2 2 200 1 2 1970s Earlyamber 1 1 1 2 1 1 3 2 3 50 1 3 1950s FKxNP 6 1 1 1 2 1 ? 1 250 1 2 ? Flordabelle 1 1 1 2 1 1 1 1 150 1 1 1960s Flordabest 1 1 1 2 1 1 1 1 250 1 2 2000s Flordacrest 1 1 1 2 1 1 2 1 425 2 4 1980s Flordadawn 1 1 1 2 1 1 1 1 300 1 3 1970s Flordaglo 1 1 1 1 1 1 1 1 200 1 2 1980s Flordagold 1 1 1 2 1 1 2 1 325 1 3 1960s Flordaguard 1 1 1 2 1 1 1 1 300 1 3 1980s 221

PAGE 222

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu NormPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden Flordaking 1 1 1 2 1 1 2 2 400 2 4 1960s Flordaprince 1 1 1 2 1 1 1 1 150 1 1 1970s Flordastar 1 1 1 2 1 1 2 1 225 1 2 1970s Gulfcrest 2 1 2 2 1 1 2 2 525 2 5 1990s Gulfcrimson 2 1 2 2 1 1 1 1 400 2 4 2000s Gulfking 2 1 2 2 1 1 1 1 350 1 3 1990s Gulfprince 2 1 2 2 1 1 1 1 450 2 4 1990s Late_Arkansas 3 1 1 ? ? ? ? ? 950 3 7 ? Legend 4 1 1 2 1 1 1 2 850 3 6 1980s Martha_Jane 1 2 1 2 1 1 1 1 475 2 5 1980s N onpareil 6 1 1 1 2 1 2 1 350 1 3 ? Okinawa 5 1 1 1 1 1 1 1 100 1 1 1940s OkXPK_hap 5 ? ? ? ? ? ? ? ? ? ? ? Oro_Ac 1 1 2 2 2 1 1 1 250 1 2 1980s P_kansuensis 7 1 1 1 2 1 1 1 450 2 4 ? Rayon 1 1 1 2 1 1 1 1 200 1 2 1970s Red_Ceylon 5 1 1 1 1 1 1 1 50 1 1 1890s Strickland 5 1 1 ? ? ? ? ? ? ? ? ? Sun_Fre 1 2 1 2 1 1 1 1 500 2 5 1970s Sunbes t 1 2 1 2 1 1 1 2 350 1 3 1990s Sunblaze 1 2 1 2 1 1 1 2 250 1 2 1970s Suncoast 1 2 1 2 1 1 1 2 400 2 4 1980s Sunhome 1 2 1 2 1 1 1 1 300 1 3 1970s Sunlite 1 2 1 2 1 1 1 1 450 2 4 1960s Sunmist 1 2 1 1 1 1 2 1 300 1 3 1980s Sunraycer 1 2 1 2 1 1 1 2 275 1 3 1980s Sunred 1 2 1 2 1 1 1 1 250 1 2 1960s Sunsplash 1 2 1 2 1 1 2 1 400 2 4 1980s Tropic_Snow 1 1 1 1 1 1 1 1 200 1 2 1980s 222

PAGE 223

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu NormPeentot RenGloEglas S howNonshor Chillq Chillrange1p Chillrange2o Decaden Tropic_Sweet 1 1 1 2 1 1 1 1 175 1 2 1970s Tropicbeauty 1 1 1 2 1 1 1 1 150 1 1 1980s UF2000 1 1 2 2 1 1 1 1 300 1 3 1990s UFBeauty 1 1 2 2 1 1 2 1 150 1 1 1990s UFBlaze 1 1 2 2 1 1 1 1 250 1 2 1990s UFGold 1 1 2 2 1 1 1 1 400 2 4 1990s UFO 1 1 2 2 1 2 1 2 250 1 2 1990s UFQueen 1 2 2 2 1 1 1 2 300 1 3 1990s UFRoyal 1 2 2 2 1 1 1 1 250 1 2 1990s UFSharp 1 1 2 2 1 1 2 1 350 1 3 1990s UFSun 1 2 2 2 1 1 1 1 150 1 1 1990s 9730c_UFOne 1 1 2 2 1 1 2 1 100 1 1 1990s 9742c_UFFlair 1 1 2 2 1 1 1 2 150 1 1 1990s 9747c 1 1 1 2 1 1 1 2 150 1 1 1990s 9748c 1 1 2 2 1 1 1 2 250 1 2 1990s 9903cn 1 2 2 2 1 1 1 1 250 1 2 1990s 0804 1 1 1 2 1 1 1 1 250 1 2 2000s 0809 1 1 1 2 1 1 1 ? ? ? ? 2000s 0912 1 1 1 2 1 1 1 1 300 1 3 2000s 0915 1 1 1 2 1 1 1 2 300 1 3 2000s 0102c 1 1 2 2 1 1 2 2 250 1 2 2000s 0201c_hap 1 1 2 2 1 1 1 1 350 1 3 2000s 0202cn 1 2 2 2 1 1 1 1 250 1 2 2000s 0207cw 1 1 2 1 1 1 1 1 450 2 4 2000s 0209cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0209cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0211CW 1 1 2 1 1 1 1 1 350 1 3 2000s 0213c 1 1 2 2 1 1 2 1 250 1 2 2000s 0301orn 1 2 1 2 1 1 1 1 250 1 2 2000s 223

PAGE 224

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu Nor mPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden 0309c 1 1 2 2 1 1 1 1 350 1 3 2000s 0316c 1 1 2 2 1 1 1 2 250 1 2 2000s 0401orn 1 2 1 2 1 1 1 1 350 1 3 2000s 0404cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0406cn 1 2 2 2 1 1 2 1 250 1 2 2000s 0501cnh 1 2 2 2 2 1 1 2 350 1 3 2000s 0502cnh 1 2 2 2 2 1 1 1 200 1 2 2000s 0512c 1 1 2 2 1 1 1 1 250 1 2 2000s 0601dw 1 2 1 2 1 1 1 1 350 1 3 2000s 0602dw 1 2 1 2 1 1 1 1 350 1 3 2000s 0603cn 1 2 2 2 1 1 1 1 350 1 3 2000s 0604orn 1 2 1 2 1 1 1 1 350 1 3 2000s 0605orn 1 2 1 2 1 1 1 1 400 2 4 2000s 0606ch 1 1 2 2 2 1 1 1 250 1 2 2000s 0607cw 1 1 2 1 1 1 1 1 300 1 3 2000s 0608c 1 1 2 2 1 1 1 1 250 1 2 2000s 0608c 1 1 2 2 1 1 1 1 250 1 2 2000s 0610pc 1 1 2 2 1 2 1 2 250 1 2 2000s 0612cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0612cw 1 1 2 1 1 1 1 1 200 1 2 2000s 0613cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0615cw 1 1 2 1 1 1 1 1 300 1 3 2000s 0616c 1 1 2 2 1 1 1 1 250 1 2 2000s 0617cnwh 1 2 2 1 2 1 1 2 250 1 2 2000s 0618pc 1 1 2 2 1 2 1 2 250 1 2 2000s 0619new 1 1 2 1 1 2 1 2 300 1 3 2000s 0619pcw 1 1 2 1 1 2 1 1 250 1 2 2000s 0619pcw 1 1 2 1 1 2 1 2 300 1 3 2000s 0620c 1 1 2 2 1 1 2 1 300 1 3 2000s 224

PAGE 225

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw W hiteYellowv NormHiliu NormPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden 0621c 1 1 2 2 1 1 1 1 450 2 4 2000s 0701dw 1 ? 1 2 1 1 1 1 250 1 2 2000s 0703dw 1 ? 1 2 1 1 1 1 250 1 2 2000s 0705c_UFBest 1 1 2 2 1 1 1 1 200 1 2 2000s 0706cw 1 1 2 1 1 1 1 1 300 1 3 2000s 0707c 1 1 2 2 1 1 1 1 200 1 2 2000s 0708cnw 1 2 2 1 1 1 1 2 150 1 1 2000s 0709cnw 1 2 2 1 1 1 1 ? 300 1 3 2000s 0711c 1 1 2 2 1 1 1 1 250 1 2 2000s 0712cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0712cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0713c 1 1 2 2 1 1 1 1 450 2 4 2000s 0714cw 1 1 2 1 1 1 1 1 200 1 2 2000s 0715cn 1 2 2 2 1 1 2 2 300 1 3 2000s 0716cw 1 1 2 1 1 1 1 1 350 1 3 2000s 0718c 1 1 2 2 1 1 1 1 300 1 3 2000s 0719c 1 1 2 2 1 1 1 1 250 1 2 2000s 0720c 1 1 2 2 1 1 1 1 200 1 2 2000s 0721cw 1 1 2 1 1 1 1 1 300 1 3 2000s 0801cnw 1 2 2 1 1 1 1 ? 250 1 2 2000s 0802cn 1 2 2 2 1 1 1 1 350 1 3 2000s 0805c 1 1 2 2 1 1 1 1 250 1 2 2000s 0806c 1 1 2 2 1 1 1 1 200 1 2 2000s 0808cnh 1 2 2 2 2 1 1 2 300 1 3 2000s 0810cn 1 2 2 2 1 1 1 1 200 1 2 2000s 0811cn 1 2 2 2 1 1 1 ? 350 1 3 2000s 0812cn 1 2 2 2 1 1 ? 1 ? ? ? 2000s 0813c 1 1 2 2 1 1 1 ? 200 1 2 2000s 0815c 1 1 2 2 1 1 1 ? ? ? ? 2000s 225

PAGE 226

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu NormPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden 0816c 1 1 2 2 1 1 1 ? ? ? ? 2000s 0818cn 1 2 2 2 1 1 ? 1 250 1 2 2000s 0819cn 1 2 2 2 1 1 ? 2 250 1 2 2000s 0821pcnh 1 2 2 2 2 2 1 2 250 1 2 2000s 0822c 1 1 2 2 1 1 ? 1 200 1 2 2000s 0825cn 1 2 2 2 1 1 2 1 250 1 2 2000s 0826cnh 1 2 2 2 2 1 ? 1 350 1 3 2000s 0827cwh 1 1 2 1 2 1 ? 1 250 1 2 2000s 0828cw 1 1 2 1 1 1 1 1 250 1 2 2000s 0829pcw 1 1 2 1 1 2 1 1 300 1 3 2000s 0902c 1 1 2 2 1 1 1 1 350 1 3 2000s 0903cw 1 1 2 1 1 1 2 1 350 1 3 2000s 0904ny 1 2 3 2 1 1 1 1 350 1 3 2000s 0905c 1 1 2 2 1 1 1 1 350 1 3 2000s 0907w 1 1 1 1 1 1 1 1 350 1 3 2000s 0908pc 1 1 2 2 1 2 1 2 300 1 3 2000s 0909c 1 1 2 2 1 1 1 1 250 1 2 2000s 0910ch 1 1 2 2 2 1 1 1 250 1 2 2000s 0913y 1 1 3 2 1 1 1 1 250 1 2 2000s 0914c 1 1 2 2 1 1 1 1 300 1 3 2000s 0916c 1 1 2 2 1 1 2 2 450 2 4 2000s 1001c 1 1 2 2 1 1 1 1 300 1 3 2010s 1002cn 1 2 2 2 1 1 1 1 250 1 2 2010s 1003c n 1 2 2 2 1 1 1 1 ? ? ? 2010s 1004cn 1 2 2 2 1 1 1 ? ? ? ? 2010s 1005c 1 1 2 2 1 1 1 ? ? ? ? 2010s 1006cw 1 1 2 1 1 1 1 1 ? ? ? 2010s 1007pc 1 1 2 2 1 2 2 ? ? ? ? 2010s 1008y 1 1 3 2 1 1 1 ? ? ? ? 2010s 226

PAGE 227

Table A 1. Continued. SelectionID#z Programy PeachNecx MeltNMeltCrisw WhiteYellowv NormHiliu NormPeentot RenGloEglas ShowNonshor Chillq Chillrange1p Chillrange2o Decaden 1009cnwh 1 2 2 1 2 1 1 ? ? ? ? 2010s 1010a 8 ? ? 2 1 1 ? ? ? ? ? 2010s 1011a 8 ? ? 2 1 1 ? ? ? ? ? 2010s 1012cn 1 2 2 2 1 1 1 ? ? ? ? 2010s 1013cn 1 2 2 2 1 1 1 ? ? ? ? 2010s 1014c 1 1 2 2 1 1 1 ? ? ? ? 2010s 1015cn 1 2 2 2 1 1 1 ? ? ? ? 2010s 1016cn 1 2 2 2 1 1 1 ? ? ? ? 2010s 1017cnh 1 2 2 2 2 1 1 ? ? ? ? 2010s 1018c 1 1 2 2 1 1 1 ? ? ? ? 2010s 1019cw 1 1 2 1 1 1 1 ? ? ? ? 2010s 1020cnh 1 2 2 2 2 1 1 ? ? ? ? 2010s 1021c 1 1 2 2 1 1 1 1 ? ? ? 2010s 1022ch 1 1 2 2 2 1 2 ? ? ? ? 2010s 1023ch 1 1 2 2 2 1 2 1 ? ? ? 2010s 1024c 1 1 2 2 1 1 2 1 ? ? ? 2010s 1025cw 1 1 2 1 1 1 1 ? ? ? ? 2010s 1026c 1 1 2 2 1 1 1 1 ? ? ? 2010s 1027cnwh 1 2 2 1 2 1 1 ? ? ? ? 2010s 1028cw 1 1 2 1 1 1 1 ? ? ? ? 2010s 1029cw 1 1 2 1 1 1 1 ? ? ? ? 2010s 1030c 1 1 2 2 1 1 1 ? ? ? ? 2010s 1031wc 1 1 2 1 1 1 1 ? ? ? ? 2010s 1032c 1 1 2 2 1 1 1 ? ? ? ? 2010s 1033cnw 1 2 2 1 1 1 1 ? ? ? ? 2010s zSelectionID# = cultivar name or genotype selection number. yProgram = origin of the germplasm material: 1) University of Florida, 2) UF UGA USDA, 3) High chill landrace cultivar unknown origin, 4) NCSU (North Carolina State University) varieties, 5) Landrace cultivars, 6) Prunus dulcis 7) Prunus kansuensis and 8) UF apricot selections. 227

PAGE 228

xPeachNec= fruit type: 1) P each and 2) Nectarine. wMeltNMeltCris= fruit flesh: 1) Melting (freestone), 2) N on melting (clingstone) and 3) C ri spy. vWhiteYellow= fruit flesh color: 1) W hite and 2) Y ellow. uNormHili= fruit anthocyanin content: 1) Normal and 2) H ighlighter (reduced anthocyanin) tNormPeento = fruit shape: 1) Round and 2) P eento. sRenGloEgla = leaf gland types: 1) R enifor m, 2) G lobose and 3) E glandular. rShowNonsho = flower type: 1) Showy and 2) N o n showy qChill = chilling hour requirement (hours estimate). pChillrange1 = range chilling hour requirement: 1) Low <400 chill hours, 2) Medium 400700 chill hours, and 3) High >700 chill hours. oChillrange2 = range chilling hour requirement: 1) 0 150 chill hours, 2) 151250 chill hours, 3) 251350 chill hours, 4) 351450 chill hours, 5) 451550 chill hours, 6) 751850 chill hours, 7) 851950 chill hours, 8) 9511050 chill hours, and 9) 1051 1150 chill hours. nDecade = decade in which genotype was selected. ?=U nknown. 228

PAGE 229

Table A 2. Summary statistics of 36 simple sequ ence repeat (SSR) markers for 195 peach germplasm representatives of the genetic pools utilized in breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Marker Genotype No. A z Ae Ho He F PIC BPPCT006A 27 13 4.51 0.62 0.78 0.2 0.75 BPPCT006B 16 9 3.21 0.49 0.69 0.29 0.63 BPPCT008 37 y 18 5.94 0.59 0. 83 0.29 0.81 BPPCT013 13 11 1.58 0.26 0.37 0.31 0.35 BPPCT014 14 9 1.97 0.42 0.49 0.14 0.45 BPPCT017 20 12 3.51 0.67 0.71 0.06 0.67 BPPCT023 19 12 2.67 0.59 0.63 0.06 0.56 BPPCT025 30 14 5.1 0.65 0.8 0.19 0.78 BPPCT026 19 14 1.86 0.35 0.46 0.24 0.43 BPPCT027 12 7 2.53 0.18 0.6 0.7 0.53 BPPCT028 11 9 1.41 0.18 0.29 0.36 0.27 BPPCT029 12 9 1.51 0.26 0.34 0.23 0.33 BPPCT030 19 14 2.08 0.39 0.52 0.25 0.47 BPPCT038 23 15 2.43 0.46 0.59 0.22 0.51 BPPCT039 10 11 1.4 0.23 0.29 0.18 0.26 CPDCT025 18 10 2.4 0.41 0.58 0.3 0.55 CPDCT027 10 10 1.11 0.04 0.1 0.64 0.1 CPDCT038 36 22 4.84 0.65 0.79 0.18 0.76 CPPCT005 28 16 3.17 0.64 0.68 0.07 0.65 CPPCT008 13 14 1.12 0.06 0.11 0.48 0.11 CPPCT015 10 8 2.06 0.45 0.52 0.12 0.42 CPPCT022 33 16 3.65 0.51 0.73 0.3 0.69 CPPCT026 27 15 3.91 0.64 0.74 0.14 0.71 CPPCT033 21 11 2.94 0.54 0.66 0.18 0.62 CPSCT004 3 2 1.07 0.06 0.06 0.12 0.06 CPSCT008 10 11 1.99 0.03 0.5 0.94 0.41 CPSCT034 16 13 1.92 0.34 0.48 0.29 0.45 EPDC3832 10 8 2.13 0.88 0.53 0.65 0.42 EPD CU3117 3 2 1.35 0.15 0.26 0.41 0.23 EPPISF002 11 8 2.24 0.47 0.55 0.15 0.48 EPPISF032 11 7 1.67 0.29 0.4 0.28 0.37 PMS2 11 8 1.33 0.18 0.25 0.27 0.24 UDP96 001 10 7 1.57 0.28 0.36 0.22 0.32 UDP96 003 30 16 4.52 0.63 0.78 0.19 0.75 UDP96 019 25 14 4.1 3 0.53 0.76 0.31 0.72 UDP98 025 22 15 1.97 0.34 0.49 0.31 0.47 UDP98 412 26 13 2.54 0.51 0.61 0.16 0.58 Average 18 11.43 2.58 0.4 0.52 0.25 0.48 zA = number of observed alleles, Ae = effective number of alleles, Ho = observed heterozygosity, He = expec ted heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = polymorphism information content. yNumbers in bold represent highest and lowest values for each variable. 229

PAGE 230

T able A 3. Summary of the population stratification simulation r esults of 36 simple sequence repeat (SSR) markers for 195 germplasm representatives of University of Floridas genetic pools. File name K Run Est. Ln prob. of data z Mean value of Ln likelihood Variance of Ln likelihood bypop_run_1_rep2_f 2 1 15145.6 149 76.5 338.3 bypop_run_1_rep3_f 2 2 15129.3 14956.3 345.8 Struct_run_1_f 2 3 15120.6 14955.5 330.2 Bypop_run_1_f 2 4 15149 14976.5 344.9 bypop_run_2_rep3_f 3 1 14301.1 14056.2 489.9 Bypop_run_2_f 3 2 14331.4 14095.2 472.3 Struct_run_3_f 3 3 14324.8 14104.7 440.1 bypop_run_2_rep2_f 3 4 14328.5 14099 458.9 Struct_run_4_f 4 1 13855.8 13585.5 540.6 bypop_run_3_rep3_f 4 2 14010.8 13702.3 617.1 Bypop_run_3_f 4 3 13977.2 13685.6 583.3 bypop_run_3_rep2_f 4 4 13855.3 13592.3 525.9 byp op_run_4_rep2_f 5 1 13539.6 13187.7 703.8 Bypop_run_4_f 5 2 13615.3 13190.6 849.4 bypop_run_4_rep3_f 5 3 13574.6 13221.1 707 Struct_run_5_f 5 4 13595.9 13217.4 757.1 Bypop_run_5_rep2_f 6 1 13345.4 12876.8 937.1 bypop_run_5_rep2o_f 6 2 13560 .2 13051.8 1016.9 Bypop_run_5_rep3_f 6 3 13344.4 12880.3 928.2 bypop_run_5_rep3o_f 6 4 13357.4 12875.4 963.9 Bypop_run_6_rep2_f 7 1 13191.2 12596.8 1188.8 Bypop_run_6_rep3_f 7 2 13374 12748.1 1251.9 Bypop_run_6_f 7 3 13325.2 12589.6 1471.3 bypop_run_6_rep3o_f 7 4 13192.6 12649.1 1087 bypop_run_7_rep3_f 8 1 12928.6 12368.5 1120.1 Bypop_run_7_f 8 2 13119.8 12448.4 1342.7 bypop_run_7_rep2_f 8 3 12790.7 12258 1065.2 Struct_run_8_f 8 4 13234.8 12305.5 1858.5 Bypop_run_8_f 9 1 127 29.4 12146.9 1165 bypop_run_8_rep2o_f 9 2 12655.3 12061.7 1187.3 Bypop_run_8_rep2_f 9 3 12968.6 12121.5 1694.2 Bypop_run_8_rep3_f 9 4 13184.4 12202.8 1963.2 Bypop_run_9_f 10 1 12639.3 11941.8 1395 bypop_run_9_rep3o_f 10 2 12303 11632.7 1340 .5 Bypop_run_9_rep2_f 10 3 12713.6 12069.3 1288.6 Bypop_run_9_rep3_f 10 4 12461.4 11821.4 1280 Bypop_run_10_rep3_f 11 1 13172.5 11932.1 2480.8 Bypop_run_10_rep4_f 11 2 12479.9 11764.6 1430.4 Bypop_run_10_rep2_f 11 3 12364.9 11682.8 1364.2 B ypop_run_10_f 11 4 12386.8 11679.3 1414.9 Bypop_run_11_f 12 1 12302.8 11540.5 1524.7 230

PAGE 231

Table A 3. Continued. File name K Run Est. Ln prob. of data z Mean value of Ln likelihood Variance of Ln likelihood Bypop_run_11_rep3_f 12 2 12386.9 11542 1689.8 Bypop_run_11_rep4_f 12 3 12187.8 11511.8 1352.1 Bypop_run_11_rep2_f 12 4 13131.9 11771.5 2720.9 Bypop_run_12_f 13 1 13659 11409 4500 Bypop_run_12_rep2_f 13 2 12311.1 11445.8 1730.7 Bypop_run_12_rep3_f 13 3 13218.2 11548.6 3339 Bypop_run_12_ rep4_f 13 4 12656.1 11644.9 2022.3 Bypop_run_13_rep4_f 14 1 11941.9 11150.9 1581.8 Bypop_run_13_f 14 2 12238.9 11433 1611.7 Bypop_run_13_rep2_f 14 3 11988.1 11202.1 1572.1 Bypop_run_13_rep3_f 14 4 14843.6 11407.8 6871.5 zParameters of 105 in teractions after a burnin of 104 interactions. Table A 4. Analysis of the population structure results for 195 germplasm representatives of the University of Florida genetic pool using the Evanno method (Evanno et al., 2005) implemented in the Structure Harvester software (Earl and VonHoldt, 2012). K z Runs Mean LnP(K) Stdev LnP(K) Ln'(K) |Ln''(K)| Delta K 2 4 15136.1 13.5 3 4 14321.5 13.8 814.7 418.0 30.2 4 4 13924.8 81.1 396.7 53.3 0.7 5 4 13581.4 32.4 343.4 163.9 5.1 6 4 13401.9 105.7 179.5 48.4 0.5 7 4 13270.8 93.2 131.1 121.2 1.3 8 4 13018.5 197.5 252.3 118.2 0.6 9 4 12884.4 240.6 134.1 221.1 0.9 10 4 12529.3 184.3 355.1 426.8 2.3 11 4 12601 384.2 71.7 170.4 0.4 12 4 12502.4 427.6 98.7 557.4 1.3 13 4 12961.1 596.9 458.8 666 .7 1.1 14 4 12753.1 1399.7 208.0 zK=3 and K=5, highest delta K values 231

PAGE 232

Table A 5. Summary of the population stratification simulation results of 36 simple sequence repeat (SSR) markers for 168 peach UF varieties and advanced materials (20002010 selections). File name K Run Est. Ln prob. of data z Mean value of Ln likelihood Variance of Ln likelihood ResultsNec_run_1_f 2 1 11305.5 11182.6 245.8 Resultspar_run_1_f 2 2 11303.3 11182.5 241.6 ResultsPrun_1_f 2 3 10992.3 10871.6 241.3 Resultsrun _1_f 2 4 11303.0 11181.8 242.3 ResultsNec_run_2_f 3 1 10956.5 10778.1 356.9 Resultspar_run_2_f 3 2 10942.8 10757.9 369.9 ResultsPrun_2_f 3 3 10556.6 10409.1 295.0 Resultsrun_2_f 3 4 10871.1 10723.3 295.6 ResultsNec_run_3_f 4 1 10516.7 1030 6.0 421.4 Resultspar_run_3_f 4 2 10661.0 10444.3 433.4 ResultsPrun_3_f 4 3 10241.1 10024.2 433.8 Resultsrun_3_f 4 4 10523.1 10305.6 435.0 Resultsrun_4_f 4 5 10518.2 10330.8 374.9 Resultsrun_run_4_f 4 6 10520.2 10305.3 429.7 Resultspar_run_4 _f 5 1 10244.2 10002.0 484.3 ResultsPrun_4_f 5 2 9985.4 9736.2 498.5 Resultsrun_5_f 5 3 10241.8 10001.0 481.5 Resultsrun_run_5_f 5 4 10247.3 10001.8 490.9 Resultspar_run_5_f 6 1 10069.8 9799.5 540.6 ResultsPrun_5_f 6 2 9814.3 9522.8 582.9 Resultsrun_6_f 6 3 10142.7 9841.0 603.2 Resultsrun_run_6_f 6 4 10070.1 9790.6 559.0 Resultspar_run_6_f 7 1 9969.6 9634.5 670.2 ResultsPrun_6_f 7 2 9724.6 9406.9 635.6 Resultsrun_7_f 7 3 9969 9649.9 638.2 Resultsrun_run_7_f 7 4 9896.8 9577 .1 639.3 Resultspar_run_7_f 8 1 9813.9 9402.5 822.8 ResultsPrun_7_f 8 2 9573.1 9196.3 753.6 Resultsrun_8_f 8 3 9771.6 9399 745.2 Resultsrun_run_8_f 8 4 9781.9 9413.5 736.9 Resultspar_run_8_f 9 1 9692.3 9293.5 797.6 ResultsPrun_8_f 9 2 9489 .9 9064 851.8 Resultsrun_9_f 9 3 9696.6 9271.7 849.8 Resultsrun_run_9_f 9 4 9717.5 9285.4 864.2 Resultspar_run_9_f 10 1 9510.8 9075.9 869.9 ResultsPrun_9_f 10 2 9307.9 8847 922 Resultsrun_run_10_f 10 3 9531 9078.1 905.6 zParameters of 105 interactions after a burnin of 104 interactions. 232

PAGE 233

Table A 6 Analysis of the population structure results for 168 peach UF varieties and advanced materials (20002010 selections) using the Evanno method (Evanno et al., 2005) implemented in the Structure Harvester software (Earl and VonHoldt, 2012). K z Run s Mean LnP(K) Stdev LnP(K) Ln'(K) |Ln''(K)| Delta K 2 4 11226.03 155.82 3 4 10831.75 187.22 394.28 59.24 0.32 4 6 10496.72 137.43 335.03 17.99 0.13 5 4 10179.68 129.54 317.04 161.59 1.25 6 4 10024.23 144.09 155.45 21.23 0.15 7 4 9890.00 115.44 134.23 20.65 0.18 8 4 9735.13 109.51 154.88 68.83 0.63 9 4 9649.08 106.69 86.05 113.13 1.06 10 3 9449.90 123.39 199.18 zK=5 highest delta K value. 233

PAGE 234

Table A 7 Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earlygold peach (T E) reference map. Chromosome Position Marker Fluorophore Ta(C) z Forward sequence Reverse sequence 1 9 CPSCT008 HEX 62 TGGATCCAATCCAAGAGTCTG GCAGCAAGTTGTTCTTGGTTC 1 25.8 CPDCT038 HE X 62 ATCACAGGTGAAGGCTGTGG CAGATTCATTGGCCCATCTT 1 33.9 CPPCT026 HEX 55 AGACGCAGCACCCAAACTAC CATTACATCACCGCCAACAA 1 47.3 BPPCT027 HEX 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 1 55.2 BPPCT016 HEX 57 GATTGAGAGATTGGGCTGC GAGGATTCTCATGATTTGTGC 1 65.1 CPPC T029 HEX 55 CCAAATTCCAAATCTCCTAACA TGATCAACTTTGAGATTTGTTGAA 1 77.4 BPPCT028 HEX 57 TCAAGTTAGCTGAGGATCGC GAGCTTGCCTATGAGAAGACC 2 9.6 UDP98 025 FAM 57 GGGAGGTTACTATGCCATGAAG CGCAGACATGTAGTAGGACCTC 2 25 BPPCT013 HEX 57 ACCCACAAATCAAGCATATCC AGCTTCAGCCACCAA GC 2 38 BPPCT030 HEX 57 AATTGTACTTGCCAATGCTATGA CTGCCTTCTGCTCACACC 2 48.6 CPSCT034 HEX 62 AGGTGGACAATAGCCGTGAT TTTCCAGACCCTGAGAAAGC 3 18 BPPCT039 FAM 57 ATTACGTACCCTAAAGCTTCTGC GATGTCATGAAGATTGGAGAGG 3 36.4 CPDCT025 HEX 62 GACCTCATCAGCATCACCAA TTCCCTAA CGTCCCTGACAC 3 46.4 CPDCT027 HEX 62 TGAGGAGAGCACTGGAGGAG CAACCGATCCCTCTAGACCA 4 10.4 CPPCT005 FAM 52 CATGAACTCTACTCTCCA TGGTATGGACTCACCAAC 4 28.3 UDP96 003 FAM 57 TTGCTCAAAAGTGTCGTTGC ACACGTAGTGCAACACTGGC 4 34.1 EPDC3832 FAM 57 CTTTTGAAGGCCCAATACCA AT CACTGCTTCGCCTTCATT 4 45.4 BPPCT023 HEX 57 TGCAGCTCATTACCTTTTGC AGATGTGCTCGTAGTTCGGAC 4 52.7 EPPISF032 HEX 57 TCCCCCACAGATATTTCAGC GTCGAGGAGAGAGGGCTTTT 5 5.2 BPPCT026 FAM 57 ATACCTTTGCCACTTGCG TGAGTTGGAAGAAAACGTAACA 5 20.1 BPPCT017 HEX 57 TTAAGAGTTTGTGA TGGGAACC AAGCATAATTTAGCATAACCAAGC 5 32.9 BPPCT038 FAM 57 TATATTGTTGGCTTCTTGCATG GAGCTTGCCTATGAGAAGACC 5 44 BPPCT014 HEX 57 TTGTCTGCCTCTCATCTTAACC CATCGCAGAGAACTGAGAGC 6 8.7 CPPCT008 FAM 59 GAGCTCTCACGCATTAGTTT TTTGACTGCATAACAAAACG 6 17.5 UDP96 001 HEX 57 AGTTTGATTTTCTGATGCATCC TGCCATAAGGACCGGTATGT 6 30.1 BPPCT008 FAM 57 ATGGTGTGTATGGACATGATGA CCTCAACCTAAGACACCTTCACT 6 35.8 CPPCT015 FAM 50 TGGAGTGCCAATACTATTTA CATATGCATGGTTATGGT 6 41 EPPISF002 FAM 56 CGACGTGTGACCAAAGGAC GCAACTCCATCCACATTTCTC 6 56.4 B PPCT025 FAM 57 TCCTGCGTAGAAGAAGGTAGC CGACATAAAGTCCAAATGGC 6 72 UDP98 412 HEX 57 AGGGAAAGTTTCTGCTGCAC GCTGAAGACGACGATGATGA 7 9.5 CPSCT004 FAM 62 GCTCTGAAGCTCTGCATTGA TTTGAAATGGCTATGGAGTACG 7 18.6 CPPCT022 FAM 50 CAATTAGCTAGAGAGAATTATTG GACAAGAAGCAAGTAGTT TG 7 29.6 BPPCT029 FAM 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 7 38.9 CPPCT033 HEX 50 TCAGCAAACTAGAAACAAACC TTGCAATCTGGTTGATGTT 7 47.8 PMS2 FAM 55 CACTGTCTCCCAGGTTAAACT CCTGAGCTTTTGACACATGC 7 64.7 EPDCU3392 HEX 57 CTTTTCATGGGTTCCTCACC ATCAACCAGTTCA CGCACAA 8 14.1 BPPCT006 HEX 57 GCTTGTGGCATGGAAGC CCCTGTTTCTCATAGAACTCACAT 234

PAGE 235

Table A 7 Continued. Chromosome Position Marker Fluorophore Ta(C) z Forward sequence Reverse sequence 8 20.8 UDP96 019 HEX 57 TTGGTCATGAGCTAAGAAAACA TAGTGGCACAGAGCAACACC 8 24. 8 CPPCT006 HEX 59 AATTAACTCCAACAGCTCCA ATGGTTGCTTAATTCAATGG 8 42.6 CPDCT023 FAM 62 GTGGCAAATGTTGGCAAAG AACACAAAGCAGCACCAAGA 8 54.7 EPDCU3117 FAM 57 CAGAGGGAACAGTGTGAGCA TGTTGTTGTCGACCCTGAAA zTa = annealing temperature. 235

PAGE 236

Table A 8 Primer sequences of cpDNA as described by Shaw et al. (2005, 2007) and Morris et al. (2008). Region Primer name and sequence (5 3) 3trnV ndhC trnV (UAC) x2: GTC TAC GGT TCG ART CCG TA ndhC: TAT TAT TAG AAA TGY CCA RAA AAT ATC ATA TTC ndhJ trnF ndhJ: ATG CCY GAA AGT TGG ATA GG TabE: GGT TCA AGT CCC TCT ATC CC (Taberlet et al., 1991) trnL trnF and trnL intron 5trnL (UAA) F: CGA AAT CGG TAG ACG CTA CG trnF(GAA): ATT TGA ACT GGT GAC ACG AG (Taberlet et al., 1991) trnT trnL trnT (UGU) F: CAT TAC AAA TGC GAT GCT CT 5trnL(UAA)R: TCT ACC GAT TTC GCC ATA TC (Taberlet et al., 1991) trnQ 5rps16 trnQ (UUG) : GCG TGG CCA AGY GGT AAG GC rpS16x1: GTT GCT TTY TAC CAC ATC GTT T trnH psbA trnH (GUG) : CGC GCA TGG TGG ATT CAC AAT CC (Tate and Simpson, 2003) psbA: GTT ATG CAT GAA CGT AAT GCT C (Sang et al., 1997) ndhF rpl32 rpL32 R: CCA ATA TCC CTT YYT TTT CCA A ndhF: GAA AGG TAT KAT CCA YGM ATA TT atpB rbcL atpB 1: ACATCKARTACKGGACCAATAA Chiang et al. (1998) rbcL 1: AACACCAGCTTTRAATCCAA 236

PAGE 237

Table A 9 Primers for candidate genes associated with flowering time response, branching plant architecture and isozymes in peach. Gene symbol Gene description Expected size (bp) Ta (C) z Forward Primer Reverse Primer Reference AGL20 SOC1 AGAMOUS like 20 / SUPPRESOR OF OVEREXPRESSION of CONSTANS 1 569 55 ATGCAGACAAC CATAGAACG CTATTGAAAAA TGCAGCACA This study AGL24 AGAMOUSlike 24 653 55 GGATGAGCTGC TGAAGTTAG TTTTTCTCCAG TGCCATAAT This study AXR1 AUXIN RESISTANT 1 419 59 AGCAAGGACAG ACAGCCCTA CTTGGCCTTA ACAGCATCGT Carrillo Mendoza (2012) BFT BROTHER of FLO WERING LOCUS T and TERMINAL FLOWER 1 570 55 GAGCCATTGAC TCTTGGTAG GCAACCCTAT TTCTTGTCCT This study BRC1 BRANCHED 1 342 57 CACAAGCACCA CCCCTTACT CCTGAATGAG CAACCACTCA Carrillo Mendoza (2012) BRC2 BRANCHED 2 692 57 TCAAGCGGGAA TAAGGACAG GCTCCACTGG GAATTGTTGT Ca rrillo Mendoza (2012) BRM BRAHMA 704 55 AATCAACAGGA AGTGGAAGA GTTTCCAGTC CGTAACATTG This study CO CONSTANS 567 55 AAGGCTGCCTA CAGTTACAA ACAAAGGGAA TGAAACACTG This study CRY1 Cryptochrome 1 523 55 CACCAAGAGGT CCACTGA CTGAATCTGTT CCCACTGTT This study CRY2 Cry ptochrome 2 554 55 ATTGAAGCCAT TGGAGCTA GTGACATGTG CCTTCAACTA This study CUC1 CUP SHAPED COTYLEDON 1 364 57 CTTGACAGCAG CTTCACTGG AGCTCTGCCA CGGTAGAAAA Carrillo Mendoza (2012) CUC2 CUP SHAPED COTYLEDON 2 741 57 GTTTTCTACCG TGGCAGAGC TTGCTCATTCG GGTCTTCTT Carr illo Mendoza (2012) CUC3 CUP SHAPED COTYLEDON 3 751 57 GAGTGAGCTGA GTGGGGAAG TGGCCCTGTT GGTTCTTAAC Carrillo Mendoza (2012) ELF6 EARLY FLOWERING 6 850 55 AATTTCACCTG AGGTTGTTG CTACTTCAAG CCAGTGTGGT This study FD FERREDOXIN 701 55 CTGTTCTCAGC TTGACCAAT ATCGA TGAAG ATTGTGCTTT This study FD1 FERREDOXIN 1 518 55 GTAACCAAATA TGGCAGCTC CAGCAGCTAT CAAAACATGA This study F P F 1 FLOWERING PROMOTING FACTOR 1 335 55 CCAAAGCACTA GCTACTACCA GATGTTGGGG TTTTTAATGA This study FG 505 55 TCCAAGGTCTT CTACCTCAA CCCATCCATA ATTCTACCAA T his study 237

PAGE 238

Table A 9 Continued. Gene symbol Gene description Expected size (bp) Ta (C) z Forward Primer Reverse Primer Reference FLC FLF FLOWERING LOCUS C / FLOWERING LOCUS F 540 55 AAATTGGAAGA GGCAAGATT TTCTAGGGCC ACTTTACTCA This study FRIGIDA FRIGIDA 7 51 55 AATACCAATGG TTTCACGTC TTAACTGCATC AATTTGTCG This study FT TSF FLOWERING LOCUS T / TWIN SISTER of FLOWERING LOCUS T 558 55 CAAGGTCTGTT TCTCTCAGG TTGAACTCGA AACCTCAACT This study GI FB GIGANTEA 692 55 AAGTTAATTGC AGCTCCTGA CTGGTTGATT CTCCACATCT This study HOS1 HIGH EXPRESSION OF OSMOTICALLY RESPONSIVE GENES 1 568 55 CGCCTTATATC CAAGAAGTG ATGAATCCAG AACATCAAGC This study LAS LATERAL SUPPRESSOR 402 57 ATGCGCCAATT GCTCATTAC AAAGTCGAGG ATGTGGATGG Carrillo Mendoza (2012) LFY LEAFY 795 55 GTGATTTCTTG ATCCAGGTC ATC TTGGGCT TGTTGATGTA This study MAF1 MAF3 AGL31 MADS AFFECTING FLOWERING 1 / MADS AFFECTING FLOWERING 3 / AGAMOUSlike 31 594 55 AGATGGGGAG AGGAAAGATT CTGCAGCATG TACTAGAGGA This study MAF2 MADS AFFECTING FLOWERING 2 545 55 GATCGAAAACA CAACAAACC ACCCATTAAAA GCAC ATACA This study MAF4 MADS AFFECTING FLOWERING 4 568 55 TGTGACATTGA CATTGCTCT ATCATCGTCAA CATGCATAA This study MAF5 MADS AFFECTING FLOWERING 5 354 55 GCTTCAGATCA AGAGGATTG ATTAACATAGC GTGCCAGAA This study MAX1 MORE AXILLARY BRANCHES 1 331 57 TTACGAGCATC TCCT TGCTG TTGCAACTAAT GGGGAAACC Carrillo Mendoza (2012) MAX2 MORE AXILLARY BRANCHES 2 441 57 AAAACTTGGAT GCTGCTGCT TCCGAATCTC CTCCAGATTG Carrillo Mendoza (2012) MAX3 MORE AXILLARY BRANCHES 3 387 57 TTGCTTCCTCG GTCTCCTAA GAGGTAGCGT CTTGCTTTGG Carrillo Mendoza (2 012) MAX4 MORE AXILLARY BRANCHES 4 341 57 CGGTCATTGCA GATTGTTGT CAACCCTCTT CCATGTTCGT Carrillo Mendoza (2012) MFT MOTHER OF FLOWERING LOCUS T 529 55 AACATGTCCGT CTACTTTGG TTTACCTCTAA CAGGGCTTG This study PGDH PHOSPHOGLYCERATE DEHYDROGENASE 614 55 ATCTCACT GCA TATGCCTCT GCTATTTCAAT GGCTACACC This study PGI PHOSPHOGLUCOSE ISOMERASE 1 528 55 CACCTGAAACT TTGGAGAAG CCAAGATCCA CTTCGAATAC This study 238

PAGE 239

Table A 9 Continued. Gene symbol Gene description Expected size (bp) Ta (C) z Forward Primer Reverse Primer Referenc e PHYA PHYTOCHROME A 520 55 TAAAGCAATTG GAAACCATT GATTCCACTC AGAACACCAG This study PHYB PHYD PHYTOCHROME B / PHYTOCHROME D 600 55 ACTTCTTTTCC GCGCTTTA TGTGTTCCAAA TCATCAAAA This study PHYE PHYTOCHROME E 573 55 TGAAATTGTTT GTGAGGACT GCCCTGCTAA AGGAGTATTT This study PIN PIN FORMED 447 57 ATTTCAGCTTT GGCAACAGG GCCAAGTCCT GCATCAGATA G Carrillo Mendoza (2012) RAX1 REGULATOR OF AXILLARY MERISTEMS 1 666 55 CTTAACTACCT GAGGCCAAA TGATGAGAGG AAGAAGGAAA This study RAX2 RAX3 REGULATOR OF AXILLARY MERISTEMS 2 / REGULATOR OF AXILLARY MERISTEMS 3 638 55 AAGACAGGATC ATCTGCAAC GCTCCAAAGC ACTAGAGAAA This study REV REVOLUTA 480 57 CGCCAGTATGT TCGAAGTGT CCAAGGAATC AGGTCTCAGC Carrillo Mendoza (2012) RGA RGA1 REPRESSOR OF GIBERELIN / REPRESSOR OF GIBERELIN 1 478 55 CCACTCATTCT CCGATATG T CATTTGCTTAA CCACAGACA This study RGL1 RGL2RGL3 RGA like 1 / RGA like 2 / RGA like 3 542 55 CAGTCCACTAC AACCCATCT AGCTTCAGCA AAGTAAGTGG This study s6pdh SORBITOL 6 PHOSPHATE DEHYDROGENASE 800 57 AGAATAAGGTG TTGGACATAGA CG AGAGTGGTCC TGGATTTCTTA TCTA Bortiri e t al. (2002) SPS SUPERSHOOT 485 57 ACTCAAAGACG CCAGTGGTC GCCCTTGGGG ATGAAGTAAT Carrillo Mendoza (2012) SPY SPINDLY 633 55 ATTTGACAGTT TCACGGAAG CAGTCCCTCT TGAGTATTGC This study TFL1ATC TERMINAL FLOWER 1 / CENTRORADIALIS 562 55 TGGTCATGAGC TATTTCCTT TTAGACT CAA GGAGAGCATC This study TLF2 TERMINAL FLOWER 2 587 55 TGAGCTCCAGT ATCTCATCA AGAGGGAAAT GGAAGAAAAC This study VRN1 REDUCED VERNALIZATION RESPONSE 1 621 55 TTCAATTTAATG TAACAGAGCTT CCTGGCTTAA TTTTGCTCTA This study VRN2 REDUCED VERNALIZATION RESPONSE 2 770 55 TTCAACTA T AG GGACTACAACA ACATTGCATCA AGCAAAAG This study zTa = annealing temperature. 239

PAGE 240

Table A 10 Internal primers used to improve sequence coverage, quality and basecalling accuracy of gene regions associated with axillary meristem formation, dormancy re sponse and isozyme expression. Gene symbol z Ta (C) y Sequence CUC2int F 55 GAACTTTAGTGATCTCATGCTA CUC3int R 55 CAAGCTGAAGAAGTACCACT RAX 2 RAX3int F 55 TCAATGTTTGTTCTCTCACA PGIint F 55 TCAAATTAGCAGAGGTGGG PGIint R 55 CAGAGAACTTTTGGATCTTG PGDHint F 55 C AAAGGTGTTCAATGATGAA PGDHint R 55 GTGTGAGAGTGACATTCTCA sp6dhint F 55 CAACTACGAGCTCTTTCTGA AGL20 SOC1int F 55 TGACCAAAATATGCAGGTGA AGL20 SOC1int R 55 CTCTCCAACTGTTGCTCAAT BRMint F 55 CTGGCCTCTTAATTTGTC COint F 55 CTTTGCTTAATGCTAGCTCA COint R 55 CTGAAG ATACTCCAAGAATATT ELF6int F 55 CAATTCCTTTGTGAACAT ELF6int R 55 TCAATAAGGCATTGGCAGAG FDint R 55 GTTCTTGCTGCCTTTTGAG FRIGIDAint R 55 CTATGTTCGCCATGTTGGT FT TSFint F 55 ATTAAGTGCATACGTACGGG FT TSFint R 55 AATACACACCAATGCAAATA GI FBint F 55 ATGTTAGGTCGTT GATGCAT GI FBint R 55 TTAGACCTGAAGAACCACTG HOS1int F 55 GACGTCCAACGCCTTTATTC HOS1int R 55 CATTAGATATGCCAGCCAAT LFYint F 55 ACCAAGGTACGGAGTTTACC LFYint R 55 TGGAATCACTATATATTTCTAACAA LFYint2 F 55 ATATAGTGATTCCAATAATGATG MAF2Aint R 55 TCTCAACTCAACTGAG TACCAT MAF4int F 55 TGATCATGTGTGGGAACTT MAF4int R 55 AAGAAGTTCCCACACATGAT TFL2int F 55 AGAGGGAAATGGAAGAAAAC zF= forward primer, R= reverse primer. yTa = annealing temperature. 240

PAGE 241

Table A 1 1 Summary statistics of candidate gene regions based on exon and intron boundaries for 11 Prunus species. Region ID Exon(s) Intron(s) Total % var L (bp)z Substy Indelsx % varw L (bp) Subst Indels % var L (bp) Subst Indels % var Subst Indels AGL20SOC1 151 5 0 3.31 293 23 20 14.68 444 28 20 10.81 6.31 4.50 AGL24 494 48 15 12.75 494 48 15 12.75 9.72 3.04 AXR1 181 6 0 3.31 82 5 0 6.10 263 11 0 4.18 4.18 0.00 BRM 95 4 0 4.21 271 14 3 6.27 366 18 3 5.74 4.92 0.82 BRC1 248 4 0 1.61 51 2 0 3.92 299 6 0 2.01 2.01 0.00 BRC2 298 15 0 5.03 298 15 0 5.03 5.03 0.00 BFT 218 9 0 4.13 240 14 9 9.58 458 23 9 6.99 5.02 1.97 CO 419 28 6 8.11 419 28 6 8.11 6.68 1.43 CRY1 73 2 0 2.74 427 27 4 7.26 500 29 4 6.60 5.80 0.80 CRY2 450 33 5 8.44 450 33 5 8.44 7.33 1.11 CUC1A 210 6 0 2.86 95 10 2 12.63 3 05 16 2 5.90 5.25 0.66 CUC1B 210 6 0 2.86 95 10 2 12.63 305 16 2 5.90 5.25 0.66 CUC2 33 0 0 0.00 544 43 9 9.56 257 43 9 9.01 16.73 3.50 CUC3 31 0 0 0.00 474 39 17 11.81 505 39 17 11.09 7.72 3.37 ELF6 63 2 0 3.17 530 42 24 12.45 593 44 24 11.47 7.42 4.05 FD 359 15 8 6.41 177 21 3 13.56 536 36 11 8.77 6.72 2.05 FD1 481 17 2 3.95 481 17 2 3.95 3.53 0.42 FG 55 2 0 3.64 339 27 4 9.14 394 29 4 8.38 7.36 1.02 FLC FLF 110 2 0 1.82 338 19 25 13.02 448 21 25 10.27 4.69 5.58 FT TSF 139 1 0 0.72 302 15 21 11.92 441 16 21 8.39 3.63 4.76 FPF1 329 6 3 2.74 329 6 3 2.74 1.82 0.91 FRIGIDA 630 35 8 6.83 630 35 8 6.83 5.56 1.27 GI FB 76 3 0 3.95 501 31 19 9.98 577 34 19 9.19 5.89 3.29 HOS1 135 1 0 0.74 63 7 0 11.11 198 8 0 4.04 4.04 0.00 L AS 378 25 1 6.88 378 25 1 6.88 6.61 0.26 LFY 615 86 25 18.05 615 86 25 18.05 13.98 4.07 MAF1 MAF3 AGL31 133 4 0 3.01 363 28 9 10.19 496 32 9 8.27 6.45 1.81 MAF2A 102 3 1 3.92 167 10 15 14.97 269 13 16 10.78 4.83 5.95 241

PAGE 242

Table A 11 Continu ed. Region ID Exon(s) Intron(s) Total % var L (bp)z Substy Indelsx % varw L (bp) Subst Indels % var L (bp) Subst Indels % var Subst Indels MAF2B 335 14 9 6.87 335 14 9 6.87 4.18 2.69 MAF4 474 54 38 19.41 474 54 38 19.41 11.39 8.02 MAF 5 125 5 0 4.00 185 20 11 16.76 310 25 11 11.61 8.06 3.55 MAX1A 140 5 0 3.57 187 11 5 8.56 327 16 5 6.42 4.89 1.53 MAX1B 221 8 0 3.62 102 9 1 9.80 323 17 1 5.57 5.26 0.31 MAX2 376 16 1 4.52 376 16 1 4.52 4.26 0.27 MAX3 164 7 0 4.27 147 23 5 19.0 5 311 30 5 11.25 9.65 1.61 MAX4 94 6 0 6.38 145 25 3 19.31 239 31 3 14.23 12.97 1.26 MFT 114 2 0 1.75 281 28 3 11.03 395 30 3 8.35 7.59 0.76 PGI 74 6 0 8.11 225 16 13 12.89 299 22 13 11.71 7.36 4.35 PGDH 191 8 0 4.19 286 22 2 8.39 477 30 2 6.71 6.29 0. 42 PHYA 338 4 0 1.18 155 6 0 3.87 493 10 0 2.03 2.03 0.00 PHYBPHYD 500 30 4 6.80 500 30 4 6.80 6.00 0.80 PHYE 49 5 1 12.24 450 26 4 6.67 499 31 5 7.21 6.21 1.00 PIN 165 1 1 1.21 178 8 4 6.74 343 9 5 4.08 2.62 1.46 VRN1 424 16 0 3.77 168 19 0 11.31 592 35 0 5.91 5.91 0.00 VRN2 418 25 11 8.61 418 25 11 8.61 5.98 2.63 RAX1 47 1 0 2.13 398 22 4 6.53 445 23 4 6.07 5.17 0.90 RAX2 RAX3 215 6 0 2.79 288 14 7 7.29 503 20 7 5.37 3.98 1.39 RGA RGA1 454 15 0 3.30 454 15 0 3.30 3.30 0.0 0 REV 105 10 0 9.52 273 20 6 9.52 378 30 6 9.52 7.94 1.59 RGL1RGL2RGL3 510 36 1 7.25 510 36 1 7.25 7.06 0.20 s6pdh 339 7 0 2.06 290 26 18 15.17 629 33 18 8.11 5.25 2.86 SPS 215 7 0 3.26 148 14 5 12.84 363 21 5 7.16 5.79 1.38 SPY 100 3 0 3.00 406 20 5 6.16 506 23 5 5.53 4.55 0.99 TFL1ATC 92 2 0 2.17 365 14 11 6.85 457 16 11 5.91 3.50 2.41 TFL2 77 2 0 2.60 422 25 7 7.58 499 27 7 6.81 5.41 1.40 242

PAGE 243

Table A 11 Continued. Region ID Exon(s) Intron(s) Total % var L (bp)z Substy Indelsx % varw L (bp) Subst Indels % var L (bp) Subst Indels % var Subst Indels TOTAL 9072 330 28 4 14451 1094 412 11 23203 1424 440 8 6.14 1.90 zL= total aligned sequence length. ySubst =substitutions. xIndels = number of insertion / deletion events. w% var = [(Subst+In dels)/L] 100. 243

PAGE 244

Table A 1 2 Summary results of the phylogenetic analyses for diploid sequence data. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Trees (No) L ength CI RI RC L (bp) Inf (No) lnL AGL20 SOC1 443 13 27 28 1.000 1.000 1.000 464 21 11 62 0.790 0.740 0.585 443 13 772.842 AGL24 493 24 1 42 0.976 0.984 0.960 584 56 1 134 0.985 0.989 0.974 493 24 936.960 AXR1 263 8 24 10 1.000 1.000 1.000 263 8 406.360 BRM 221 4 10 10 1.000 1.000 1.000 240 4 10 29 1.000 1.000 1.000 221 4 332.595 BRC1 299 1 2 4 1.000 1.000 1.000 299 1 480.679 BRC2 298 5 1 12 1.000 1.000 1.000 298 5 470.726 BFT 458 9 3 20 1.000 1.000 1.000 477 25 3 39 1.000 1.000 1.000 458 9 817.102 CO 416 13 1 20 1.000 1.000 1.000 420 16 3 25 0.960 0.974 0.935 416 13 713.811 CRY1 500 12 5 28 0.964 0.960 0.926 549 40 5 77 0.987 0.981 0.969 500 12 881.916 CRY2 450 15 6 30 1.000 1.000 1.000 485 24 3 6 6 0.985 0.982 0.968 450 15 759.729 CUC1A 430 2 6 11 1.000 1.000 1.000 431 2 6 12 1.000 1.000 1.000 430 2 644.941 CUC1B 305 2 26 10 1.000 1.000 1.000 306 2 26 11 1.000 1.000 1.000 305 2 460.375 CUC2 577 20 36 50 0.920 0.922 0.848 630 47 6 104 0.952 0. 965 0.919 577 20 1076.250 CUC3 503 23 11 38 1.000 1.000 1.000 561 67 11 96 1.000 1.000 1.000 503 23 909.547 ELF6 585 26 4 53 0.868 0.894 0.776 627 45 6 112 0.786 0.789 0.620 585 26 1054.082 FD 536 16 16 37 0.973 0.980 0.953 551 19 16 53 0.962 0.966 0 .929 536 16 924.255 FD1 481 5 3 12 1.000 1.000 1.000 520 5 3 51 1.000 1.000 1.000 481 5 730.555 FG 394 16 3 32 0.969 0.974 0.943 418 40 542 70 0.786 0.885 0.696 394 16 754.585 FLC FLF 459 14 64 23 1.000 1.000 1.000 535 82 1 106 0.943 0.965 0.910 459 14 762.478 FTTSF 503 14 75 18 1.000 1.000 1.000 561 54 3 82 0.927 0.950 0.880 503 14 756.323 FPF1 329 1 1 5 1.000 1.000 1.000 336 5 1 12 1.000 1.000 1.000 329 1 470.312 FRIGIDA 522 11 12 32 1.000 1.000 1.000 579 41 1 104 0.856 0.810 0.693 522 11 907.176 GI FB 573 16 12 31 0.968 0.976 0.945 644 45 3 106 0.953 0.965 0.920 573 16 954.621 HOS1 198 5 1 7 1.000 1.000 1.000 198 5 312.347 LAS 378 8 52 18 1.000 1.000 1.000 381 11 52 21 1.000 1.000 1.000 378 8 648.781 LFY 611 48 1 106 0.858 0.873 0.749 670 80 2 175 0.857 0.866 0.743 611 48 1465.959 244

PAGE 245

Table A 1 2 Continued. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) lnL MAF1 MAF3 AGL31 494 15 6 30 0.933 0.470 0.884 524 36 8 64 0.906 0.929 0.842 494 15 846.279 MAF2A 265 6 36 15 0.933 0.944 0.881 283 17 5 31 0.839 0.884 0.741 265 6 462.861 MAF2B 335 8 3 18 1.000 1.000 1.000 400 23 22 88 0.943 0.821 0.775 335 8 533.146 MAF4 471 20 155 58 0.948 0.933 0.885 572 35 24 164 0.951 0.895 0.851 471 20 962.605 MAF5 310 16 1 23 1.000 1.000 1.000 361 44 1 74 1.000 1.000 1.000 310 16 555.134 MAX1A 327 7 1 14 1.000 1.000 1.000 341 10 2 29 0.966 0.966 0.932 327 7 523.030 MAX1B 323 4 1 16 1.000 1.000 1.000 324 4 1 17 1.000 1.000 1.000 323 4 540.059 MAX2 376 10 3 11 1.000 1.000 1.000 382 16 3 17 1.000 1.000 1.000 376 10 579.672 MAX3 311 13 24 30 0.967 0.971 0.938 335 27 24 59 0.898 0.895 0.804 311 13 595.726 MAX4 238 14 2 25 1.000 1.000 1.000 240 15 2 27 1.000 1.000 1.000 238 14 455.622 MFT 394 10 4 24 1.000 1.000 1.000 398 12 4 28 1.000 1.000 1.000 394 10 671.169 PGI 297 13 6 24 0.958 0.969 0.928 309 19 9 37 0.946 0.955 0.903 297 13 583.154 PGDH 4 77 14 44 22 1.000 1.000 1.000 478 14 44 23 1.000 1.000 1.000 477 14 784.529 PHYA 493 1 6 10 1.000 1.000 1.000 493 1 750.176 PHYBPHYD 500 16 18 28 0.964 0.974 0.940 546 48 18 74 0.986 0.993 0.979 500 16 825.974 PHYE 499 14 6 26 1.000 1. 000 1.000 517 29 6 44 1.000 1.000 1.000 499 14 835.046 PIN 343 4 10 7 1.000 1.000 1.000 349 5 10 13 1.000 1.000 1.000 343 4 546.355 VRN1 592 14 1 26 1.000 1.000 1.000 592 14 973.535 VRN2 418 7 14 18 1.000 1.000 1.000 465 9 14 65 1.000 1 .000 1.000 418 7 654.134 RAX1 445 15 4 23 1.000 1.000 1.000 450 17 4 28 1.000 1.000 1.000 445 15 734.114 RAX2 RAX3 503 5 2 18 1.000 1.000 1.000 547 9 8 64 0.969 0.913 0.885 503 5 810.005 RGA RGA1 454 2 29 10 1.000 1.000 1.000 454 2 720.811 REV 377 15 10 27 1.000 1.000 1.000 386 19 10 36 1.000 1.000 1.000 377 15 710.356 RGL1 RGL2 RGL3 510 17 1 32 1.000 1.000 1.000 513 17 1 35 1.000 1.000 1.000 510 17 929.111 s6pdh 625 15 12 33 0.879 0.897 0.789 649 28 1 61 0.869 0.886 0.770 625 15 1051.341 SPS 363 8 11 14 1.000 1.000 1.000 370 13 3 21 1.000 1.000 1.000 363 8 583.152 245

PAGE 246

Table A 1 2 Continued. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) lnL SPY 506 8 1 20 1.000 1.000 1.000 524 25 1 38 1.000 1.000 1.000 506 8 838.247 TFL1 ATC 457 8 212 15 0.867 0.882 0.765 495 37 20 53 0.962 0.982 0.945 457 8 717.034 TFL2 499 7 6 24 0.917 0.900 0.825 511 9 8 38 0.921 0.870 0.801 499 7 785.199 3'trnV ndhC 567 29 1082 67 0.896 0.821 0.735 906 96 75 437 0.913 0.775 0.708 567 29 998.774 trnL trnF and trnL intron 836 3 1 12 1.000 1.000 1.000 836 3 1167.722 trnQ 5'rps16 479 8 10 17 0.882 0.889 0. 784 509 19 3 50 0.900 0.848 0.764 479 8 724.826 trnH psbA 335 4 2925 10 1.000 1.000 1.000 384 17 675 61 0.967 0.955 0.923 335 4 446.838 ndhF rpl32 598 1 1 7 1.000 1.000 1.000 625 7 1 34 1.000 1.000 1.000 598 1 780.124 atpB rbcL 769 3 1 13 1.000 1.000 1.000 779 3 1 23 1.000 1.000 1.000 769 3 1085.826 ITS 610 7 2 54 0.981 0.889 0.872 646 16 2 94 0.947 0.783 0.741 610 7 1099.593 zL=total aligned sequence length. Inf=parsimony informative sites. CI=consistency index. RI=retention index. RC=rescaled co nsistency index. 246

PAGE 247

Table A 1 3 Summary results of the phylogenetic analyses for phased haplotype sequence data. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Tr ees (No) Length CI RI RC L (bp) Inf (No) lnL AGL20 SOC1 444 29 229 36 0.889 0.972 0.864 469 54 871 93 0.613 0.877 0.537 444 29 852.61 AGL24 494 48 8 64 0.922 0.971 0.895 590 107 15 167 0.928 0.975 0.904 494 48 1067.87 AXR1 263 8 1 11 1.000 1.000 1.000 263 8 441.17 BRM 366 15 1 22 0.955 0.963 0.919 386 35 8 45 0.911 0.918 0.837 366 15 662.89 BRC1 299 3 1 6 1.000 1.000 1.000 299 3 440.99 BRC2 298 9 1 15 1.000 1.000 1.000 298 9 516.30 BFT 458 14 1 25 1.000 1.000 1.000 478 32 8 48 0.938 0.980 0.919 458 14 803.13 CO 419 23 3 30 1.000 1.000 1.000 427 31 48 42 0.905 0.957 0.865 419 23 777.69 CRY1 500 19 6 35 0.971 0.972 0.944 549 47 6 84 0.988 0.989 0.977 500 19 913.50 CRY2 450 31 6 39 0.949 0.975 0.925 4 85 66 7 79 0.911 0.955 0.870 450 31 865.10 CUC1A 430 12 2494 22 0.773 0.844 0.652 431 13 2494 23 0.783 0.848 0.664 430 12 732.94 CUC1B 305 5 3 17 0.941 0.889 0.837 306 5 3 18 0.944 0.889 0.840 305 5 534.29 CUC2 577 35 4 53 0.925 0.957 0.885 630 62 54 112 0.911 0.953 0.868 577 35 1123.05 CUC3 505 31 1 49 1.000 1.000 1.000 566 82 28 112 0.973 0.985 0.959 505 31 995.94 ELF6 593 50 3 55 0.927 0.979 0.908 643 98 312 149 0.678 0.897 0.608 593 50 1177.99 FD 536 31 2 41 0.976 0.992 0.968 564 52 4 73 0.932 0.969 0.903 536 31 1023.21 FD1 481 10 1 20 1.000 1.000 1.000 520 12 1 59 1.000 1.000 1.000 481 10 817.46 FG 394 17 3 38 0.974 0.978 0.952 418 41 4684 76 0.803 0.891 0.715 394 17 732.76 FLC FLF 448 25 48 27 0.926 0.979 0.906 547 118 1796 162 0.765 0.936 0.716 448 25 805.14 FTTSF 507 18 150 20 1.000 1.000 1.000 573 79 1 102 0.843 0.951 0.802 507 18 783.89 FPF1 329 1 1 6 1.000 1.000 1.000 336 5 1 13 1.000 1.000 1.000 329 1 511.56 FRIGIDA 630 36 2 40 0.950 0.971 0.922 681 87 3 99 0.899 0.945 0.850 630 36 1116.61 GI FB 577 39 48 45 0.956 0.984 0.940 662 119 80 153 0.837 0.953 0.798 577 39 1076.97 HOS1 198 6 3 9 0.889 0.929 0.825 198 6 366.16 LAS 378 15 1 25 1.000 1.000 1.000 381 18 1 28 1.000 1.000 1.000 378 15 666.32 LFY 615 101 318 149 0.732 0.914 0.669 693 179 352 249 0.751 0.924 0.694 615 101 1681.58 247

PAGE 248

Table A 13 Continued. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Trees ( No) Length CI RI RC L (bp) Inf (No) lnL MAF1 MAF3 AGL31 496 27 1 43 0.860 0.906 0.780 531 59 7 87 0.828 0.882 0.730 496 27 997.52 MAF2A 268 13 3 17 0.941 0.972 0.915 283 23 15 41 0.756 0.863 0.653 268 13 465.01 MAF2B 335 9 1 19 1.000 1.000 1.000 393 2 4 3 82 0.939 0.839 0.788 335 9 584.64 MAF4 474 69 48 95 0.758 0.907 0.687 584 178 148731 244 0.746 0.898 0.670 474 69 1228.68 MAF5 310 26 1 29 1.000 1.000 1.000 361 71 1 80 1.000 1.000 1.000 310 26 554.88 MAX1A 327 11 1 18 1.000 1.000 1.000 342 15 3 34 0.971 0.977 0.949 327 11 579.13 MAX1B 323 10 1 20 1.000 1.000 1.000 324 11 1 21 0.857 0.850 0.729 323 10 567.47 MAX2 376 10 1 16 1.000 1.000 1.000 382 16 1 22 1.000 1.000 1.000 376 10 583.46 MAX3 311 22 24 40 0.925 0.963 0.890 335 45 24 64 0.953 0 .979 0.933 311 22 648.98 MAX4 239 22 1 34 0.941 0.962 0.906 242 24 1 37 0.946 0.966 0.914 239 22 588.05 MFT 395 17 1 31 1.000 1.000 1.000 398 18 1 35 0.971 0.979 0.951 395 17 728.50 PGI 299 19 18 25 1.000 1.000 1.000 317 36 12 49 0.878 0.952 0.835 299 19 562.10 PGDH 477 25 12 35 0.886 0.952 0.843 478 26 68 40 0.800 0.909 0.727 477 25 899.96 PHYA 493 3 3 12 1.000 1.000 1.000 493 3 746.53 PHYBPHYD 500 22 12 35 0.943 0.962 0.907 545 63 44 86 0.907 0.952 0.864 500 22 841.73 PHYE 499 16 1 31 1.000 1.000 1.000 514 31 1 46 1.000 1.000 1.000 499 16 880.12 PIN 343 8 3 10 1.000 1.000 1.000 348 13 4 19 0.789 0.857 0.677 343 8 547.09 VRN1 592 25 1 36 0.972 0.984 0.956 592 25 1122.53 VRN2 418 20 1 27 1.000 1.000 1.000 481 45 1 91 0.989 0.986 0.975 418 20 780.61 RAX1 445 18 304 28 0.929 0.957 0.888 450 23 96 36 0.861 0.915 0.788 445 18 795.43 RAX2 RAX3 503 18 1 21 0.952 0.980 0.933 548 63 20 71 0.915 0.966 0.885 503 18 823.37 RGA RGA1 454 12 1 17 0.941 0.957 0.900 454 12 807.33 REV 378 20 1 32 0.969 0.982 0.951 388 30 2 43 0.953 0.972 0.927 378 20 702.94 RGL1 RGL2 RGL3 510 22 1 39 1.000 1.000 1.000 513 22 1 42 1.000 1.000 1.000 510 22 956.73 s6pdh 629 30 12 42 0.810 0.925 0.748 658 48 19962 90 0.700 0.861 0.603 629 30 1139.50 SPS 363 11 1 22 1.000 1.000 1.000 370 16 4 31 0.935 0.957 0.895 363 11 639.51 248

PAGE 249

T able A 13 Continued. Gene IDz Maximum parsimony Maximum parsimony (including gaps) Maximum likelihood L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) Trees (No) Length CI RI RC L (bp) Inf (No) lnL SPY 506 15 1 27 0.963 0.964 0.929 524 33 1 46 0.957 0.976 0.933 506 15 864.70 TFL1 ATC 457 14 82 21 0.857 0.949 0.814 496 53 180 65 0.877 0.974 0.854 457 14 700.82 TFL2 499 24 21 32 0.875 0.956 0.836 513 33 33824 55 0.764 0.897 0.685 499 24 814.74 3'trnV ndhC 567 29 1082 67 0.896 0.821 0.735 906 96 75 437 0.913 0.775 0.708 567 29 1078.61 trnL trnF and trnL intron 836 3 1 12 1.000 1.000 1.000 836 3 1187.19 trnQ 5'rps16 479 15 16 17 0.882 0.961 0.848 508 43 72 54 0.815 0.910 0.741 479 15 691.73 trnH psbA 335 4 2925 10 1.000 1.000 1.000 385 18 42147 62 0.968 0.963 0.932 335 4 478.89 ndhF rpl32 598 1 1 7 1.000 1.000 1.000 625 7 1 34 1.000 1.000 1.000 598 1 779.03 atpB rbcL 769 3 1 13 1.000 1.000 1.000 779 3 1 23 1.000 1.000 1.000 769 3 1086.50 ITS 611 33 2 49 0.959 0.958 0.919 653 70 9 109 0.817 0.847 0.692 611 33 1141.13 zL=total aligned sequence length. Inf=parsimony informative sites. CI=consistency index. RI=r etention index. RC=rescaled consistency index. 249

PAGE 250

Table A 1 4 Models of sequence evolution of cpDNA sequence data using JModelTest for 11 Prunus species. cpDNA Region 3'trnV ndhC trnLtrnF and trnL intron trnQ 5'rps16 trnH psbA ndhF rpl32 atpB rbcL Criterio n AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TVM+G F81+G HKY F81 F81+G F81 TrN TrN F81 F81 TPM1uf HKY Partition 012314 000000 010010 000000 000000 000000 010020 010020 000000 000000 012210 010010 lnL 998.83 1003.29 1193.21 1195.26 721.90 724.05 463.63 463.63 791.38 791.38 1089.65 1092.93 K 28 24 24 23 24 23 25 25 23 23 25 24 freqA 0.42 0.42 0.36 0.36 0.32 0.32 0.44 0.45 0.36 0.36 0.34 0.34 freqC 0.06 0.06 0.16 0.16 0.13 0.13 0.13 0.13 0.12 0.12 0.15 0.15 freqG 0.11 0.11 0.15 0.15 0.12 0. 12 0.08 0.08 0.11 0.11 0.14 0.14 freqT 0.41 0.42 0.33 0.33 0.43 0.43 0.35 0.34 0.40 0.40 0.37 0.37 R(a) [AC] 0.30 1.00 1.00 R(b) [AG] 0.32 1.86 1.51 R(c) [AT] 0.51 1.00 0.00 R(d) [CG] 0.00 1.00 0.00 R(e) [CT] 0.32 0.11 1.51 R(f) [GT] 1.00 1.00 1.00 gamma shape 0.015 0.017 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 8 4 4 3 4 3 5 5 3 3 5 4 Likelihood ratio test (LRT) 8.92 4.10 4.31 6.55 P value 0.06 0.04 0.04 0.01 zAkaike information criterion (AIC; Akaike, 1974) and Bayesian information criteria (BIC; Schwarz, 1978). 250

PAGE 251

Table A 1 5 Models of sequence evolution of diploid ITS and candidate gene regions sequence data using JModelTest for 11 Prunus species. Region ITS PGDH PGI s6pdh AXR1 BRC1 BRC2 Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TIM1 TPM1uf TIM3 F81 HKY HKY TPM1uf+G TPM1uf+G TIM1 K80 JC JC TPM3uf F81 Partition 012230 012210 012032 000000 010010 010010 012210 012210 012230 010010 000000 000000 012012 000000 lnL 1143.87 1144.91 800.94 806.57 554.32 554.32 1067.34 1067.34 420.73 431.00 445.69 445.69 476.01 480.33 K 26 25 30 27 26 26 26 26 30 25 24 24 29 27 freqA 0.19 0.19 0.28 0.2 8 0.27 0.27 0.27 0.27 0.24 0.35 0.35 freqC 0.31 0.31 0.18 0.19 0.18 0.18 0.21 0.21 0.18 0.15 0.14 freqG 0.31 0.31 0.20 0.20 0.18 0.18 0.20 0.20 0.25 0.27 0.27 freqT 0.19 0.18 0.34 0.34 0.37 0.37 0.32 0.32 0.33 0.23 0.24 R(a) [AC] 1 .00 1 .00 3.34 1 .00 1 .00 1 .00 0 .00 R(b) [AG] 2.72 2.16 1.43 2.45 2.45 1.49 1.77 R(c) [AT] 0.40 0.40 1 .00 0.18 0.18 0 .00 1 .00 R(d) [CG] 0.40 0.40 3.34 0.18 0.18 0 .00 0 .00 R(e) [CT] 1.60 2.16 6.23 2.45 2.45 6.01 1.77 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 0.01 0.01 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 6 5 6 3 4 4 6 6 6 1 0 0 5 3 Likelihood ratio test (LRT) 2.08 11.26 20.55 8.66 P value 0.15 0.0008 0.001 0.01 251

PAGE 252

Table A 15 Continued. Region CUC1A CUC1B CUC2 CUC3 LAS MAX1A MAX1B Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM1 JC TPM1uf JC TIM2+G TPM2uf+G HKY F81 TIM1 HKY HK Y HKY HKY JC Partition 012210 000000 012210 000000 010232 010212 010010 000000 012230 010010 010010 010010 010010 000000 lnL 670.90 675.76 483.54 493.47 1086.88 1088.04 901.80 903.84 622.74 625.70 534.79 534.79 552.48 562.59 K 24 22 29 24 31 30 28 27 3 0 28 28 28 28 24 freqA 0.28 0.28 0.31 0.31 0.24 0.23 0.26 0.26 0.28 freqC 0.18 0.19 0.21 0.21 0.37 0.38 0.19 0.19 0.19 freqG 0.24 0.17 0.14 0.14 0.16 0.15 0.20 0.20 0.21 freqT 0.30 0.36 0.34 0.34 0.23 0.24 0.36 0.36 0.32 R(a) [AC] 1 .00 1 .00 0.38 0.43 1 .00 R(b) [AG] 490.29 629.03 3.92 2.90 3.40 R(c) [AT] 278.81 401.92 0.38 0.43 4.38 R(d) [CG] 278.81 401.92 1 .00 1 .00 4.38 R(e) [CT] 490.29 629.03 2.17 2.90 17.22 R(f) [GT] 1 .00 1 .00 1 00 1 .00 1 .00 gamma shape 0.02 0.06 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 2 0 5 0 7 6 4 3 6 4 4 4 4 0 Likelihood ratio test (LRT) 9.71 19.86 2.32 4.09 5.92 20. 23 P value 0.008 0.001 0.13 0.04 0.05 0.0005 252

PAGE 253

Table A 15 Continued. Region MAX2 MAX3 MAX4 PIN RAX1 RAX2 RAX3 REV Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM1uf F81 TPM3uf TPM3uf TrN TrN TPM1uf JC HKY HKY TPM3uf F81 HKY+G JC Partition 012210 000000 012012 012012 010020 010020 012210 000000 010010 010010 012012 000000 010010 000000 lnL 574.49 579.74 598.92 598.92 462.86 462.86 518.29 526.33 736.74 736.74 796.63 802.36 685.16 696.10 K 29 27 29 29 29 29 29 24 28 28 29 27 29 24 freqA 0.24 0.24 0.30 0.30 0.30 0.30 0.27 0.33 0.33 0.30 0.30 0.25 freqC 0.15 0.15 0.18 0.18 0.13 0.13 0.23 0.17 0.17 0.23 0.23 0.23 freqG 0.34 0.34 0.18 0.18 0.19 0.19 0.20 0.15 0.15 0.14 0.15 0.21 freqT 0.28 0.28 0.34 0.34 0.38 0.38 0.31 0.34 0.34 0.32 0.32 0.32 R(a) [AC] 1 .00 3.87 3.87 1 .00 1 .00 1 .00 13.26 R(b) [AG] 487.95 6.48 6.48 4.84 4.84 2.32 10.39 R(c) [AT] 900.49 1 .00 1 .00 1 .00 1 .00 0 .00 1 .00 R(d) [CG] 900.49 3.87 3.87 1 .00 1 .00 0 .00 13.26 R( e) [CT] 487.95 6.48 6.48 2.04 2.04 2.32 10.39 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 0.73 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 5 3 5 5 5 5 5 0 4 4 5 3 5 0 Likelihood ratio test (LRT) 10.50 16.09 11.48 21.87 P value 0.005 0.007 0.003 0.0006 253

PAGE 254

Table A 15 Continued. Region SPS AGL24 AGL20SOC1 BFT BRM CO CRY1 Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TrNef JC TPM3uf TPM3uf F81 F81 TPM3uf HKY F81 F81 TPM3uf F81 TrN HKY Partition 010020 000000 012012 012012 000000 000000 012012 010010 000000 000000 012012 000000 010020 010010 lnL 598.18 602.68 919.44 919.44 793.33 793.33 754.18 755.39 349. 82 349.82 672.79 675.91 834.37 835.51 K 26 24 29 29 23 23 29 28 27 27 29 27 29 28 freqA 0.32 0.32 0.28 0.28 0.30 0.30 0.23 0.23 0.26 0.26 0.22 0.22 freqC 0.17 0.17 0.18 0.18 0.18 0.18 0.15 0.15 0.19 0.20 0.15 0.15 freqG 0.19 0.19 0.20 0.20 0.18 0 .18 0.16 0.16 0.15 0.15 0.20 0.20 freqT 0.33 0.33 0.34 0.34 0.34 0.34 0.45 0.45 0.40 0.39 0.43 0.43 R(a) [AC] 1 .00 0.33 0.33 3.70 4.22 1 .00 R(b) [AG] 1.54 3.78 3.78 9.88 2.88 3.90 R(c) [AT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 R(d) [CG] 1 .00 0.33 0.33 3.70 4.22 1 .00 R(e) [CT] 5.91 3.78 3.78 9.88 2.88 1.75 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 2 0 5 5 3 3 5 4 3 3 5 3 5 4 Likelihood ratio test (LRT) 9.00 2.43 6.25 2.29 P value 0.01 0.12 0.04 0.13 254

PAGE 255

Table A 15 Continued. Region CRY2 ELF6 FD FD1 FG FLC FLF FPF1 Criterion AICz BIC AIC BIC AIC BIC AIC BIC AI C BIC AIC BIC AIC BIC Model TPM2uf F81 TrN+G TrN+G TPM1uf+G F81 TPM3uf K80 TPM2uf HKY F81+G F81 TPM3 JC Partition 010212 000000 010020 010020 012210 000000 012012 010010 010212 010010 000000 000000 012012 000000 lnL 783.52 788.42 1123.01 1123.01 969.91 975.19 743.21 750.84 737.69 738.73 746.41 748.07 487.81 489.89 K 29 27 30 30 30 27 29 25 29 28 22 21 26 24 freqA 0.28 0.27 0.30 0.30 0.27 0.27 0.19 0.26 0.26 0.26 0.26 freqC 0.18 0.18 0.18 0.18 0.19 0.20 0.25 0.19 0.19 0.23 0.23 freqG 0.17 0.17 0.16 0.16 0.19 0.20 0.28 0.22 0.22 0.13 0.13 freqT 0.38 0.38 0.36 0.36 0.34 0.34 0.28 0.33 0.33 0.38 0.38 R(a) [AC] 0.29 1 .00 1 .00 1 .00 0.001 0.45 494.00 R(b) [AG] 1.23 2.73 2.73 1.41 1.99 1.89 656.79 R(c) [AT] 0.29 1 .00 1 .00 0.45 1 .00 0.45 1 .00 R(d) [CG] 1 .00 1 .00 1 .00 0.45 0.001 1 .00 494.00 R(e) [CT] 1.23 5.88 5.88 1.41 1.99 1.89 656.79 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 1 .00 1 .00 0.12 0.01 pinv(I) alph a(G) pinv(IG) alpha(IG) Free parameters 5 3 6 6 6 3 5 1 5 4 4 3 2 0 Likelihood ratio test (LRT) 9.79 10.55 15.25 2.08 3.33 4.17 P value 0.007 0.01 0.004 0.15 0.07 0.12 255

PAGE 256

Table A 15 Continued. Region FRI GIDA FT TSF GI FB HOS1 LFY MAF1 MAF3 AGL31 MAF2A Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM3uf HKY TPM3uf F81 TPM3uf F81 F81 F81 HKY+G HKY+G TPM3uf +G TPM3uf F81+G F81 Partition 012012 010010 012012 000000 012012 000000 000000 000000 010010 010010 012012 012012 000000 000000 lnL 888.68 890.74 787.17 790.60 952.97 957.07 309.35 309.35 1448.28 1448.28 867.65 870.21 438.59 440.61 K 19 18 27 25 29 27 27 27 29 29 30 29 24 23 freqA 0.27 0.27 0.24 0.24 0.25 0.25 0.27 0.27 0 .32 0.32 0.29 0.29 0.21 0.21 freqC 0.16 0.16 0.25 0.26 0.18 0.18 0.14 0.14 0.13 0.13 0.21 0.21 0.18 0.18 freqG 0.17 0.17 0.14 0.14 0.16 0.16 0.20 0.20 0.20 0.20 0.17 0.17 0.16 0.16 freqT 0.40 0.40 0.37 0.36 0.41 0.41 0.39 0.39 0.35 0.35 0.33 0.33 0.46 0 .46 R(a) [AC] 0.18 5.29 3.08 0.15 0.15 R(b) [AG] 2.21 5.29 3.23 2.18 2.17 R(c) [AT] 1 .00 1 .00 1 .00 1 .00 1 .00 R(d) [CG] 0.18 5.29 3.08 0.15 0.15 R(e) [CT] 2.21 5.29 3.23 2.18 2.17 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 0.329 0.329 0.011 0.015 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 5 4 5 3 5 3 3 3 5 5 6 5 4 3 Likelihood ratio test (LRT) 4.11 6.87 8.20 5. 13 4.04 P value 0.04 0.03 0.02 0.02 0.04 256

PAGE 257

Table A 15 Continued. Region MAF2B MAF4 MAF5 MFT PHYA PHYBPHYD PHYE Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TIM2 F81 TPM2uf HKY TPM1uf F81 TPM2uf F81 TrN+G F81 TrN+G TrN TPM3uf HKY Partition 010232 000000 010212 010010 012210 000000 010212 000000 010020 000000 010020 010020 012012 010010 lnL 536.00 539.81 935.31 937.25 532.52 534.91 699.02 701.87 738.31 743.35 820.51 821.78 843.32 845.60 K 22 19 29 28 27 25 29 27 30 2 7 28 27 29 28 freqA 0.22 0.21 0.28 0.28 0.29 0.29 0.20 0.20 0.27 0.26 0.32 0.32 0.24 0.24 freqC 0.17 0.18 0.15 0.15 0.17 0.17 0.24 0.24 0.18 0.18 0.15 0.15 0.22 0.21 freqG 0.18 0.17 0.14 0.13 0.15 0.15 0.21 0.21 0.22 0.22 0.16 0.16 0.17 0.17 freqT 0.43 0.43 0.43 0.43 0.39 0.39 0.35 0.35 0.33 0.33 0.36 0.36 0.37 0.37 R(a) [AC] 2.96 2.77 1 .00 0.32 1 .00 1 .00 1 .00 0.51 R(b) [AG] 1.32 6.73 1.10 1.27 0.01 9.45 9.32 3.16 R(c) [AT] 2.96 2.77 0.35 0.32 1 .00 1 .00 1 .00 1 .00 R(d) [CG] 1 .00 1 00 0.35 1 .00 1 .00 1 .00 1 .00 0.51 R(e) [CT] 6.55 6.73 1.10 1.27 5.69 2.31 2.30 3.16 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 0.01 0.15 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 6 3 5 4 5 3 5 3 6 3 6 5 5 4 Likelihood ratio test (LRT) 7.63 3.88 4.77 5.70 10.08 2.54 4.56 P value 0.05 0.05 0.09 0.06 0.02 0.11 0.03 2 57

PAGE 258

Table A 15 Continued. Region RGA RGA1 RGL1 RGL2 RGL3 SPY TFL1 ATC TFL2 VRN1 VRN2 Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TIM2 JC TPM2+G K80 TVM F81 TPM2uf+I TPM2uf HKY+G HKY TPM1uf F81 TPM1uf HKY Partition 010232 000000 010212 010010 012314 000000 010212 010212 010010 010010 012210 000000 012210 010010 lnL 698.85 706.84 908.84 911.10 821.93 828.25 697.93 700.83 814.55 816.35 985.54 989.35 652.40 653.49 K 30 24 27 25 31 27 30 29 29 28 29 27 29 28 freqA 0.21 0.31 0.30 0.29 0.29 0.33 0.33 0.31 0.31 0.32 0.32 freqC 0.28 0.18 0.18 0.17 0.17 0.17 0.17 0.17 0.17 0.13 0.13 freqG 0.27 0.18 0.18 0.17 0.17 0.17 0.17 0.23 0.23 0.18 0.18 freqT 0.24 0.33 0.33 0.37 0.37 0.33 0.33 0.30 0.30 0.37 0.37 R(a) [AC] 0 .00 0.41 1.23 0 .00 0.01 1 .00 1 .00 R(b) [AG] 1.03 1.94 1.48 2.41 1. 62 3.78 585.90 R(c) [AT] 0 .00 0.41 0.09 0 .00 0.01 1.60 54.37 R(d) [CG] 1 .00 1 .00 1.20 1 .00 1 .00 1.60 54.37 R(e) [CT] 3.50 1.94 1.48 2.41 1.62 3.78 585.90 R(f) [GT] 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 1 .00 gamma shape 0.18 0.01 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 6 0 3 1 7 3 6 5 5 4 5 3 5 4 Likelihood ratio test (LRT) 15.98 4.53 12.63 5.80 4.00 7.64 2.18 P value 0.01 0.10 0.01 0.02 0.05 0 .02 0.14 zAkaike information criterion (AIC; Akaike, 1974) and Bayesian information criteria (BIC; Schwarz, 1978). 258

PAGE 259

Table A 1 6 Models of s equence evolution of cpDNA sequence data (classified by haplotypes) using JModelTest for 11 Prunus species. Region 3' trnV ndhC trnL trnF and trnL intron trnQ 5'rps16 trnH psbA ndhF rpl32 atpB rbcL Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TVM+G F81+G HKY F81 F81+G F81 TrN F81 F81 F81 TPM1uf HKY Partition 012314 000000 010010 000000 000000 000000 010020 000000 000000 000000 012210 010010 lnL 998.85 1003.29 1193.21 1195.26 721.90 724.05 463.62 465.72 791.38 791.38 1089.65 1092.93 K 28 24 24 23 46 45 35 33 23 23 25 24 freqA 0.42 0.42 0.36 0.36 0.32 0.32 0.44 0.45 0.36 0.36 0.34 0.34 freqC 0.06 0.06 0.16 0.16 0.13 0.13 0.13 0.13 0.12 0.12 0.15 0.15 freqG 0.11 0.11 0.15 0.15 0.12 0.12 0.08 0.08 0.11 0.11 0.14 0.14 freqT 0.41 0.42 0.33 0.33 0.43 0.43 0.35 0.34 0.40 0.40 0.37 0.37 R(a) [AC] 0.30 1.00 1.00 R(b) [AG] 0.32 1.86 1.51 R(c) [AT] 0.51 1.00 0.00 R(d) [CG] 0.005 1.00 0.00 R(e) [CT] 0.32 0.11 1.51 R(f) [GT] 1.00 1.00 1.00 gamma shape 0.01 0.02 0.02 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 8 4 4 3 4 3 5 3 3 3 5 4 Likelihood ratio test (LRT) 8.89 4.10 4.31 4.20 6.55 P value 0.06 0.04 0.04 0.12 0.01 zAkaike information criterion (AIC; Akaike, 1974) and Bayesian information criteria (BIC; Schwarz, 1978). 259

PAGE 260

Tab le A 1 7 Models of sequence evolution of phased haplotype ITS and candidate gene regions sequence data using JModelTest for 11 Prunus species. Region ITS PGDH PGI s6pdh AXR1 BRC1 BRC2 Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Mod el TPM1uf+G HKY+G TIM3+G F81+G TPM1uf HKY HKY+G HKY+G TIM1ef TIM1ef JC JC HKY F81 Partition 012210 010010 012032 000000 012210 010010 010010 010010 012230 012230 000000 000000 010010 000000 lnL 1135.44 1137.35 895.18 902.98 568.67 569.76 1143.91 1143.91 435.51 435.51 461.29 461.29 500.29 501.32 K 44 43 57 54 43 42 75 75 33 33 32 32 30 29 freqA 0.19 0.20 0.29 0.28 0.27 0.27 0.26 0.26 0.35 0.35 freqC 0.31 0.31 0.18 0.18 0.18 0.18 0.22 0.22 0.15 0.15 freqG 0.31 0.31 0.20 0.20 0.18 0.18 0.20 0.20 0.27 0.27 freqT 0.19 0.19 0.34 0.34 0.37 0.37 0.32 0.32 0.23 0.23 R(a) [AC] 1.00 3.13 1.00 1.00 1.00 R(b) [AG] 2.43 1.98 2.18 1.32 1.32 R(c) [AT] 0.35 1.00 0.37 0.11 0.11 R(d) [CG] 0.35 3.13 0.37 0.11 0.11 R(e) [CT] 2.43 6.79 2.18 4.04 4.04 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 gamma shape 0.18 0.18 0.12 0.02 0.02 0.02 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 6 5 7 4 5 4 5 5 3 3 0 0 4 3 Likelihood ratio test (LRT) 3.83 15.60 569.76 2.06 P value 0.05 0.001 0.14 0.15 260

PAGE 261

Table A 17 Continued. Region CUC1A CUC1B CUC2 CUC3 LAS MAX1A MAX1B Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC B IC Model TIM2ef+I+G TPM2+G TPM1uf TPM1uf F81+G F81 HKY HKY TVM+G TVM+G HKY HKY TPM2uf+G JC Partition 010232 010212 012210 012210 000000 000000 010010 010010 012314 012314 010010 010010 010212 000000 lnL 739.45 742.28 536.25 536.25 500.88 501.94 989.86 989.86 678.71 678.71 570.11 570.11 580.15 594.09 K 51 49 37 37 32 31 56 56 44 44 38 38 38 32 freqA 0.28 0.28 0.24 0.24 0.31 0.31 0.23 0.23 0.26 0.26 0.28 freqC 0.18 0.18 0.24 0.23 0.21 0.21 0.37 0.37 0.19 0.19 0.19 freqG 0.24 0.24 0.11 0.11 0.14 0.14 0.15 0.15 0.20 0.20 0.20 freqT 0.30 0.30 0.41 0.41 0.34 0.34 0.24 0.24 0.35 0.35 0.33 R(a) [AC] 296.59 248.70 1 1 0.84 0.84 0.29 R(b) [AG] 477.79 704.68 5.77 5.77 5.02 5.02 1.24 R(c) [AT] 296.59 248.70 2.53 2.53 1.32 1.32 0.29 R(d) [CG] 1.00 1.00 2.53 2.53 1.92 1.92 1.00 R(e) [CT] 1252.17 704.68 5.77 5.77 5.02 5.02 1.24 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 1.00 1.00 gamma shape 0.05 0.02 0.35 100.00 100.00 0.02 pinv(I) 0.50 al pha(G) pinv(IG) alpha(IG) Free parameters 5 3 5 5 4 3 4 4 8 8 1 1 6 0 Likelihood ratio test (LRT) 5.66 2.11 27.88 P value 0.06 0.15 0.0001 261

PAGE 262

Table A 17 Continued. Region MAX2 MAX3 MAX4 PIN R AX1 RAX2 RAX3 REV Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM1uf F81 TPM3uf+G TPM3uf+G TPM1uf TPM1uf TPM1uf+G K80+G TPM1uf+G HKY+I TPM3uf F81 HKY K80 Partition 012210 000000 012012 012012 012210 012210 012210 010010 012210 010010 012012 000000 010010 010010 lnL 612.80 617.01 663.49 663.49 518.06 518.06 558.84 563.27 1307.50 1310.27 820.23 825.04 722.69 729.16 K 35 33 44 44 41 41 48 44 46 45 39 37 42 39 freqA 0.23 0.24 0.30 0.30 0.31 0.31 0.27 0.34 0.34 0.30 0.30 0.25 freqC 0.15 0.15 0.18 0.18 0.12 0.12 0.23 0.17 0.17 0.23 0.23 0.22 freqG 0.34 0.34 0.18 0.18 0.19 0.19 0.20 0.14 0.14 0.14 0.14 0.20 freqT 0.28 0.28 0.33 0.33 0.39 0.39 0.30 0.34 0.34 0.32 0.32 0.32 R(a) [AC] 1.00 2.80 2.80 1.00 1.00 1.00 1.00 6.44 R(b) [AG] 6.47 6.14 6.14 3.48 3.48 3.01 3.09 6.27 R(c) [AT] 10.20 1.00 1.00 0.45 0.45 0.31 0.50 1.00 R(d) [CG] 10.20 2.80 2.80 0.45 0.45 0.31 0.50 6.44 R(e) [CT] 6.47 6.14 6.14 3.48 3.48 3.01 3.09 6.27 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 gamma shape 0.25 0.25 0.02 0.01 0.11 pinv(I) 0.27 alpha(G) pinv(IG) alpha(IG) Free parameters 5 3 4 4 5 5 6 4 6 5 5 3 4 1 Likelihood ratio test (LRT) 8.41 8.86 5.54 9.61 12.92 P value 0.01 0.01 0.02 0.008 0.005 262

PAGE 263

Table A 17 Continued. Region SPS AGL24 AGL20SOC1 BFT BRM CO CRY1 Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model HKY+G HKY HKY+G F81+G TPM3uf HKY F81+I F81+I TPM3uf+I TPM2uf HKY HKY Partition 010010 010010 010010 000000 012012 010010 000000 000000 012012 010212 010010 010010 lnL 1078.98 1080.23 857.93 858.93 788.11 789.30 643.07 643.07 757.22 758.79 886.09 886.09 K 59 58 69 68 49 48 44 44 52 51 36 36 freqA 0.33 0.33 0.29 0.29 0.30 0.30 0.26 0.26 0.26 0.26 0.22 0.22 freqC 0.16 0.16 0.17 0.18 0.18 0.18 0.14 0.14 0.20 0.20 0.14 0.14 freqG 0.18 0.18 0.20 0.20 0.18 0.18 0.20 0.20 0.15 0.15 0.20 0.20 freqT 0.33 0.33 0.34 0.34 0.34 0. 33 0.39 0.39 0.39 0.39 0.43 0.43 R(a) [AC] 3.65 2.51 0.70 R(b) [AG] 13.31 2.09 1.12 R(c) [AT] 1.00 1.00 0.70 R(d) [CG] 3.65 2.51 1.00 R(e) [CT] 13.31 2.09 1.12 R(f) [GT] 1.00 1.00 1.00 gamma shape 0.71 0.02 0.02 pinv(I) 0.00 0.00 0.00 alpha(G) pinv(IG) alpha(IG) Free parameters 5 4 5 4 5 4 4 4 6 5 4 4 Likelihood ratio test (LRT) 2.51 2.00 2.39 3.14 P value 0.11 0.16 0.12 0.08 263

PAGE 264

Table A 17 Continued. Region CRY2 ELF6 FD FD1 FG FLC FLF FPF1 Criterion AIC z BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM3 K80 HKY+G HKY TPM1uf F81 TrN K80 TPM2uf HKY HKY F81 TPM3 JC Partition 012012 010010 010010 010010 012210 000000 010020 010010 010212 010010 010010 000000 012012 000000 lnL 1476.11 1478.05 1170.56 1172.17 1006.54 1009.15 803.56 810.82 780.92 782.73 769.82 771.44 495.07 497.78 K 52 51 73 72 47 45 41 37 39 38 62 61 28 26 freqA 0.30 0.3 0 0.27 0.27 0.19 0.26 0.26 0.25 0.25 freqC 0.17 0.17 0.19 0.19 0.24 0.19 0.19 0.23 0.24 freqG 0.16 0.16 0.20 0.20 0.28 0.22 0.22 0.12 0.12 freqT 0.36 0.36 0.34 0.34 0.28 0.33 0.33 0.39 0.40 R(a) [AC] 0.56 1.00 1.00 0.37 353.19 R(b) [AG] 1.60 1.20 3.65 1.89 577.61 R(c) [AT] 1.00 0.48 1.00 0.37 1.00 R(d) [CG] 0.56 0.48 1.00 1.00 353.19 R(e) [CT] 1.60 1.20 8.14 1.89 577.61 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 gamma shape 0.38 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 2 1 5 4 5 3 5 1 5 4 4 3 Likelihood ratio test (LRT) 3.88 3.22 5.21 14.52 3.61 3.23 P value 0.05 0.07 0.07 0.006 0.06 0.07 264

PAGE 265

Tab le A 17 Continued. Region FRIGIDA FT TSF GI FB HOS1 LFY MAF1 MAF3 AGL31 MAF2A Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TPM3uf TPM3uf TPM3uf F81 TPM3uf+G HKY F81 F81 HKY+G HKY+G TPM2uf+G HKY+G TIM1+G F81 Partition 012012 012012 012012 000000 012012 010010 000000 000000 010010 010010 010212 010010 012230 000000 lnL 1092.00 1092.00 807.88 812.06 1071.66 1074.71 324.16 324.16 1757.52 1757.52 975.20 976.40 456.56 461.43 K 31 31 49 47 70 68 33 33 79 79 42 41 35 31 freqA 0. 29 0.29 0.23 0.23 0.25 0.25 0.27 0.27 0.33 0.33 0.29 0.29 0.21 0.21 freqC 0.16 0.16 0.25 0.26 0.18 0.18 0.14 0.14 0.14 0.14 0.20 0.20 0.18 0.18 freqG 0.17 0.17 0.14 0.14 0.16 0.16 0.20 0.20 0.19 0.19 0.17 0.17 0.15 0.16 freqT 0.39 0.39 0.37 0.36 0.41 0. 41 0.39 0.39 0.34 0.34 0.33 0.33 0.46 0.46 R(a) [AC] 0.13 0.13 6.08 2.35 0.46 1.00 R(b) [AG] 1.73 1.73 5.89 3.29 1.97 1.67 R(c) [AT] 1.00 1.00 1.00 1.00 0.46 0.33 R(d) [CG] 0.13 0.13 6.08 2.35 1.00 0.33 R(e) [CT] 1.73 1 .73 5.89 3.29 1.97 0.33 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 1.00 gamma shape 0.44 0.17 0.17 0.02 0.02 0.16 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 5 5 5 3 6 4 3 3 5 5 6 5 7 3 Likelihood ratio test (LRT) 8.37 6.09 2.39 9.75 P value 0.02 0.05 0.12 0.04 265

PAGE 266

Table A 17 Continued. Region MAF2B MAF4 MAF5 MFT PHYA PHYBPHYD PHYE Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model TIM2 F81 TrN+G HKY+G TPM3uf F81 GTR+G JC TPM2uf+I+G F81 TrN+G TrN HKY HKY Partition 010232 000000 010020 010010 012012 000000 012345 000000 010212 000000 010020 010020 010010 010010 lnL 547.23 551.12 1243.79 1245.46 572.91 575.79 748.08 766.73 753.31 75 9.70 877.01 878.87 886.48 886.48 K 24 21 84 83 35 33 47 38 37 33 44 43 34 34 freqA 0.22 0.22 0.28 0.29 0.28 0.28 0.20 0.27 0.27 0.32 0.32 0.24 0.24 freqC 0.17 0.18 0.15 0.14 0.17 0.17 0.24 0.18 0.18 0.15 0.15 0.21 0.21 freqG 0.18 0.17 0.14 0.14 0.15 0.15 0.21 0.22 0.22 0.17 0.17 0.17 0.17 freqT 0.43 0.43 0.43 0.42 0.40 0.39 0.35 0.33 0.33 0.36 0.36 0.37 0.37 R(a) [AC] 3.37 1.00 2.16 0.42 0.19 1.00 1.00 R(b) [AG] 1.35 4.14 2.84 0.49 1.64 7.10 6.37 R(c) [AT] 3.37 1.00 1.00 0.14 0 .19 1.00 1.00 R(d) [CG] 1.00 1.00 2.16 1.51 1.00 1.00 1.00 R(e) [CT] 6.32 2.24 2.84 1.49 1.64 1.34 1.43 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 1.00 1.00 gamma shape 0.11 0.12 100.00 0.01 0.14 pinv(I) 0.55 alp ha(G) pinv(IG) alpha(IG) Free parameters 6 3 6 5 5 3 9 0 7 3 6 5 Likelihood ratio test (LRT) 7.78 3.34 5.75 37.29 12.78 3.73 P value 0.05 0.07 0.06 0.00 0.01 0.05 266

PAGE 267

Table A 17 Continued. Region RGA RGA 1 RGL1RGL2RGL3 SPY TFL1 ATC TFL2 VRN1 VRN2 Criterion AICz BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC AIC BIC Model K80+G K80+G TPM2 K80 TPM1uf+G F81 TIM1+G F81 TPM3uf+G TPM3uf+G TPM1uf TPM1uf HKY HKY Partition 010010 010010 010212 010010 012210 00000 0 012230 000000 012012 012012 012210 012210 010010 010010 lnL 758.79 758.79 958.45 960.06 872.92 877.19 751.39 759.55 895.99 895.99 1059.85 1059.85 718.19 718.19 K 42 42 30 29 40 37 59 55 70 70 39 39 44 44 freqA 0.30 0.30 0.29 0.28 0.33 0.33 0.31 0 .31 0.31 0.31 freqC 0.18 0.18 0.17 0.17 0.17 0.17 0.17 0.17 0.14 0.14 freqG 0.18 0.18 0.18 0.18 0.16 0.16 0.23 0.23 0.19 0.19 freqT 0.33 0.33 0.37 0.37 0.34 0.34 0.30 0.30 0.36 0.36 R(a) [AC] 0.40 1.00 1.00 0.20 0.20 1.00 1.00 R(b) [AG] 1.68 1.09 4.43 4.42 4.42 4.64 4.64 R(c) [AT] 0.40 0.30 0.83 1.00 1.00 1.47 1.47 R(d) [CG] 1.00 0.30 0.83 0.20 0.20 1.47 1.47 R(e) [CT] 1.68 1.09 5.60 4.42 4.42 4.64 4.64 R(f) [GT] 1.00 1.00 1.00 1.00 1.00 1.00 1.00 gamma shape 0.02 0.02 0.20 0.01 0.02 0.02 pinv(I) alpha(G) pinv(IG) alpha(IG) Free parameters 2 2 2 1 6 3 7 3 6 6 5 5 4 4 Likelihood ratio test (LRT) 3.23 8.54 16.32 P value 0.07 0.04 0.003 zAkaike information criterion (AIC; Akaike, 1974) and Bayesian information criteria (BIC; Schwarz, 1978). 267

PAGE 268

Table A 18 List of Prunus accessions collected and available at the Florida Museum of Natural History, Gainesville, FL. Vou cher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 1 Prunus umbellata PumbFl01 US FL 29 36.690 N 82 21.271 W 2 Prunus umbellata PumbFl02 US FL 29 36.666 N 82 21.257 W 3 Prunus umbellata PumbFl03 US F L 29 38.082 N 82 21.321 W 4 Prunus umbellata PumbFl04 US FL 29 38.071 N 82 21.316 W 5 Prunus umbellata PumbFl05 US FL 29 37.104 N 82 20.426 W 6 Prunus umbellata PumbFl06 US FL 29 31.245 N 82 18.005 W 7 Prunus angustifolia PangFl07 U S FL 29 38.882 N 82 29.756 W 8 Prunus umbellata PumbFl08 US FL 29 31.191 N 82 18.739 W 9 Prunus umbellata PumbFl09 US FL 29 31.183 N 82 18.817 W 10 Prunus umbellata PumbFl10 US FL 29 31.020 N 82 18.624 W 11 Prunus umbellata PumbFl11 US FL 29 31.266 N 82 18.595 W 12 Prunus americana PameFL12 US FL 29 28.965 N 82 14.525 W 13 Prunus umbellata PumbGA13 US GA 30 52.768 N 84 33.876 W 14 Prunus umbellata PumbGA14 US GA 30 52.768 N 84 33.876 W 15 Prunus pumila PpumIL15 var. susquehanue US IL 41 20.972 N 95 59.167 W 16 Prunus pumila PpumIL16 var. susquehanue US IL 41 20.972 N 95 59.167 W 17 Prunus umbellata PumbFl17 US FL 29 01.771 N 82 09.549 W 18 Prunus umbellata PumbFl18 US FL 28 45.237 N 82 05.840 W 19 Prunus umbellata PumbFl19 US FL 28 39.404 N 82 03.279 W 20 Prunus umbellata PumbFl20 US FL 28 39.415 N 82 03.280 W 21 Prunus umbellata PumbFl21 US FL 28 18.163 N 82 08.330 W 22 Prunus angustifolia PangFl22 US FL 30 20.970 N 83 04 .476 W 23 Prunus umbellata PumbFl23 US FL 30 21.178 N 83 09.957 W 24 Prunus umbellata PumbFl24 US FL 30 36.013 N 84 25.068 W 25 Prunus angustifolia PangGA25 US GA 30 43.451 N 84 25.695 W 26 Prunus americana PameGA26 US GA 30 44.310 N 84 27.565 W 27 Prunus umbellata PumbFl27 US FL 30 34.705 N 84 21.526 W 28 Prunus umbellata PumbFl28 US FL 30 29.296 N 84 19.133 W 29 Prunus angustifolia PangFl29 US FL 30 28.341 N 83 39.564 W 268

PAGE 269

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 30 Prunus angustifolia PangFl30 US FL 30 28.198 N 83 37.010 W 31 Prunus umbellata x angustifolia PumbxangFl31 US FL 30 28.471 N 83 33.678 W 32 Prunus umbellata PumbF l32 US FL 30 29.330 N 83 30.021 W 33 Prunus umbellata PumbFl33 US FL 30 28.156 N 83 24.614 W 34 Prunus umbellata PumbFl34 US FL 30 23.505 N 83 11.592 W 35 Prunus umbellata PumbFl35 US FL 30 23.505 N 83 11.592 W 36 Prunus umbellata PumbFl36 US FL 27 59.255 N 82 19.553 W 37 Prunus umbellata PumbFl37 US FL 28 04.565 N 82 21.177 W 38 Prunus umbellata PumbFl38 US FL 28 25.191 N 82 16.707 W 39 Prunus umbellata PumbFl39 US FL 28 24.809 N 82 14.515 W 40 Prunus umbellata PumbFl40 US FL 28 23.413 N 82 13.255 W 41 Prunus umbellata PumbFl41 US FL 28 28.530 N 82 15.417 W 42 Prunus umbellata PumbFl42 US FL 28 35.077 N 82 12.621 W 43 Prunus umbellata PumbFl43 US FL 28 54.114 N 82 06.243 W 44 Prunus umb ellata PumbFl44 US FL 28 59.051 N 82 08.241 W 45 Prunus umbellata PumbFl45 US FL 29 00.482 N 82 09.095 W 46 Prunus umbellata PumbFl46 US FL 29 06.315 N 82 11.070 W 47 Prunus umbellata PumbFl47 US FL 30 22.203 N 83 14.816 W 48 Prunus angustifolia PangGA48 US GA 30 42.393 N 84 23.527 W 49 Prunus umbellata PumbGA49 US GA 30 45.922 N 84 28.905 W 50 Prunus umbellata PumbGA50 US GA 30 45.748 N 84 28.717 W 51 Prunus umbellata PumbGA51 US GA 30 45.729 N 84 28.656 W 52 Prunus umbellata PumbGA52 US GA 30 45.972 N 84 28.913 W 53 Prunus umbellata PumbGA53 US GA 30 45.981 N 84 28.922 W 54 Prunus umbellata PumbGA54 US GA 30 45.986 N 84 28.926 W 55 Prunus umbellata PumbGA55 US GA 30 45.740 N 84 28.804 W 56 Prunus angustifolia PangGA56 US GA 30 46.326 N 84 28.845 W 57 Prunus angustifolia PangGA57 US GA 30 42.656 N 84 23.928 W 58 Prunus umbellata PumbFl58 US FL 30 33.039 N 84 22.883 W 269

PAGE 270

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 59 Prunus umbellata PumbFl59 US FL 30 22.566 N 83 16.991 W 60 Prunus angustifolia PangFl60 US FL 30 22.578 N 83 17.037 W EX1 Prunus angustifolia PangFlEX1 US FL 30 40.203 N 84 15.018 W EX2 Prunus angustifolia PangFlEX2 US FL 30 40.203 N 84 15.018 W 61 Prunus sargentii PsarPA61 US PA 40 48.356 N 77 52.136 W 62 Prunus sargentii PsarPA62 US PA 40 48.356 N 77 52.136 W 63 Prunus virginiana PvirPA63 US PA 40 48.341 N 77 52.102 W 64 Prunus serotina PseroPA64 US PA 40 48.026 N 77 52.004 W 65 Prunus serotina PseroPA65 US PA 40 48.033 N 77 52.001 W 66 Prunus sargentii PsarPA66 US PA 40 47.841 N 77 51.977 W 67 Prunus subhirtella PsubtPA67 US PA 40 47.847 N 77 51.679 W 68 Prunus subhirtella PsubtPA68 US PA 40 47.847 N 77 51.679 W 69 Prunus serrulata PserrPA69 US PA 40 47.914 N 77 51.647 W 70 Prunus x. yedoensis PxyedPA70 US PA 40 47.788 N 77 52.231 W 71 Prunus americana PamePA71 US PA 40 47.788 N 77 51.534 W 72 Prunus serotina PseroPA72 US PA 40 18.960 N 79 34.760 W 73 Prunus americana PamePA73 US PA 40 18.960 N 79 34.760 W 74 Prunus umbellata PumbFl74 US FL 30 23.222 N 82 52.401 W 75 Prunus umbellata PumbFl75 US FL 30 29.414 N 83 1.123 W 76 Prunus americana PameGA76 US GA 32 39.366 N 83 42.657 W 77 Prunus alleghaniensis PallNY77 US NY 40 51.508 N 73 52.862 W 78 Prunus alleghaniensis PallMI78 US MI 79 Prunus americana x mariti ma PamexmarMN79 US MN 80 Prunus americana PameGA80 US GA 32 29.186 N 83 35.194 W 81 Prunus americana x maritima PamexmarMN81 US MN 82 Prunus hortulana PhorWA82 US WA 35 34.982 N 82 32.019 W 83 Prunus angustifolia PangKS83 US KS 84 Prunus rivularis PrivLA84 US LA 85 Prunus mexicana PmexLA85 US LA 270

PAGE 271

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 86 Prunus maritima PmarNY86 US NY 40 51.508 N 73 52.862 W 87 Prunus maritima PmarNY87 US NY 40 51.508 N 73 52.862 W 88 Prunus munsoniana PmunTX88 US TX 89 Prunus hortulana PhorMO89 US MO 90 Prunus angustifolia PangGA90 US GA 91 Prunus andersonii x salicina PandxsalCA91 US CA 92 Prunus pumila PpumDK92 var. susquehanue US DK 93 Prunus triloba PtriCH93 US CH 94 Prunus umbellata PumbFl94 US FL 27 30.024 N 81 25.513 W 95 Prunus umbellata PumbFl95 US FL 30 28.428 N 84 17.143 W 96 Prunus umbellata PumbGA96 US GA 97 Prunus pumila PpumPA97 var. depressa US PA 98 Prunus umbellata PumbFl98 US FL 30 24.572 N 83 58.010 W 99 Prunus angustifolia PangFl99 US FL 30 24.694 N 83 55.763 W 100 Prunus angustifolia PangFl100 US FL 30 23.121 N 83 49. 859 W 101 Prunus angustifolia PangFl101 US FL 30 10.269 N 83 37.808 W 102 Prunus angustifolia PangFl102 US FL 30 4.882 N 83 30.2081 W 103 Prunus umbellata PumbFL103 US FL 30 6.986 N 83 15.572 W 104 Prunus umbellata PumbFl104 US FL 30 3.173 N 83 8.949 W 105 Prunus umbellata PumbFl105 US FL 29 59.598 N 83 0.648 W 106 Prunus umbellata PumbFl106 US FL 29 57.136 N 82 52.952 W 107 Prunus umbellata PumbFl107 US FL 29 55.615 N 82 43.269 W 108 Prunus angustifolia PangFl108 US FL 29 39.688 N 82 36.395 W 109 Prunus umbellata x angustifolia PumbxangFL109 US FL 29 37.939 N 82 36.312 W 110 Prunus angustifolia PangFl110 US FL 29 37.939 N 82 36.312 W 111 Prunus umbellata PumbFl111 US FL 29 33.417 N 82 29.262 W 112 Prunus angustifolia x umbellata PangxumbFl112 US FL 29 35.358 N 82 26.348 W 113 Prunus armeniaca cv. Canino Parm113 US 114 Prunus geniculata PgenFl114 US FL 27 56.242 N 81 34.564 W 271

PAGE 272

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 115 Prunus geniculata PgenFl115 US FL 27 56.242 N 81 34.564 W 116 Prunus mume PmumJP116 US FL 117 Prunus mume PmumTW117 US FL 118 Prunus mume PmumJP118 US FL 1 19 Prunus mume PmumJP119 US FL 120 Prunus persica cv. Okinawa PperJP120 US FL 121 Prunus persica cv. Okinawa PperJP121 US FL 122 Prunus angustifolia PangFl122 US FL 29 38.484 N 81 41.278 W 123 Prunus angustifolia PangFl123 US FL 29 37.713 N 81 52.153 W 124 Prunus umbellata PumbFl124 US FL 29 37.435 N 81 55.040 W 125 Prunus umbellata PumbFl125 US FL 29 36.575 N 82 0.687 W 126 Prunus umbellata PumbFL126 US FL 29 35.924 N 82 7.054 W 127 Prunus umbellata PumbFl127 US FL 29 37.290 N 82 15.408 W 130 Prunus fasciculata PfasCA130 var. fasciculata US CA 33 5.643 N 116 27.857 W 131 Prunus fasciculata PfasCA131 var. fasciculata US CA 33 5.604 N 116 29.073 W 132 Prunus fasciculata PfasCA132 var. fasciculata US CA 33 5.819 N 116 28.477 W 133 Prunus avium PaviCA133 US CA 33 5.621 N 116 37.742 W 134 Prunus ilicifolia PiliCA134 US CA 32 45.266 N 117 11.777 W 135 Prunus ilicifolia PiliCA135 US CA 32 45.266 N 117 11.777 W 136 Prunus cerasifer a PcerCA136 US CA 33 30.222 N 117 7.313 W 137 Prunus cerasifera PcerCA137 US CA 33 30.222 N 117 7.313 W 138 Prunus cerasifera PcerCA138 US CA 33 30.222 N 117 7.313 W 139 Prunus glandulosa PglaCA139 US CA 34 7.206 N 116 25.707 W 140 Prunus ilicifolia PiliCA140 var. integrifolia US CA 34 7.206 N 116 25.707 W 141 Prunus eremophila PereCA141 US CA 35 02.727 N 115 08.650 W 142 Prunus umbellata PumbFl142 US FL 29 38.395 N 82 25.372 W 143 Prunus umbellata PumbFl143 US FL 29 50.147 N 82 15.809 W 144 Prunus umbellata PumbFl144 US FL 28 45.292 N 82 6.134 W 145 Prunus umbellata PumbFl145 US FL 28 32.842 N 81 43.265 W 272

PAGE 273

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 146 Prunus geniculata PgenFl146 US FL 28 32.709 N 81 41.861 W 147 Prunus geniculata PgenFl147 US FL 28 32.703 N 81 41.869 W 148 Prunus geniculata PgenFl148 US FL 28 32.696 N 81 41.880 W 149 Prunus geniculata PgenFl149 US FL 28 32.706 N 81 41.884 W 150 Prunus geniculata PgenFl150 US FL 28 32.706 N 81 41.888 W 151 Prunus umbellata PumbFL151 US FL 28 32.825 N 81 39.535 W 152 Prunus umbellata PumbFl152 US FL 28 37.523 N 81 44.640 W 153 Prunus americana x umbellat a Pame x umbFl153 US FL 29 36.829 N 82 23.633 W 154 Prunus angustifolia PangGA154 US GA 32 48.148 N 83 48.459 W 155 Prunus umbellata PumbGA155 US GA 32 48.111 N 83 48.418 W 156 Prunus umbellata PumbGA156 US GA 32 47.962 N 83 47.542 W 157 Prun us umbellata PumbGA157 US GA 32 47.962 N 83 47.542 W 158 Prunus angustifolia PangGA158 US GA 32 15.431 N 83 44.941 W 159 Prunus angustifolia PangGA159 US GA 31 55.687 N 83 44.810 W 160 Prunus angustifolia PangGA160 US GA 31 39.553 N 83 36.191 W 161 Prunus angustifolia PangGA161 US GA 30 58.241 N 83 22.914 W 162 Prunus angustifolia PangFl162 US FL 30 3.947 N 82 37.687 W 163 Prunus umbellata PumbFL163 US FL 28 32.842 N 81 43.265 W 164 Prunus umbellata PumbFL164 US FL 28 32.842 N 81 43.265 W 165 Prunus umbellata PumbFL165 US FL 28 32.842 N 81 43.256 W 166 Prunus umbellata PumbFL166 US FL 28 32.842 N 81 43.242 W 167 Prunus umbellata PumbFL167 US FL 28 32.839 N 81 43.227 W 168 Prunus umbellata PumbFL168 US FL 28 32.834 N 81 43.225 W 169 Prunus umbellata PumbFL169 US FL 28 32.831 N 81 43.234 W 170 Prunus umbellata PumbFL170 US FL 28 32.828 N 81 43.244 W 171 Prunus umbellata PumbFL171 US FL 28 32.828 N 81 43.254 W 172 Prunus umbellata PumbFL172 US FL 28 32.829 N 81 43.257 W 173 Prunus umbellata PumbFL173 US FL 28 32.827 N 81 43.267 W 174 Prunus geniculata PgenFl174 US FL 28 29.538 N 81 39.328 W 273

PAGE 274

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Lat itude Longitude 175 Prunus geniculata PgenFl175 US FL 28 29.396 N 81 39.367 W 176 Prunus geniculata PgenFl176 US FL 28 29.378 N 81 39.361 W 177 Prunus geniculata PgenFl177 US FL 28 29.373 N 81 39.345 W 178 Prunus geniculata PgenFl178 US FL 28 29.365 N 81 39.346 W 179 Prunus geniculata PgenFl179 US FL 28 29.410 N 81 39.386 W 180 Prunus geniculata PgenFl180 US FL 28 29.395 N 81 39.432 W 181 Prunus geniculata PgenFl181 US FL 28 29.365 N 81 39.433 W 182 Prunus geniculata PgenFl182 US FL 28 32.712 N 81 41.859 W 183 Prunus geniculata PgenFl183 US FL 28 32.716 N 81 41.863 W 184 Prunus geniculata PgenFl184 US FL 28 32.719 N 81 41.875 W 185 Prunus geniculata PgenFl185 US FL 28 32.714 N 81 41.889 W 186 Prunus geniculata Pg enFl186 US FL 28 32.707 N 81 41.896 W 187 Prunus geniculata PgenFl187 US FL 28 32.716 N 81 41.850 W 188 Prunus geniculata PgenFl188 US FL 27 51.529 N 81 26.675 W 189 Prunus geniculata PgenFl189 US FL 27 45.751 N 81 27.234 W 190 Prunus genic ulata PgenFl190 US FL 27 45.751 N 81 27.234 W 191 Prunus geniculata PgenFl191 US FL 27 45.751 N 81 27.234 W 192 Prunus geniculata PgenFl192 US FL 27 45.906 N 81 26.990 W 193 Prunus geniculata PgenFl193 US FL 27 45.903 N 81 26.985 W 194 Prun us geniculata PgenFl194 US FL 27 41.608 N 81 29.539 W 195 Prunus geniculata PgenFl195 US FL 27 41.607 N 81 29.509 W 196 Prunus geniculata PgenFl196 US FL 27 41.609 N 81 29.113 W 197 Prunus geniculata PgenFl197 US FL 27 41.607 N 81 29.479 W 198 Prunus geniculata PgenFl198 US FL 27 48.828 N 81 28.905 W 199 Prunus geniculata PgenFl199 US FL 27 48.853 N 81 28.916 W 200 Prunus geniculata PgenFl200 US FL 27 48.707 N 81 28.732 W 201 Prunus geniculata PgenFL201 US FL 27 48.709 N 81 28. 731 W 202 Prunus geniculata PgenFL202 US FL 27 48.719 N 81 28.726 W 203 Prunus geniculata PgenFL203 US FL 27 48.719 N 81 28.726 W 274

PAGE 275

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 204 Prunus geniculata PgenFL204 US FL 27 48.724 N 81 28.724 W 205 Prunus geniculata PgenFL205 US FL 27 49.822 N 81 28.357 W 206 Prunus geniculata PgenFL206 US FL 27 49.822 N 81 28.357 W 207 Prunus geniculata PgenFL207 US FL 27 49.848 N 8 1 28.168 W 208 Prunus geniculata PgenFL208 US FL 27 49.530 N 81 27.787 W 209 Prunus geniculata PgenFL209 US FL 27 49.288 N 81 28.109 W 210 Prunus geniculata PgenFL210 US FL 27 49.313 N 81 28.276 W 211 Prunus geniculata PgenFL211 US FL 27 56.001 N 81 34.607 W 212 Prunus geniculata PgenFL212 US FL 27 55.972 N 81 34.673 W 213 Prunus geniculata PgenFL213 US FL 27 55.957 N 81 34.696 W 214 Prunus geniculata PgenFL214 US FL 27 55.957 N 81 34.696 W 215 Prunus geniculata PgenFL215 US FL 27 56.151 N 81 34.783 W 216 Prunus geniculata PgenFL216 US FL 27 56.182 N 81 34.796 W 217 Prunus geniculata PgenFL217 US FL 27 56.415 N 81 34.702 W 218 Prunus geniculata PgenFL218 US FL 27 56.398 N 81 34.618 W 219 Prunus geniculata PgenFL219 US FL 27 56.394 N 81 34.610 W 220 Prunus geniculata PgenFL220 US FL 27 56.285 N 81 34.419 W 221 Prunus geniculata PgenFL221 US FL 27 56.278 N 81 34.421 W 222 Prunus geniculata PgenFL222 US FL 27 56.250 N 81 34.380 W 223 Prunus geniculata PgenFL223 US FL 27 56.239 N 81 34.386 W 224 Prunus geniculata PgenFL224 US FL 27 56.251 N 81 34.349 W 225 Prunus geniculata PgenFL225 US FL 27 56.260 N 81 34.347 W 226 Prunus geniculata PgenFL226 US FL 27 56.270 N 81 34.378 W 227 Prunus geniculata PgenFL227 US FL 27 56.198 N 81 34.321 W 228 Prunus geniculata PgenFL228 US FL 27 55.981 N 81 34.178 W 229 Prunus geniculata PgenFL229 US FL 27 55.753 N 81 34.163 W 230 Prunus geniculata PgenFL230 US FL 27 55.731 N 81 34.164 W 231 Prunus geniculata PgenFL231 US FL 27 55.743 N 81 34.146 W 232 Prunus geniculata PgenFL232 US FL 27 55.726 N 81 34.088 W 275

PAGE 276

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 233 Prun us geniculata PgenFL233 US FL 27 55.701 N 81 34.079 W 234 Prunus geniculata PgenFL234 US FL 27 55.686 N 81 34.067 W 235 Prunus geniculata PgenFL235 US FL 27 55.696 N 81 34.097 W 236 Prunus geniculata PgenFL236 US FL 27 55.687 N 81 34.087 W 237 Prunus americana PameGA237 US GA 30 44.260 N 84 27.512 W 238 Prunus geniculata PgenFL238 US FL 27 8.524 N 81 22.030 W 239 Prunus geniculata PgenFL239 US FL 27 8.524 N 81 22.030 W 240 Prunus geniculata PgenFL240 US FL 27 8.535 N 81 22.014 W 241 Prunus geniculata PgenFL241 US FL 27 31.262 N 81 24.513 W 242 Prunus geniculata PgenFL242 US FL 27 31.263 N 81 24.518 W 243 Prunus geniculata PgenFL243 US FL 27 31.260 N 81 24.512 W 244 Prunus geniculata PgenFL244 US FL 27 33.534 N 81 24.375 W 245 Prunus geniculata PgenFL245 US FL 27 32.678 N 81 27.209 W 246 Prunus geniculata PgenFL246 US FL 27 32.678 N 81 27.209 W 247 Prunus umbellata PumbGA247 US GA 30 45.748 N 84 28.717 W 248 Prunus angustifolia PangFL248 US FL 30 20.582 N 83 5.148 W 249 Prunus munsoniana PmunGA249 US GA 30 45.748 N 84 28.717 W 250 Prunus mexicana PmexKS250 US KS 37.7395 N 97.2651 W 251 Prunus americana PameKS251 US KS 37.7401 N 97.2639 W 252 Prunus mexicana PmexKS252 US KS 37.7429 N 97.2678 W 253 Prunus virginiana PvirKS253 US KS 37.7430 N 97.2708 W 254 Prunus americana PameKS254 US KS 37.7427 N 97.2719 W 255 Prunus mexicana PmexKS255 US KS 37.7425 N 97.2728 W 256 Prunus angustifolia PangKS256 US KS 37.7440 N 97.2678 W 257 Prunus mexicana PmexKS257 US KS 37.7246 N 97.2002 W 258 Prunus americana PameKS258 US KS 37.7279 N 97.2014 W 259 Prunus andersonii PandCA259 US CA 38 3.324 N 119 7.321 W 260 Prunus andersonii PandCA260 US CA 38 3.314 N 119 7.335 W 261 Prunus andersonii PandCA261 US CA 38 4.167 N 119 4.731 W 276

PAGE 277

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 262 Prunus andersonii PandCA262 US CA 37 44.908 N 118 55.764 W 263 Prunus andersonii PandCA263 US CA 37 8.054 N 118 20.042 W 264 Prunus fasciculata PfasCA264 var. punctata US CA 35 5.558 N 118 18.444 W 265 Prunus fasciculata PfasCA265 var. punctata US CA 35 5.558 N 118 18.444 W 266 Prunus fasciculata PfasCA266 v ar. punctata US CA 35 4.752 N 118 19.880 W 267 Prunus fasciculata PfasCA267 var. punctata US CA 35 5.478 N 118 18.569 W 268 Prunus fasciculata PfasCA268 var. fasciculata US CA 33 5.823 N 116 28.464 W 269 Prunus fasciculata PfasCA269 var. fasciculata US CA 33 5.774 N 116 28.376 W 270 Prunus fasciculata PfasCA270 var. fasciculata US CA 33 5.754 N 116 28.332 W 271 Prunus fremontii PfreCA271 US CA 33 5.533 N 116 29.327 W 272 Prunus fremontii PfreCA272 US CA 33 5.498 N 116 29.466 W 273 Prunus f remontii PfreCA273 US CA 33 5.517 N 116 29.445 W 274 Prunus fremontii PfreCA274 US CA 33 12.855 N 116 27.234 W 275 Prunus havardii PhavTX275 US TX 30 40.561 N 101 40.378 W 276 Prunus havardii PhavTX276 US TX 30 40.556 N 101 40.396 W 277 Pru nus havardii PhavTX277 US TX 30 40.607 N 101 40.368 W 278 Prunus rivularis PrivpubTX278 var. pubescens US TX 279 Prunus rivularis PrivpubTX279 var. pubescens US TX 280 Prunus rivularis PrivpubTX280 var. pubescens US TX 281 Prunus rivularis PrivpubTX281 var. pubescens US TX 282 Prunus rivularis PrivpubTX282 var. pubescens US TX 283 Prunus rivularis PrivpubTX283 var. pubescens US TX 284 Prunus rivularis PrivpubTX284 var. pubescens US TX 285 Prunus rivularis PrivpubTX285 var. pubescens US TX 286 Prunus rivularis PrivpubTX286 var. pubescens US TX 287 Prunus rivularis PrivrivTX287 var. rivularis US TX 30 10.566 N 98 29.586 W 288 Prunus rivularis PrivrivTX288 var. rivularis US TX 30 10.566 N 98 29.586 W 289 Prunus rivula ris PrivrivTX289 var. rivularis US TX 30 10.566 N 98 29.586 W 290 Prunus rivularis PrivrivTX290 var. rivularis US TX 30 14.541 N 98 23.171 W 277

PAGE 278

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 291 Prunus minutiflora PminTX291 US TX 30 35.567 N 98 17.935 W 292 Prunus minutiflora PminTX292 US TX 30 35.576 N 98 17.971 W 293 Prunus minutiflora PminTX293 US TX 30 35.589 N 98 17.971 W 294 Prunus texana PtexTX294 US TX 30 45.070 N 98 22.479 W 295 Prunus texana PtexTX295 US TX 30 45.070 N 98 22.479 W 296 Prunus texana PtexTX296 US TX 30 45.070 N 98 22.479 W 297 Prunus texana PtexTX297 US TX 30 42.809 N 98 49.547 W 298 Prunus texana PtexTX298 US TX 30 42.809 N 98 49.547 W 299 Prunus rivularis PrivrivTX299 var. rivularis US TX 30 49.395 N 99 22.971 W 300 Prunus rivularis PrivpubTX300 var. pubescens US TX 31 11.851 N 100 30.038 W 301 Prunus rivularis PrivpubTX301 var. pubescens US TX 30 33.601 N 100 38.141 W 302 Prunus angustifolia PangTX302 US TX 30 17.209 N 97 13.947 W 303 Prunus gracilis PgraTX303 US TX 30 16.544 N 97 12.221 W 304 Prunus gracilis PgraTX304 US TX 30 16.551 N 97 12.225 W 305 Prunus gracilis PgraTX305 US TX 30 42.661 N 96 58.682 W 306 Prunus angustifolia x gracilis PangxgraTX306 US TX 30 42.512 N 96 58.618 W 307 Prunus gracilis PgraTX307 US TX 30 42.512 N 96 58.618 W 308 Prunus gracilis PgraTX308 US TX 30 42.522 N 96 58.624 W 309 Prunus angustifolia PangTX309 US TX 3 2 51.143 N 94 41.676 W 310 Prunus americana PrunTX309 US TX 32 57.616 N 94 27.320 W 311 Prunus angustifolia PangAR311 US AR 33 20.735 N 93 11.198 W 312 Prunus mexicana PmexAR312 US AR 33 20.803 N 93 11.137 W 313 Prunus mexicana PmexAR313 US AR 33 20.912 N 93 11.077 W 314 Prunus mexicana PmexAR314 US AR 35 20.214 N 93 8.239 W 315 Prunus mexicana PmexAR315 US AR 35 18.088 N 93 9.792 W 316 Prunus hortulana PhorAR316 US AR 36 10.715 N 94 20.738 W 317 Prunus americana PameAR317 US AR 36 10.745 N 94 20.636 W 318 Prunus hortulana PhorAR318 US AR 36 10.547 N 94 21.495 W 319 Prunus hortulana PhorAR319 US AR 36 12.611 N 94 27.963 W 278

PAGE 279

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country St ate Latitude Longitude 320 Prunus hortulana PhorAR320 US AR 36 12.619 N 94 27.965 W 321 Prunus angustifolia PangAR321 US AR 35 33.529 N 94 13.530 W 322 Prunus americana PameAR322 US AR 34 47.950 N 91 49.842 W 323 Prunus pumila PpumsusMA323 va r. susquehanue US MA 42 33.442 N 72 31.428 W 324 Prunus pumila PpumsusMA324 var. susquehanue US MA 42 33.722 N 72 31.732 W 325 Prunus pumila PpumdepMA325 var. depressa US MA 42 35.702 N 72 34.804 W 326 Prunus pumila PpumdepMA326 var. depressa US MA 4 2 35.701 N 72 34.817 W 327 Prunus pumila PpumsusMA327 var. susquehanue US MA 42 33.865 N 71 35.785 W 328 Prunus pumila PpumsusMA328 var. susquehanue US MA 42 33.858 N 71 35.810 W 329 Prunus pumila PpumsusMA329 var. susquehanue US MA 41 44.383 N 70 35.049 W 330 Prunus maritima PmarMA330 var. maritima US MA 41 33.098 N 70 30.552 W 331 Prunus maritima PmarMA331 var. maritima US MA 41 33.098 N 70 30.608 W 332 Prunus maritima PmarMA332 var. maritima US MA 41 33.092 N 70 30.709 W 333 Prunus maritima PmarMA333 var. maritima US MA 41 33.092 N 70 30.709 W 334 Prunus maritima PmarMA334 var. maritima US MA 41 33.054 N 70 30.987 W 335 Prunus nigra Pnig335 Canada MB 336 Prunus cerasifera Pcer336 US 337 Prunus americana PamelanMO337 var. lanata US MO 338 Prunus alleghaniensis PallePA338 US 339 Prunus hortulana PhorKS339 US 340 Prunus maritima PmarMA340 var. flava US 341 Prunus cerasifera Pcer341 Russia 342 Prunus americana PamelanMO342 var. lanata US MO 343 Prunus munsoniana Pmun343 US 344 Prunus americana Pame344 Canada MB 345 Prunus nigra PnigVT345 US VT 44 18.845 N 72 23.604 W 346 Prunus americana PameVT346 US VT 44 16.395 N 73 14.038 W 347 Prunus avium PaviPA347 US PA 40 34.224 N 75 15.492 W 348 Prunus avium PaviPA348 US PA 40 32.604 N 75 15.200 W 279

PAGE 280

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 349 Prunus avium PaviPA349 US PA 40 25.418 N 75 4.071 W 350 Prunus alleghaniensis PallePA350 US PA 40 49.976 N 78 0.516 W 351 Prunus alleghaniensis PallePA351 US PA 40 50.930 N 77 58.285 W 352 Prunus alleghaniensis PallePA352 US PA 40 52.487 N 77 55.570 W 353 Prunus avium PaviPA353 US PA 40 25.573 N 79 33.026 W 354 Prunus avium PaviPA354 US PA 40 28.870 N 79 38.882 W 355 Prunus alleghaniensis PallePA355 US PA 40 2.568 N 80 5.210 W 356 Prunus umbellata PumbAR356 US AR 34.58145 N 92.37143 W 357 Prunus umbellata PumbAR357 US AR 34.57521 N 92.27229 W 358 Prunus mexicana x rivularis PmexrivTX358 US TX 359 Prunus angustifolia x texana PangxtexTX359 US TX 360 Prunus angustifolia x texana Pangxtex TX360 US TX 361 Prunus texana PtexTX361 US TX 362 Prunus angustifolia x texana PangxtexTX362 US TX 363 Prunus texana PtexTX363 US TX 364 Prunus angustifolia x texana PangxtexTX364 US TX 365 Prunus angustifolia x texana PangxtexTX365 US TX 366 Prunus mexicana PmexTX366 US TX 367 Prunus mexicana Pme xTX367 US TX 368 Prunus alleghaniensis PalleMI368 US MI 42 49.098 N 83 39.868 W 369 Prunus alleghaniensis PalleMI369 US MI 42 49.111 N 83 40.813 W 370 Prunus pumila PpumMI370 var. susquehanue US MI 44 32.196 N 84 22.234 W 371 Prunus pumila PpumMI371 var. susquehanue US MI 49 33.042 N 84 22. 239 W 372 Prunus americana PameMI372 US MI 43 54.048 N 85 10.340 W 373 Prunus americana PameIN373 US IN 41 13.220 N 86 53.494 W 374 Prunus angustifolia PangIN374 US IN 41 11.200 N 86 54.081 W 375 Prunus americana PameIN375 US IN 41 9.311 N 87 11.947 W 376 Prunus angustifolia PangIL376 US IL 40 42.741 N 88 41.88 W 377 Prunus americana PameIL377 US IL 40 42.256 N 88 4.785 W 280

PAGE 281

Table A 18 Continued. Voucher Collection (No). Genus Speci es IDz Rank1 Infraspecific Country State Latitude Longitude 378 Prunus americana PameIL378 US IL 40 31.769 N 88 4.009 W 379 Prunus americana PameIL379 US IL 40 6.789 N 88 19.165 W 380 Prunus hortulana PhorIL380 US IL 39 18.813 N 87 40.322 W 3 81 Prunus americana PameIL381 US IL 39 18.268 N 87 40.236 W 382 Prunus americana PameIL382 US IL 39 18.248 N 87 40.230 W 383 Prunus hortulana PhorIL383 US IL 39 14.139 N 87 40.041 W 384 Prunus americana PrunIL384 US IL 39 14.189 N 87 40.041 W 385 Prunus hortulana PhorIL385 US IL 38 2.942 N 88 14.259 W 386 Prunus americana PrunIL386 US IL 37 33.311 N 88 24.265 W 387 Prunus hortulana PhorIL387 US IL 37 16.210 N 89 6.556 W 388 Prunus nigra PhorIL388 US IL 37 38.209 N 89 28.901 W 389 Prunus hortulana PhorIL389 US IL 37 41.687 N 89 28.453 W 390 Prunus PrunIL390 US IL 37 49.181 N 89 19.334 W 391 Prunus angustifolia PangIL391 US IL 37 49.181 N 89 19.334 W 392 Prunus americana PameIL392 US IL 38 2.426 N 89 22.606 W 393 Prunus angustifolia PangIL393 US IL 38 27.182 N 89 33.293 W 394 Prunus hortulana PhorMO394 US MO 38 49.582 N 91 0.138 W 395 Prunus hortulana PhorMO395 US MO 39 4.734 N 92 19.817 W 396 Prunus hortulana PhorMO396 US MO 39 4.734 N 92 19.817 W 397 Prunus americana PameMO397 var. lanata US MO 40 17.626 N 92 34.481 W 398 Prunus americana PameIO398 US IO 40 37.022 N 92 31.425 W 399 Prunus americana PameIO399 US IO 41 18.019 N 92 48.537 W 400 Prunus munsoniana Pmun400 US 401 Prunu s munsoniana Pmun401 US 402 Prunus subcordata PsubCA402 US CA 403 Prunus subcordata PsubCA403 US CA 404 Prunus subcordata PsubCA404 US CA 405 Prunus umbellata PumbFL405 US FL 30.714168 N 85.93558 W 406 Prunus umbellata PumbFL406 US FL 30.755095 N 85.328411 W 281

PAGE 282

Table A 18 Continued. Voucher Collection (No). Genus Species IDz Rank1 Infraspecific Country State Latitude Longitude 407 Prunus umbellata PumbFL407 US FL 408 Prunus umbellata PumbFL408 US FL 409 Prunus v irginiana PvirFL409 US FL 410 Prunus umbellata PumbAL410 US AL 30.545704 N 87.883336 W zID = first letter represented the genus ( Prunus =P), next three letters represented the species ( americana=ame, etc.), and the following letters represented the state of origin and collection number (Florida collection 12=FL12, etc.). 282

PAGE 283

Table A 19 List of the Prunus accessions submitted to the USDA National Clonal Germplasm Repository in Davis, CA. ID z Collection (No.) Genus Species Seeds (No.) PallePA350 350 P runus alleghaniensis 12 PallePA355 355 Prunus alleghaniensis 35 PallePA352 352 Prunus alleghaniensis 7 PameGA26 26 Prunus americana 18 PameGA237 237 Prunus americana 10 PameGA26 26 Prunus americana 10 PaviPA347 347 Prunus avium 6 PaviPA348 348 Prunu s avium 32 PaviPA353 353 Prunus avium 10 PaviPA349 349 Prunus avium 7 PameIL384 384 Prunus americana 10 PameIL379 379 Prunus americana 8 PameIL386 386 Prunus americana 8 PameAR317 317 Prunus americana 15 PameIL381 381 Prunus americana 1 PameIA399 3 99 Prunus americana 10 PandCA263 263 Prunus andersonii 19 PangGA154 154 Prunus angustifolia 57 PangGA158 158 Prunus angustifolia 34 PangFl162 162 Prunus angustifolia 23 PangTX309 309 Prunus angustifolia 9 PangTX302 302 Prunus angustifolia 28 PangTX 306 306 Prunus angustifolia 6 PangxtexTX364 364 Prunus angustifolia x texana 29 PangxtexTX365 365 Prunus angustifolia x texana 14 PangxtexTX359 359 Prunus angustifolia x texana 6 PfasCA267 267 Prunus fasciculata 20 PfasCA264 264 Prunus fasciculata 25 PgraTX307 307 Prunus gracilis 10 PgraTX308 308 Prunus gracilis 5 PgraTX305 305 Prunus gracilis 11 PhorAR320 320 Prunus hortulana 34 PhorAR316 316 Prunus hortulana 1 PhorAR319 319 Prunus hortulana 40 PhorAR316 316 Prunus hortulana 11 PhorIL383 383 Prunus hortulana 10 PhorMO394 394 Prunus hortulana 12 PhorMO396 396 Prunus hortulana 10 PhorIL380 380 Prunus hortulana 1 PhorIL389 389 Prunus hortulana 10 283

PAGE 284

Table A 19 Continued. ID z Collection (No.) Genus Species Seeds (No.) PhorIL385 385 Prunus ho rtulana 2 PhorAR318 318 Prunus hortulana 9 PmarMA333 333 Prunus maritima 29 PmarMA332 332 Prunus maritima 17 PmarMA334 344 Prunus maritima 10 PmarMA330 330 Prunus maritima 33 PmarMA331 331 Prunus maritima 11 PmexAR314 314 Prunus mexicana 38 Pmexri vTX358 358 Prunus mexicana x rivularis 20 PmunGA249 249 Prunus munsoniana 71 PpumsusMA328 328 Prunus pumila var. susquehuane 16 PrivpubTX282 282 Prunus rivularis var. pubescens 8 PrivpubTX300 300 Prunus rivularis var. pubescens 10 PrivrivTX289 289 Pru nus rivularis var. rivularis 10 PrivrivTX299 299 Prunus rivularis var. rivularis 9 PrivrivTX288 288 Prunus rivularis var. rivularis 6 PrivrivTX290 290 Prunus rivularis var. rivularis 18 PrivrivTX287 287 Prunus rivularis var. rivularis 6 PtexTX297 297 Prunus texana 2 PtexTX295 295 Prunus texana 14 PtexTX363 363 Prunus texana 12 PtexTX295 295 Prunus texana 19 PtexTX294 294 Prunus texana 5 PumbFL166 166 Prunus umbellata 17 PumbFL172 172 Prunus umbellata 27 PumbFL170 170 Prunus umbellata 39 PumbFL1 63 163 Prunus umbellata 57 PumbGA155 155 Prunus umbellata 25 PumbFl17 17 Prunus umbellata 8 PumbGA13 13 Prunus umbellata 16 PumbFl59 59 Prunus umbellata 12 PumbGA53 53 Prunus umbellata 20 PumbAR357 357 Prunus umbellata 30 PumxamFl153 153 Prunus umbe llata x americana 15 zID = first letter represented the genus ( Prunus =P), next three letters represented the species ( americana=ame, etc.), and the following letters represented the state of origin and collection number (Florida collection 12=FL12, etc.). 284

PAGE 285

Table A 20 Summary statistics of PGI MAX4 PHYE, and VRN1 using introns haplotype phased sequence information. PGI MAX4 PHYE VRN1 Aligned L. (bp) z 316 224 505 170 Substitutions (no.) 61 53 69 57 Indels (no.) 24 15 15 6 PICs y (no.) 85 68 84 63 % Variability x 26.90 30.36 16.63 37.06 Eta w (no.) 71 62 76 68 G+C content 0.39 0.28 0.35 0.36 Hap v (no.) 66 68 45 49 Hd u 0.926 0.929 0.931 0.870 (per site) 0.018 0.040 0.014 0.028 (per sequence) from Eta 11.749 10.512 13.394 11.946 from Eta 0.052 0.085 0.034 0.074 D r 1.990 1.619 1.779 1.895 p value <0.05 >0.1 <0.05 <0.05 zComparisons using haplotype phased sequence data. yPICs = potentially informative characters. PICs = substitutions + indels. x% variability = [(Substitution+ Indels)/L] 100. wEta = number of mutations. vHap = number of haplotypes (Nei, 1987). Gaps not considered. uHd = haplotype diversity (Nei, 1987). Gaps not considered. t = nucleotide diversity (Nei, 1987). s = Watterson estimator of population mutation rate (Watterson, 1975). rD = Tajimas D (Tajima, 1989). 285

PAGE 286

Table A 21 Summary statistics of PGI MAX4 PHYE, and VRN1 using exon haplotype phased sequence information PG I MAX4 PHYE VRN1 Aligned L. (bp) z 143 205 48 392 Substitutions (no.) 24 36 9 56 Indels (no.) 1 0 0 0 PICs y (no.) 25 36 9 56 % Variability x 17.48 17.56 18.75 14.29 Eta w (no.) 25 39 9 66 G+C content 0.43 0.42 0.49 0.40 Hap v (no.) 31 41 9 57 Hd u 0.84 5 0.790 0.532 0.908 (per site) 0.013 0.021 0.016 0.011 (per sequence) from Eta 4.137 6.613 1.586 11.595 0.029 0.048 0.033 0.030 D r 1.541 1.659 1.171 1.929 p value >0.1 >0.1 >0.1 <0.05 zComparisons using haplotype phased sequence data. yPIC s = potentially informative characters. PICs = substitutions + indels. x% variability = [(Substitution+Indels)/L] 100. wEta = number of mutations. vHap = number of haplotypes (Nei, 1987). Gaps not considered. uHd = haplotype diversity (Nei, 1987). Gaps n ot considered. t = nucleotide diversity (Nei, 1987). s = Watterson estimator of population mutation rate (Watterson, 1975). rD = Tajimas D (Tajima, 1989). 286

PAGE 287

Table A 2 2 Simple sequence repeat (SSR) markers selected from the Prunus Texas almond Earlygold peach (T E) reference map. Chromosome Position Marker Fluorophore Ta(C) z Forward sequence Reverse sequence 1 9 CPSCT008 HEX 62 TGGATCCAATCCAAGAGTCTG GCAGCAAGTTGTTCTTGGTTC 1 25.8 CPDCT038 HEX 62 ATCACAGGTGAAGGCTGTGG CAGATTCATTGGCCCATCTT 1 33.9 CPPCT026 HEX 55 AGACGCAGCACCCAAACTAC CATTACATCACCGCCAACAA 1 47.3 BPPCT027 HEX 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 1 55.2 BPPCT016 HEX 57 GATTGAGAGATTGGGCTGC GAGGATTCTCATGATTTGTGC 1 65.1 CPPCT029 HEX 55 CCAAATTCCAAATCTCCTAACA TGATCAACTTTGAGATTTGTTGAA 1 77.4 BP PCT028 HEX 57 TCAAGTTAGCTGAGGATCGC GAGCTTGCCTATGAGAAGACC 2 9.6 UDP98 025 FAM 57 GGGAGGTTACTATGCCATGAAG CGCAGACATGTAGTAGGACCTC 2 25 BPPCT013 HEX 57 ACCCACAAATCAAGCATATCC AGCTTCAGCCACCAAGC 2 38 BPPCT030 HEX 57 AATTGTACTTGCCAATGCTATGA CTGCCTTCTGCTCACACC 2 48.6 CPSCT034 HEX 62 AGGTGGACAATAGCCGTGAT TTTCCAGACCCTGAGAAAGC 3 18 BPPCT039 FAM 57 ATTACGTACCCTAAAGCTTCTGC GATGTCATGAAGATTGGAGAGG 3 36.4 CPDCT025 HEX 62 GACCTCATCAGCATCACCAA TTCCCTAACGTCCCTGACAC 3 46.4 CPDCT027 HEX 62 TGAGGAGAGCACTGGAGGAG CAACCGATCCCT CTAGACCA 4 10.4 CPPCT005 FAM 52 CATGAACTCTACTCTCCA TGGTATGGACTCACCAAC 4 28.3 UDP96 003 FAM 57 TTGCTCAAAAGTGTCGTTGC ACACGTAGTGCAACACTGGC 4 34.1 EPDC3832 FAM 57 CTTTTGAAGGCCCAATACCA ATCACTGCTTCGCCTTCATT 4 45.4 BPPCT023 HEX 57 TGCAGCTCATTACCTTTTGC AGATGT GCTCGTAGTTCGGAC 4 52.7 EPPISF032 HEX 57 TCCCCCACAGATATTTCAGC GTCGAGGAGAGAGGGCTTTT 5 5.2 BPPCT026 FAM 57 ATACCTTTGCCACTTGCG TGAGTTGGAAGAAAACGTAACA 5 20.1 BPPCT017 HEX 57 TTAAGAGTTTGTGATGGGAACC AAGCATAATTTAGCATAACCAAGC 5 32.9 BPPCT038 FAM 57 TATATTGTTGGC TTCTTGCATG GAGCTTGCCTATGAGAAGACC 5 44 BPPCT014 HEX 57 TTGTCTGCCTCTCATCTTAACC CATCGCAGAGAACTGAGAGC 6 8.7 CPPCT008 FAM 59 GAGCTCTCACGCATTAGTTT TTTGACTGCATAACAAAACG 6 17.5 UDP96 001 HEX 57 AGTTTGATTTTCTGATGCATCC TGCCATAAGGACCGGTATGT 6 30.1 BPPCT008 FAM 57 ATGGTGTGTATGGACATGATGA CCTCAACCTAAGACACCTTCACT 6 35.8 CPPCT015 FAM 50 TGGAGTGCCAATACTATTTA CATATGCATGGTTATGGT 6 41 EPPISF002 FAM 56 CGACGTGTGACCAAAGGAC GCAACTCCATCCACATTTCTC 6 56.4 BPPCT025 FAM 57 TCCTGCGTAGAAGAAGGTAGC CGACATAAAGTCCAAATGGC 6 72 UDP98 412 HEX 57 AGGGAAAGTTTCTGCTGCAC GCTGAAGACGACGATGATGA 7 9.5 CPSCT004 FAM 62 GCTCTGAAGCTCTGCATTGA TTTGAAATGGCTATGGAGTACG 7 18.6 CPPCT022 FAM 50 CAATTAGCTAGAGAGAATTATTG GACAAGAAGCAAGTAGTTTG 7 29.6 BPPCT029 FAM 57 GGACGGACAGAAATGAAGGT CCTTAACCCACGCAACTCC 7 38.9 CPPCT033 HEX 50 TCAGCAAACTAGAAACAAACC TTGCAATCTGGTTGATGTT 7 47.8 PMS2 FAM 55 CACTGTCTCCCAGGTTAAACT CCTGAGCTTTTGACACATGC 7 61.8 CPPCT017 FAM 60 TGACATGCATGCACTAAACAA TGCAAATGCAATTTCATAAAGG 7 64.7 EPDCU3392 HEX 57 CTTTTCATGGGTTCCTCACC ATCAACCAGTTCAC GCACAA 287

PAGE 288

Table A 22 Continued. Chromosome Position Marker Fluorophore Ta(C) z Forward sequence Reverse sequence 8 7.8 CPPCT019 FAM 50 AATTCAATGTCAAGACACA TCATCAAAATAAATATCCAGT 8 14.1 BPPCT006 HEX 57 GCTTGTGGCATGGAAGC CCCTGTTTCTCATAGAACTCACAT 8 20.8 UD P96 019 HEX 57 TTGGTCATGAGCTAAGAAAACA TAGTGGCACAGAGCAACACC 8 24.8 CPPCT006 HEX 59 AATTAACTCCAACAGCTCCA ATGGTTGCTTAATTCAATGG 8 31.6 PS1H3 HEX 55 8 42.6 CPDCT023 FAM 62 GTGGCAAATGTTGGCAAAG AACACAAAGCAGCACCAAGA 8 54.7 EPDCU3117 FAM 57 CAGAGGGAACAGTGTGAG CA TGTTGTTGTCGACCCTGAAA zTa = annealing temperature. 288

PAGE 289

Table A 2 3 P henotypic information available for the southeastern US plums specimens SelectionID#z Speciesy Statex Chillw Chillrange1v Chillrange2u PameFL12 1 1 418 2 3 PameGA237 1 2 804 3 7 PameGA 26 1 2 804 3 7 PameGA76 1 2 1110 4 8 PameGA80 1 2 1110 4 8 PangFl07 2 1 626 2 5 PangFl162 2 1 631 2 5 PangGA25 2 2 804 3 7 PangGA90 2 2 1460 4 9 PbesIL15 6 3 1400 4 9 PfasCA131 7 3 600 2 5 PgenFl115 3 1 230 1 1 PgenFl147 3 1 282 1 1 PgenFl150 3 1 282 1 1 PgenFl182 3 1 282 1 1 PgenFl183 3 1 282 1 1 PgenFl184 3 1 282 1 1 PgenFl185 3 1 282 1 1 PgenFl186 3 1 282 1 1 PgenFl187 3 1 282 1 1 PgenFL234 3 1 230 1 1 PhorMD89 8 3 1400 4 9 PmarNY87 9 3 1800 4 9 PmexLN85 10 3 600 2 5 PmunTX88 11 3 1 100 3 8 PperJP121 12 3 250 1 1 PrunFl163 4 1 282 1 1 PrunFl164 4 1 282 1 1 PrunFl165 4 1 282 1 1 PrunFl166 4 1 282 1 1 PrunFl167 4 1 282 1 1 PrunFl168 4 1 282 1 1 PrunFl169 4 1 282 1 1 PrunFl170 4 1 282 1 1 PrunFl171 4 1 282 1 1 PrunFl172 4 1 282 1 1 PrunFl173 4 1 282 1 1 PumbFl01 5 1 600 2 5 PumbFl02 5 1 600 2 5 PumbFl03 5 1 600 2 5 PumbFl04 5 1 600 2 5 289

PAGE 290

Table A 2 3 Continued. SelectionID#z Speciesy Statex Chillw Chillrange1v Chillrange2u PumbFl05 5 1 600 2 5 PumbFl06 5 1 500 2 4 PumbFl 08 5 1 500 2 4 PumbFl09 5 1 500 2 4 PumbFl10 5 1 500 2 4 PumbFl104 5 1 631 2 5 PumbFl105 5 1 631 2 5 PumbFL105b 5 1 631 2 5 PumbFl106 5 1 631 2 5 PumbFl107 5 1 631 2 5 PumbFl11 5 1 500 2 4 PumbFl111 5 1 563 2 4 PumbFl124 5 1 610 2 5 PumbFl125 5 1 610 2 5 PumbFL126 5 1 500 2 4 PumbFl127 5 1 600 2 5 PumbFl142 5 1 600 2 5 PumbFl142alle 5 1 600 2 5 PumbFl143 5 1 600 2 5 PumbFl144 5 1 350 1 2 PumbFl152 5 1 428 2 3 PumbFl17 5 1 418 2 3 PumbFl18 5 1 418 2 3 PumbFl19 5 1 400 2 3 PumbFl20 5 1 4 00 2 3 PumbFl21 5 1 250 1 1 PumbFl23 5 1 631 2 5 PumbFl24 5 1 804 3 7 PumbFl27 5 1 804 3 7 PumbFl28 5 1 804 3 7 PumbFl32 5 1 879 3 7 PumbFl33 5 1 879 3 7 PumbFl34 5 1 631 2 5 PumbFl35 5 1 631 2 5 PumbFl36 5 1 289 1 1 PumbFl37 5 1 289 1 1 PumbFl 38 5 1 306 1 2 PumbFl39 5 1 306 1 2 PumbFl40 5 1 306 1 2 PumbFl41 5 1 306 1 2 PumbFl42 5 1 350 1 2 290

PAGE 291

Table A 2 3 Continued. SelectionID#z Speciesy Statex Chillw Chillrange1v Chillrange2u PumbFl43 5 1 418 2 3 PumbFl44 5 1 418 2 3 PumbFl45 5 1 418 2 3 PumbFl46 5 1 418 2 3 PumbFl47 5 1 631 2 5 PumbFl58 5 1 820 3 7 PumbFl59 5 1 631 2 5 PumbFl74 5 1 631 2 5 PumbFl75 5 1 700 2 6 PumbFl94 5 1 230 1 1 PumbFl95 5 1 804 3 7 PumbFl98 5 1 823 3 7 PumbGA13 5 2 804 3 7 PumbGA14 5 2 804 3 7 PumbGA247 5 2 805 3 7 PumbGA49 5 2 804 3 7 PumbGA50 5 2 804 3 7 PumbGA51 5 2 804 3 7 PumbGA52 5 2 804 3 7 PumbGA53 5 2 804 3 7 PumbGA54 5 2 804 3 7 PumbGA55 5 2 804 3 7 PumbGA96 5 2 1110 4 8 PumbxangFL109 5 1 626 2 5 zSelectionID# = genotype collection number and ID. ySpecies: 1) Prunus americana, 2) P. angustifolia, 3) P. geniculata, 4) P. umbellatalike, 5) P.umbellata, 6) P. pumila 7) P. fasciculata 8) P. hortulana 9) P. martima 10) P. mexicana, 11) P. munsoniana, and 12) P. persica xState = collecti on site: 1) Florida and 2) Georgia. wChill = chilling hour requirement estimate based on peach standards. vChillrange1 = range chilling hour requirement: 1) Low <400 chill hours, 2) Medium 400700 chill hours, and 3) Medium high 7011100 chill hours, and 4) High >1100 chill hours. uChillrange2 = range chilling hour requirement: 1) 201300 chill hours, 2) 301400 chill hours, 3) 401500 chill hours, 4) 501600 chill hours, 5) 601700 chill hours, 6) 701800 chill hours, 7) 801900 chill hours, 8) 11001200 c hill hours, and 9) > 1400 chill hours. 291

PAGE 292

Table A 2 4 Summary statistics of 33 simple sequence repeat (SSR) markers for 106 genotypes of the North American Plums germplasmz available in the Stone Fruit Genetics and Breeding Program at University of Florida, Gainesville, FL. Marker Genotype No. A y Ae Ho He F PIC BPPCT006A 54 18 11.57 0.66 0.91 0.27 0.91 BPPCT006B 62 32 16.68 0.62 0.94 0.34 0.94 BPPCT008A 21 12 2.52 0.37 0.60 0.39 0.58 BPPCT008B 61 26 13.81 0.73 0.93 0.22 0.92 BPPCT014 75 33 15.87 0.73 0.9 4 0.22 0.93 BPPCT016 7 x 7 1.26 0.03 0.21 0.86 0.19 BPPCT017 77 30 16.36 0.78 0.94 0.17 0.93 BPPCT025A 71 40 13.80 0.76 0.93 0.18 0.92 BPPCT026 81 36 22.10 0.76 0.95 0.20 0.95 BPPCT027 17 13 2.19 0.24 0.54 0.57 0.51 BPPCT028 68 23 15.71 0.74 0.94 0.21 0.93 BPPCT029 22 14 3.19 0.41 0.69 0.41 0.65 BPPCT030 12 11 1.39 0.25 0.28 0.10 0.28 BPPCT039 83 37 20.47 0.80 0.95 0.16 0.95 CPDCT023 75 34 16.23 0.69 0.94 0.27 0.93 CPDCT025 78 31 18.05 0.82 0.94 0.13 0.94 CPDCT027 76 37 16.96 0.82 0.94 0.13 0.94 CPPCT005 76 35 15.43 0.80 0.94 0.15 0.93 CPPCT008 45 26 8.34 0.41 0.88 0.54 0.87 CPPCT019 72 38 19.51 0.77 0.95 0.19 0.95 CPPCT022 78 41 19.94 0.75 0.95 0.21 0.95 CPPCT026 76 33 16.44 0.81 0.94 0.14 0.93 CPPCT033 63 26 12.01 0.76 0.92 0.17 0.91 CPSC T004 17 10 3.13 0.60 0.68 0.11 0.64 CPSCT034 32 17 4.56 0.67 0.78 0.14 0.75 EPDCU3117 23 12 5.00 0.44 0.80 0.45 0.77 EPDCU3392 78 29 14.89 0.84 0.93 0.10 0.93 EPPISF002 52 20 9.95 0.76 0.90 0.15 0.89 EPPISF032 40 19 4.94 0.68 0.80 0.15 0.78 PMS2 70 3 0 15.14 0.83 0.93 0.12 0.93 UDP96 001 23 15 3.22 0.56 0.69 0.19 0.66 UDP96 003 13 11 1.58 0.17 0.37 0.54 0.36 UDP96 019 19 15 3.22 0.48 0.69 0.30 0.65 UDP98 025 29 16 3.93 0.51 0.75 0.32 0.71 UDP98 412 80 33 15.63 0.90 0.94 0.04 0.93 Average 52.17 24 .57 11.00 0.63 0.81 0.25 0.80 zGermplasm available include d P. americana, P. angustifolia, P. geniculata, P. hortulana, P. martima P. mexicana, P. munsoniana, P. persica P. pumila P. umbellata, and P. fasciculata species. yA = number of observed allel es, Ae = effective number of alleles, Ho = observed heterozygosity, He = expected heterozygosity, F = Wrights fixation index [ F =( He Ho )/ He =1 ( Ho / He )], and PIC = p olymorphism information content. xNumbers in bold represent highest and lowest values for eac h variable. 292

PAGE 293

Table A 2 5 Summary of the population stratification simulation results of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species File name K Run Est. Ln prob of data z Mean value of Ln likelihood Variance of Ln likelihood run_run_1_frep1_f 1 1 17881.5 17696.3 370.4 run_run_1_frep2_f 1 2 17886.4 17696.3 380.1 run_run_1_frep3_f 1 3 17879.3 17696.6 365.5 run_run_1_frep4_f 1 4 17881.6 17696.3 370.6 ru n_run_2_frep1_f 2 1 17350.7 16777.3 1146.9 run_run_2_frep2_f 2 2 17285.2 16778.6 1013.2 run_run_2_frep3_f 2 3 17260.8 16778.5 964.6 run_run_2_frep4_f 2 4 17238.7 16785.1 907 .0 run_run_3_frep1_f 3 1 16780.7 16025.3 1510.8 run_run_3_frep2_f 3 2 16556.5 16159.2 794.8 run_run_3_frep3_f 3 3 16571.3 16001.9 1138.9 run_run_3_frep4_f 3 4 16558.5 16160.2 796.7 run_run_4_frep1_f 4 1 16311.6 15532.7 1557.9 run_run_4_frep2_f 4 2 16274 .0 15537.2 1473.8 run_run_4_frep3_f 4 3 16290.3 15555. 7 1469.3 run_run_4_frep4_f 4 4 16265.4 15547.8 1435.2 run_run_5_frep1_f 5 1 15865.5 15115.8 1499.2 run_run_5_frep2_f 5 2 15788.7 15139.1 1299.3 run_run_5_frep3_f 5 3 15883.3 15126.9 1512.8 run_run_5_frep4_f 5 4 15955.7 15116.8 1677.9 run_ru n_6_frep1_f 6 1 16119.5 14751 .0 2736.9 run_run_6_frep2_f 6 2 16055.3 14706.7 2697.3 run_run_6_frep3_f 6 3 16516.1 14715.9 3600.5 run_run_6_frep4_f 6 4 16310 .0 14748.7 3122.7 run_run_7_frep1_f 7 1 15866.5 14508.4 2716.2 run_run_7_frep2_f 7 2 15775.6 14578.3 2394.5 run_run_7_frep3_f 7 3 15702.2 14571.5 2261.4 run_run_7_frep4_f 7 4 15830 .0 14498.4 2663.3 run_run_8_frep1_f 8 1 15546.7 14412.1 2269.1 run_run_8_frep2_f 8 2 15887.1 14317.5 3139.1 run_run_8_frep3_f 8 3 17146.5 14291. 4 5710.1 run_run_8_frep4_f 8 4 15814.7 14410.7 2808 .0 run_run_9_frep1_f 9 1 15793.4 14203.7 3179.4 run_run_9_frep2_f 9 2 15734.9 14183 .0 3103.8 run_run_9_frep3_f 9 3 15836.3 14165 .0 3342.5 run_run_9_frep4_f 9 4 16363.8 14109.6 4508.4 run_ru n_10_frep1_f 10 1 15640.2 14054.1 3172.2 run_run_10_frep2_f 10 2 15663.1 14077.3 3171.6 run_run_10_frep3_f 10 3 15489 .0 14020.3 2937.5 run_run_10_frep4_f 10 4 15733.2 14020.5 3425.4 zParameters of 105 interactions after a burnin of 104 interac tions. 293

PAGE 294

Table A 2 6 Analysis of the population structure results for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species using the Evanno method (Evanno et al., 2005) implemented in the Structure Harvester softwar e (Earl and VonHoldt, 2012). K z Runs Mean LnP(K) Stdev LnP(K) Ln'(K) |Ln''(K)| Delta K 1 4 17882.20 2.99 2 4 17283.85 48.44 598.35 68.75 1.42 3 4 16616.75 109.50 667.10 335.68 3.07 4 4 16285.33 20.33 331.43 80.60 3.96 5 4 15873.30 68.58 412.03 788.95 11.50 6 4 16250.23 207.64 376.93 833.58 4.01 7 4 15793.58 71.45 456.65 761.83 10.66 8 4 16098.75 713.68 305.18 471.83 0.66 9 4 15932.10 290.79 166.65 134.08 0.46 10 4 15631.38 102.83 300.73 zK= 5 and 7 were the highest delta K value 294

PAGE 295

Figure A 1 Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Cavalli Sforza chord distance (Cavalli Sforza and Edwards, 1967) of 36 simple sequence repeat (SSR) markers for 195 germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram ro oted with apricots at the base. 295

PAGE 296

Figure A 2. Neighbor Joining (NJ) cluster analysis based on CavalliSforza chord distance (Cavalli Sforza and Edwards, 1967) of 36 simple sequence repea t (SSR) markers for 195 germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 296

PAGE 297

Figure A 3 Neighbor Joining (NJ) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 297

PAGE 298

Figur e A 4. Peach (green) and nectarine (black) fruit types genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 g ermplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 298

PAGE 299

Figure A 5. Whi te (green) and yellow (black) fruit flesh color genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germplas m representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 299

PAGE 300

Figure A 6. Normal (gr een) and highlighter /reduce d anthocyanin (black) genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 300

PAGE 301

Figure A 7. Round (gr een) and peento (black) fruit shape genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germplasm representa tives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 301

PAGE 302

Figure A 8. Reniform (turquoise), globose (green), and eglandular (black) leaf glands genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germ plasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 302

PAGE 303

Figure A 9. Showy (green) and nonshowy (black) flower type genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple sequence repeat (SSR) markers for 195 germplasm repr esentatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apricots at the base. 303

PAGE 304

Figure A 10. L ow chill <400 hours (turquoise), medium chill 400700 hours (green), and high chill >700 hours (black) genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 36 simple seque nce repeat (SSR) markers for 195 germplasm representatives of the genetic pools utilized for breeding and selection at the University of Florida stone fruit breeding and genetics program since its creation in 1952 until present Cladogram rooted with apric ots at the base. 304

PAGE 305

Figure A 11. Genotypes with 0150 chill hours (dark blue), 151250 chill hours (powder blue), 251350 chill hours (turquoise), 351450 chill hours ( sea foam green), 451550 chill hours (green), 751850 chill hours (yellow), 851950 chill hours (orange), and 9511050 chill hours (black) traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) for 195 germplasm representatives of the genetic pools utilized for breeding and selection at University of Florida. 305

PAGE 306

Figure A 12. Introduction and/or genotypes selection decades of 1890s (dark bl ue), 1940s (powder blue), 1950s (turquoise), 1960s (light blue), 1970s (sea foam green), 1980s (green), 1990s (yellow), 2000s (orange), and 2010s (black) traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) for 195 germplasm representatives of the genetic pools utilized for breeding and selectio n at the University of Florida. 306

PAGE 307

Figure A 13. Analysis of the population structure results for 168 peach germplasm representatives of the UF varieties and advanced materials (20002010 selections) using the Evanno method (Evanno et al., 2005) implemented by the Structure Harvester software (Earl and VonHoldt, 2012). A) Second order change in the log likelihood delta K ( ). B) Rate of change of the log l ikelihood distribution (mean). C) Absolute value of the second order change in the log l ikelihood distribution (mean). D) The average log likelihood and the standard error of two to five reps per run. A B C D 307

PAGE 308

Figure A 14. Unwei ghted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) and population structure results for k=2 to k=10 (each color representing a different population structure) of 36 simple sequence repeat (SSR) markers for 168 peach UF varieties and advanced materials (20002010 selections). Clade names described on Figure 21. 308

PAGE 309

K=2 K =3 K=4 K=5 K=6 K=7 K=8 K=9 K=10 309

PAGE 310

Fig ure A 15. Majority rule consensus trees for 3'trnV ndhC cpD NA region diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and sequence data (with haplotype number prefix) using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maxim um likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 310

PAGE 311

Figure A 16. Majority ru le consensus trees for trnL trnF and trnL intron cpDNA regions diploid sequence d ata using: A) maximum parsimony and B ) maximum likelihood; and sequence data (with haplotype number prefix) using : C) maximum parsimony and D ) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 311

PAGE 312

Figure A 17. Majority rule consensus trees for trnQ 5rps16 cpDNA region diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and sequence data (with haplotype number prefix) using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 312

PAGE 313

Figure A 18. Majority rule c onsensus trees for trnH psbA cpDNA region sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) ma ximum likelihood; and sequence data (with haplotype number prefix) using : D) maximum parsimony (including g aps) and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E 313

PAGE 314

Figure A 19. Majority rule consensus trees for ndhF rpL32 cpDNA region diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and sequence data (with haplotype number prefix) using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 314

PAGE 315

Figure A 20. Majority rule consensus trees for atpB rbcL cpDNA region diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and sequence data (with haplotype number prefix) using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Tr ees are rooted with P. fasciculata A B C D E F 315

PAGE 316

Figure A 21. Majority rule consensus trees for ITS diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequenc e data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 316

PAGE 317

Figure A 22. Majority rule consensus trees for PGDH diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 317

PAGE 318

Figure A 23. Majority rule consensus trees for PGI diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence da ta using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 318

PAGE 319

Figure A 24. Majority rule consensus trees for S6PDH diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence dat a using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 319

PAGE 320

Figure A 25. Majority rule consensus trees for AXR1 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using : C) maximum parsimony and D) maxim um likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 320

PAGE 321

Figure A 26. Majority rule consensus trees for BRC1 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using : C) maximum parsimony and D) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 321

PAGE 322

Figure A 27. Majority rule consensus trees for BRC2 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using C) maximum parsimony and D) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 322

PAGE 323

Figure A 28. Majority rule consensus trees for CUC1A diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 323

PAGE 324

Figure A 29. Majority rule consensus trees for CUC1B diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata. A B C D E F 324

PAGE 325

Figure A 30. Majority rule consensus trees for CUC2 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic seq uence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. T rees are rooted with P. fasciculata A B C D E F 325

PAGE 326

Figure A 31. Majority rule consensus trees for CUC3 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Tr ees are rooted with P. fasciculata A B C D E F 326

PAGE 327

Figure A 32. Majority rule consensus trees for LAS diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 327

PAGE 328

Figure A 33. Majority rule consensus trees for MAX1A diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 328

PAGE 329

Figure A 34. Majority rule consensus trees for MAX1B diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees ar e rooted with P. fasciculata A B C D E F 329

PAGE 330

Figure A 35. Majority rule consensus trees for MAX2 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence dat a using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 330

PAGE 331

Figure A 36. Majority rule consensus trees for MAX3 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are ro oted with P. fasciculata A B C D E F 331

PAGE 332

Figure A 37. Majority rule consensus trees for MAX4 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data us ing : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are root ed with P. fasciculata A B C D E F 332

PAGE 333

Figure A 38. Majority rule consensus trees for PIN diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 333

PAGE 334

Figure A 39. Majority rule consensus trees for RAX1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted wi th P. fasciculata A B C D E F 334

PAGE 335

Figure A 40. Majority rule consensus trees for RAX2 RAX3 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data usi ng : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are roote d with P. fasciculata A B C D E F 335

PAGE 336

Figure A 41. Majority rule consensus trees for REV diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted w ith P. fasciculata A B C D E F 336

PAGE 337

Figure A 42. Majority rule consensus trees for SPS diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 337

PAGE 338

Figure A 43. Majority rule consensus trees for AGL24 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D ) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted wit h P. fasciculata A B C D E F 338

PAGE 339

Figure A 44. Majority rule consensus trees for AGL20 SOC1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 339

PAGE 340

Figure A 45. Majority rule consensus trees for BFT diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted wi th P. fasciculata A B C D E F 340

PAGE 341

Figure A 46. Majority rule consensus trees for BRM diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 341

PAGE 342

Figure A 47. Majority rule consensus trees for CO diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maxi mum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. f asciculata A B C D E F 342

PAGE 343

Figure A 48. Majority rule consensus trees for CRY1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximu m parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fas ciculata A B C D E F 343

PAGE 344

Figure A 49. Majority rule consensus trees for CRY2 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasci culata A B C D E F 344

PAGE 345

Figure A 50. Majority rule consensus trees for ELF6 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fascicu lata A B C D E F 345

PAGE 346

Figure A 51. Majority rule consensus trees for FD diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsim ony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 346

PAGE 347

Figure A 52. Majority rule consensus trees for FD1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 347

PAGE 348

Figure A 53. Majority rule consensus trees for FG diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 348

PAGE 349

Figure A 54. Majority rule consensus trees for FLC FLF diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E ) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 349

PAGE 350

Figure A 55. Majority rule consensus trees for FPF1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 350

PAGE 351

Figure A 56. Majority rule consensus trees for FRIGIDA diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 351

PAGE 352

Figure A 57. Majority rule consensus trees for FT TSF diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 352

PAGE 353

Figure A 58. Majority rule consensus trees for GI FB diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 353

PAGE 354

Figure A 59. Majority rule consensus trees for HOS1 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using : C) maximum parsimony and D) maximum likelihood. Bootstrap values gr eater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 354

PAGE 355

Figure A 60. Majority rule consensus trees for LFY diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values great er than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 355

PAGE 356

Figure A 61. Majority rule consensus trees for MAF1 MAF3 AGL3 1 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap val ues greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 356

PAGE 357

Figure A 62. Majority rule consensus trees for MAF2A diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 357

PAGE 358

Figure A 63. Majority rule consensus trees for MAF2B diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 358

PAGE 359

Figure A 64. Majority rule consensus trees for MAF4 di ploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 359

PAGE 360

Figure A 65. Majority rule consensus trees for MAF5 dipl oid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values gr eater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 360

PAGE 361

Figure A 66. Majority rule consensus trees for MFT diploi d sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A& B C D E F 361

PAGE 362

Figure A 67. Majority rule consensus trees for PHYA diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using : C) maximum parsimony and D) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 362

PAGE 363

Figure A 68. Majority rule consensus trees for PHYB PHYD diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 363

PAGE 364

Figure A 69. Majority rule consensus trees for PHYE diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using : D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities v alues are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 364

PAGE 365

Figure A 70. Majority rule consensus trees for RGA RGA1 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using C) maximum parsimony and D) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata. A B C D 365

PAGE 366

Figure A 71. Majority rule consensus trees for RGL1 RGL2 RGL3 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using D) ma ximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 366

PAGE 367

Figure A 72. Majority rule consensus trees for SPY diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fas ciculata A B C D E F 367

PAGE 368

Figure A 73. Majority rule consensus trees for TFL1 ATC diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using D) maxim um parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fa sciculata A B C D E F 368

PAGE 369

Figure A 74. Majority rule consensus trees for TFL2 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasci culata A B C D E F 369

PAGE 370

Figure A 75. Majority rule consensus trees for VRN1 diploid sequence data using: A) maximum parsimony and B) maximum likelihood; and haplotypic sequence data using C) maximum parsimony and D) maximum likelihood. Bootst rap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D 370

PAGE 371

Figure A 76. Majority rule consensus trees for VRN2 diploid sequence data using: A) maximum parsimony, B) maximum parsimony (including gaps), and C) maximum likelihood; and haplotypic sequence data using D) maximum parsimony, E) maximum parsimony (including gaps), and F) maximum likelihood. Bootstrap values greater than 50% are described above the branches for MPT. Posterior probabilities values are described above the branches for MLT. Trees are rooted with P. fasciculata A B C D E F 371

PAGE 372

Figure A 77. Phylogenetic analyses of combined i sozymes diploid sequence data (1399 bp). Majority rule consensus trees using A) maximum parsimony (38 trees, 82 steps, CI=0.902, RI=0.922, RC=0.832) and B) maximum parsimony (including gaps) (4 trees, 127 steps, CI=0.874, RI=0.890, RC=0.778) analyses. Boot strap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL=2464.46). Branch lengths described above the branches and posterior probabilities values below the branches. Trees are rooted with P. fasciculata A B C 372

PAGE 373

Figure A 78. A non standarized multi locus combined isozymes genes sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B 373

PAGE 374

Figure A 79. Phylogenetic analyses of combined branching genes diploid sequence data (6659 bp). Majority rule consensus trees using A) maximum parsimony (10 trees, 363 steps, CI=0.917, RI=0.927, RC=0.850) and B) maximum parsimony (including gaps) (2 trees, 560 steps, CI=0.914, RI=0.920, RC=0.842) analyses. Bootstrap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL=11668.52). Branch lengths described above the branches and posterior probabilities values below the branches. Trees are rooted with P. fasciculata A B C 374

PAGE 375

Figure A 80. A non standarized multi locus combined branching genes sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B 375

PAGE 376

Figure A 81. Phylogenetic analyses of combined dormancy related genes diploid sequence data (15371 bp). Majority rule consensus trees using A) maximum parsimony (2 trees, 943 steps, CI=0.924, RI=0.929, RC=0.859) and B) maximum parsimony (including gaps) (2 trees, 1876 steps, CI=0.875, RI=0.876, RC=0 .766) analyses. Bootstrap values greater than 50% are described above the branches. C) Maximum likelihood tree ( lnL=27249.04). Branch lengths described above the branches and posterior probabilities values below the branches. Trees are rooted with P. fasc iculata A B C 376

PAGE 377

Figure A 82. A non standarized multi locus combined dormancy related genes sequence data network using: A) gaps as missing characters and B) gaps as 5th characters. A B 377

PAGE 378

Figure A 83. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data ( 885 characters ) including gaps (971 trees, 546 steps, CI=0.85, RI=0.93). Bootstrap values greater than 50% are described above the branches. 378

PAGE 379

Figure A 84. Ma jority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data (553 bp) without gaps (1000 trees, 136 steps, CI=0.96, RI=0.98). Bootstrap values greater than 50% are described above the bra nches. 379

PAGE 380

Figure A 85. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI diploid sequence data ( 525 characters ) including gaps ( 1000 trees, 175 steps, CI=0.88, RI=0.96 ). Bootstrap values greater than 50% are described above the branches. 380

PAGE 381

Figure A 86. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI diploid sequence data ( 457 bp) without gaps (1000 trees, 121 steps, CI=0. 72, RI=0.92 ). Bootstrap values greater than 50% are described above the branches. 381

PAGE 382

Figure A 87. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 diploid sequence data (393 characters ) including gaps (1000 trees, 199 steps, CI=0.78, RI=0.9 2 ). Bootstrap values greater than 50% are described above the branches. 382

PAGE 383

Figure A 88. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 diploid sequence data (329 bp) without gaps (1000 trees, 1 18 steps, CI=0. 78, RI=0.9 3 ). Bootstrap values greater than 50% are described above the branches. 383

PAGE 384

Figure A 89. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysi s of PHYE diploid sequence data ( 659 characters ) including gaps (1000 trees, 188 steps, CI=0. 98, RI=0.9 9 ). Bootstrap values greater than 50% are described above the branches. 384

PAGE 385

Figure A 90. Majority rule consensus tree of the MPTs using the parsimony ratc het method for rapid parsimony analysis of PHYE diploid sequence data (553 bp) without gaps (1000 trees, 82 steps, CI=0. 95, RI=0.99). Bootstrap values greater than 50% are described above the branches. 385

PAGE 386

Figure A 91. Majority rule consensus tree of the MP Ts using the parsimony ratchet method for rapid parsimony analysis of VRN1 diploid sequence data (564 characters ) including gaps (1000 trees, 73 steps, CI=0.96 RI=0.99). Bootstrap values greater than 50% are described above the branches. 386

PAGE 387

Figure A 92. M ajority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 diploid sequence data (560 bp) without gaps (1000 trees, 6 9 steps, CI=0. 96, RI=0.99). Bootstrap values greater than 50% are described above the branches. 387

PAGE 388

Figure A 93. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined nuclear genes diploid sequence data (2101 characters ) including gaps ( 838 trees, 644 steps, CI=0.79, RI=0.94). Bo otstrap values greater than 50% are described above the branches. 388

PAGE 389

Figure A 94. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined nuclear genes diploid sequence data (1859 bp) without gaps (957 trees, 400 steps, CI=0.79, RI=0.94). Bootstrap values greater than 50% are described above the branches. 389

PAGE 390

Figure A 95. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined diploid sequ ence data total evidence approach (3026 characters ) including gaps ( 916 trees, 993 steps, CI=0.7 6 RI=0.9 2 ). Bootstrap values greater than 50% are described above the branches. 390

PAGE 391

Figure A 96. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of combined diploid sequence data total evidence approach ( 2452 bp) without gaps ( 9 84 trees, 436 steps, CI=0. 81, RI=0.9 5 ). Bootstrap values greater than 50% are described above the branches. 391

PAGE 392

Figure A 97. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA sequence data with haplotype number prefix ( 885 characters ) including gaps (200 trees, 547 steps, CI=0.84 RI=0.9 3 ). Bootstrap values greater than 50% are described above the branches. 392

PAGE 393

Figure A 98. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of trnH psbA haplotype phased sequence data ( 553 bp) without gaps (938 trees, 136 steps CI=0. 96, RI=0.9 8 ). Bootstrap values greater than 50% are described above the branches. 393

PAGE 394

Figure A 99. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (550 chara cters ) including gaps ( 169 trees, 4 59 steps, CI=0. 46, RI=0.84 ). Bootstrap values greater than 50% are described above the branches. 394

PAGE 395

A 395

PAGE 396

A 396

PAGE 397

Figure A 100. Majority rule consensus tree of the MPTs using t he parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (549 characters ) after removing minor frequency haplotypes and including gaps ( 196 trees, 284 steps, CI=0.75, RI=0.93). Bootstrap values greater than 50% are des cribed above the branches. 397

PAGE 398

398

PAGE 399

Figure A 101. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (459 bp) without gaps (889 trees, 252 steps, CI=0.48, RI=0.88). Boots trap values greater than 50% are described above the branches. Red boxes represented samples with haplotypes in two different clades. 399

PAGE 400

A 400

PAGE 401

A 401

PAGE 402

Figure A 1 02. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PGI haplotype phased sequence data (459 characters ) after removing minor frequency haplotypes ( 2 00 trees, 163 steps, CI=0.75, RI=0.94). Bootstrap values greater than 50% are described above the branches. 402

PAGE 403

403

PAGE 404

Figure A 103. Maximum likelihood tree using RAxML for PGI haplotype phased sequence data (459 bp) (lnL= 1967.55). Maximum likelihood posterior probabilities values are located below the branches. 404

PAGE 405

A A 405

PAGE 406

Figure A 104. Ma jority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 haplotype phased sequence data (466 characters ) including gaps ( 200 trees, 4 18 steps, CI=0. 61, RI=0.89 ). Bootstrap values greater than 50% are described above the branches. 406

PAGE 407

A 407

PAGE 408

A 408

PAGE 409

Figure A 105. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 haplotype phased sequence data (466 cha racters ) after removing minor frequency haplotypes and including gaps (184 trees, 307 steps, CI=0.72, RI=0.93). Bootstrap values greater than 50% are described above the branches. 409

PAGE 410

410

PAGE 411

Figure A 106. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 haplotype phased sequence data (364 bp) without gaps ( 930 trees, 304 steps, CI=0.51 RI=0. 84). Bootstrap values greater than 50% are described above the branches. Red boxes represented samples with hapl otypes in two different clades. 411

PAGE 412

A 412

PAGE 413

A 413

PAGE 414

Figure A 107. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of MAX4 haplotype phased sequence data (364 bp) after removing minor frequency haplotypes ( 189 trees, 237 steps, CI=0.50, RI=0.85). Bootstrap values greater than 50% are described above the branches. 414

PAGE 415

A 415

PAGE 416

A 416

PAGE 417

Figure A 108. Maximum likelihood tree using RAxML for MAX4 haplotype phased sequence data ( 364 bp) (lnL= 1902.66). Maximum likelihood posterior probabilities values are located below the branches. 417

PAGE 418

A 418

PAGE 419

A 419

PAGE 420

F igure A 109. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE haplotype phased sequence data (661 characters ) including gaps ( 200 trees, 241 steps, CI=0. 87, RI=0.98). Bootstrap values greater than 50% are described above the branches. 420

PAGE 421

421

PAGE 422

Figure A 110. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE haplotype phased sequence data (648 characters ) after removing minor frequency haplotypes and inclu ding gaps (200 t rees, 210 steps, CI=0.93, RI=0.99). Bootstrap values greater than 50% are described above the branches. 422

PAGE 423

423

PAGE 424

Figure A 111. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE haplotype phased sequence data ( 553 bp) without gaps ( 2 00 trees, 114 steps, CI=0.88 RI=0.9 8 ). Bootstrap values greater than 50% are described above the branches. Red boxes represented samples with haplotypes in two different clades. 424

PAGE 425

425

PAGE 426

Figure A 112. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of PHYE haplotype phased sequence data ( 553 bp) after removing minor frequency haplotypes ( 2 00 trees, 112 steps, CI=0.90, RI=0.98 ). Bootstrap values greater than 50% are described above the branches. 426

PAGE 427

427

PAGE 428

Figure A 113. Maximum likelihood tree using RAxML for PHYE haplotype phased sequence data ( 553 bp) (lnL= 1550.06). Maximum likelihood posterior probabilities values are located below the branches. 428

PAGE 429

A 429

PAGE 430

A 430

PAGE 431

Figure A 114. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 haplotype phased sequence data (570 characters ) including gaps ( 2 00 tr ees, 212 steps, CI=0.67, RI=0.91). Bootstrap values greater than 50% are described above the branches. 431

PAGE 432

432

PAGE 433

Figure A 115. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 haplotype phased sequence data ( 570 characters ) after removing minor frequency haplotypes and including gaps (200 trees, 185 steps, CI=0. 77, RI=0.9 4 ). Bootstrap values greater than 50% are described above the branches. 433

PAGE 434

434

PAGE 435

Figure A 116. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 haplotype phased sequence data (562 bp) without gaps ( 2 00 trees, 190 steps, CI=0. 71 RI=0.93). Bootstrap values greater than 50% are described above the branches. Red boxes represented samples with haplotypes in two different clades. 435

PAGE 436

436

PAGE 437

Figure A 117. Majority rule consensus tree of the MPTs using the parsimony ratchet method for rapid parsimony analysis of VRN1 haplotype phased sequence data (562 bp) after re moving minor frequency haplotypes ( 200 trees, 172 steps, CI=0.78 RI=0.94 ). Bootstrap values greater than 50% are described above the branches. 437

PAGE 438

438

PAGE 439

Figure A 118. Maximum likelihood tree using RAxML for VRN1 haplotype phased sequence data (562 bp) (lnL= 2 094.19). Maximum likelihood posterior probabilities values are located below the branches. 439

PAGE 440

A 440

PAGE 441

A 441

PAGE 442

Figure A 119. Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) of 3 3 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. 442

PAGE 443

Figure A 120. Neighbor Joining (NJ) cluster analysis based on Neis genetic distance (Nei and Takezaki, 1983) of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus co re collection and southeastern US species. 443

PAGE 444

Figure A 121. Prunus americana ( dark blue), P. angustifolia ( p owder blue), P. geniculata (turquoise), P. umbellatalike (sea foam green), P. umbellata (yellow) and other Prunus species (black) traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and T akezaki, 1983) of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. 444

PAGE 445

Figure A 122. L ow chill <400 hours (dark blue), medium chill 400700 hours (green), medium high chill 700 1110 hours (yellow), and high chill >1110 hours (black) genotypes traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster based on Neis genetic distance (Nei and Takezaki, 1983) of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. 445

PAGE 446

Figure A 123. Genotypes with 201300 chill hours (dark blue), 301400 chill hours (powder blue), 401500 chill hours (light blue), 501600 chill hours (turquoise), 601700 chill hours (sea foam green), 701800 chill hours (green), 801900 chill hours (yellow), 11001200 chill hours (orange), and >1400 chill hours (black) traced over the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) cluster b ased on Neis genetic distance (Nei and Takezaki, 1983) of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. 446

PAGE 447

Figure A 124. A non standarized neighbor net network of 33 simple sequence repeat (SSR) markers for 99 Prunus specimens representatives of the southestern US species. Colors represent population structure results with k=5. 447

PAGE 448

P. angustifolia P. geniculata P. americana P. umbellata like 448

PAGE 449

Figure A 125. Population structure result for k=5 and k= 7 separated by species of 33 simple sequence repeat (SSR) markers for 106 Prunus specimens representatives of the Prunus core collection and southeastern US species. 449

PAGE 450

Figure A 126. Fstoutlier detection using the Beaumont and N icholss modified frequentist method (Beaumont and Nichols, 1996) of 33 simple sequence repeat (SSR) markers for 69 Prunus umbellata specimens calculated for pair wise comparisons between genotype groups with similar phenotypic traits classes for A) Chilling range 1 and B) Chill range 2. Outlier locus are shown. A B 450

PAGE 451

Figure A 127. Posterior probability calculations to selection using a Bayesian method and a reversiblejump MCMC approach of 33 simple sequence repeat (SSR) markers for 69 Prunus umbellata specimens calculated using pair wise comparisons between genotype groups with similar phenotypic traits classes: Chill range 1 with prior odds A) 10, B) 1 and C) 0.1; and chill range 2 with prior odds D) 10, E) 1 and F) 0.1. A B C D E F 451

PAGE 452

LIST OF RE FERENCES Akaike H. 1974. A new look at the statistical model identification criteria. IEEE T. Automat. Cont. 19: 716 723. Andersen, P.C., W.B. Sherman, and J.G. Williamson. 2001. Low chill peach and nectarine cultivars from the University of Florida breedi ng program: 50 years of progress. Proc. Fla. State Hort. Soc. 114:3336. Antao, T., A. Lopes, R.J. Lopes, A. BejaPereira, and G. Luikart. 2008. LOSITAN: A workbench to detect molecular adaptation based on a Fstoutlier method. BMC Bioinformatics 9:323328 Aranzana, M.J., E. K Abbassi, W. Howad, and P. Ars. 2010. Genetic variation, population structure and linkage disequilibrium in peach commercial varieties. BMC Genetics 11:6979. Aranzana, M.J., J. Carbo, and P. Ars. 2003a. Microsatellite variability i n peach ( Prunus persica (L.) Batsch): cultivar identification, marker mutation, pedigree inferences and population structure. Theor. Appl. Genet. 106:13411352. Aranzana, M.J., J. GarcaMas, J. Carbo, and P. Ars. 2002. Development and variability analysi s of microsatellite markers in peach. Plant Breeding 121:8792. Aranzana, M.J., A. Pineda, P. Cosson, E. Dirlewanger, J. Ascasibar, G. Cipriani, C.D. Ryder, R. Testolin, A. Abbott, G.J. King, A.F. Iezzoni, and P. Ars. 2003b. A set of simple sequence repeat (SSR) markers covering the Prunus genome. Theor. Appl. Genet. 106:819825. Ars, P., R. Messeguer, M. Viruel, K. Tobbut, E. Dirlewanger, F. Santi, R. Quarta, E. Ritter. 1994. The European Prunus mapping project. Euphytica 77:97100. Badenes, M.L. and D.E Parfitt. 1995. Phylogenetic relationships of cultivated Prunus species from an analysis of chloroplast DNA variation. Theor. Appl. Genet. 90:10351041. Beaumont, M.A. and R.A. Nichols. 1996. Evaluating loci for use in the genetic analysis of population. Proc. R Soc. Lond. B. Biol. Sci. 263:16191626. Beckman, T.G. and W.R. Okie. 1994. Native North American plum species potential for variety and rootstock development. Acta Hort. 359:4148. Beckman, T.G., W.R. Okie, A.P. Nyczepir, P.L. Pusey, and C.C. Reill y. 1998. Relative susceptibility of peach and plum germplasm to Armillaria root rot. HortScience 33:10621065. Benson, D.A., M. Cavanaugh, K. Clark, I. KarschMizrachi, D.J. Lipman, J. Ostell, and E.W. Sayers. 2013. GenBank. Nucleic Acids Res. 41:D3642. 452

PAGE 453

B entham, G. and J.D. Hooker. 1865. Genera plantarum Vol. 1. London: Lovell Reed & Co. Blaker, K. 2010. Detection of seed dormancy QTL in three F2 populations of peach. MS Thesis. University of Florida. Bortiri, E., S.H. Oh, F. Y. Gao, and D. Potter. 2002. The phylogenetic utility of nucleotide sequences of sorbitol 6phosphate dehydrogenase in Prunus (Rosaceae). Am. J. Bot. 89:1697 1708. Bortiri, E., S.H. Oh, J. Jiang, S. Baggett, A. Granger, C. Weeks, M. Buckingham, D. Potter, and D.E. Parfitt. 2001. Phy logeny and systematic of Prunus (Rosaceae) as determined by sequence analysis of ITS and the chloroplast trnL trnF spacer DNA. Systematic Botany 26:797807. Bortiri, E., B. Vanden Heuvel, and D. Potter. 2006. Phylogenetic analysis of morphology in Prunus r eveals extensive homoplasy. Pl. Syst. Evol. 259:5371. Britton, N.L. and J.A. Shafer. 1908. North American trees being descriptions and illustrations of the trees growing independently of cultivation in North America, North Mexico and the West Indies. New York, New York, USA. Carrillo Mendoza, O. 2012. Genetics of tree architecture in peach ( Prunus persica (L.) Batsch). Ph.D. Dissertation. University of Florida. CavalliSforza, L.L. and A.W.F. Edwards. 1967. Phylogenetic analysis: models and estimation proc edures. Am. J. Hum. Gen. 19: 233257. Chaparro, J.X., D.J. Werner, D. OMalley, R.R. Sederoff. 1994. Targeted mapping and linkage analysis of morphological, isozyme, and RAPD markers in peach. Theor. Appl. Genet. 87:805815. Chavez, D.J. and J.X. Chaparro. 2011. Identification of markers linked to seedlessness in Citrus kinokuni hort. ex Tanaka and its progeny using bulked segregant analysis. HortScience 46:693697. Cheng, Z., K. Gasic, and Z. Wang. 2013. Population structure, linkage disequilibrium and sel ective loci in natural populations of Prunus davidiana. J. Agri. Sci. 5: 69 85. Cipriani, G., G. Lot, W. G. Huang, M.T. Marrazzo, E. Peterlunger, and R. Testolin. 1999. AC/GT and AG/CT microsatellite repeats in peach [ Prunus persica (L) Batsch]: isolation, characterisation and cross species amplification in Prunus Theor. Appl. Genet. 99:65 72. Clement, M., D. Posada, and K.A. Crandall. 2000. TCS: a computer program to estimate gene genealogies. Mol. Eco. 9:16571659. 453

PAGE 454

Cox, A.C., A.F. Johnson, S. Kane, and L .G. Chafin. 2004. Status survey for scrub plum, Prunus geniculata Florida Natural Areas Inventory, Tallahassee, Florida, USA. Dai, W. and J.A. Walla. 2007. Development of segregating populations for molecular and genetic analysis of X disease in resistanc e chokecherry ( Prunus virginiana L.). Abstract. PAG XV conference, San Diego, CA. De Candolle, A.P. 1825. Prodromus systematis naturalis regni vegetabilis Vol. 2. Paris: Treuttel et Wrtz. Depyrere, L., P. Chaerle, P. Breyne, K.V. Mijnsbrugge, and P. Goet ghebeur. 2009. A combined morphometric and AFLP based diversity study challenges the taxonomy of the European members of the complex Prunus L. section Prunus Plant Syst. Evol. 279:219231. Dickson, E.E., K Arumuganathan, S. Kresovich, and J.J. Doyle. 1992. Nuclear DNA content variation within Rosaceae. Ame. J. Bot. 79:10811086. Dirlewanger, E. and C. Bodo. 1994. Molecular genetic mapping of peach. In: Progress in temperate fruit breeding. Eds. Schmidt, H., and M. Kellerhals. Kluwer, The Netherlands, pp 309 311. Dirlewanger, E., P. Cosson, M. Tavaud, M.J. Aranzana, C. Poizat, A. Zanetto, P. Ars, and F. Laigret. 2002. Development of microsatellite markers in peach ( Prunus persica (L.) Batsch) and their use in genetic diversity analysis in peach and sweet ch erry (Prunus avium L.). Theor. Appl. Genet. 105:127138. Dirlewanger, E., E. Graziano, T. Joobeur, F. GarrigaCaldere, P. Cosson, W. Howad, and P. Arus. 2004. Comparative mapping and marker assisted selection in Rosaceae fruit crops. PNAS 101:98919896. Dirlewanger, E., J. Quero Garca, L.L. Dantec, P. Lambert, D. Ruiz, L. Dondini, E. Illa, B. Quilot Turion, J M. Audergon, S. Tartarini, P. Letourmy, and P. Ars. 2012. Comparison of the genetic determinism of two key phonological traits, flowering and maturi ty dates, in three Prunus species: peach, apricot, and sweet cherry. Heredity 109:280292. Durham, R.E., G.A. Moore, and W.B. Sherman. 1987. Isozyme banding patterns and their usefulness as genetic markers in peach. J. Am. Soc. Hort. Sci. 112:10131018. Ea rl, D.A. and B.M. VonHoldt. 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4:359361. Edgar, R.C. 2004. MUSCLE: multiple sequence alignment with high acc uracy and high throughput. Nucleic Acids Res. 32:17921797. 454

PAGE 455

Eduardo, I., G. Chietera, R. Pirona, I. Pacheco, M. Troggio, E. Banchi, D. Bassi, L. Rossini, A. Vecchietti, and C. Pozzi. 2012. Genetic dissection of aroma volatile compounds from the essential oil of peach fruit: QTL analysis and identification of candidate genes using dense SNP maps. Tree Genet. Geno. DOI 10.1007/s112950120546z. Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of clusters of individuals using the software STR UCTURE: a simulation study. Mol. Eco. 14:26112620. Fan, S., D.G. Bielenberg, T.N. Zhebentyayeva, G.L. Reighard, W.R. Okie, D. Holland, and A.G. Abbott. 2009. Mapping quantitative trait loci associated with chilling requirement, heat requirement and bloom date in peach ( Prunus persica ). New Phytologist 185:917930. FAOSTAT data, 2010, http://faostat3.fao.org/home/index.html#DOWNLOAD Farris, J.S., M. Kllersj, A.G. Kluge, and C. Bult. 19 94a. Permutations. Cladistics 10:6576. Farris, J.S., M. Kllersj, A.G. Kluge, and C. Bult. 1994b. Testing significance of incongruence. Cladistics 10:315319. Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolut ion 39:783791. Fischer, M.C., M. Foll, L. Excoffier, and G. Heckel. 2011. Enhanced AFLP genome scans detect local adaptation in highaltitude populations of a small rodent ( Microtus arvalis ). Mol. Ecol. 20:1450 1462. Floyd, W.F. 1920. Peach growing in Florida. Fla. Coop. Ext. Bul. 27:58. Foll, M., M.C. Fischer, G. Heckel, and L. Excoffier. 2010. Estimating population structure from AFLP amplification intensity. Mol. Ecol. 19:46384647. Foll, M. and O. Gaggiotti. 2008. A genomescan method to identify selected loci appropriate for both dominant and codominant markers: a Bayesian perspective. Genetics 180:977993. Germa i n Aubrey, C.C. 2012. Historical biogeography and conservation genetics of central Florida scrub endemics. Ph.D. Dissertation. University of Florida. Goicoechea, P.G., R.J. Petit, and A. Kremer. 2012. Detecting the footprints of divergent selection in oaks with linked markers. Heredity 109:361371. 455

PAGE 456

Goodstein, D.M., S. Shu, R. Howson, R. Neupane, R.D. Hayes, J. Fazo, T. Mitros, W. Dirks, U. He llsten, N. Putnam, and D.S. Rokhsar. 2012. Phytozome: a comparative platform for green plant genomics. Nucleic Acid Res. 40:D1178D1186. Harper, R. 1911. A new plum from the Lake Region of Florida. Torreya 11:6467. Hedrick, U.P. 1911. The plums of New Yor k. N.Y. Dept. Agr. 18th Ann. Rpt. v. 3, pt. 2. Horvath, A., H. Christmann, and F. Laigret. 2008. Genetic diversity and relationships among Prunus cerasifera (cherry plum) clones. Botany 86:13111318. Hutchinson, J. 1964. The genera of the flowering plants, Vol. 1. Dicotyledones Oxford: Clarendon Press. Huson, D.H. and D. Bryant. 2006. Application of phylogenetic networks in evolutionary studies. Mol. Biol. Evol. 23:254267. Joly, S. and A. Bruneau. 2006. Incorporating allelic variation for reconstructing t he evolutionary history of organisms from multiple genes: an example from Rosa in North America. Syst. Biol. 55:623636. Jung, S., M. Staton, T. Lee, A. Blenda, R. Svancara, A. Abbott, and D. Main. 2008. GDR (Genome Database for Rosaceae): integrated webd atabase for Rosaceae genomics and genetic data. Nucleic Acids Res. 36:10341040. Kaneko, T., T. Terachi, and K. Tsunewaki. 1986. Studies on the origin of crops species by restriction endonuclease analysis of organellar DNA. II. Restriction analysis of ctDNA of 11 Prunus species. Jpn. J. Genet. 61:157168. Katayama, H. and C. Uematsu. 2005. Structural analysis of chloroplast DNA in Prunus (Rosaceae): evolution, genetic diversity and unequal mutations. Theor. Appl. Genet. 111:14301439. Keulemans, J. 1994. Pollination and fruit set in self incompatible plum cultivars. Acta Hort. 359:260268. Koehne, E. 1893. Deutsche dendrology Sttugart: Verlag von Ferdinand Enke. Layne, D.R. and D. Bassi. 2008. The peach. Botany, production and uses. CABI International. Cambridge, MA. 615 pp. Layne, R.E. and W.B. Sherman. 1986. Interspecific hybridization of Prunus HortScience 21:4851. Lee, S. and J. Wen. 2001. A phylogenetic analysis of Prunus and the Amygdaloideae (Rosaceae) using ITS sequences of nuclear ribosomal DNA. A m. J. Bot. 88:150160. 456

PAGE 457

Li, M. H., T. Iso Touru, L. Hannele, and J. Kantanen. 2010. A microsatellitebased analysis for the detection of selection on BTA1 and BTA20 in northern Eurasian cattle ( Bos taurus ) populations. Gen. Sel. Evol. 42:3245. Librado, P. and J. Rozas. 2009. DnaSP v5: A software for comprehensive analysis of DNA polymorphism. Bioinformatics 25:14511452. Linnaeus, 1737. Hortus cliffortianus Amsterdam, Holland: published by the author. Linnaeus, 1754. Genera plantarum 5th ed. Stockholm: Im pensis Laurentti Salvii. Liu, K. and S.V. Muse. 2005. PowerMarker: an integrated analysis environment for genetic marker analysis. Bioinformatics 21:21282129. Maddison, W. P. and D.R. Maddison. 2011. Mesquite: a modular system for evolutionary analysis. V ersion 2.75 http://mesquiteproject.org. Martnez Garca, P.J., D.E. Parfitt, E.A. Ogundiwin, J. Fass, H.M. Chan, R. Ahmad, S. Lurie, A. Dandekar, T.M. Gradziel and C.H. Crisosto. 2012. High density SNP mapping and QTL analysis for fruit quality characteris tics in peach ( Prunus persica L.). Tree Genet. Geno. Doi: 10.1007/s11295012 05227. Mason, S.C. 1913. The pubescent fruited species of Prunus of the Southwestern states. J. Agri. Res. 1:147179. Messeguer, R., P. Ars, and M. Carrera. 1987. Identification of peach cultivars with pollen isozymes. Sci. Hort. 31:107117. Miller, M.A., W. Pfeiffer, and T. Schwartz. 2010. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. Proc. Gateway Computing Env. Workshop. Morris, A.B., S.M. Icke rt Bond, D.B. Brunson, D.E. Soltis, and P.S. Soltis. 2008. Phylogeographic structure and temporal complexity in American sweetgum ( Liquidambar styraciflua). Molecular Ecology 17: 3889 3900. Mowrey, B.D. and D.J. Werner. 1990. Phylogenetic relationships among species of Prunus as inferred by isozyme markers. Theor. Appl. Genet. 80:129133. Mowrey, B., D. Werner, and D. Byrne. 1990a. Inheritance of isocitrate dehydrogenase, malate dehydrogenase, and shikimate dehydrogenase in peach and peach almond hybrids. J. Am. Soc. Hort. Sci. 115:312 319. Mowrey, B., D. Werner, and D. Byrne. 1990b. Isozyme survey of various species of Prunus in the subgenus Amygdalus Sci. Hort. 44:251 260. Myers, R. and R. Myers. 1990. Scrub and high pine, Ecosystems of Florida, 150194. 457

PAGE 458

Nei, M and N. Takezaki. 1983. Estimation of genetic distances and phylogenetic trees from DNA analysis. Proc. 5 the World Cong. Genet. Appl. Livstock Prod. 21: 405412. Nei, M. 1987. Molecular evolutionary genetics. New York: Columbia University Press. Ne wbigin, E., M.A. Anderson, A.E. Clarke. 1993. Gametophytic self incompatibility systems. Plant Cell 5:13151324. Nixon, K.C. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15:407414. Nixon, K.C. and J.M. Carpenter. 1996 On simultaneous analysis. Cladistics 12:221241. Oatley, G., G. Voelker, T.M. Crowe, and R.C.K. Bowie. 2012. A multi locus phylogeny reveals a complex pattern of diversification related to climate and habitat heterogeneity in southern African whiteeyes. Mol. Phyl. Evol. 64:633644. Okie, W.R. 1998. Handbook of peach and nectarine varieties. Agricultural Research Service. Okie, W.R. and J.H. Weinberger. 1996. Plums. In: Janick J. Moore JN (eds) Fruit Breeding, vol. 1. Tree and tropical fruits. John Wiley and Sons, Inc., New York. Pairon, M., B. Petitpierre, M. Campbell, A. Guisan, O. Broennimann, P.V. Baret, A. L. Jacquemart, and G. Besnard. 2010. Multiple introductions boosted genetic diversity in the invasive range of black cherry ( Prunus serotina ; Rosac eae). Annals of Botany 105:881890. Peakall, R. and P.E. Smouse, 2006. GENALEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Mol. Eco. Notes 6, 288 295. Peakall, R. and P.E. Smouse, 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research an update. Bioinformatics 28:25372539. Posada, D. 2008. JModeltest: Phylogenetic Model Averaging. Mol. Bio. Evol. 25:12531256. Pritchard J.K., M. Stephens, and P. Donnelly. 2000. Inferenc e of population structure using multilocus genotype data. Genetics 155:945959. Quarta, R., M.T. Dettori, I. Verde, Z. Broda, and A. Gentile. 1998. Genetic analysis of agronomic traits and genetic linkage mapping in a BC1 population using RFLPs and RAPDs. Acta horticulturae 465:5159. 458

PAGE 459

Rajapakse, S., L.E. Belthoff, G. He, A.E. Estager, R. Scorza, I. Verde, R.E. Ballard, W.V. Baird, A. Callahan, R. Monet, and A.G. Abbott. 1995. Genetic linkage mapping in peach using morphological, RFLP and RAPD markers. Theor Appl. Genet. 90:503 510. Reales, A., D.J. Sargent, K.R. Tobutt, and D. Rivera. 2010. Phylogenetics of Eurasian plums, Prunus L. section Prunus (Rosaceae), according to coding and noncoding chloroplast DNA sequences. Tree Genet. Geno. 6:3745. Rehder, A. 1940. Manual of cultivated trees and shrubs hardy in North America. 2nd ed. New York: Macmillan. Rohrer, J.R. 2009. In prep. Rosaceae. For: Flora of North America Editorial Committee, eds. 1993+. Flora of North America North of Mexico. 16+ vols. New York and Oxford. Vol. 9. Rohrer, J.R., R. Ahmad, S.M. Southwick, and D. Potter. 2004. Microsatellite analysis of relationships among North American plums ( Prunus sect. Prunocerasus Rosaceae). Plant Syst. Evol. 244:6975. Rohrer, J.R., M.A. OBrien, and J.A. An derson. 2008. Phylogenetic analysis of North American plums ( Prunus sect. Prunocerasus : Rosaceae) based on nuclear LEAFY and S6PDH sequences. J. Bot. Res. Inst. Texas 2:401414. Rosenberg, N.A. 2004. Distruct: a program for the graphical display of populat ion structure. Mol. Eco. Notes 4:137138. Saitou N. and M. Nei. 1987. The neighbor joining method: a new method for reconstructing phylogenetic trees. Mol. Bio. Evol. 4:406425. Snchez Prez, R., F. Dicenta, and P. Martnez G mez. 2012. Inheritance of chi lling and heat requirements for flowering in almond and QTL analysis. Tree Gen. Geno. 8:379389. Sargent, C.S. 1905. Manual of the trees of North America. The Riverside press, Cambridge. Sargent, C.S. 1905. Manual of the trees of North America. The Riverside press, Cambridge. Schwarz, G. 1978. Estimating the dimension of a model. Annals Stat. 6:461464. Scorza, R. and W.R. Okie. 1991. PEACHES ( PRUNUS ). Acta Hort. 290:177234. Sharpe, R.H. 1957. Okinawa peach shows promising resistance to root knot nematodes. Proc. Fla. State Hort. Soc. 70:320 322. Sharpe, R.H. 1961. Developing new peach varieties for Florida. Proc. Fla. State Hort. Soc. 74:348352. 459

PAGE 460

Sharpe, R.H. 1969. Subtropical peaches and nectarines. Proc. Fla. State Hort. Soc. 82:302306. Sharpe, R.H. and J.B. Aitken. 1971. Progress of the nectarine. Proc. Fla. State Hort. Soc. 84:338345. Sha r pe, R.H., T.E. Webb, and H.W. Lundy. 1954. Peach Variety Tests. Proc. Fla. State Hort. Soc. 67:245247. Shaw, J., E.B. Lickey, J.T. Beck, S.B. Farmer., W. Liu, J. Mi ller, K.C. Siripun, C.T. Winder, E.D. Schilling, and R.L. Small. 2005. The tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Amer. J. Bot. 92:142166. Shaw, J., E.B. Lickey, E.E. Schilling, and R.L. Small. 2007. Comparison of whole chloroplast genome sequences to choose noncoding regions for phylogenetic studies in Angiosperms: the tortoise and the hare III. Amer. J. Bot. 94:275288. Shaw, J. and R.L. Small. 2004. Addressing the Hardest puzzle i n the American pomology: Phylogeny of Prunus sect. Prunocerasus (Rosaceae) based on seven noncoding chloroplast DNA regions. Am. J. Bot. 91:985996. Shaw, J. and R.L. Small. 2005. Chloroplast DNA phylogeny and phylogeography of the North American Plums ( P runus subgenus Prunus section Prunocerasus Rosaceae). Am. J. Bot. 92:20112030. Sherman, W.B. and P.M. Lyrene. 2003. Low chill breeding of deciduous fruits at the University of Florida. Proc. XXIV IHC Genetics and Breeding Tree Fruits and Nuts. Ed. J. J anick. Acta Hort. 622:599605. Sherman, W.B., P.M. Lyrene, and R.H. Sharpe. 1996. Low chill peach and nectarine breeding at the University of Florida. Proc. Fla. State Hort. Soc. 109:222223. Sherman, W.B., J. Rodriguez, and E.P. Miller. 1984. Progress in low chill peaches and nectarines from Florida. Proc. Fla. State Hort. Soc. 97:320 322. Sherman, W.B., B.L. Topp and P.M. Lyrene. 1990. Nonmelting flesh for fresh market peaches. Proc. Fla. State Hort. Soc. 103:293 294. Sikes, D.S. and P.O. Lewis. 2001. Beta software, version 1. PAUP* implementation of the parsimony ratchet. Distributed by authors. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, USA. Small, R.L., J.A. Ryburn, R.C. Cronn, T. Seelanan, and J.F. Wendel. 1998. The tortoise and the hare: choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. Am. J. Bot. 85:13011315. 460

PAGE 461

Sokal R. and C. Michener. 1958. A statistical method for evaluating system atic relationships. University of Kansas Science Bulletin 38:1409 1438. Soltis, D.E., and P.S. Soltis. 1998. Choosing and approach and an appropriate gene for phylogenetic analysis. In Molecular Systematics of Plants II: DNA sequencing, eds. D.E. Soltis, P .S. Soltis, and J.J. Doyle, pp. 1 42. Kluwer Academic Publishers, Norwell, Massachusetts. Stamatakis, A. 2006. RAxMLVI HPC: Maximum Likelihood based Phylogenetic Analyses with Thousands of Taxa and Mixed Models. Bioinformatics 22:26882690. Stamatakis, A. P. Hoover, and J. Rougemont. 2008. A fast bootstrapping algorithm for the RAxML webservers. Syst. Biol. 57:758 771. Sutherland, B.G., T.P. Robbins, and K.R. Tobutt. 2004. Primers amplifying a range of Prunus S alleles. Plant Breeding 123:582584. Swoffo rd, D.L. 2002. PAUP*: phylogenetic analysis using parsimony (* and other methods) version 4.0b10. Sinauer Associates, Sunderland, Massachusetts, USA. Tajima, F. 1989. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Gene tics 123:585595. Testolin, R., T. Marrazzo, G. Cipriani, R. Quarta, I. Verde, M.P. Dettori, M. Pancaldi, and S. Sansavini. 2000. Microsatellite DNA in peach ( Prunus persica L. Batsch) and its use in fingerprinting and testing the genetic origin of cultivars. Genome 43:512520. Tsumura, Y., K. Uchiyama, Y. Moriguchi, S. Ueno, and T. IharaUjino. 2012. Genome scanning for detecting adaptive genes along environmental gradients in the Japanese conifer, Cryptomeria japonica. Heredity 109:349360. Turner, W.R., D.S. Wilcove, and H.M. Swain. 2006. State of the scrub: Conservation progress, management responsibilities, and land acquisition priorities for imperiled species of Floridas Lake Wales Ridge. Archbold Biological Station, Lake Placid, Florida, USA. Uematsu, C., T. Sasakuma, and Y. Ogihara. 1991. Phylogenetic relationships in the stonefruit group of Prunus as revealed by restriction fragment analysis of chloroplast DNA. Jpn. J. Genet. 66:5969. Untergasser, A., I. Cutcutache, T. Koressaar, J. Ye, B.C. Fairc loth, M. Remm, and S.G. Rozen. 2012. Primer3new capabilities and interfaces. Nucleic Acids Res. 40:e115. USDA. Plants Profile database (PLANTS) [Online Database]. URL: http://plants.usda.gov/index.html (07 April 2010). 461

PAGE 462

Verde I., N. Bassil, S. Scalabrin, B. Gilmore, C.T. Lawley, et al. 2012. Development and Evaluation of a 9K SNP Array for Peach by Internationally Coordinated SNP Detection and Validation in Breeding Germplasm. PLoS ONE 7: e35668. Wang, Y. 1985. Peach growing and germplasm in China. Acta Hort. 173:5155. Wang, Y., L.L. Georgi, T.N. Zhebentyayeva, G.L. Reighard, R. Scorza, and A.G. Abbott. 2002. Highthroughput targeted SSR marker development in peach ( Prunus persica ). Genome 45:319328. Watkins, R. 1976. Cherry, plum, peach, apricot, and alm ond. In N.W. Simmons [ed.], Evolution of crop plants. Longman, London, UK. Watterson, G.A. 1975. On the number of segregating sites in genetical models without recombination. Theor. Popul. Biol. 7:256276. Waugh, F.A. 1899. The pollination of plums. Vermont Agricultural Experimentation Station, Annual Report 12:189209. Waugh, F.A. 1901. Plums and plum culture. Orange Judd, New York, New York, USA. Webb, S. and R. Myers. 1990. Historical biogeography, Ecosystems of Florida, 70103. Weekley C.W., D.N. Zaya, E.S. Menges, and A.E. Faivre. 2010. Mutliple causes of seedling rarity in scrub plum, Prunus geniculata (Rosaceae), an endangered shrub of the Florida scrub. Am. J. Bot. 97:144155. Wen, J., S.T. Berggren, C. H. Lee, S. IckertBond, T. S. Yi, K.O. Yoo, L. Xie, J. Shaw, and D. Potter. 2008. Phylogenetic inferences in Prunus (Rosaceae) using chloroplast ndhF and nuclear ribosomal ITS sequences. J. Syst. Evol. 46:322332. Wen, J. and E.A. Zimmer. 1996. Phylogeny and biogeography of Panax L. (the ginseng genus Araliaceae): inferences from ITS sequences of nuclear ribosomal DNA. Mol. Phylo. Evol. 6:166 177. Wight, W.F. 1915. Native American species of Prunus Bull. US Dept. Agr. 179:1 75. Wnsch, A. and J.I. Hormaza. 2002. Cultivar identification and genetic fi ngerprinting of temperate fruit tree species using DNA markers. Euphytica 125:5667. Zhang, Y. X., C.X. Zeng, and D. Z. Li. 2012. Complex evolution in Arundinarieae (Poaceae: Bambusoideae): incongruence between plastid and nuclear GBSSI gene phylogenies. Mol. Phylogenet. Evol. 63:777797. Zhu, Q., X. Zheng, J. Luo, B.S. Gaut, and S. Ge. 2007. Multilocus analysis of nucleotide variation of Oryza sativa and its wild relatives: severe bottleneck during domestication of rice. Mol. Biol. Evol. 24:875 888. 462

PAGE 463

Zwick l, D. J. 2006. Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Ph.D. dissertation, The University of Texas at Austin 463

PAGE 464

BIOGRAPHICAL SKETCH Dario Chavez, born in Riobamba, Ecuador, in 1983, is the youngest son of Carlos Chavez and Ximena Velasquez. His three older brothers taught him several lessons of life that allowed him to look for new ideas and goals. From elementary to high school, he studied under the Salesians a nd like many of us; he was undecided about what was next after high school. His father, a farmer by heritage and pass ion, taught him to love farming. Y et Dario never thought about relating the farm to a field of study. Alex, his older brother, challenged him to apply to Zamorano University, a well known agricultural school in Honduras that has the educational basis of learning by doing. His studies started in 2002 and in four years he earned his bachelor of science in agricultural science and production. In Zamorano, Dario made new friends from all over Latin America. After graduation, he worked at Ohio State University for six months as a short term scholar. In August 2006, he began his Master of Science degree in plant breeding and genetics under the di rection of Dr. Lyrene at the University of Florida. In 2008, Dario began his Ph.D. with Dr. Eileen Kabelka in squach breeding and genetics. After Dr. Kabelkas departure from UF, he began his Ph.D. dissertation project in the stone fruit breeding and genetics program under the direction of Dr. Jos X. Chaparro. Dario met several people who changed his perception of life. He also met many who have taught him to l ove and have passion for what they do: his beautiful wife, his father, his mother, his brothers and finally his mentors, Dr. Lyrene, Dr. Jos X. Chaparro, and Dr. Wayne Sherman. He received his Ph.D. from the University of Florida in the fall of 2013. Dari os new goal will be to obtain a position in plant breeding and genetics 464